PROCEEDINGS OF THE

TWENTY' SEVENTH ANNUAL CONGRESS

19 5 3 OF THE

SOUTH AFRICAN SUGAR TECHNOLOGISTS'

ASSOCIATION

THE SOUTH AFRICAN

SUGAR TECHNOLOGISTS'

A S S O C I A T I O N

The South African Sugar Technologists' Association was

founded in 1926, It is an independent, self-constituted

organisation of technical workers and others directly

interested in the technical aspect of the South African

Sugar Industry. It operates under the aegis of the

South African Sugar Association, but is governed under

its own constitution by a Council elected by its own

members.

The office of the Association is situated on premises

kindly made available to it by the South African Sugar

Association at the latter's Experiment Station at Mount

Edgecombe.

iii

PRINCIPAL CONTENTS

PAGE

OFFICIAL O P E N I N G : D R . W. S. RAPSON'S A D D R E S S 1

PRESIDENTIAL A D D R E S S : M R . G. C. D Y M O N D 6

CANE SUGAR IN E A S T AFRICA, by A. C. B A R N E S , C.M.G 8

T W E N T Y - E I G H T H A N N U A L SUMMARY OF CHEMICAL LABORATORY R E P O R T S , by CHS .

G. M. P E R K 16

ANNUAL W E A T H E R R E P O R T FOR 1952, by B. E. B E A T E R 28

CENTRIFUGAL MACHINES, by CHS . G. M. P E R K 34

I N H E R E N T STEAM LOSSES F O U N D IN W A T E R T U B E B O I L E R S , by W M . C LINDEMANN 48

A R E V I E W OF BAGASSE F U E L QUALITY T R E N D S AND OF R E C E N T B O I L E R E F F I C I E N C Y

T E S T S , by A. F. MCCULLOCH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

SCALING OF EVAPORATOR T U B E S OF NATAL FACTORIES, by K. D O U W E S DEKKER.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

SOME N O T E S ON AUTOMATIC CONTROL IN SOUTH AFRICAN SUGAR FACTORIES, by H.

R O H L O F F 74

CLARIFICATION AND MISCELLANEOUS DATA, by G. C. DYMOND 81

A B R I E F R E V I E W OF A PORTION OF THE LITERATURE D E A L I N G WITH S U L P H U R

B U R N I N G AND T H E FORMATION AND DISSOCIATION OF SULPHUR T R I O X I D E , by

R. J. HOGARTH AND C. T. J E N K I N 87

T H E LABORATORY DETERMINATION OF MOISTURE PER CENT BAGASSE U S I N G T H E

D I E T E R T MOISTURE T E L L E R , by G. D. THOMPSON 103

T H E E F F E C T O F T H E CITRIC SOLUBILITY O F LANGEBAAN PHOSPHORITE (LANGEBAAN

PHOSPHATE ROCK) W H E N COMPOSTED WITH F I L T E R CAKE, by H. E. KRUMM.. . . . . . . . . . . . . . . . . . . . . . . . . . . 108

MEALYBUG AND I T S E F F E C T ON SUGAR CANE, by J . D I C K 113

R E S U L T S OF MOSAIC TOLERANCE T R I A L S , by N. C. K I N G 120

LOOKING A H E A D , by W. F. C. J E X 129

SOME OBSERVATIONS IN CONNECTION WITH CANE GROWING, by H. L. GARLAND .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

CANE PRACTICES AND PRODUCTIVITY, by C. T. W I S E 133

SUGAR CANE GROWING IN NATAL. F I E L D M E T H O D S AND PRACTICES IN OPERATION

IN 1953, by C H. O. PEARSON 135

iv

OFFICERS 1953-1954

President. Vice-President.

G. C. DYMOND J. B. GRANT

Hon. Secretary.

(Mrs.) N. BOYD-SMITH

Former Presidents.

1926-27 M. MCMASTER 1935-36 G.C.WILSON 1944-45 G. S. MOBERLY

1927-28 M. MCMASTER 1936-37 G.C.WILSON 1945-46 G. S. MOBERLY

1928-29 H. H. DODDS 1937-38 J. RAULT 1946-47 W. BUCHANAN

1929-30 H. H. DODDS 1938-39 P. MURRAY 1947-48 W. BUCHANAN

1930-31 G. S. MOBERLY 1939-40 P.MURRAY 1948-49 J. L. DU TOIT

1931-32 G. C, DYMOND 1940-41 E. P. HEDLEY 1949-50 H. H. DODDS

1932-33 G. C. DYMOND 1941-42 F. W. HAYES 1950-51 A. MCMARTIN

1933-34 B .E .D . PEARCE 1942-43 A. MCMARTIN 1951-52 G. C. DYMOND

1934-35 E. CAMDEN-SMITH 1943-44 G. BOOTH 1952-53 G. C. DYMOND

Former Vice-Presidents.

1926-27 L. E. ROUILLARD 1935-36 E. CAMDEN-SMITH 1944-45 G. BOOTH

1927-28 H.'H. DODDS 1936-37 J. RAULT 1945-46 W. BUCHANAN

1928-29 G. S. MOBERLY 1937-38 P. MURRAY 1946-47 G. C DYMOND

1929-30 G. S. MOBERLY 1938-39 E. P. HEDLEY 1947-48 G. C. DYMOND

1930-31 G. C. DYMOND 1939-40 E. P. HEDLEY 1948-49 G. C. DYMOND

1931-32 A. C. WATSON 1940-41 F . W . H A Y E S 1949-50 J. L. DU TOIT

1932-33 A. C. WATSON 1941-42 A. MCMARTIN 1950-51 O. W. M. PEARCE

1933-34

1934-35

G. C. DYMOND

E. CAMDEN-SMITH

B. E. D. PEARCE

1942-43 G. BOOTH 1951-52 O. W. M. PEARCE

1943-44 F. B. MACBETH 1952-53 K. DOUWES DEKKER

Council of the Association.

J. P. N. BENTLEY K. DOUWES DEKKER W. G. GALBRAITH O. W. M. PEARCE

L. F. CHIAZZARI O. A. FELTHAM M. HILL J. RAULT

W. O. CHRISTIANSON J. L. DU TOIT A. MCMARTIN G. E. SEYMOUR

v.

South African Sugar Technologists' Association.

Twenty-Seventh Annual Congress

The Twenty-seventh Annual Congress of the South African Sugar Technologists' Association was held at the M.O.T.H. Centre, Old Fort Road, Durban, on 17th, 18th and 19th March, 1953, and at the South African Sugar Association's Experi-ment Station, Mount Edgecombe, on 20th March, 1953.

The following members and-visitors were present :

G. C. DYMOND, President.

ADAMS, B. N.

A L E X A N D E R , J . B .

A L E X A N D E R , K. E . F .

ALMOND, F. L.

ALMOND, Mrs. N.

A S H E , G.

B A R N E S , A. C.

B E A T E R , B . E .

B E C K E T T , B . H .

B E N T L E Y , J . P . N .

BOAST, W. B.

BONFA, N.

B O O T H , C. E.

B O O T H , M R S . C. S.

B O Y D - S M I T H , M R S . N .

BRASSEY, B .

BRIGGS, T .

BROGLIO, A. DE

B R O K E N S H A , G. J .

BROMLEY, C. K.

BUCHANAN, W. K.

B U C K , L.

C H E V E S , R.

CHRISTIANSON, W. O.

C H R I S T I E , j .

COIGNET, I. J. P. CUNNINGHAM, J . D.

DAMANT, E. L.

DAVIDSON, M I S S H .

D E D E K I N D , E . T . J .

D E N T , C. E.

D I C K , J .

D I C K , J . McD.

D O D D S , M R S . E . A .

D O D D S , H . H .

D O U W E S D E K K E R , K .

DUCHENNE, J . O.

D R A E G E R , P . L .

D Y M O N D , G. C,

D Y M O N D , K.

D Y M O N D , M R S . M.

E V A N S , C. S.

FARQUHARSON, J. C.

F I E L D I N G , W. L.

F I E L D I N G , M R S . W. L.

F I L M E R , M I S S P. N.

F O R D , R. L.

F O U R I E , D .

F O W L I E , P .

GALBRAITH, W. G.

GARLAND, H. L.

GAZZARD, H. T.

GLEASHIN, W.

GOLDING, C. D.

GOODWIN, A. H.

G R I E G , M.

GROOM, M R S . F .

G U N N , J . R .

H A C K L A N D , B.

H A L L , R. M.

H A R D Y , R. L.

HEDGCOCK, E. W.

HEDGCOCK, G. W.

H E D L E Y , E . P .

H E E N A N , B. C.

H E M P S O N , P .

H I B B E R D , A .

H I L L , M.

H I L L - L E W I S , C.

H I N D S H A W , W. A.

HOGARTH, R. J.

H O L T , C. H. K.

D 'HOTMAN DE V I L L I E R S , O.

H U N T L E Y , J . K .

JACKSON, L. F .

J E X , W. F . C.

K E N N Y , A.

K I N G , N. C.

K I N G , M R S . N .

KINSMAN, L. R.

L E E , G. L I S L E , D . DE

L I S L E , M R S . D . D E

L I N D E M A N N , W. C.

L I N T N E R , J .

L L O Y D , A. A.

LOGAN, F. A.

L U D O R F , P .

M A I N , J . W.

MATHEVV, G. E.

MACBETH, F . B .

MCCULLOCH, C. F.

M C K E N N A , H. G.

MCMARTIN, A.

MCMARTIN, S.

MILLAR, J . D.

M U R P H Y , E. S.

MURRAY, P .

N A R B E T H , B. N.

NICKSON, G.

O D D I N G , J .

O W E N , M R S . M. H.

P A L A I R E T , H. E . H .

P A R R I S H , J . R .

PATERSON, A.

P E A R C E , O. W. M.

PEARSON, C. H. O.

P E R K , C. G. M.

P H I P S O N , E . H .

P O R T E O U S , J .

P O U G N E T , J . F .

R A P S O N , W. S.

R A U L T , J .

R E N A U D , C. L.

R I C - H A N S E N , R . W.

RISHWORTII , A. H.

R I S K I N , B. V.

R O H L O F F , H.

ROSSOUW, G. S. H.

R O W E , A.

R U D D , C. M.

SARGENT, M R S . M.

SCHMELZ, G.

SEYMOUR, G. E.

SIMPSON, J. K.

S N E D D E N , J . W.

SOUCHON, A. C. H.

S P E N C E , L. A.

STARLING, C. N.

S T E W A R D , E .

S T E Y N , C, L.

S T E P H E N S O N , R. A.

TAYLOR, M R S . R. G.

THOMPSON, G. D.

THOMPSON, G. M.

T O N N E R , D .

T W I N C H , J . F .

W A G N E R , C. L.

W A L S H , W. H.

W A T S O N , R. G. T.

W E B E R , L . B .

W E S T O N , N .

W H E E L E R , F . D .

W I L L I A M S , A. R.

W I L L I A M S , B. M.

W I L M O T , G. L.

W R I G H T , J . K .

W Y A T T , W.

V E R I N D E R , H. N.

VERNE, J. C.

VILJOEN, E. J.

vi.

TWENTY-SEVENTH ANNUAL CONGRESS

Proceedings of the Twenty-Seventh Annual Congress of the South African Sugar

Technologists ' Association, held at the M.O.T.H. Centre, Old For t Road, Durban ,

on March 17th, 18th and 19th, and at t he Exper iment Stat ion, Mount Edgecombe,

on March 20th, 1953.

G. C. DYMOND (President) was in the Chair.

OPENING CEREMONY

The President: Ladies and Gentlemen—It is twenty-seven years since this Association of ours was formed. In that time we have had two children —the Experiment Station, born twenty-five years ago, and the Sugar Milling Research Institute, born in 1949. The Council for Scientific and Industrial Research has had a great deal to do with the S.M. R J. since its inception, and it is actively associated with it to-day. It is for this reason that it. gives me particular pleasure to welcome here this morning Dr. W. S. Rapson, who is to perform the opening ceremony. Dr. Rapson is, as you probably all know, director of the National Chemical Research Labora­tory of the C.S.I.R. He is in the front rank of research workers in this country and I am sure that we will listen with considerable attention to what he has to say to us.

It now gives me great pleasure to call on Dr. Rapson to open this, our Twenty-seventh Annual Congress.

Dr. W. S. Rapson: About the time when I re­ceived from your secretary the invitation to address you to-day, I first became acquainted with a story of the Doge of Genoa who, in 1684, had his city bombarded by the fleets of France, and was subse­quently summoned to Versailles by Louis XIV to present his apologies and generally to explain himself. In order to make this unpleasant task easier, he was taken round that huge establishment and shown the sights. When it was all over, someone asked him what he considered most remarkable in Versailles. His answer was "That I should be here at all." I confess to a similar feeling of astonishment that I am here to-day for, although greatly honoured by your invitation, I can lay claim to a knowledge of the Sugar Industry which is at the most very limited, and to an acquaintance with sugar technology that has been concerned almost exclusively with that por­tion of its activities which is essentially chemical.

It is upon this aspect of sugar technology—namely the subject of chemistry in its application to the Sugar Industry—that I shall for the greater part dwell to-day.

Before beginning, however, I have a very pleasant duty to perform, and that is to convey to you from the Council for Scientific and Industrial Research its greetings and its best wishes for the success of your Twenty-seventh Annual Congress.

One of the main aims of C.S.I.R. policy is to pro­mote co-operative research by industry. The Sugar Industry, it should be said, was an exponent of this policy long before the C.S.I.R. entered upon the scene in 1946. For, from as early as 1925, the South African Sugar Association has been operating its Sugar Experiment Station at Mount Edgecombe. More recently also, in 1949, with the co-operation of the C.S.I.R., it extended the range of its co-operative research activities by the establishment of a Sugar Milling Research Institute.

These progressive developments have their origin in definite practical needs of the Industry, but I do not think it would be unfair to say that they would not have occurred were it not for appreciation at the technological level of the benefits which flow to industry as a result of planned research effort. Because of this role played by the technologist in interpreting and stressing the potentialities of science and of research for future developments in his in­dustry, the C.S.I.R. has a very keen interest in the activities of associations such as yours, which give scope for technical discussions of the problems of industry.

In opening your Congress last year, Dr. G. S. H. Rossouw laid considerable emphasis on the rapidity with which domestic demands for sugar have in­creased over the past ten to fifteen years. These demands have now risen to a point at which the South African market can absorb almost 700,000 tons of sugar per year, so that, together with the

export quota to the United Kingdom, the production target of almost 900,000 tons of sugar per year has been set up. We can be sure that long before this level of production has been reached, new and higher targets will have been raised.

Dr. Rossouw discussed two methods of approach to the problem of meeting this production level. On the one hand, he referred to the need for expansion of cane production and milling facilities, and on the other to the need for more vigorous application of science and technology to the problem of increasing the efficiency of production. I have a feeling that when Dr. Rossouw referred to these problems of increased production and efficiency, that he was thinking in terms of increased production of cane and of increased efficiency in obtaining sugar from cane. Now, it would be a presumption on my part —in the light of the records of the Sugar Industry— to expound on these particular matters, which are the object of your every-day attention. In seeking to expand on this theme, therefore, I shall adopt a somewhat wider interpretation of the word "Pro­duction" that is conventional in sugar circles. For, if there is one feature of modern industry which has been made abundantly clear above all others in recent years, it is the fact that no industry can afford to be content either with existing methods of production, or with existing products and markets. In this respect, the sugar industries throughout the world suffer from an inherent disadvantage in that they operate—under present conditions at. least—to supply an assured market with one specific product. As a result, all other potential activities and develop­ments tend to be subordinate to this end, and there have, on the whole, been no determined attempts to establish markets for products of the industry other than sugar.

Let. us for a moment consider the nature of some of these other potential activities and developments. First among them is the question of bagasse utilisa­tion. The prospect of continuing difficulties in the supply of softwood timbers for newsprint in the U.S.A. has recently led to a critical study of other possible raw materials. Of these, bagasse has proved perhaps the most attractive, and in recent pilot plant experiments carried out at the National Bureau of Standards in Washington, bagasse was converted to newsprint in sufficient quantities for a sample issue of the "Congressional Record" to be run off on it. The samples were considered acceptable by a panel of newsprint executives of leading newspapers. The series of trials involved in this work was a most exacting one, and a wide range of processes—some of them patented or proprietary—were explored, and the products submitted to practical trials. An aspect of interest is that the Bureau's studies failed to sub­stantiate the common claims that pentosans should be extracted from the raw pulp, and that the presence

of pith is undesirable if a pliable paper suitable for newsprint is to be obtained. As a result, it has now been claimed that it is now technically feasible to manufacture from whole bagasse, a paper which is satisfactory in all important respects for newsprint purposes.

Now, the quantity of newsprint imported into South Africa was 12,000 short tons valued at £365,000 in 1937, and in 1952 this figure had in­creased to 73,000 tons and £3,750,000 respectively. Since it takes approximately two tons of bagasse to make one ton of newsprint, and the annual pro­duction of bagasse by the South African Sugar Industry exceeds 2,000,000 tons, the technical possi­bility exists for this industry to produce the whole of our domestic requirements of newsprint.

You will note that I have stressed the fact that it is the technical possibility which exists for this major development. Much more needs to be done, however, before the economics of the development can be assessed, and it is here that activity and initiative on the part of the Sugar Industry would probably pay handsome dividends. The economic problems involved are many. They include, for example, the problems of existing capital investment in bagasse fired boilers, of alternative fuel supplies, of treatment, storage, baling and transport of bagasse. These are not small problems, but the possibilities of increased production, and therefore of increased revenue to both millers and growers are such that they justify detailed investigation. Even if imme­diate processing of bagasse is not possible, at least no new developments should be planned which do not take the possibility into account. There is a need to move quickly in this matter, because of the rapidly increasing interest in the growing of timber for pulping purposes in the Union and in Swaziland. Since wood is competitive with bagasse not only for paper production, but also for other major applica­tions such as wallboard manufacture, and since there is a limit to the number of producing units which our markets can support, the Sugar Industry runs the risk that if it delays an assessment of its oppor­tunities for too long, it may find that it has lost these opportunities altogether.

The second potential development to which I should like to refer is the production of cane wax. South Africa pioneered the commercial extraction of cane wax in 1915, but for various reasons production ceased in 1925. The potential production which is possible, namely about 4,000 to 5,000 tons per annum, is sufficiently great, however, to justify continuing research and a continuing review of the economic possibilities of resuming production. The problems involved are many, since it seems unlikely that most economic recovery could be achieved bvT

extraction of filter cakes as currently available, and some modifications of what is conventional mill

2

3

practice may be necessary in order to get a filter cake from which wax can be obtained with maxi­mum efficiency. In other words, we are up against a similar problem to that which exists in the case of bagasse utilisation—namely that the flowsheet for sugar recovery may need modification before the final goal is achieved. A threat to this development also exists, namely that unless the Sugar Industry can get production under way in time, the existing domestic markets may be filled from an alternative source. In the present case, this threat is a very real one since the S.A. Coal, Oil and Gas Corporation (SASOL) is planning to produce considerable quanti­ties of high melting waxes at its new plant in the Transvaal.

A third matter to which I should like to refer concerns the question of molasses utilisation. There has in the past been much talk of the application of ion exchange techniques in the Sugar Industry, but to date these techniques have found no large scale application. I should like to present for your con­sideration, however, the possibilities of applying a new method of de-salting in the Sugar Industry. This new method of de-salting acqueous solutions involves electrodialysis through ion exchange mem­branes and is under intensive study both in the United States of America and in Holland. We in the C.S.I.R. are planning work in this field in the near future also. In Holland this process has been studied primarily in its application to the removal of mineral salts from saline underground waters, but it has also been applied to the removal of salts from whey and from beet molasses, with most promising results.

Now at the present time the markets—and there­fore the prices—for the many millions of gallons of molasses produced by the Industry every year, are restricted because of its high content of mineral salts and because of the other impurities which it contains. The present development holds out the possibility, however, of processing some portion at. least of this molasses for animal and for human consumption. An immediate result of this, of course, would be that molasses would become less available for fermenta­tion industries or would become available to these industries in a more refined and higher priced form. This need not necessarily curtail production of large scale fermentation products such as alcohol, how­ever, since methods have already been worked out overseas for the production of molasses from wood, of which Natal is endowed with an increasing abundance. Within the C.S.I.R. we are already interesting ourselves in the possibility of such a development being an economic one in the Union. In addition, the availability of molasses in refined form would be a stimulus to the production of other fermentation products, such as citric acid, itaconic acid, etc.

This brings me to the fourth and final point which I should like to discuss—namely the industrial use of sugar as a chemical, or as a substrate for fermenta­tion processes. Whatever may be the world position, the circumstances in South Africa at present are such that the industry will be hard pressed to meet gradually increasing demands for food sugar for some years to come. There is but little prospect of imme­diate advantage to be seen, therefore, in seeking to develop or extend large scale industrial uses for sugar. I feel, however, that your Industry would be unwise to leave all initiative in regard to possible developments of this type to others, and that there should be some work in progress aimed at deter­mining the possibilities of starting production of fermentation and similar products within the Sugar Industry. That there arc others who are thinking along these lines is indicated by an increasing number of reports in the literature of investigations of fer­mentation processes using cane juice itself as raw material.

My main thesis to-day, therefore, is the need for the Sugar Industry to depart from its traditional acceptance of sugar production as its one and only goal, and to seek to develop all possible methods for the exploitation of each of its products. Only by so doing can it contribute the maximum to our national welfare and return the greatest benefits to its members.

I now declare this Twenty-seventh Annual Con­gress of the South African Sugar Technologists open.

Dr. Douwes Dekker: Mr. President, ladies and gentlemen—One of the striking non-technical features of Dr. Rapson's address is that in the beginning we find him confessing to a feeling of astonishment at being here in our midst. Striking, because this con­fession is followed by an excellent, expert discussion of a problem of the utmost importance to the Sugar Industry, which proves that there is no reason at all for such astonishment. On the contrary, I am sure that everyone present here agrees with me that the idea of inviting Dr. Rapson to open this Twenty-seventh S.A.S.T.A. Congress was a most fortunate one, and that we should be grateful for his acceptance of our invitation. Dr. Rapson, by virtue of his being Director of the National Chemical Research Labora­tory of the C.S.I.R., has a thorough knowledge of the plans for development of chemical industries in the Union and we should be glad that he has taken the trouble to include the Sugar Industry in his search for possible projects for new industries which may benefit this country as a whole.

Dr. Rapson has conveyed to us the greetings and best wishes of the Council for Scientific and Industrial Research. I am sure I speak on behalf of our Asso­ciation and all those present here when I thank him for these kind words.

These words, however, are not the only kind ones spoken by Dr. Rapson, but the kindness of the main purpose of his address is of a different nature. In Holland there is a proverb stating that he is your best friend who exposes your shortcomings, and it is in this way I think that Dr. Rapson has endeavoured to demonstrate his feelings of friendliness to the Sugar Industry.

I refer to that part of his address in which Dr. Rapson states that the main purpose of the sugar industries throughout the world is to supply an assured market with one specific product, and that this circumstance has been the cause that on the whole no determined attempts to establish markets for products of the industry other than sugar have been made.

Dr. Rapson has formulated this statement very carefully, referring merely to the establishment of markets, but the latter part of his paper and the fact that this address has been read before the Tech­nologists' Association Congress, both suggest that Dr. Rapson did not intend to confine his words to the commercial field only.

As I see it, Dr. Rapson reproaches the Sugar In­dustry in general for not having paid sufficient attention to the production of secondary products. More attempts should have been made economically to convert products of little value, such as bagasse, filter cake, molasses, and even sugar, into high-priced secondary products.

Perhaps my interpretation of Dr. Rapson's words is simplifying too greatly their meaning, but let us assume that this interpretation is sufficiently correct to be used as a basis for a further discussion.

The desirability, even necessity, for the cane sugar industries all over the world to produce secondary products is a subject which has from time to time been in the minds of I think everyone engaged in sugar. It cannot, however, be denied that for a long time there has been a strong feeling among sugar manufacturers that cobblers should stick to their lasts. Remarkably enough we see, for example, that in many countries even the production of alcohol from molasses is not always in the hands of the sugar companies.

But nowadays this feeling seems to be found in isolated cases only: as a whole the Sugar Industry in genera] is I think very much aware of the potential possibilities offered by the production of secondary products. Two factors have contributed greatly to this change of attitude, t.w. the overproduction and low price of sugar in the thirties, and the example set by the oil industry, the number of valuable secondary products of which is increasing daily.

As a consequence of this change of attitude, the majority of cane growing countries in the world are

able to point either to industries where the by­products of the Sugar Industry—bagasse, filter cake and molasses—are converted into more valuable secondary products, or to institutions of scientific and technical research where such conversions are studied. I do not propose to give you a survey of everything that has been done on this subject, for that would take the whole morning at least, but I should like to refer to the work done in another part of the Commonwealth, i.e. the British West Indies. I single out this work in particular because many of us will attend the International Congress in Trinidad and some pre-information may be useful. For this purpose I have drawn freely on publications by Dr. L. F. Wiggins, the Director of the Sugar Technology Dcpartment of the Imperial College of Tropical Agriculture of that island.

In 1944 a visit was paid to the B.W.T. by Sir Robert Robinson, Professor of Chemistry at Oxford University, and Sir John Simonsen, Director of Research of the Colonial Products Research Council, who clearly saw the need for two research institutions to be set up in the British Colonial Empire, and pre­ferably in Trinidad. These were (i) a Microbiological Research Institute to deal with fundamental micro­biology associated with various crops or products grown or made throughout the British Empire, and (ii) a Sugar Research Institute which would serve in particular the British West Indies Sugar Industry.

The latter after much discussion between the Sugar Industry and the Colonial Development and. Welfare Organization found its shape in the British West Indies Sugar Research Scheme which was inaugurated on September 1st, 1947.

The Sugar Technology Department of the Imperial College of Tropical Agriculture was used as a nucleus for the new scheme and renamed the Department of Sugar Chemistry and Technology. It may be interesting to know that the running expenses are shared on an equal footing between the Sugar Industry of the British West Indies and the Colonial Development and Welfare Funds of H.M. Treasury, similar to the system followed in this country for the financing of the Sugar Milling Research Institute to which the Natal Sugar Millers' Association and the Council for Scientific and Industrial Research contribute. The terms of reference of the scheme exclude the agricultural aspect of the Sugar Industry. Its main function is to study the technology of sugar manufacture, methods of utilization of by-products and some fundamental chemical and biochemical aspects of the sugar cane. The utilization of sugar itself for purposes other than food, which had been the concern of the Colonial Products Research Council also became involved, inasmuch as the Director of Research of the scheme, Dr. Wiggins, superintended this work.

4

5

One of the most important problems of the pro­gramme of work to be carried out under the B.W.I. Sugar Research Scheme—and now I come back to the subject of the paper of Dr. Rapson, who I hope will forgive my short digression from his topic—is the study of the utilization of the by-products of the Sugar Industry. The utilization of filter mud and bagasse have, for the last few years, already been subjects of investigation. Dr. Wiggins states, for example, that paper has been produced from bagasse which is comparable with the newsprint which was used for printing the Manchester Guardian, although the increased beating time did not increase the strength of the paper to a degree comparable with that obtained with wood pulp. He realises, however, that the most valuable product that can be made is pure cellulose for chemical industry. In this respect it may be interesting to mention that more than twenty years ago in Java a considerable quantity of pure cellulose from bagasse was produced on pilot plant scale and was converted in Holland into a good quality rayon.

Work on the utilization of molasses by the Sugar Research Scheme involves investigations into the production by chemical means of lactic acid and levulinic acid.

It cannot be denied that the British West Indies are very much concerned with the utilization of by­products, vide also Mr. Walter Scott's report to the Caribbean Commission on the Industrial Utilization of Sugar Cane By-products.

And the same can be said of many other countries.

In the United States much attention has been paid to the utilization of bagasse. Dr. Rapson has referred to recent pilot plant experiments carried out at. the National Bureau of Standards in Washington where bagasse was converted to newsprint of a quality which was considered acceptable by a panel of news­print executives.

I should, however, also like to draw your attention to an extremely interesting paper by Dr. Joseph E. Atchison, head of the Pulp and Paper Branch of the Mutual Security Agency, Washington, U.S.A., which at my request was reprinted in the South African Sugar Journal, February 1953.

In this paper Dr. Atchison discusses most thorough­ly the production of paper from bagasse and con­cludes that once the pith is removed, there are no technical difficulties in using the bagasse fibre for production of high quality bleached pulp. There are nowadays a few processes available for the reparation of pith and fibre, but whether they are already quite satisfactory is a point on which experts seem to differ. Assuming, however, that this problem has already been solved, the next question is: what are the economics of the production of bleached pulp

from bagasse ? On this point Dr. Atchison is very outspoken. His conclusion No. 7 reads:

"(7) In spite of the number of mills scattered over the world which are now using bagasse as a raw material, no really modern mass production mill has yet come into operation for large scale manufacture of high quality pulp and paper from bagasse. Those mills which are in operation are all located in countries which have low labour costs, high tariffs, import restrictions, or shortages of foreign currency. These factors make it possible for pulp and paper mills in these countries to com­pete in the local markets with low quality products which are sold at high prices. Thus it is possible for small, inefficient mills to operate successfully. Such mills could not possibly compete on either quality or price if they were located in the U.S. or its territories, or if they depended upon the United States for markets. In fact, an analysis of the type operations carried on by commercial mills using bagasse in various parts of the world indi­cates that not a single one of these mills would have much chance of success if they were operated in the same manner in the United States."

Fortunately Dr. Atchison continues in his con­clusion No. 8:

"(8) In spite of the above facts, there is no reason whatsoever that larger highly efficient mills cannot be constructed for using bagasse as a raw material. By combining modern U.S. pulp and paper mill engineering experience with technical know-how of European straw pulp manufacturers, a large scale bagasse pulp mill operation in the U.S., Puerto Rico, Cuba or Hawaii should be highly successful and its costs should be com­petitive in every way with those mills which use wood pulp."

Mr. President, we have come back to Dr. Rapson's statement about the establishment of markets. When one scrutinises technical literature it is appar­ent that technical problems have been solved to a large extent or will be solved before long. The point, however, is that—as demonstrated by Dr. Atchison for bleached pulp—to compete in the open market large and efficient factories, requiring a very con­siderable lay-out of capital are necessary. The alternative is to produce on a smaller scale for the local market. In this case certain conditions which are outside the direct control of the Sugar Industry have to be fulfilled. This path, however, seems to be increasingly followed in many sugar producing countries and the question may be asked what South Africa is doing in this respect. To answer this question one can point to the bagasse working board factory now being erected at Felixton in Zululand, and also to a cane wax extraction unit recently erected at another mill on the North Coast.

The production of products other than alcohol from molasses is increasing.

Other projects have been studied, from which I mention the production of furfural from bagasse.

Whether all this is enough is a point which cannot be answered in a few words. The fact that South Africa was probably the first country in the world to produce cane wax on a commercial scale proves that the leaders of our Industry have an eye open for the potentialities of the production of secondary products.

We should, however, be extremely grateful to Dr. Rapson for again bringing the subject to our notice. His paper will draw the attention it deserves. In my reply I have tried to bring some other aspects of the problem to the fore. To get a proper insight, how­ever, much more is required. In this connection I should mention a committee which was recently appointed by the Hawaiian sugar industry to study the situation as it is now in that territory. Various experts were called in to co-operate in an investiga­tion to find out what could and should be done with respect to secondary products. A similar committee might also be desirable for this country.

Mr. President, the importance of the problem introduced by Dr. Rapson is my excuse for having taken so much of your time. I now propose a hearty vote of thanks to Dr. Rapson for his excellent address.

THE PRESIDENT'S ADDRESS

Mr. G. C. Dymond (President): There is an old Yorkshire saying to the effect that one gets nothing for nothing and precious little for sixpence.

That is a truism that is even more accentuated to-day but, in the case of our Association, I feel that much has been done for little enough. The stalwarts of the past are going and it is sometimes with a feeling of dismay that we wonder who among the younger generation will take their places in the technical side of our Industry.

Can some long-sighted policy be evolved in time, not only to encourage young South African brains to enter the Sugar Industry, but to ensure adequate financial assistance in a progressive plan of co­operative research, or must we forever draw on experts from other sugar countries, who are given educational facilities and financial incentives that our own kind have so sadly lacked in the past ?

Next month the largest South African delegation ever to attend an International Sugar Congress will be on its way to the B.W.I. There they will learn of development plans for research, the evolution and use of by-products, methods of increasing production in field and factory—and they will learn that all this can be achieved only by intelligent planning, co­

operation and the final lavish expenditure of money when required.

The usual procedure is to vote a sum of money, with which organisations are set up, committees formed, buildings constructed and facilities made available for a nebulous object called research. My own view is different. Before spectacular edifices are built, before expensive equipment is purchased and before a staff is engaged, find the man or the Association. Then, wherever he may be, build around him. The rest will follow.

The highlights of history, in whatever sphere of activity you may choose, always centre on individ­uals—individuals who have scorned red tape, musty methods and rusty bureaucracy. Many have been burnt up. A few have succeeded. It will be argued that in these days of mass production and meticulous costing, companies and countries dare not release scientists to pursue their hunches without a tight hold on the reins and the money bags. A few do set aside a yearly amount for this purpose or budget, and immediately write it off as a dead loss. They are the minority.

So, by frustration, are often lost those bright thoughts and ideas which flourish in serenity and die in committee and board rooms. To those young scientists entering the Sugar Industry as routine technologists at the bottom of the ladder, or as qualified scientists in research organisations, I say, render to the organisation the bits and piece-work of the organisation, however dull it may be, but unto yourselves stimulate your ideas, back your hunches, and carry out your individual work outside the forty-eight-hour week even if that means Christmas Day. Or as John Keats says in the Fall of Hyperion: "Fanatics have their dreams wherewith they weave a paradise for a sect."

In accomplishing this you will have to acquire a frame of mind impervious to jealousy, frustration and obstruction, and for an ideal, think of the mes­sage of our Queen who told us to keep alive the spirit of adventure. So in industry in general and in sugar in particular, there is a vast field of unknowns in which the adventurous mind may still find an in­finity of new things.

Our delegates to the B.W.I. should be instructed to investigate the immediate workings and the long-term planning of research, and what is being spent by other sugar countries. This will not be difficult, for the International Sugar Association is a unique organisation in a world of financial and trade barriers. They symbolise the "Three No's:" NO secrecy, NO politics and NO colour or creed bars.

Views and data are freely interchanged and lasting friendships made. There our delegates will find that sugar covers every conceivable scientific activity

6

7

connected with the Industry and that there are no artificial financial barriers between sections of the composite whole.

To be more specific, what are our major problems? Here I would stress that there is no scientific dividing line between field and factory, but only a fictitious barrier. Sucrose, fibre and non-sugars are grown in the fields and the final product fabricated in the factories and refineries. Up to now the cane breeder has confined his attention to the development of higher sucrose canes for longer periods under all conditions of soil, altitude and climate, with no thought as to their possible effect on manufacture. In order to overcome the unknown odds of new varieties in slip, millability, trash content and steam raising, the only answer would appear to be the maintenance of seed beds on each farm and estate and inter-row planting of the various varieties, so as to secure a mixed supply of raw material for crushing.

Another problem is not only the maintenance of yields per acre, but the age-old endeavour of attempt-ting to grow two stalks where but one grew before. The problem of more food for the rapidly increasing populations of the world would seem to be in the same category as a rabbit attempting to catch its own tail.

The International Society is moving a step further in attempting to establish a central collection of all the sugar canes of the world, which may be studied, properly named and made available for research along cytogenetical lines. In this important project South Africa will be asked to join. Slowly but in­exorably we are moving not only to the establish­ment of an international garden for sugar canes, but

to closer collaboration and research on what is one of the most important foods and sources of by­products in the world. As a cheap, highly concen­trated food sugar is unrivalled, its production endless and the number and value, of its by-products almost limitless. Fermentation products, protein foods, waxes and allied compounds, paper and fertilisers all have been touched but barely organised.

The exchange of knowledge is being achieved, but there are many problems which can only be studied and compared by experts who, knowing their own conditions, can pick the bones from the successes achieved on the same problems overseas.

This costs money and it is up to our delegates to bring back a true comparative picture of the results being achieved, and at what cost. Only in this way can the expenditure involved in their presence at the I.S.S.C.T. be justified.

And, finally, I wish to pay tribute to the workers of our own Association, which has in its life of twenty-seven years achieved much with so little. I believe that our Association could be made the pivot and the guiding hand in scientific planning in the South African Sugar Industry. It is the natural link in the chain of research, as between the agricultural and the milling sides of the Industry. It is in close and constant contact with every facet of the Indus­try's progress. It is able to judge objectively and to report factually on research projects. Now, as never before in the history of our Association, there is need for us to help in planning and co-ordinating research in the Sugar Industry. The will is there, and the voluntary enthusiasm. These, and their translation into action, are the salt of progress.

CANE SUGAR IN EAST AFRICA By A. C. BARNES, C.M.G.

Introduction.

The expression "East Africa" for the purpose of this presentation connotes the Kenya Colony and Protectorate, Tanganyika Territory, and the Uganda Protectorate, comprising a total land area of 642,878 square miles, slightly more than four-fifths that of the Union of South Africa. The territories include Lake Victoria and the whole or portions of other lakes amounting in all to 38,901 square miles, of which Lake Victoria situated at an elevation of 3,720 feet above sea level accounts for 26,828 square miles. The gross area is thus 681,029 square miles. Kenya and Tanganyika are bounded on the east by the Indian Ocean. Uganda lies in the interior with access to the sea through the other two. The climate ranges from tropical heat and humidity in the coastal and lake areas to cool and temperate conditions in the highlands, with extensive arid belts in the interior of Kenya and Tanganyika. The population of the three territories is estimated at 18,400,000 of all races. There are 59,000 Europeans and 259,000 Asians. The annual net rate of increase is estimated to be one and a half per cent, for the African population, two per cent, for the Asian, and one to one and a half per cent, for the European.

East Africa generally is in an active state of de­velopment, both agricultural and industrial. Within the memory of many persons, some of whom are still actively engaged in varied pursuits in Eastern Equa­torial Africa, the region has emerged from a state of primitive savagery and slavery to become the source of considerable quantities of varied and greatly needed commodities, including cotton and copper, sisal and sugar, coffee and tea, wheat and flour, pyre-thrum and pineapples. Though electric power has been available in all the larger centres for many years, special mention should be made of the Owen Falls hydro-electric scheme which will make use of the enormous power of the lake water as it passes through a gorge nearly half a mile wide at Jinja and becomes the river Nile.

Man Mau atrocities dominate Kenya news at the moment, but this subversive criminal pagan out­break is localised, and though calling for stern sup­pression, directly affects but a small fraction of the population and total land area.

Historical.

Sugar cane has long been a native crop grown on the edges of swamps or in moist ground, used for chewing and the preparation by crude methods of alcoholic drinks. The places selected by the Africans for cane have had much to do with the local fallacy

that it is a swamp crop. Choice of land has really been dictated by the fact that cane can only survive if supplied with ample water during its period of growth, and most of East Africa experiences little or no rainfall for the greater part of the year. Because of this some of the earlier sugar ventures were established in unsuitable areas.

Cane sugar production by modern methods was commenced in 1924 at Miwani near Kisumu in Kenya. Up to that time locally produced cane sugar was confined to "jaggery," usually made then, as to-day, by direct concentration of the cane juice in open pans. A refinement of that simple method, copied from the West Indies in 1917, was the partial separation of impurities by defecation with lime and heat, followed by concentration to saturation by stages, using open pans. At the strike point the massecuite was ladled into wooden boxes and stirred with paddles until it crystallized. The sugar which was golden yellow in colour was afterwards dried in the sun. The product sold readily at £40 per long ton. This was the first serious attempt at organised cane growing and sugar manufacture.

The cane was grown under irrigation, water being brought by canal from the Athi River over a distance of two miles. The three-roller mill was water-driven. It may be of interest to record that the earliest attempts at chemical control in East Africa were made in connection with this venture. This and other investigational work on sugar cane conducted at that time provided a basis for the later establish­ment, of larger enterprises. Among the significant facts observed were the richness of the canes then grown and the high purities of the juices. Details are to be found in the Annual Reports of the Govern­ment Analyst and Director of Chemical Research, Nairobi, for 1917 and 1918.

Production and Distribution.

To-day there are five estates with vacuum pan factories in operation, two in Kenya, one in Tan­ganyika, and two in Uganda. Jaggery continues to be made, particularly in Uganda, and in western Kenya near to the Kavirondo Gulf, where more than forty units were working in August, 1952. In one district alone 2,000 acres of cane are solely used for jaggery production, the yield being reported as 200 lbs. per long ton of cane, sold then at 10s. per frasila (35 lbs.).

The five factories produce direct consumption sugar by the sulphitation process. The Miwani organisation in Kenya, near Lake Victoria, is the only one which accepts cane from a considerable

8

Fig. 1 irrigated fields of sugar cane. Kiboko Flats, Kenya. 1917

Fig. 2. The first organised attempt at sugar manufacture, Kenya, 1917

9

10

number of farmers, who in almost all cases own jaggery factories, and sell their cane when it suits them. There is no regulation of farmers' production and deliveries, nor any control of the price paid for cane by the manufacturer. One company in Uganda purchases cane from adjoining estates engaged for the main part in other agricultural ventures. Prices and distribution of sugar are controlled by the Director of Produce Disposal, East Africa High Commission. The retail price is the same for all grades of vacuum pan sugar, locally produced and imported, and at present there is no regulation of quality. The price at the factory is based on the Ministry of Food price for direct consumption sugar, the profits allowed to main agents, wholesalers and retailers being controlled.

Details of annual production for the past six years for the periods 1st July to 30th June are given in Table 1.:

TABLE 1 Production by territories in long tons commercial sugar

Kenya Tanganyika Uganda Total

1946-47 ... 4,845 6,565 46,185 57,595 1947-48 ... 10,498 7,206 66,130 83,834 1948-49 ... 11,050 5,593 53,012 69,665 1949-50 ... 11,682 8,546 49,593 69,821 1950-51 ... 13,196 7,585 52,576 73,357 1951-52 ... 14,301 9,952 50,936 75,189

The companies operating the estates and factories are: Kenya: Miwani Sugar Mills (Kenya) (MIWA)

Ltd., Miwani, near Kisumu. Kenya Sugar Ltd., Ramisi, (KENY) via Mombasa.

Tanganyika: Tanganyika Planting Co. Ltd. (TANG) Arusha Chini Estate, Moshi.

Uganda: Madhvaninagar Sugar Works (MADH) Ltd., Kakira, near Jinja. Uganda Sugar Factory Ltd., (UGAN) Lugazi, near Jinja.

Abbreviations used in tables which follow are indicated after the name of each company.

TABLE 2

Production by factories in long tone, 1st July to 31th June MIWA KENY TANG MADH UGAN

1946-47 ... 1,423 3,422 6,565 32,615 13,570 1947-48 ... 6,332 4,166 7,206 40,159 25,971 1948-49 ... 5,384 5,666 5,593 30,535 22,477 1949-50 ... 5,460 6,222 8,546 30,591 19,002 1950-51 ... 8,248 4,948 7,585 33,987 18,589 1951-52 ... 10,428 3,873 9,952 28,481 22,455

Miwani Sugar Mills (Kenya) Ltd. are successors to the Victoria Nyanza Sugar Co. Ltd., the original company operating in Kenya, having purchased the estates and factory in June, 1947.

Details of sugar consumption for the past three years are given in Table 3, which includes the figures for Zanzibar, military requirements, and industry.

T A B L E 3

Long tons, for annual periods 1st July to 31th June

Tangan- Military, Kenya yika Uganda Zanzibar etc. TOTAL

1949-50 . . . 42 ,485 20,297 22,411 4 ,989 348 90,530

1950-51 . . . 39,561 24,316 30,848 4,400 200 99,325

1951-52 . . . 40,747 25,773 33,555 3,900 325 104,300

The difference between local production and con­sumption has been provided by importation through the British Ministry of Food, the annual quantities being:

1949-50 26,665 tons 1950-51 21,650 tons 1951-52 33,932 tons

Consumption.

Estimated consumption for the year ended 31st December, 1952, was 109,030 tons. Excluding Zanzibar and military requirements the consumption of sugar per person on the basis of the figures quoted was at the rate of only 13.2 lbs. per annum. For Kenya alone, where the non-African population is greater than in Tanganyika or Uganda it was 16 lbs. per person. It is evident that demand exceeds supply, and that were more sugar available it would readily sell. During mid-1952 retail purchasers in many large centres, particularly Nairobi, could not obtain their full requirements, while much of the sugar that was on offer was of sub-standard quality and unfit for human consumption. This position continued for some weeks, and was another indica­tion of the inability of the local industry to satisfy consumptive demand. In fairness to the East Africa producers it should be stated that the low quality sugar had been imported.

The Need for Expansion.

A conservative estimate of the present annual shortfall between production and demand is 40,000 tons. This will undoubtedly increase year by year until by 1957 it will be more than twice as much even if the existing producers expand their output from an annual average total of 71,590 tons to 100,000 tons.* The three companies in Kenya and Tanganyika have embarked on schemes for progres­sive increases in production which should reach a

* Weight figures are in terms of long tons of 2,240 lbs.

Fig. 3. Irrigated cane at Arusha Chini, Tanganyika, 1952

Fig. 4. A typical labour "camp'

11

total of 25,000 tons a year in five or six years. Present indications are that annual output when that expansion has occurred may be expected to remain at an annual average of 100,000 to 110,000 tons for the five operating factories unless increased capacity is provided at those in Kenya and Tangan­yika, where the companies concerned have reserves of land capable of providing more cane. Assuming that this took place, a further 25,000 tons a year might be produced by 1959-60, but annual demand would by that time reach or exceed 200,000 tons.

The inevitable conclusion is that East Africa needs an immediately effective development programme of sugar production to overtake the shortfall, keep pace with expanding demand, and make the region self-supporting in respect of sugar. An exportable sur­plus for the growing needs of the Eastern Belgian Congo, the Sudan, and other adjacent countries might be attained with advantage.

In addition to the present expansion projects, there is great need of new enterprises, one initially planned to produce 40,000 tons of sugar per annum, capable of future extension to double that capacity, and one or possibly two of the order of 20,000 tons

output, similarly to be doubled. This may at first glance appear to be an optimistic interpretation of the position, but the evidence of careful enquiries and close observation in the course of an extensive tour of the region rather points to the probable inadequacy of a sugar development scheme even of that magnitude.

Outline of Existing Industry. Climate and soil are favourable to sugar cane

grown under natural rainfall in parts of Uganda and Kenya near Lake Victoria. The deposition of dew-plays an important part in providing water for cane in these areas. The three projects within this zone now provide 80 per cent, of the total annual pro­duction of East Africa. At the coast some forty miles south of Mombasa is the only other enterprise with satisfactory rainfall. Irrigation schemes are how­ever being put in on the two Kenya estates for part of the cane lands. The one company in Tanganyika is operating in an area with an average annual rainfall of 16.73 inches, and the cane lands are irrigated by a gravity supply.

Agricultural operations are highly mechanized on all estates. It should be noted that agricultural

Notes. 1 Area being extended. 2 From official returns. 3 Approximate figures. 4 Farmers supplied cane from 1,650 acres additional. Recorded figures apply to estate operations. 5 Lowest and highest of four stations,

12

13

machinery and equipment, including tractors, imple­ments, and tools essential for their maintenance, are imported free of customs duty from all countries. Imports from hard currency countries are con­trolled. Crop data and other information are given in Table 4.

The cane variety position is not generally satis­factory and though attention has been given to the matter, the selection and testing of new introductions are poorly organised and inadequate for the needs of the estates. A collection of ninety is maintained by the Kenya Department of Agriculture in Kisumu. Among them are seven members of the N :Co. series including 310. The wide range of conditions in East Africa makes it impossible to centralise experimental and varietal investigation. Such work should there­fore be undertaken on the estates' land in such a manner as to study the requirements of each area. The absence of an association of cane planters and sugar manufacturers is an undoubted drawback to co-ordinated investigations as well as to the resolu­tion of industrial problems. The failure to form such a body appears not to be related to economic interests but to some unconnected conflict of views. Thus at any time it is possible to bring four out of the five companies into union, but the fifth, which may be one or the other of two, is consistently in opposition. The solution may be found in some form of legal sanction, as the industry is unquestionably of high importance in the general economy of the region.

Labour.

The employment of labour in East Africa is con­trolled and regulated by law, the principles of the legislation being basically similar in each territory. Labour departments administer these laws as well as those concerning health, safety and welfare in factories. The establishment of industrial relations between employers and employees by the formation of staff associations and works committees is en­couraged. Official annual reports are published in respect of labour administration.

The provision of housing by the employer is a legal obligation unless employees can return to their homes at night, and on the sugar estates labour camps provide accommodation for most, if not all, of the labour employed. The site, design, and con­duction of labour camps are subject to approval. The number of persons to the acre and to each room are regulated. Health and welfare services are com­pulsory, and rations must be supplied for unskilled labour in accordance with approved scales. The recruitment of labour is conducted by licensed re­cruiters who may operate only in particular districts which are specified from time to time.

Communications. The map at Fig. 5 shows existing and projected

railways and the routes of lake and river steamer services within East Africa. Four of the five sugar enterprises are on or near the railway. The other depends on road transport, though communication with other coastal areas by water could be developed. The Arusha Chini factory is connected with Kahe Junction by an E.A. standard gauge tramline eight miles in length over which the estate traffic is manually propelled. The road system of the terri­tories is extensively used for heavy traffic. The East African Railways and Harbours is the controlling authority for rail and steamer services, as well as for bus services on certain main routes. The map reveals projected railways of which that from Kampala to Kasese is under construction. Another important link is the proposed new line in Tanganyika which will run from Korogwe south-west to join the Central Line. The route has not been finally deter­mined, but if constructed, the line will pass through a promising new area for a cane sugar project near Turiani. The possibility of a rail link with Rhodesia has been explored.

The success of cane sugar ventures depends upon ease and cheapness of communication and transport, as well as upon the climatic, agricultural, and tech­nical requirements of successful cane production and sugar extraction. The locations suitable for new enterprises are limited by these considerations, and by the availability of sufficient areas of land within economic distance of satisfactory sites for factories. Though large scale agricultural undertakings in East Africa are of modern origin, they have been estab­lished in most instances by private enterprise with no regional crop planning schemes. This has brought about a form of ribbon development along the lines of communication, especially the railways, in zones where soil and climate are suitable for particular crops. A striking example of this is sisal in the mid-Pangani valley of Tanganyika where a chain of monocrop cultivations follows the track of the Tanga-Moshi railway line. Excellent conditions for large scale sugar production under irrigation are presented in this area, but they arc also highly suited to sisal, and it seems unlikely that there will be a change to sugar.

Factors Affecting New Enterprises.

There are, however, extensive runs of land where conditions are favourable to the success of new sugar enterprises, though with the exception of the north­eastern lake area irrigation would be desirable, and in some cases essential. Apart from the advantages of the lake climate, the availability of ample supplies of water of excellent quality is important. The average temperature of lake water calculated on the

14

mean of figures obtained weekly at four depths ranges between 24.1°C in August to 26.7°C in March.

An important feature is that in all the developed and undeveloped cane areas, reaping and manu­facturing operations can be conducted throughout the year. Factories can count on not less than three hundred days' gross grinding time. It is only during the season of heavy rain that there is any serious interruption of operations, and this is of short duration. These factors exercise a highly favourable effect on the organisation of plantation work and factory economy. Staff and labour can be contin­uously employed on the same operations, with im­proved efficiency and the absence of seasonal peaks.

The governments of the East African territories have carried out extensive studies of climate, hydro­logy, and soils, the results of which are being used to make preliminary selections of land suitable for various forms of agricultural development including sugar cane. Further intensive investigations are then made in relation to the conditions of the particular enterprise decided upon for each area.

New cane sugar ventures of economic size and output require heavy capital outlay, which is much greater than was needed prior to the last war. Despite this the exceptionally favourable conditions in East Africa, which are here described, support the view that well planned and properly conducted enterprises will prove successful and profitable.

Mr. Dymond said that he thought everybody would agree that Mr. Barnes had given an interesting picture of conditions in East Africa. He was particularly interested in Mr. Barnes' description of the large number of jaggery mills in a small area. This reminded him of Brazil where fifteen years ago there were over 40,000, most of them of the type described by Mr. Barnes. Under economic pressure and development these were absorbed into the larger centrals.

Mr. Walsh paid tribute to Mr. Barnes for his fine description of the East African sugar industry. No one was more capable of doing this than Mr. Barnes and he paid tribute to the work done by him in East Africa. Mr. Walsh said he had visited East Africa several times in the past ten years and had been struck by the lack of advancement and the lack of co-ordination. Much of the operation was done by officials who had little experience. He said that South Africa's procedure was being very closely followed in East Africa and there was a great deal of jealousy at the co-ordination and activity in South Africa. South Africa had many links with East Africa. In the Miwani plant was the milling plant from Reunion and equipment from Verulam and Sezela. There were many South Africans in the area who welcomed a link with South Africa.

Mr. Walsh said he was sure Mr. Barnes could go further into the possibilities of East Africa. It was fantastic to think of factories that could crush for fourteen months at a stretch. He felt more could have been done in East Africa if they had had the advantage there of an organisation such as South Africa had, together with better canes and better technical supervision. If this were done, the future of East Africa would probably be illimitable.

Dr. Dodds endorsed Mr. Walsh's tribute to Mr. Barnes' paper. He said that some years ago at the Experiment Station a letter of enquiry had been received from East Africa asking about the effect of cold water on sugar. Some of the water used in East Africa was very cold, just a few degrees above freezing point.

Mr. Barnes, replying to Dr. Dodds, said that he knew of no cane subject to irrigation in such con­ditions. The only concern using irrigation was the one in Tanganyika situated at 4,000 ft., where it was intensely hot during the day and where the water temperature was probably not less than 35 degrees; this in spite of the fact that the water came from the snows of Kilimanjaro. He thought this query might have come from one of the remoter parts of the territory. One venture near Nairobi had failed entirely because of the slow growth of the cane.

Dr. McMartin said one point occurred to him in connection with the varieties grown. He asked whether it was known whether the early canes grown by the natives were grown before the white people went to East Africa. A moot point was whether sugar cane in Natal preceded the Europeans or the Europeans preceded the canes. The point was held by some people that the cane cultivation came only after the Europeans while others felt that sugar cane was distributed in the very early days, mainly by the Arabs, and that the natives continued this distribution southwards into Natal.

Mr. Barnes replied that it was a very interesting speculative subject. He would say unquestionably that the canes were grown in East Africa before the advent of the Europeans. David Livingstone men­tioned cane in some of his writings. He said Uganda had been in contact for a thousand years via the Nile with Egyptain civilisation, while Arabs had penetrated what is now Tanganyika. Some of the canes in East Africa had local names. A remarkable thing was that many of the locally grown cane had persisted through a long period of years without being affected by mosaic. He expressed the view that sugar cane was grown in these territories and was used by natives long before the European arrived.

Mr. Dymond congratulated Mr. Barnes and asked that a hearty vote of thanks be accorded to him for his paper.

Fig. 5. Railway and steamer commuuications map.

15

16

TWENTY-EIGHTH ANNUAL SUMMARY OF CHEMICAL LABORATORY REPORTS

SOUTH AFRICAN SUGAR FACTORIES, SEASON 1952-1953

By CHS. G. M. PERK.

The "Final Manufacturing Results" have this year again been compiled in the form of three tables:

Table I comprises the following data:

(i) The data for the beginning and the end of the crushing season.

(ii) The weights of cane crushed.

(iii) The quality of the cane crushed.

(iv) The percentages of the cane varieties crushed.

(v) The weights and analyses of the sugars made, and

(vi) An account of time lost by stoppage.

Table II contains data concerning:

(vii) The sucrose balance.

(viii) Lost absolute juice per cent, fibre, sucrose extraction, boiling house performance and recoveries; and the analyses of—

(ix) Final bagasse, (x) First expressed juice,

(xi) Last expressed juice, (xii) Mixed juice, (xiii) Clarified juice, (xiv) Filter cake, and (xv) Syrup.

Table III gives data concerning:

(xvi) Analyses, quantites and crystal content of massecuites.

(xvii) Data on final molasses.

(xviii) Lbs. of chemicals used per ton of cane, per ton of sugar, and per 1,000 lbs. of brix in mixed juice.

In addition to thse tables three more tables are shown:

Table IV giving a comparison of important final data of the last decade.

Table V showing the monthly results of the last crushing season, and

Table VI comparing Natal's data with those of some other countries.

A.—General.

During the 1952-53 season nineteen mills were again in operation, seventeen of which contributed to this Annual Summary. Our discussions will include data and results of these seventeen mills which crushed 98.9 per cent, of all cane crushed, and produced 99.3 per cent, of all sugar produced; the total cane crop of the nineteen mills being 5,722,583 tons and the quantity of sugar produced being 670,188 tons.

Since an additional 137,672 tons of sugar were produced, the above results compare favourably with the results of the previous season; in 1951-52 the total cane crop being 4,805,249-tons and the sugar produced 532,505-tons. Compared with the 1950-51 season last season's results show only a slight deficit in sugar production; viz. only 15,621 tons less. The results of the 1950-51 season were 5,721,390 tons of cane crushed and 685,798 tons of sugar made.

The aggregate number of hours actual crushing of the seventeen mills during the 1952-53 season amounted to 70,274 hours, against 60,305 hours during the 1951-52 season, and 71,145 hours during the 1950-51 season. These figures amount, to an average of 172 days per mill during the 1952-53 season; 148 days during the 1951-52 season and 174 days during the 1950-51 season.

The average throughput figures based on cane, fibre, brix and sugar for the last three crushing seasons are shown below. Attention is drawn to the fact that not the same milling tandems were in operation in all three seasons.

Per Hour Actual Crushing. Season.

1952 1951 1950

Tons of cane crushed ... 80.56 78.96 79.57 Tons of fibre crushed ... 12.97 12.85 12.57 Tons of brix processed ... 12.05 11.53 12.20 Tons of sugar made ... 9.47 8.79 9.56

of 96° pol sugar, against 8.76 tons in 1951-52 and 8.11 tons in 1950-51.

B.—Milling Control Data.

Lost Absolute Juice in Final Bagasse % Fibre.

1952 1951 1950

UF 52 49 46 ZM 47 42 42 FX 47 42 43 EN 48 44 43 AK 37 35 37 DK ... ... 38 42 37 DL 55 54 53 GL 31 41 41 MV 48 45 45 CK 41 41 39 TS 32 40 37 NE 36 33 30 IE 38 36 36 RN 33 31 33 ES 36 38 35 SZ 38 36 34 UK 38 34 35

Arithmetical averages ... 40.8 40.2 39.3

Since brix includes sucrose as well as nonsucrose, both of which materials are to be processed, "tons of brix" is a better yardstick for the load imposed upon the boiling house than "tons of sugar made". The same can be said about "tons of fibre" as a yardstick for the load imposed on the milling trains compared with "tons of cane". To sub­stantiate this opinion we have only to point to the difference in fibre content of the cane crushed by Umfolozi (12.95 per cent, fibre) and by Amatikulu and Zululand (17.71 and 17.36 per cent, respec­tively). There are, however, also appreciable differences, from the point of view of processing, between cane from one factory and another; not only in so far as the quantity of juice is concerned but still more in regard to the purity. Thus "tons of cane" is not only impracticable as a base for evaluating the load imposed on the milling trains, it. is also impracticable for indicating the load imposed on the boiling house. The numbers of cu.ft. of massecuites boiled and of tons of steam consumed are linearly proportionate to "tons of brix in mixed juice"; such a relation does not exist, however, between the two and "tons of cane crushed".

The amount of N:Co.310 crushed during the 1952-53 season increased to 38 per cent, against 21 per cent, in 1951-52, and 15 per cent, in 1950-51. Generally speaking, when a new variety of cane has arrived at the point when more than fifty per cent, of all cane crushed is the new variety, complaints about difficulties in the crushing of it and in burning of its bagasse cease to be heard. It will be worth­while to observe whether this will also be the case with N:Co.310, when more than fifty per cent, of the cane crop will consist of this variety.

Except for the high fibre content of the cane (partly due to adhering trash) the quality of the cane crushed last season was far better than that crushed in 1951-52; last year's cane quality approxi­mating closely to that of the 1950-51 season. Firstly, the average purity of mixed juice amounted to 86.25 while in 1951-52 the average was only 84.9, and in 1950-51 86.4. Secondly, the average sucrose content of the cane was 13.87 per cent, during last season as compared with only 13.33 per cent, in 1951-52; the average of the 1950-51 season, however, amounted to 14.19 per cent.

Since the number of tons of cane required to produce one ton of 96° pol sugar is dependent (a.o.) on the fibre and the sucrose content of the cane crushed and on the purity of its juice, this number is a fairly accurate yardstick for evaluating the quality of the cane. In the 1952-53 season 8.27 tons of cane were required to produce one ton

This year only Gledhow and Tongaat show con­siderably improved figures compared with the previous two seasons. Most mills, however, show less favourable figures. On the whole the mill performance as indicated by the Lost Absolute Juice figure has declined over the period 1950 to 1952. On the other hand, the fibre throughput has steadily increased from 1950 to 1952. Very likely the imbibition was raised to counteract the effect of the increasing throughput; however, it could not prevent the Lost Absolute Juice figure from rising.

The joint influence of the increased throughput and imbibition is shown in a higher moisture content of the final bagasse.

17

1952 1951 19S0

Tons of fibre crushed per hour actual crushing 12,97 12.85 12.57

Imbibition % Fibre ... 217 215 206 Moisture % bagasse ... 52.53 51.71 51.22

The figures calculated for Imbibition Efficiency (see the 27th Annual Summary for the calculation and the meaning of this figure) are as follows:

UF ZM FX EN AK DK DL GL MV CK TS NE IL RN ES SZ UK

Arithmetical average

Imbibition Efficiency.

1952

43 44 49 27 51 80 55 76 44 64 49 42 49 59 41 53 47

51.4

1951

46 39 36 27 43 65 62 57 52 54 48 39 41 43 40 64 41

46.9

1950

30 44 38 35 42 92 65 64 45 55 45 43 41 55 42 46 39

48.3

C.—Boiling House Performance Data.

During the 1952-53 season some factories showed monthly Boiling House Performance figures quite near to one hundred per cent. These figures gave rise to enquiries in regard to how it is possible to attain a B.H.P. figure of one hundred per cent, or even more than one hundred per cent. It is because of these enquiries that we want to give here the following explanation.

Boiling House Performance indicates the per­centage ratio between (a) crystallized sucrose or crystal in sugar, and (b) crystallizable sucrose (i.e. sucrose available for crystallization) in mixed juice.

In order to account for the differing compositions of sugars of different pol, instead of pol in sugar as is done when Boiling House Recovery is cal­culated), crystal in sugar is chosen as the basis of comparison in the case of Boiling House Perform­ance.

To calculate the crystallizable sucrose in mixed juice with only the aid of the purity a few assump­tions have to be made:

(a) the difference between the total quantity of sucrose in mixed juice and the quantity of crystallizable sucrose in mixed juice is the quantity of sucrose which is expected not to be recovered as crystal in sugar. It is the

quantity of sucrose which is expected to be lost (i) in final molasses, (ii) in filter cake, and (iii) undetermined;

(b) to simplify matters, however, somewhat in­correctly it is assumed that the losses (ii) and (iii) are of the same natureas the losses in final molasses, which in any case constitute 75 per cent, of the total losses. Since the losses in final molasses comprise the greatest part of the total losses it is assumed that the. total losses of sucrose are proportionate to the amount of nonsucrose in final molasses as are the losses of sucrose in final molasses. They are, however, also slightly affected by the quality of the nonsucrose (see (c) );

(c) it is further assumed that the quantity of nonsucrose in final molasses is proportionate to the quantity of nonscurose in mixed juice (N.B.—Making this assumption is virtually setting a standard for the clarification effect.);

(d) the nature of the nonsucrose in final molasses is determined by the nature of the nonsucrose in mixed juice since it is assumed that the clarification effect will always be the same;

(e) lower purity mixed juice corresponds to lower purity final molasses, due to a higher reducing sugar/ash ratio. (The sucrose retaining power of nonsucrose in final molasses originating from low purity mixed juice is less than the sucrose retaining power of nonsucrose from high purity mixed juice.)

Were it not for this last assumption (e), total sucrose expected to be lost would be proportionate to nonsucrose in mixed juice, and the original Winter formula with a fixed factor (i.e. 0.4) could be applied. Since, however, the nature of the non­sucrose is to be taken into account, a Winter co­efficient increasing with mixed juice purity should be used.

It is obvious that a Boiling House Performance — 100 will be attained by those mills which have sucrose losses equal to the losses anticipated in the formula. If these losses are less, the Boiling House Performance will be more than 100. For example, this will be the case when the quantity of final molasses is less owing to a better clarification effect, or when the final molasses is better exhausted. Since variations in purity of mixed juice and sugar are accounted for in the calculation of the Boiling House Performance figure, the figure can be used as a yardstick for performance. It is a qualitative figure, in contrast with Boiling House Recovery which is purely a quantitative figure, as is Ex­traction. Since Boiling House Performance is a

18

19

qualitative figure, it is possible to attain a qualific­ation of one hundred per cent, and even more when the actual performance has exceeded the target. Such an achievement is not possible in the case of Boiling House Recovery, since it is impossible to recover more sucrose (pol) in sugar than was present in mixed juice.

We shall elucidate the above by comparing the B.H.P. figures with the B.H.R. figures obtained in two cases with different mixed juice purities and sugars with different pols. In the first case we start from 100 tons of sucrose present in mixed juice of 84 purity, and in the second case from 100 tons of sucrose present in mixed juice of 87 purity. In the first case the average pol of the processed sugars is 99.4°V. and in the second case 98.3° Ventzke. In both cases it is assumed that a Boiling House Pei-formance of one hundred per cent, is obtained.

I II

Sucrose in mixed juice (A) ... 100 100 tons Crystallizable sucrose in mixed

juice 90.9 92.4 tons Roiling House Performance ... 100 100 per cent Available Crystal in sugar (B) ... 90.9 92.4 tons Crystal in sugar (C) 99.1 97.6 per cent. Sugar made (B x 100)/C = D ... 91.7 94.7 tons Pol of sugars (E) 99.4° 98.3 Ventzke Available pol in sugars (D x E) /

100 = F 91.2 93.1 tons Boiling House Recovery (F x

100)/A 91.2 93.1 per cent.

Notwithstanding that the quality of the work done was the same (the performances in both cases came up to standard) the second case shows a figure nearly 2 per cent, higher for Boiling House Recovery.

Purity of Purity of Boiling House Mixed Juice. Filial Molasses. Performance.

1952 1951 1950 1952 1951 1950 1952 1951 1950

UF ... 84.0 84.4 83.7 40.0 41.8 40.3 95.55 94.85 96.96 ZM ... 86.0 84.9 85.8 40.9 40.2 40.9 95.75 96.16 96.38 FX ... 84.7 83.8 84.6 37.8 37.8 38.6 98.06 98.23 97.67 EN ... 87.3 87.0 87.9 42.1 43.5 42.2 95.67 93.79 95.66 AK ... 85.9 85.4 86.7 36.8 40.1 41,7 98.72 96.94 97.48 DK ... 86.2 85.0 85.9 39.4 38.5 37.6 95.Jl 96.19 93.86 DL ... 86.9 85.6 86.3 37.6 39.3 38 8 97.97 97.16 96.63 GL ... 86.7 85.8 85.8 39.1 40.1 41.0 98.43 96.70 96.76 MV ... 86.9 85.0 86.2 40.8 40.6 30.1 97.08 94.70 96.03 CK ... 86.5 85.3 87.1 39.5 38.5 40.4 98.47 97.71 97.92 TS ... 87.2 85.2 87.9 39.0 40.5 41.1 97.82 97.17 96.50 NE ... 86.6 85.1 86.2 47.2 43.7 47.3 96.49 97.22 96.92 IL ... 85.2 82.9 85.0 42.0 42.7 42.4 96.82 94.94 06.56 RN ... 86.9 83.2 86.5 40.7 39.0 40.1 94.79 95.07 96.61 ES ... 87.4 86.5 88.7 39.0 37.8 38.8 97.89 97.26 97.31 SZ ... 87.0 85.0 87.1 37.7 38.0 38.2 97.77 97.51 97.47 UK ... 86.6 85.0 86.6 39.0 40.9 40.1 97.70 96.65 97.50

-Means ... 86.25 84.9 86.4 39.3 40.3 40.5 97.17 96.66 96.88

When we peruse the table of Boiling House Per­formance figures we see that in most cases the improvement in performance is due to a lower purity of the final molasses and that the drop in performance is caused by a less exhausted molasses. In some cases, however, a reduction or an increase in undetermined sucrose losses has also had an effect on the efficiency.

Since the average figures are weighted means we may conclude that the sugar industry as a whole has improved its performance, mainly due to a lower purity of the final molasses.

Since the reducing sugar content of mixed juice of lower purity is usually higher than the reducing sugar content of mixed juice of higher purity, in general a lower molasses purity may be expected when mixed juice purity is lower. A lower purity of mixed juice may correspond with a better ex-haustibility; however, this also implies a greater quantity of molasses to be handled by the low grade equipment. A greater quantity of molasses to be handled means a reduction of the spinning time of the low grade centrifugals. On the other hand a molasses of lower purity requires a longer spinning time. These two characteristics will con­flict when the capacity of the low grade centrifugals is insufficient. In such a case it can occur that the molasses purity has to be raised in order to reduce the spinning time.

D.—Recovery Data.

When compared with the 1951-52 season, the average Overall Recovery of 1952-53 season has improved more than one unit, viz. from 82.5 to 83.7. Since the average Extraction is the same for both years, this improvement is caused only by the rise in average Boiling House Recovery. The improvement of this average figure is due for the greater part to the increase in the average purity of the mixed juice (from 84.9 to 80.25) arid to a lesser extent by the reduction of the purity of the final molasses (from 40.3 to 39.3).

When compared with the 1950-51 season the average Overall Recovery remained the same, since the average Extraction decreased to the same extent as the average Boiling House Recovery increased. The decrease in Extraction is caused by the increase in fibre content of the cane. The increase in Boiling House Recovery is caused in this case by a lower molasses purity (reduced from 40.5 to 39.3) since the purities of the mixed juices arc about the same (86.4 and 86.25) for both years.

E.—Sucrose Balance.

Since it is impossible to obtain an insight into the sucrose losses occurring in the boilinghouse, without knowing the sucrose loss in final molasses, the average sucrose loss in final molasses had to be estimated in order to obtain the average sucrose balance of the seventeen factories. The estimation of the sucrose loss in final molasses for those factories which do not yet weigh their molasses is based on the assumption that 83 per cent, of the nonsucrose in mixed juice is to be found in final molasses.

This supposition assumes, however, that no ab­normal losses of nonsucrose have occurred, and that exactly 17 per cent, of the nonsucrose in mixed juice is expelled by the filters. The value of this estimation is thus limited. However, it is the only way to procure an average sucrose balance.

Keeping the restricted value of the estimated average sucrose loss in final molasses in mind, the slight increase of undetermined sucrose losses com­pared with the average figure of the 1951-52 season (when the loss was calculated in the same way) is less than the possible error in the determination of the figures.

The average sucrose losses in bagasse for the last three years are fairly proportionate to the fibre contents of the cane during those years. The average performances of the milling trains in the last three seasons, as indicated by the average Lost Absolute Juice figures, do not differ to the extent that they have affected to an appreciable degree the sucrose losses in bagasse; the noticeable differ­ences being due to the difference in fibre content in cane.

F.—Final Molasses Data.

Since 1936 the average purity of the final Natal molasses has been lower than 14; since 1945 lower than 42, and since 1950 lower than 41; the average figures obtained during the last three seasons being 40.5 (1950); 40.3 (1951) and 39.3 (1952). It may be presumed that before 1936 it was assumed that the average purity of the Natal molasses could never be reduced below 42 or even below 44, and even now some will wonder whether the average purity can be pushed below 39, but time will show that 39 is not the ultimate goal.

G.—Massecuite Data.

Just as in the case of the previous Summary the average massecuite and run-off figures do not include the data from Natal Estates and Illovo, since these factories use boiling schemes which are quite different from the general ABC boiling scheme of the Natal factories.

Cubic Feet of Massecuites pet Ton of Brix in Mixed Juice.

1052 1951 1950

Crystal Content of Massecuites.

1952 1951 1950

A massecuite B massecuite C massecuite

Total Ditto per ton of sugar

... 22.2

... 10.1 6.5

... 38.9

... 49.5

21.1 11.5

7.5 40.2 52.9

34 6

7.1 41.7 53.2

A massecuite

B massecuite

C massecuite

Purity C massecuite

52.1 51.3 56.1

43.8 43.0 46.9

33.9 32.6 35.7

60.5 60.8 61.8

The number of cu.ft. of C massecuite depends on (a) the purity of the mixed juice, (b) the purity of the C massecuite and (c) the purity of the pre-cured C sugar; the latter controlling the amount of non­sucrose which, by being re-cycled, will pass the C massecuite once more. The lower number of cu.ft. of C massecuite boiled during the 1952 season is caused by (a) the higher purity of mixed juice, (b) the lower purity of the C massecuite and pro­bably also by (c) a higher purity of the pre-cured C sugar.

When a C massecuite does not contain sufficient crystal it cannot be concentrated to a high brix without false grain occurring. It is therefore essential that sufficient crystal is available to obtain a good exhaustion of the C massecuite. However, a too high crystal content makes the massecuite too stiff a material to handle properly.

Let us assume that we make artificial C masse­cuites by mixing 30, 35 and 40 parts of crystals with 70, 65 and 60 parts of final molasses respectively. When the purity of molasses used is 39 then the purities of the masscuites will be 57.30; 60.35 and 63.40 respectively. On the other hand when we try to boil, cool and purge real massecuites of these compositions, the following will appear:

The first type of C massecuite cannot be con­centrated without false grain occurring because it does not contain sufficient crystal surface area. This phenomenon is not a result of the low purity, as is assumed sometimes, but of a too low crystal content.

The third type of C massecuite will become stiff and unyielding when it is concentrated sufficiently to make the drop in purity from 63.4 (massecuite) to 39.0 (molasses) possible; and when concentrated it cannot be cooled because when cooled it cannot be stirred: the crystal content is too high.

The second type of C massecuite can be con­centrated and cooled without any difficulty, because it contains just about the right amount of crystal. When re-heated after cooling it will purge easily.

20

21

H.—Chemicals. Per 1000 parts of brix in Mixed Juice the following

quantities of chemicals have been consumed during the last three seasons:

Part. Of Farts of Lime. Parts ol Sulphur. Phosphoric Paste.

1052 1951 1!)50 1952 1951 1950 1951

19.3 24.7 12.8

20.1 27.6 l 15.5

0.8 7.0 6.0 10.7 2.7 5.0

7.9 7 10.6 10

4.4 i

Doornkop and Darnall. in particular reduced their consumption of chemicals considerably during last season.

I.—Fuel Data.

In Table II under "Final Bagasse" the quantities of bagasse indicated as percentages of cane are shown again together with the analyses and the lower calorific values of the bagasse. The average quantity of bagasse during the 1952-53 season amounts to 36.59 per cent, on cane and the average L.C.V. to 3,064 Btu./lb. Per 100 lbs. of cane 36.59 x 3,064 = 112,112 Btu's were available. If all this bagasse was burnt with a boiler efficiency of 70 per cent., 81 lbs. of steam from and at 212°F. would have been produced, which is even more than required for the manufacture of 100 per cent, of plantation white.

The additional fuel consumed by the boiler furnaces of the factories, excluding fuel consumed by locomotives, workshops, for irrigation purposes, etc. is shown below. The table shows the tons of

coal and wood combusted in the 1952-53 season; and also the equivalent tons of bagasse consumed in the 1952 season compared with the equivalent tons of bagasse consumed in the previous two seasons.

For the conversion the low Calorific Value of Natal coal is the assumed to be 33/4 times, and air-dry wood 1.2 times the L.C.V. of average Natal bagasse

Mr. Dymond said this type of paper was very valuable, especially to members of mill staffs. It was a careful, factual study that contained much valuable data. He asked that Mr. Perk be accorded a hearty vote of thanks.

Sugar Milling Research Institute, Durban.

Table I.—CANE CRUSHED, CANE QUALITY, VARIETIES, SUGARS PRODUCED, TIME ACCOUNT AND THROUGHPUT.

UF. ZM. FX. EN. AK. DK.

Crushing period fFrom To

CANE CRUSHED f Tons of 8,000 lbs. \Mttrictons

, 20.6.52 8.4.S2 2.4.52 21.5.52 7.4.32 15.5.52 2.6.52 15.5.52 23.5.52 23.4.52 16.4.52 7.5.52 29.5.52 21.5.52 5.5.52 5.5.52 6.6.52 2.4.52 9.2.53 4.11.52 31.1.53 6.12.52 26.1.53 24.12.52 16.2.53 13.1.53 9.12.52 17.1.53 16.11.52 20.12.S2 6.11.52 10.11.52 17.10.52 24.10.52 22.10.52 16.2.53

670,373 395,305 520,284 56,11)8 500,225 127,737 500,141 281,133 138,673 216,846 700,446 640,214 183,500 140,320 156,363 293,938 74,102 5,661,604 517,442 358,615 477,444 50,982 453,797 115,881 509,311 345,759 125,802 196,539 635,070 580,793 166,469 127,302 141,850 266,656 67,224 5,136,123

TOTAL RAINFALL during 1952

SUGARS White Sugar

Tons of 2,000 lbs. Government Grade Raw Sugar

j • , . / T o n s of 2,000 lbs Sugar made and estimated J SO, p.p.m. in White Sugar SO, p.p.m. in Government Grade Polarization of Government Grade Polarization of Raw Sugar Safety factor of Raw Sugar Average Polarization of all Sugars White Sugar per cent, total Sugars made

OVERALL TIME EFFICIENCY (Hours Actual Crushing per cent. Hours Mill Open)

Total Hours of Stoppage per cent. Hours Mill Open Hours of Stoppage due to Shortage of Cane per cent. Hours

Mill Open

THROUGHPUT per hour actual crushing:

Tons of Cane crushed Tons of Fibre Tons of Brix processed Tons of Sugar bagged

90 — 16,240 4,804 — 10,700 — 31,532 2,870 15,032 43,105 '2,033 1,630 13,613 3,870 387 14,492 12,800 48,226 3,118 33,635 — 44,111 — 63,724 497 1,105 63,347 40,223 57,514 6,530 58.695 14,570 64,111 46,520 16,781 57,469 41,S33 52,170 5,924 5,1,247 13,217 58,ltil 42,202 15,223

59 75 — — 88 — 112 76 — 98.17 98.23 98.27 98.29 98.29 08.58 97.79 97.88 98.28 98.24 97.75 98.52 — 98.48 — 98.06 97.88 98.38 0.38 0.32 0.35 — 0.30 — 0.31 — 0.28

98.22 98.19 98.82 99.42 98.44 99.45 98.06 99.22 98.55 — — 28 75 — 73 — 68 17

88.8 95.5 96.2 95.7 94.3 91.6 87.2 94.8 95.0 11.2 4.5 3.8 4.3 5.7 8.4 12.8 5.2 5.0

3.7 1.5 0.4 0.4 1.9 4.2 2.0 0.9 1.5

136.97 96.88 87.07 14.27 87.93 30.71 120.56 81.03 36.07 17.74 16.82 14.31 2.27 15.56 5.04 19.98 13.54 5.73 20.60 14.57 12.34 2.13 12.96 4.61 17.06 12.41 5.48 15.21 11.33 9.52 1.66 10.32 3.50 13.79 9.89 4.30

13,479 11,017

— 25,107 22,777

— 98.58

99.24

94.2 5.8

2.1

42.52 7.15

— 8,827 79,295 88,122 79,943

— 98.79 98.79

0.29 98.79

94.8 5.2

0.9

169.80 26.50 26.60 21.36

70,434 6,021

— 76,455 69,359

_ 98.44

99.78

1.3

0.2

139.32 22.10 21.58 16.64

— — 21,696 21,096 19,682

— — 96=

96° 125R*

94.6

2.2

61.79

9.21 7.30

10,545 6,479

__ 17,024 15,441

— 98.30

99.23 62

97.5 2.5

1.2

42.06 6.74 6.62

13,227 5,888

— 19,115 17,341

37 53

98.34

99.42 69

97.0 3.0

0.5

47.84 7.52

23,882 10,467

— 34,349 31,161

31 57

98.26

99.35 70

87.0 13.0

9.5

97.78 15.93 14.45 11.45

53 4,496 4.938 9,487 8,606

— 98.42 98.42

98.42

95.3

2.3

28.31 4.35

3.62

197,958 161,363 305,348 665,646 603,884

— 98.29 98.43

98.63

93.8 6.2

2.0

80.56 12.97 12.05 9.47

5,000 tons of bought i t Arithmetical average.

Table II.—SUCROSE BALANCE. RECOVERIES, BAGASSE. JUICES, FILTER CAKE AND SYRUP.

UF. ZM. FX. EN. AK. DK. DL. GL. MV. CK. TS. NE.

SUCROSE BALANCE (Sucrose % Suciose in CaneJ Sucrose in Bagasse (A) 7.08 8.80 7.99 8.15 7.30 6.83 Sacrose in Filter Cake (B) 0.63 0.33 0.38 0.88 0.47 0.52 Sucrose in Final Molasses (c) 8.72 8.69 7.81 7.26 6.94 7.75 Sucrose in Undetermined Losses (D) 2.00 1.24 1.04 2.30 0.71 3.15 Sucrose Jost in Boiling House (s)+(c) + (n) H-33 10.26 9.23 10.44 8.12 11.42 SucrosefaiTotAlLosses(A) + (B) + (e)-L.(D) 18.41 19.06 17.22 18.59 15.42 18.25

10.00 6.57 8.38 6.82 5.19 5.75 7.04 5.93 5.85 6.14 6.12 7.00 0.46 0.65 0.67 0.29 0.10 0.53 0.21 0.29 0.14 — 1.55 8.43 6.30 6.3i 7.05 — — 7.06 9.44 _ __ _ _ 7.45 1.12 1.40 1.16 — — 2.72 0.36 — — — — 1.46 7.88 8.36 8.88 8.31 8.52 10.31 0.84 11.35 8.55 9.22 8.78 9.34

17.88 13.93 17.26 15.13 13.71 16.08 16.88 17.28 14.40 15,36 14.99 16.34

LOST ABSOLUTE JUICE % FIBRE IN BAGASSE .. Imbibition water per cent. Fibre Imbibition water per cent. Cane Extraction (Sucrose in Mixed Juice % Sucrose in Cane) .. BOILING HOUSE PERFORMANCE Boiling House Recovcy (Sucrose in Sugar % Sucrose ii

Mixed Juice) Overall Recovery (Sucrose in Sugar % Sucrose in Cane) ..

"" 52.2 191 24.7 92.9 95.6 87.8

81.6

47.1 252 38.2 91.2 95.8 88.8

80.9

46.9 224 36.9 92.0 98.1 90.0

82.8

47.9 252 40.1 91.8 95.7 88.7

81.5

37.4 219 38.7 92.7 9S.7 91.3

84.6

37.9 292 33.1 93.2 95.1 87.8

81.8

54.7 229 37.9 90.0 98.0 91.2

82,1

30.7 193 32.3 94.4 98.4, 91.2

86.1

48.3 228 36.2 90.3 97.1 90.3

82.7

40.3 167 28.0 93.2 98.5 91.1

84.9

32.1 193 30.0 94.8 97.S 91.0

88.3

35.6 240 38.1 94.3 96.5 89.1

83.9

38.2 193 33.4 93.0 96.8 89.4

83.1

33.1 236 37.7 94.1 94.8 87.9

82.7

36.0 226 35.6 94.2 97.9 90.9

85.6

38.1 226 36,7 93.6 97.8 90.4

84.6

37.9 267 41.0 93.8 97.7 90.6

85.0

40.9 217 34.9 93.0 97.2 90.0

83.7

FINAL BAGASSE Sucrose per cent Moisture per cent Fibrepercent Weight per cent. Cane Lower Calorific Value (7650 - 1 8 S-

FIRST EXPRESSED JUICE Brix Purity (apparent)

LAST EXPRESSED JUICE

S6.4W Btu./lb.)..

3.09

42.25 30.66

2953

19.09 87.4

3.10 52.88 43.03 40.34

3025

21.09 87.9

2.70 53.63 42.64 38.54

2968

19.79 87.0

3.38 49.12 46.54 34.22 3345

20.64 90.1

2.39 54.51 42.40 41.75 2897

20.10 88,2

2.70 49.74 46.82 35.07

3304

20.54 88.6

3.19 57.34 38.62 42.89 2638

20.34 88.2

2.19 50.53 46.66 35.83 3245

20.59 88.8

3.31 52.16 43.66 36.42 3084

21.04 89.3

2.48 51.74 45.15 37.28 3124

20.40 88.0

2.19 51.46 45.64 34.20 3165

21.05 89.0

MIXED JUICE Brix Purity (Gravity) Reducing Sugar/Sucrose Ratio ... Purity drop from First Expressed Jui

CLARIFIED JUICE Brix Purity (apparent) Reducing Sugar/Sucrose Ratio ... PH

FILTER CAKE Weight per cent. Cane Per cent. Sucrose

SYRUP Brix Purity (apparent) Reducing Sugar/Sucrose Ratio ... Purity increase from Mixed Juice PH

15.60 84.7§

2.72

14.55 86.0

3.27

14.41 84.7

4.17

14.00 87.8J

2.93

15.20 85.8

3.82

15.31 86.I5J

—

14.89 80.9

3.24

15.88 86.7

3.72

15.24 86.9

2.3

16.08 86.5

2.24

16.34 87.2

—

4.7 1.8

5.2 0.91

7.1 0.69

6.2 1.99

5.1 0.84

6.0 1.21

6.2 1.03

4.7 1.92

5.0 1.93

4.1 0.96

5.5 0.52

11.8 0.64

4.0 0.72

6.2 0.67

54 0.37 — 0.34

3.9 5.94

6.3 0.94

2.35 2.67 2.58 51.12 48.32 48196 45.65 48.07 47.85 34.73 35.99 33.48 3191 3427 3373

20.84 20.42 20.75 88.8 87.5 90.0

2.48 3.38 3.63 73.4 74.0 81.1 15.4 13.5 8.9

14.95 15.30 15.10 86.6 85.2 86.9J

2.51 3.41 2.22

2.48 49.75 46.95 33.49 3307

20.16 90.6

2.57 75.0 15.6

14.08 86.7

3.22

2.50 50.41 48.24 35.16 3250

19.77 89.6

3.28 74.7 14.9

14.53 86.7

2.44

2.95 46.98 49.28 31.14 3522

20.76 89.2

3,49 77.55 11.7

14.61 86.6t

—

2.65 52.53 43.99 36.59 3064

20.43* 88.6*

3.20* 76.2 12.0*

15.20 86.25 2.92

15.04 15.25 13.66 13.46 14.04 16.06 14.07 16.65 14.76 16.7 15.85 86.1 86.5 85.4 88.7 86.6 87.7 87.4 88.2 88.0 88.0 88.0 2.70 3.03 4.08 2.35 3.51 — 3.11 — 2.2 — — 7.4 7.6 7.3 7.2 7.3 7.0 7.3 6.9 7.3 6.0 7.5

14. U4 93.3

1.08 6.9

15.38 87.1

2.82

—

15.95 87.4

2.12

—

15.98 SS.4-.

2.46 7.2

14.10 — 87.9 87.5

2.30 — 7.2 —

15.13f 87.4f

2.79f 7.2f

53.8f 87. «f 2.66f l . a t

"Arithmetical average. t NE not included in arithmetical average. t Apparent purity.

54.1 86.2

2.60 1.4 7.0

02.6 86.5

2.66 0.5 7.05

51.9 85.4

4.08 0.7 6.8

58.0 88.7

1.76 1.4 7.0

52.8 86.4

3.33 o'.e 6.9

51.0 88.3

— 2.1 6.8

51.2 87.4 3.10 0.5 7.1

51.1 88.4

— 1.7 6.8

51.6 88.5 2.0 1.6 7.1

51.9 88.3

— 1.8 6.7

48.9 87.9

— 0.7

—

62.4 93.4 1.06 6.8 7.1

59.6 86.8 2.88

— —

59.5 88.0

2.26 1.1

—

56.5 88.4 2.81 1.7 7.0

56.0 88.1

1.91 1.3 6.8

„ -

87.9

— 1.3

—

53.8f 87.6f 2.66f 1.2f

« t

FACTORY ...

Table III.—MASSECUITES, RUN-OFFS, FINAL MOLASSES AND CHEMICALS.

FACTORY UF. ZM. FX. EN. AK. DK. DL. GL. MV. CK. TS. NE. IL. RN. ES. SZ. UK. Averages.

FIRST MASSECUITE Brix Purity (apparent) Cubic feet per ton of brix in Mixed Juice Purity of Run-oft (apparent) Purity drop (apparent) Crystal per cent. Massecuite I

SECOND MASSECUITE Brix Purity (apparent) Cubic feet per ton of brix in Mixed Juice Purity of Run-off (apparent) Purity drop (apparent) Crystal per cent. Massecuite II

THIRD MASSECUITE Brix Purity (apparent) Cubic feet per ton of brix in Mixed Juice Purity of Run-off (apparent) Purity drop (apparent) Crystal per cent. Massecuite III

JELLY Brix Purity (apparent) Cubic feet per ton brix in Mixed Juice Purity of Run-oS (apparent) Purity drop (apparent) Crystal per cent. Jelly

CUBIC FEET OF ALL MASSECUITES Per ton of Sugar Per ton of brix in Mixed Juice

FINAL MOLASSES Brix Purity (gravity) Reducing Sugars per rent. Sulphated Ash per cent Reducing Sugar Ash Ratio

Weight per cent. Cane (at 85° brix)

CHEMICALS Lime—lbs. per ton of Cane

lbs. per ton of Sugar lbs. per 1,000 lbs. brix in Mixed Juice

Sulphur—lbs. per ton of Cane lbs. per ton of Sugar lbs. per 1,000 lbs. brix in Mixed Juice

Phosphoric— lbs. per ton of Cane

lbs. per 1,000 lbs. brix in Mixed Juice ...

•EN re lelted tt

82.0 93.2 93.8 92.7 03.5 91.5 02.2 01.5 00.9 01.0 02.5 84.9 83.5 84.3 87.0 SI.5 86.6 87.6 88.1 86.4 88.3 88.0 20.2 18.1 21.4 24.6* 21.9 23.6 20.1 25.5 22.8 — — 60.0 81.3 63.4 60.7 61.1 68,5 09.7 69.8 69.0 72.3 72.0 19.0 22.2 20.0 17.3 20.4 18.1 18.0 18.(5 16.5 16.0 16.0 51.2 53.5 53.6 52.8 49.0 52.5 54.6 56.3 49.8 52.6 52.8

95.0 07.6 97.9 05.3 06.2 94.5 96.4 04.9 03.7 94.1 05.6 73.2 71.8 09.5 71.3 68,0 75.5 70.0 70.8 74.4 73.5 76.4 10. 4 11.5 9.8 11.7 9.7 15.6 9.8 9.7 11.8 — — 50.3 47.3 44.5 49.8 43.2 52.8 43.7 40.5 50.6 50.4 53.1 22.8 24.5 25.0 21.5 25.7 22.7 26.3 24.3 23.8 22.0 23.3 43.7 45.4 44.1 40.8 43.5 45.4 45.1 43.1 44.7 43.4 47.5

08.6 99.1 98.3 06.7 97.5 97.0 08.9 95.1 96.4 96.2 08.6 61.1 62.2 59.2 59.6 58.U 63.3 58.4 58.7 60.1 01.7 62.7

7.1 0.9 6.6 7.7 6.4 8.7 6.0 6.4 6.1 — — 10.0 40.9 37.0 42.1 37.0 30.4 38.1 39.1 40.8 30.5 38.5 21.0 21.3 21.3 17.5 21.0 23.9 20.3 19.0 10.3 22.2 24.2 34.6 35.7 33.7 20.2 32.5 38.2 32.4 30.6 31.4 35.3 38.7

90.8 94.6 39.3 84.6 10.0 58.8

03.5

S.v9 65.0 19.8 53.0

98.5

7 . 6

46.6 26.2 46.9

92.7 85.1 16.8 64.1 21.1 54.3

94.7 78.0 15.5% 52.9 25.0 50.4

95.0 66.9 9.8%

42.7 24.2 40.2

95.8 87.4 22.0 70.4 17.0 53.9

97.5 73.0

52.3 21.3 43.5

100.0 00.2

7.7

40.6 19.6 33.0

92.5 84.2 26.8 68.0 16.3 47.0

96.4 71.2 9.1

46.6 23.5 43.3

09.0 60.9

0.1

38.3 22.6 36.3

92.4 84.8 26.5 66.7 18.1 50.3

05.9 70.1 9.2

40.0 24.1 42.9

98,8 61.0

5.2

40.6 20.4 34.0

01.5 86.8 24.69 69.7 17.1 51.6

95.2 74.4

55.2 19.2 40.8

98.5 59.8

7.9

39.0 20.8 33.5

92.511 86 0 22.151 67.9|| 18.1|| S2.1|]

93.7|| 72.3|! 10.081 M.9|| 23.4] 43.8 j

97.9:| 60.5;| 6.521

39.4;! 21.11| 33.911

2.20 2.99 1.87

%c; excluding NE and IL.

excluding NE and IL.

49.8 •37.7

94.1 10.0

— -6.85

61.09 23.34 2.84

25.57 9.67 0.08 0.70 0.29

36.4

92.4 40.9 10.30 14.13

3.54

5.94 50.83 19.77 2.32

19.8t> 7.73 0.83 7.07 2.7-5

48.9 37.7

90.9 37.8 12.34 17.22

3.17

5.52 50.40 10.47 2.20

20.13 7.76 0.64 5.S5 2.26

40.4*

83.3 42.lt

— -2.89

7.37 63

24

3

2 6

10

0

a

2

34

69

0 3

11

20

67

76 .

24

38.1

89.8 36. S 12.92 12.20 l.O8

0.31 53.74 21.39 2.29

19.48 7.75 0.73 6.22 2.48

63. 0 47.8

88.0 39.4

_ „

3.10

3.85 33.73 12.82 1.52

13.29 5.05 1.01 8.85 3.36

36.0

90.9 37.6 12.86 14.86 0.87 2.70

4.02 35.10 14.19 1.60

13.99 5.64 0.61 5.29 2.11

52.3 41.7

82.9 39.1 11.6

-2.06

6.35 51.99 20.72 2.19

17.91 7.39 1.12 9.20 3.67

40.7

89.0 40.8 10.14

-2.93

6.21 51.31 20.42 2.75

22.71 9.04 0.67 5.51 2.19

39.3

87.0 39.5

__. --

4.59 39.05 15.74 2.16

18.62 7.39 0.91 7.85 3.12

~ 92.4 39.0

9.60

-2.78

4.78 37.28 15.25 1.62

12.05 5.18 0.86 6.75 2.76

66.1

89.5 47.2

7.22 12.44 0.58

_ — — — — — —

55. l i 43.Ti

88.5 42.0 14.08 12.97 1.09 3.54§

6.36 53.84 21.35 2.60

21.99 4.84 0.65 5.56 2.20

37.9

90.5 40.7

— --

7.28 60

2,5

3

27

10

0

7

2

0 3

13

36

6 8

66

94

76

99

51 3

41.1

91.4 39.0

— --

6.31 51

20

2

19

7

0

4

1

5 3

63

4 4

9 1

97

57

6 8

87

5 3 .

42.4

90.1 37.7

— --

7.16 61

24

3

2 6

10

1

8

3

12

27

0 8

3 2

43

0 1

6 2

41

5 1 . 5

41.1

92.2 39.0

— — -

6.53 51

20

2

19

7

0

5

2

0 1

34

4 8

39

73

76

9 4

37

49.SIT 38.87*

90.3 39.3

— -3.09

5.75 49

19

2

1 9

7

0

6

2

0 2

30

2 6

2 9

60

7 4

3 0

48

*EN remelted the sugar originating from 41 per cent, of {Included returns from refinery. ^Arithmetical average; excluding NE and IL. its Second Massecuite.

^Included final molasses expelled from bought raws during tApparent purity. refining process. "[Weighted average, excluding NE and IL.

--— —

-

_ —

-_ —

-—

— —

_ -— —

--— —

95.4 48.6

37.6 11.0 16.9

-_.. —

_ -— —

-

— —

--— —

--—

91.3 45.3

1.6 42.0

3 . 2

5 . 2

_

— —

--— —

90.8 43.2

1.5 37.0 6 . 2

9 . 5

-— — —

--— —

Table IV.—COMPARATIVE RESULTS FOR RECENT YEARS.

COUNTRY

YEAR

Tons of Cane Crushed (Total Crop) ... Tons of Sugar Made (Total Crop) ... Tons of Cane per Ton of Sugar (Total Crop) Tons of Cane per Ton of 96° Sugar (Total Crop)

CANE Per Cent. Sucrose ... Per Cent. Fibre

JAVA RATIO

JUICES Purity of First Expressed J nice ... Purity of Last Expre sed Juice ... Purity of Mixed Juice Purity of Syrup Purity Drop First to Last Expressed Juice Purity Drop First Expressed to Mixed Juice ... Purity Drop First Expressed Juice to Syrup ... Increase in Purity from Mixed Juice to Syrup

Reducing Sugar /Sucrose Ratio of Syrup

EXTRACTION AND RECOVERIES Sucrose % Cane Lost in Manufacture ... Sucrose in Sugar % Sucrose in Cane (Overall Rec.) ... Sucrose in Mixed Juice % Sucrose in Cane (Extraction) Sucrose in Sugar % Sucrose in Mixed Juice (E-.H. Rec.) Lost Absolute Juice % Fibre in Bagasse Imbibition % Fibre Imbibition % Cane Boiling House Performance

BAGASSE Per cent. Sucrose ... Per cent. Moisture

FILTER CAKE Per cent. Sucrose ... ... Weight per cent. Cane

GRAVITY PURITY OF FINAL MOLASSES Average Polarisation of All Sugars

LOSSESS (Sucrose Balance) Sucrose in Bagasse % Sucrose in Cane (A) Sucrose in Filter Cake % Sucrose in Cane (B) ... Sucrose in Molasses % Sucrose in Cane (c) Undetermined Sucrose % Sucrose in Cane (D) Lost in Boilinghouse % Sucrose in Cane (B) -f (c) + (n) Total Losses % Sucrose in Cane (A) + (B) + (c) + (D)

1943.

4,278,914 585,392 9.02 8.78

13.14 15.26

77.78

88.7 76.4 86.6 88.1 12.3

2 . 1 0 . 6 1.6

2.61

2.16 83.5 93.0 89.8

_.... 207 32

— 2.76

50.80

1.1 5 . 1

41.8 98.59

7.03 0.36

~~ 9.45

1 16.48

1944.

5,351,945 614,158 8.71 8,48

13.67 15.83

77.38

88.4 75.8 86.2 87.8 12.6

2 . 2 0 . 5 1.6

2.89

2.30 83.1 93.1 89.3

— 213 34

— 2.73

50.23

1.2 5 . 2

42.4 98.62

6.87 0.37

• —

9.99 16.86

1945.

4,607,055 553,024 8.33 8.10

14.28 15.99

77.36

88.4 75.9 86.2 87.8 12.4 2.1 0 . 5 1.6 3.38 2.84

2.42 83-3 93.3 89.3

— 219 35

— 2.77

50.19

1.1 5 . 6

42.0 98.73

6.72 0.35

— 9.98

16.70

1946.

3,990,017 474,769 8.40 8.17

14.21 16.21

77.03

88.2 75.1 85.9 87.4 13.1

2 . 4 0 . 8 1.6 3.30 2.80

2.42 82.9 93.1 89.1

• —

217 35

— 2.79

50.32

1.0 5 . 9

41.8 98.70

6.93 0.28

— 10.13 17.06

NATAL.

1947.

4,543,255 512,005 8.87 8.62

13.32 15.80

76.99

88.5 75.0 86.2 88.0 13.4

2 . 2 0 . 5 1.8 2.95 2.62

2.16 83.7 93.4 89.6

— 218 34

— 2.54

50.46

1.1 6 . 0

41.1 98.83

6.56 0.32

— 9.71

16.27

1948.

5,216,144 607,845 8.58 8.33

13.89 15.90

76.98

88.1 75.5 85.9 87.5 12.6

2.2 0 . 6 1.6 3.67 3.07

2.33 83.2 93.3 89.1

— 214 34

— 2.67

30.53

1.3 5 . 9

41.5 98.93

6.68 0.36

— 10.13 16.81

1949.

4,929,580 561,122

8.54

13.52 16.19

76.47

88.6 76.2 86.2 87.9 12.5

2 . 4 0 . 7 1.7 3.11 2.55

2.25 83.4 92.9 89.7

— 208 34

— 2.66

50.84

1.1 5 . 9

41.4 98.84

7.06 0.34

— •

9.59 16.65

1950.

5,721,390 685,798 8.34 8.11

14.19 15.80

77.42

88.7 75.8 86.4 87.6 12.9

2 . 3 1.1 1.3 3.12 2.81

2.32 83.7 93.3 89.6 39.3

206 33 96.9

2.72 51.22

1.2 5 . 5

40.5 98.77

6.67 0.37

— 9.68

16.35

1951. 1952.

4,805,2495 722,583 532,505 9.02 8.76

13.33 16.28

76.56

87.6 74.5 84.9 86.2 13.0

2 . 6 1.4 1 .3 3.52 3.25

2.33 82.5 93.0 88.7 40.2

215 35 96.7

2.57 51.71

1 .2 6 . 0

40.3 98.79

7.01 0.52 8.61 1.36

11.28 17.50

670,188 8.50 8.27

13.87 16.10

77.04

88.6 76.2 86.2 87.6 12.0

2 . 4 1.2 1.4 2.92 2.66

2.26 83.7 93.0 90.0 40.9

217 35 97.2

2.65 52.53

0 . 9 6 . 3

39.3 98.69

7.00 0.43 7.45 1.46 9.34

16.34

Table V.—AVERAGE MANUFACTURING RESULTS BY MONTHLY PERIODS FOR NATAL SUGAR FACTORIES

REPORTING TO THE SUGAR MILLING RESEARCH INSTITUTE, SEASON 1952-1953.

Period Ended

Tons of 2,000 lbs. Cane Crushed

Tons of 2,000 lbs. Sugar made and estimated

Tons of Cane per ton of Sugar ...

Sucrose per cent. Cane

Fibre per cent. Cane

Java Ratio

Sucrose per cent. Bagasse

Moisture per cent. Bagasse

Imbibition per cent. Cane

Extraction...

Boiling House Recovery

Overall Recovery ...

Purity of Mixed Juice

Reducing Sugar/Sucrose ratio, Mixed Juice ...

Gravity Purity of Final Molasses

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

This period To date ...

3 May, 1952.

191,674

19,769

9.70

12.53

16.94

76.74

2.41

52.99

38

92.5

87.6

81.1

83.9

4.54

38.5

31 May 1952.

414,623 606,295

44,542 64,310

9.31 9.42

12.86 12.76

16.29 16.50

77.81 77.48

2.36 2.38

52.14 52.42

36 37

93.3 93.1

88.7 88.3

82.7 82.2

84.9 84.6

3.70 3.90

38.3 38.4

28 June 1952.

639,150 1,259,735

71,051 136,870

9.00 9.20

13.22 12.99

16.32 16.40

76.92 77.19

2.51 2.45

52.33 52.36

36 36

93.0 93.0

89.4 88.9

83.1 82.7

86.2 85.5

2.90 3.39

40.0 39.8

2 Aug., 1952.

894,620 2,153,915

102,533 239,403

8.73 9.00

13.54 13.22

15.96 16.40

77.46 77.31

2.56 2.49

52.35 53.36

35 36

93.2 93.1

90.0 89.4

83.9 83.2

86.6 86.0

3.05 3.25

39.5 39.9

30 Aug., 1952.

739,517 2,893,432

90,081 329,484

8.20 8.78

14,22 13.48

15.89 16.14

77.20 77.28

2.63 2.53

52.50

34 36

93.4 93.2

90.3 89.6

84.3 83.5

86.9 86.2

2.37 2.96

40.3 39.8

27 Sept., 1952.

725,044 3,618,476

92,057 421,541

7.88 8.58

14.77 13.74

15.91 16.09

76.19 77.04

2.79 2.58

52.15 52.28

35 35

93.2 93.2

91.0 89.9

84.8 83.8

86.9

2.75 2.82

39.5 39.7

1 Nov., 1952.

859,624 4,478,100

110,585 532,082

7.77 8.42

15.06 13.99

16.15 16.10

77.08 77.05

2.82 2.62

52.09

35 35

93.2 93.2

90.4 90.0

84.2 83.9

86.6 86.4

2.76 2.80

40.2 39.8

29 Nov., 1952.

462,670 4,940,770

55,994 588,076

8.26 8.40

14.38 14.03

16.01 16.08

75.02 76.86

2.74 2.64

53.90 52.41

34 35

92.7 93.1

90.6 90.0

84.0 83.9

86.6 86.4

2.50 2.78

40.7 39.9

3 Jan., 1952.

422,168 5,370,938

47,822 635,898

8.83 8.45

13.26 13.96

16.23 16.08

76.20 76.99

2.69 2.65

52.30 52.41

34 35

92.3 93.1

90.8 90.0

83.8 83.9

85.6 86.4

3.46 2.83

39.9 40.0

31 Jan., 1953.

260,694 5,631,633

26,331 654,076

9.73 8.51

12.23 13.88

16.16 16.09

77.93 77.03

2.74 2.65

54.35 52.50

32 35

91.5 93.0

91.8 90.0

82.6 83.7

83.6 86.3

4.93 2.91

37.8 39.9

—

5,661,604

665,646

8.50

13.87

16.10

77.04

2.65

52.53

35

93.0

90.0

83.7

86.2

2.92

39.3*

Table VI.—COMPARATIVE RESULTS FOR RECENT YEARS.

COUNTRY

YEAR CANE

Per cent. Sucrose Per cent. Fibre

JUICES Purity of First Expressed Juice Purity of Last Expressed Juice Purity of Mixed Juice Reducing Sugar/Sucrose ratio of Mixed Juice Purity of Syrup Purity drop First Expressed to Mixed Juice ... Purity drop First Expressed to Syrup

JAVA RATIO

BAGASSE Per cent. Sucrose

EXTRACTION Imbibition % Fibre Sucrose in Mixed Juice % Sucrose in Cane ...

FILTER CAKE Per cent. Sucrose ... Weight % Cane

PURITY OF FINAL MOLASSES ....

RECOVERIES Sucrose % Cane lost in manufacture Sucrose in Sugar % Sucrose in Cane Sucrose in Sugar % Sucrose in Mixed Juice

YIELD Tons of Cane per ton of Sugar Tons of Cane per ton of Sugar of 96° Pol.

LOSSES Sucrose in Bagasse % Sucrose in Cane (A) Sucrose in Filter Cake % Sucrose in Cane (B) Sucrose in Molasses % Sucrose in Cane (c) Undetermined Sucrose % Sucrose in Cane (D) Lost in Boiling House % Sucrose in Cane (B) + {C) + (D) Total Losses % Sucrose in Cane (A) + (B) + (C) + {D)

AVERAGE: POLARISATION OF ALL SUGARS . . . '

MAURITIUS.

1950.

14.14 11.80

89.3 76.4 86.7

3 . 5 87.1

2 . 6 2 . 2

80.99

3.17

178 94.8

7 . 4 1.8

38.2

2.09 85.2 89.8

8.15 7.97

5.20 0.94 6.46 2.20 9.60

14.80

98.30

1951.

13.03 11.77

87.4 74.4 84.9

4 . 6 85.3

2 . 5 2 . 1

81.27

2.95

113 94.6

6 . 6 1.92

37.9

2.12 84.0 88.7

8.88 8.66

5.40 0.97 8.19 1.44

10.60 16.00

' 98.40

BRITISH GUIANA.

1950.

10.92 15.10

83.1

— 80.9

— 82.2 2 . 2 0 . 9

75.52

3.15 46.43

152 ' 91.2

2 . 2 2 . 9

32.3

2.18 80.0 87.8

10.98 10.91

8.81 0.59 8.44 2.11

11.14 19.95

96.57

1951.

10.61 14.76

82.5 75.8 80.4 8.47

81.7 2 . 4 0 . 8

75.92

3.06 46.40

147 91.4

2.05 2 . 7

31.0

2.03 80.8 88.4

11.27 11.20

8.60 0.52 8.22 1.84

10.58 19.18

96.64

QUEENSLAND.

1950.

14.53 12.87

88.6

— .— — 87.9

— 0 . 7

82.45

2.46

— 95.6

3 . 3 3 . 2

45.4

2.07 85.8 89.7

7.87 7.66

4.43 0.65 6.18 2.97 9.80

14.23

98.68

1951.

15.55 13.31

88.4

— — — 87.4

— 1.1

81.71

2.69 51.49

273 95.2

3 . 6 3 . 4

44.8

2.43 84.4 88.6

7.51 7.32

4.84 0.76 6.19 3.86

10.81 15.65

98.54

PHILIPPINES.

1949-50

13.84 11.82

86.1 78.5 85.7

— 86.9 0 . 4

—0.8

80.42

3.94 49.13

63 92.6

5 . 9 2 . 1

37.7

2.08 84.8 91.6

8.44 8.31

7.45 0.89 6.97 (0.10) 7.76

15.21

97.43

1950-51

13.10 11.55

85.3 76.7 84.8

_ 86.3 0 . 5

—1.0

80.24

3.81 48.95

70 92.6

4 . 1 2 . 0

36.5

1.90 85.3 92.1

8.86 8.73

7.43 0.62 7.07

(0.41) 7.28

14.71

97.98

PUERTO RICO.

1949-50

13.36 14.12

86.4

— — __ 85.1

— 1.2

78.50

2.57 49.01

179 94.3

2 . 6 3 . 7

31.4

1.73 87.1 92.3

8.37 8.25

5.71 0.72 6.06 0.44 7.22

12.93

97.36

1951-52

12.28 13.48

83.4

80.7

.— 82.3 2 . 7 1.1

78.97

2.57 49.46

170 94.0

— — 30.8

1.86 84.8 90.3

9.34 9.22

6.01 0.87' 7.71 0.58 9.16

15.17

97.32

NATAL.

1951.

13.33 16.28

87.6 74.5 84.9

3 . 5 86.2

2 . 6 1.4

76.56

2.57 51.71

215 93.0

1.2 6 . 0

40.3

2.33 82.5 88.7

8.98 8.73

7.01 0.52 8.61 1.36

11.28 17.50

98.79

1952.

13.87 16.10

88.6 76.2 86.3

2 . 9 87.6

2 . 4 1.2

77.04

2.65 52.53

217 93.0

0 . 9 6 . 3

39.3

2.26 83.7 90.0

8.50 8.27

7.00 0.43 7.25 1.66 9.34

16.34

98.69

ANNUAL WEATHER REPORT FOR 1952 By B. E. BEATER.

The first annual report of the new weather service was introduced last year. As was pointed out then, in order to have these monthly rainfall returns on as representative a basis as possible, it was decided to allocate one recording station to approximately 2 per cent, of the total cane crop produced, and to divide the area into magisterial districts, as is done in the Government Special Census of Sugar Cane Plantations. It was, however, considered necessary to include rainfall returns from certain areas such as Hluhluwe and Mkuzi, which though not justi­fiable at present on a production basis, are in­dividual areas which are of interest to the industry. The result of this allocation of recording stations is that we have in all 54 centres which send in detailed rainfall statements every month, to be summarised here and distributed. It is believed that these monthly reports, which are truly representative of the whole industry, have been constructive, and the writer takes this opportunity of expressing appreciation for the co-operation and assistance of the 54 observers concerned.

As was done last year, the rainfall data from the original 44 recording stations which have reported to the Experiment Station over the past 24 years, are once again embodied in the report. The 44 centres reporting in the past were not selected on a production basis, but despite this limitation it is proposed to continue publishing their annual returns until such time as a reasonably accurate average ratio between the rainfall of the new and old stations can be obtained, and the data from the old stations can be used for judging the rainfall at the new centres.

This annual summary introduces for the first time the internationally accepted millimeter scale for rainfall and the centigrade scale for temper­atures, both of which are now official usage in the Union. For purposes of converting rainfall there are 25.4 mm to the inch, or multiply mm by 0.03937.

Summarised Monthly Returns from 54 Centres of the Sugar Industry.

The total rainfall for the calendar year 1952 for the 54 centres is presented below:

Magisterial District

Port Shepstone .., Umzinto

Durban Camperdown

Inanda

Lower Tugela ...

Locality

... Mehlomnyama

... Hibberdene Umtwalumi Sezela Experanza Rcnishaw ... Dumisa

... Illovo

... Umbumbulu Thornville

... Mount Edgecombe-

La Mercy Canelands ... Tongaat Tongaat Inanda Tongaat

... Maidstone ... Sinembe Upper Tongaat FraserR Estate Chaka's Kraal Chaka's Kraal Groutville Kearsney Doornkop Doornkop Gledhow Darnall Tugela Month

Reorder.

... J. B, Kippen ... G. W. Hammond ... ... B. D. Archibald ... ... Mill ... Mill ... Mill ... A. W. Barker ... Mill ... E. S. Gurney ... E. O. Mapstone ...

Milk wood Kraal Exp. Station Beach

... Gersigny Bros.

... E. L. Armstrong ... Frosterly Inyaninga ...

... G. R. Groom

... Mawine

... Mill

... H. E. Heenan

... C. E. Goble

... Manager

... Exp. Farm

... W. S. Campbell ... Cranbrook ...

... H. Balcomb and Sons

... Mill

... Sprinz

... Mill

... Mill

... Dhlogweni

871.0 690. ft 685.8 053.5 82(5.3 835.7 750.1 762.0 713.0 843.5

1870.5 871.5 864.6 900.2 802.4 833.6 819.7

1020.3 954.0 825.5 904.5

1026.9 856.5 784.9

1008.6 862.8 928.4 927.6

1084.3 937.3 876.0 935.0

766.8 870.0 811.5

1126.2 1056.9 1070.6 774.2 780.5 758.2 832.4 772.2 858.8 792.7 825.5 707.1 765.3 696.2 877.1 856.0 700.8 805.9

1083.6 808.7 672.6 834.1 799.3 888.0 836.2

1056.6 785.1 782.8 788.9

34.20 27.20 27.00 37.54 32.53 32.90 20.53 30.00 28.07 33.21 34.27 34.31 34.04 35.44 31.59 32.82 32.27 40.17 37.56 32.50 35.61 40.43 33.72 30.90 39.71 33.97 36.55 36.52 42.69 36.90 34.49 36.81

30.19 34.25 31.95 44.34 41.61 42.15 30.48 30.73 29.85 32.77 30.40 33.81 31.21 32.50 27.84 30.13 27.41 34.53 33.70 27.59 31.7 S 42.66 31.84 26.48 32.84 31.47 34.96 32.92 41.60 30.91 30.82 31.06

28.

Magristerial District. Locality. Recorder. Total Rainfall from January 1st. 1951 1952 1951 1952

E s h o w e

Lower Umfolozi . . i W e s t .

U b o m b o . . .

P ie t Re t i e f

M a n d i n i A m a t i k u l u . I n y o n i M t u n z i n i . B l a c k b u r n .

E n t u m e n i . E s h o w e N k w a l e n i .

F e l i x t o n K m p a n g e n i Em pander ' Logoza U k u l u P r o p e r t i e s . Mposa K w a m l x m a m b i E t e z a

M t u b a t u b a U .L .O .A . . . . .Nyalazi R i v e r Hiuhluvve . . .

Mkuz i

P o n g o l a

A . A d a m s Mill L . P . J o h n s o n G . V . W. R o b e r t s . . . H. C. B o a s t

Mill H . W . Brockwel l . . . G. M. R o b i n s o n

Mill \V. H. S impson Mill C. F. M. H i b b e r d . . . M a n a g e r Mrs. F. VV. M. H. Sp r ingo rum L. C. M. R a t t r a y . . . H a w o r t h Bros .

Mill M a n a g e r E. A. M. E r l a n d s o n A . K r a m e r

J . P . Brash

S u p e r i n t e n d e n t

7 9 9 . 8 7 8 8 . 4 8 7 9 . 1

1147 .1 8 4 5 . 3 7 9 5 . 3 7 8 3 . 6 6 8 3 . 8

1027.4

818.fi

935.5

851.2

793.8

805.2

1 0 0 7 . 9 8 1 2 . 8 5 7 9 . 9 9 3 3 . 4 6 8 3 . 8 4 4 8 . 1 5 7 3 . 5 6 2 3 . 0

7 1 4 . 5 8 1 3 . 1 7 9 6 . 3

1 0 3 5 . 1 81.9.7 7 9 0 . 7 7 9 4 . 5 5 1 4 . 6 8 0 8 . 5 0 2 6 . 1 7 8 7 . 1 7 7 0 . 4

782.3 829.8 770.4 582.9

1078.0 761.0 665.2 522.0 532.6

31.49 31.04 34.61 43.16 33.28 31.31 30.85 26.92 40.43 32.23 36.83 33.51 31.25 31.70 39.68 32.00 22.83 36. 75 26.92 17.64 22.58 24.55

28.13 32.01 31.35 40.75 32.27 31.13 31.28 20.26 31.83 24.65 30.99 30.33 26.25 30.80 32.67 30.33 22.95 42.44 29.96 26.19 20.55 20.97

The following table gives the rainfall for 1952 by months at the Experiment Station compared with the

average for the past 27 years.

1952 Mean 1926-1952 Inclusive.

J a n u a r y

F e b r u a r y

March

April

May

J u n e

Ju ly

August

Sep tember

October

November

December

T O T A L . . .

Total for month in

. . . 1 8 0 . 1

7.09

6 1 . 7

2.43

6 1 . 5

2.42

... 1 2 7 . 8

5.03

5 7 . 9

2 - 2 * . . . . 1 1 . 4

0.45

... 3 0 . 2

1.19

3 2 . 0

1.26

... 2 3 . 4

0.92

. . . 4 8 . 0

1.89

... 7 9 . 5

3.13

.... 1 5 0 . 4

5.92

. . . 8 6 3 . 9

34.01

Aggregate

Ja°n™ar"y (i»s.)

180 .1

7.09

2 4 1 . 8

9.52

3 0 3 . 3

11.94

4 3 1 . 0

16.97

4 8 9 . 0

19.25

5 0 0 . 4

19.70

5 3 0 . 6

20.89

5 6 2 . 6

22.15

5 8 6 . 0

23.07

6 3 4 . 0

24.96

7 1 3 . 5

28.09

8 6 3 . 9

34.01

8 6 3 . 9

34.01

aggregate

1 7 9 . 9

1 1 2 . 5

8 9 . 0

1 0 5 . 7

1 0 5 . 9

9 9 . 1

9 9 . 1

9 8 . 7

9 4 . 8

9 0 . 7

8 9 . 0

9 4 . 5

9 4 . 5

No. of

(lays.

15

12

9

8

7

2

7

4

8

12

14

18

116

' rainfall

(ins.)

1 2 . 0

0.473

5 . 2

0.203

6 . 8

0.269

1 6 . 0

0.629

8 . 3

0.326

5 . 7

0.225

4 . 3

0.170

8 . 0

0.315

2 . 9

0.115

4 . 0

0.158

5 . 7

0.224

8 . 4

0.329

7 . 4

0.293

Total for

(ins.)

1 0 0 . 1

3.94

1 1 4 . 8

4.52

1 2 6 . 0

4.96

6 6 . 8

2.63

5 4 . 1

2.13

4 2 . 9

1.69

3 0 . 5

1.20

3 4 . 8

1.37

4 8 . 3

1.90

8 1 . 0

3.19

1 0 2 . 1

4.02

1 1 2 . 8

4.44

MEAN

Aggregate

Ist january {ins.)

100 .1

3.94

2 1 4 . 9

8.46

3 4 0 . 9

13.42

4 0 7 . 7

16.05

4 6 1 . 8

18.18

5 0 4 . 7

19.87

5 3 5 . 2

21.07

5 7 0 . 0

22.44

6 1 8 . 2

24.34

6 9 9 . 3

27.53

8 0 1 . 4

31.55

9 1 4 . 1

35.99

9 1 4 . 1

35.99

days.

14

12

12

8

5

4

4

5

8

14

14

15

115

rainfall

3 . 2

0.126

4 . 1

0.160

4 . 1

0.160

2 . 2

0.088

1.8

0.069

1.4

0.056

1.0

0.039

1.1

0.044

1.6

0.063

2 . 6

0.103

3 . 4

0.134

3 . 6

0.143

2 . 5

0.099

of

0 . 4 3

0 . 4 0

0 . 3 7

0 . 2 7

0 . 1 6

0 . 1 4

0 . 1 3

0 . 1 7

0 . 2 6

0 . 4 4

0 . 7 9

0 . 4 6

0 . 3 4

Average

(im.)

7 . 1

0.281

9 . 6

0.377

1 0 . 5

0.413

8 . 4

0.329

1 0 . 8

0.426

1 0 . 7

0.423

7 . 6 .

0.300

7 . 0

0.274

6 . 0

0.238

5 . 8

0.228

7 . 3

0.287

7 . 5

0.296

8 . 2

0.323

Mtunzini

MEAN . . . 841.5 801.9 33.13 31.57

29

Below are summarised the total mean rainfalls received during the growing season, October to March

and the so-called ripening season, April to September, for the past two years:

Rainfall in inches are shown in Italic Figures. October to Much. April to September.

1951 1952 1951 1952 1951 1952 1951 1952 1951 1952 1951 1952

mm

6 0 6 . 3 6 1 4 . 1 5 9 6 . 7 5 8 4 . 0 5 9 9 . 4 5 9 3 . 5 4 7 4 . 3 4 9 7 . 6 5 4 8 . 9 5 5 4 . 7

2 8 2 5 . 6 2 8 4 3 . 9

ins"

23.87 24.18 23.49 22.99 23. SO 23.37 18.67 19.59 21.61 21.84

111.24 111.96

mm "

1 8 3 . 9 2 7 0 . 7 3 0 7 . 3 2 3 2 . 7 2 5 9 . 0 2 4 4 . 6 3 2 8 . 8 2 4 7 . 8 2 9 3 . 9 2 4 7 . 6

1 3 7 2 . 9 1 2 4 3 . 4

ins.

7.24 10.66 12.10

9.16 10.20

9.63 12.94

9.76 11.57

9.75 54.05 48.95

1. South Coast

2. North Coast

Mean of 1 and 2 . . .

3. Zululand including Piet Retief

4. General Mean for I, 2 and 3

Total of Means

From these figures it will be observed that there has been a close similarity in the summer distribution

of mean total rainfall over the past two years while a rather drier winter occurred throughout the whole

industry during 1952.

Annual Rainfall from Original 44 Centres.

Records from the 44 centres, which go back 24 years, are again presented:

Port Shepstone . . .

Umzumbi

*Esperanza...

*Renishaw . . .

Park Rynie

•Tllovo

Umbogintwini

Durban (Berea) . . .

Durban (Point) . . .

Effingham

Westbrook

•Milkwood Kraal . . .

Mount Edgecombe

*Mount Edgecombe

Cornubia

Burnside

Blackburn

*Beach

Saccharine

Ottowa

*La Mercy

*Tongaat

*Sinembe

Umhlali

Chaka's Kraal . . .

Lightkeeper, S.A.R

A. H. G. Blarney

Reynolds Bros

Crookes Bros,, Ltd.

Ellingham Estates

Illovo Sugar Estate Ltd.

African Explosives & Industries Ltd.

Botanic Gardens

S.A.R. &H

Natal Estates Ltd.

Natal Estates Ltd.

Natal Estates Ltd.

Natal Estates Ltd. Mill

S.A.S.A. Experiment Station

Natal Estates Ltd.

Natal Estates Ltd.

Natal Estates Ltd

Natal Estates Ltd.

Natal Estates Ltd.

Natal Estates Ltd.

Gersigny Bros

Tongaat Sugar Co. Ltd

H. C. Heenan

G. P. Ladlau

Waldene Sugar Estate

1069 .1 42.09

1 0 0 6 . 3 39.62

1 0 3 8 . 6 40.89

9 8 2 . 2 38.67

1 0 9 0 . 9 42.95

9 5 9 . 1 37.76

1 0 4 2 . 2 41.03

1 0 2 6 . 9 40.43

1 1 7 2 . 2 46.15

8 2 8 . 3 32.61

1 0 2 7 . 9 40.47

8 3 4 . 4 32.85

1 0 1 6 . 8 40.03

9 3 2 . 4 36. 71

1 0 5 0 . 0 41.34

9 9 7 . 0 39.25

9 1 4 . 4 36.00

1 0 6 1 . 0 41.77

9 6 8 . 8 38.14

9 4 4 . 9 37.20

1025 .1 40.36

9 7 8 . 4 38.52

1 0 4 5 . 0 41.14

1 0 9 0 . 2 42.92

9 0 9 . 1 -35.79

8 9 2 . 8 3.5.2,5

8 5 9 . 5 33.84

8 2 6 . 3 32.53

8 3 5 . 7 32.90

8 9 5 . 6 3,5.26

7 6 2 . 0 30.00

861 .1 33.90

9 8 0 . 2 38.59

1 1 6 6 . 4 45.92

9 2 0 . 5 36.24

1 0 7 7 . 0 42.40

8 7 0 . 5 34.27

8 8 4 . 2 34.81

8 7 3 . 0 34.37

9 7 1 . 6 38.25

9 4 0 . 6 37.03

9 0 0 . 7 35.46

8 6 4 . 6 34.04

8 1 9 . 7 32.27

8 0 1 . 4 31.55

9 0 0 . 2 35.44

8 2 7 . 5 32.58

9 0 4 . 5 35.61

9 3 0 . 9 3 6 . 6 5

7 8 0 . 8 30.74

8 9 9 . 9 35.43

9 7 8 . 7 38.53

1 0 5 6 . 9 41.61

1 0 7 0 . 6 42.15

1099 .1 43.27

7 8 0 . 5 30.73

9 8 3 . 2 38.71

9 1 6 . 7 36.09

9 3 5 . 7 36.84

9 1 6 . 2 36.07

9 5 9 . 4 57.77

7 8 1 . 8 30.78

8 3 7 . 2 32.96

8 6 3 . 9 34.01

9 4 1 . 3 37.06

9 4 7 . 4 37.30

7 9 0 . 0 41.10

7 9 2 . 7 31.21

9 0 3 . 2 3,5.56

8 4 7 . 1 33.35

8 2 5 . 5 32.50

7 0 0 . 8 27.59

8 0 5 . 9 31.73

9 1 1 . 1 35.87

6 9 4 . 4 27.34

1054 .6 41.52

9 9 9 . 0 39.33

1 0 3 0 . 5 40.57

9 7 9 . 7 38.57

1083 .3 42.65

9 4 3 . 6 37.15

1032 .3 40.64

1020 .3 40.17

1 1 6 2 . 3 45.76

835 .7 32.90

1027 .2 40.44

8 3 3 . 6 32.82

1003 .8 39.52

9 2 7 . 1 36.50

1042 .2 41.03

9 9 2 . 6 39.08

9 0 8 . 6 35.77

1041 .4 41.00

9 5 9 . 9 37.79

934 .7 36.80

1011.7 39.83

9 6 0 . 6 37.82

1029 .2 40.52

1 0 7 5 . 9 42.36

8 9 4 . 8 35.23

31

Station.

T i n l e y M a n o r

R i e t V a l l e y

K e a r s n e y . . .

D a r n a l l

*Darna l l

M a n d i n i . . .

* A m a t i k u l u

G i n g i n h l o v u

*Mtunz in i . . .

E s h o w e

* F e l i x t o n . . .

* E m p a n g e n i W e s t

E m p a n g e n i Rail

* E m p a n g e n i

*Kulu H a l t

*Mposa

K w a m b o n a m b i

* E t e z a

*Rive rv iew

Recorder.

S i r J . L . H u l e t t & Sons . . .

H . E . E s s e r y

Sir J . L . H u l e t t & Sons . . .

Mrs . M a n n

Sir J . L. H u l e t t & Sons . . .

S t . A n d r e w s E s t a t e s

S i r J . E . H u l e t t & S o n s . . .

F . C . L i l b u r n

G . V . W . R o b e r t s

D i s t r i c t F o r e s t Officer

S i r J . L . H u l e t t & S o n s . . .

W . I I . S i m p s o n

F . S . M a n n

Z u l u l a n d S u g a r Mil lers a n d P l a n t e r s

C . E . M. H i b b e r d

W . S p r i n g o r u m

S. La r sen

H a w o r t h Bros

Umfolozi Co-op S u g a r P l a n t e r s L t d .

MEANS

Average

1929-50.

mm

(ins.)

1 0 1 7 . 5 40.06

1 1 0 1 . 9 43.38

1 1 1 3 . 3 43.83

1 0 3 6 . 6 40.81

1 0 0 9 . 4 39.74

1 0 1 7 . 0 40.04

9 8 3 . 0 38.70

1 1 1 0 . 5 43.72

1 2 7 0 . 0 50.00

1 2 8 8 . 8 50.74

1 2 3 5 . 5 48.64

9 1 9 . 2 36.19

1 0 5 9 . 7 41.72

1 0 6 0 . 2 41.74

1105 .4 43.52

9 8 4 . 3 3 8 . 7 5

1 0 3 3 . 5 40.69

9 7 8 . 7 38.53

8 6 8 . 4 34.19

1 0 2 7 . 4 40.45

1951

mm

(ins.)

9 0 7 . 3 35.72

9 3 3 . 2 36.74

9 4 5 . 6 37.23

9 6 2 . 7 37.90

8 7 6 . 0 34.49

7 9 9 . 8 31.49

7 8 8 . 4 31.04

9 7 7 . 6 38.49

1 1 4 7 . 1 45.16

9 7 7 . 4 38.48

1 0 2 7 . 4 40.45

8 1 8 . 6 32.23

9 2 5 . 8 36.45

9 3 5 . 5 36.83

8 5 1 . 2 33.51

8 0 5 . 2 31.70

8 0 6 . 2 31.74

8 1 2 . 8 32.00

5 7 9 . 9 22.83

8 9 1 . 5 35.10

1952

mm

(ins.)

6 7 5 . 1 26.58

9 1 4 . 1 35.99

8 4 1 . 8 33.14

8 3 3 . 4 32.81

7 8 2 . 8 30.82

7 1 4 . 5 28.13

8 1 3 . 1 32.01

8 8 1 . 6 34.71

1 0 3 5 . 1 40.75

9 4 2 . 3 37.10

8 0 8 . 5 31.83

6 2 6 . 1 24.65

7 8 4 . 4 3 0 . 8 8

7 8 7 . 1 30.99

7 7 0 . 4 30.33

7 8 2 . 3 30.80

7 6 0 . 5 29.94

7 7 0 . 4 30.33

5 8 2 . 9 22.95

8 4 8 . 8 33.42

Average

1929-52.

mm

(ins.)

9 9 8 . 5 39.31

1 0 8 6 . 9 42.79

1 0 9 5 . 0 43.11

1 0 2 5 . 1 40.36

9 9 4 . 4 39.15

9 9 5 . 2 39.18

9 6 7 . 7 38.10

1 0 9 5 . 5 43.13

1 2 5 5 . 0 49.41

1 2 6 1 . 4 49.66

1 2 0 8 . 8 47.59

9 0 3 . 0 35.55

1 0 4 2 . 7 41.05

1 0 4 3 . 7 41.09

1 0 8 0 . 8 42.55

9 6 8 . 2 38.12

1012 .7 39.87

9 6 2 . 9 37.91

8 4 4 . 6 33.25

1 0 1 4 . 2 . 39.93

*Records also represented in the 54 stations weather service.

Mean Monthly Rainfall for South Coast, North Coast and Zululand for 1951 and 1952.

Section.

S o u t h C o a s t 1951

1952

N o r t h C o a s t 1951

1952

M e a n of 1 a n d 2 1 9 5 1 . . .

1952. . .

Z u l u l a n d , i n c l u d i n g P i e t R e t i e f 1 9 5 1 . . .

1952 . . .

G e n e r a l M e a n of 1, 2 a n d 3 . . . 1 9 5 1 . . .

1952. . .

Jan mm (ins.).

1 2 6 . 2 4.97

1 8 7 . 2 7.37

1 0 1 . 1 3.98

1 3 5 . 4 5.33

1 0 9 . 0 4.29

1 5 1 . 6 5.97

8 4 . 6 3.33 7 0 . 9 2.79

9 9 . 1 3.90

1 1 8 . 6 4.67

Feb. mm (ins.).

6 1 . 5 2.42 6 2 . 2 2.45 5 6 . 4 2.22 6 4 . 3 2.53 5 7 . 9 2.28 6 3 . 8 2.51

4 6 . 5 1.83 6 1 . 7 2.43

5 3 . 3 2.10 6 3 . 2 2.49

Mar. mm (ins.).

1 3 8 . 7 5.46 8 1 . 5 3.21

1 3 2 . 8 5.23 9 1 . 2 3.59

1 3 4 . 6 5.30 8 8 . 1 3.47

9 9 . 1 3.90 6 9 . 6 2.74

1 2 0 . 4 4.74 8 0 . 5 3.17

Apr. mm (ins.).

1 6 . 5 0.65

1 1 4 . 8 4.52 4 1 . 2 1.62 6 7 . 6 2.66 3 3 . 5 1.32 8 2 . 3 3.24

5 1 . 6 2.03 4 9 . 8 1.96

4 0 . 9 1.61 6 9 . 3 2.73

May. mm (ins.).

1 3 . 2 0.52 6 4 . 5 2.54 1 2 . 4 0.49 7 4 . 9 2.95 1 2 . 7 0.50 7 1 . 6 2.82

2 6 . 7 1.05 9 3 . 7 3.69

1 8 . 3 0.72 8 0 . 5 3.17

June. mm (ins.).

4 . 1 0.16 1 3 . 7 0.54 2 7 . 2 1.07

9 . 9 0.39

9 . 9 0.39 1 1 . 2 0.44

3 5 . 1 1.38 2 3 . 9 0.94

2 6 . 2 1.03 1 6 . 5 0.65

July. mm (ins.).

1.0 0.04 1 8 . 0 0.71

3 . 0 0.12 2 9 . 5 1.16

2 . 5 0.10 2 5 . 9 1.02

1 8 . 8 0.74 6 2 . 7 2.47

8 . 9 0.35 4 1 . 1 1.62

Aug. mm (ins.).

8 1 . 8 3.22 1 1 . 9 0.47

1 5 1 . 1 5.95 3 0 . 5 1.20

1 2 9 . 5 5.10 2 4 . 6 0.97

1 3 8 . 7 5.46

5 . 8 0.23

1 3 4 . 1 5.28 1 7 . 3 0.68

Sept. mm (ins.).

6 7 . 3 2.65 4 7 . 8 1.88 7 2 . 4 2.85 2 0 . 3 0.80 7 0 . 9 2.79 2 9 . 0 1.14

5 7 . 9 2.28 1 1 . 9 0.47

6 5 . 5 2.58 2 2 . 9 0.90

Oct. mm (ins.).

1 0 7 . 7 4.24 4 6 . 7 1.84. 9 2 . 5 3.64 4 7 . 0 1.85 9 7 . 3 3.83 4 7 . 0 1.85

9 7 . 0 3.82 4 0 . 4 1.59

9 7 . 0 3.82 4 4 . 2 1.74

Nov. mm (ins.).

2 4 . 9 0.98 8 9 . 9 3.54 2 9 . 5 1.16

110 .7 4.36 2 7 . 9 1.10

1 0 4 . 1 4.10

1 1 . 7 0.46

1 2 1 . 9 4.80

21.6 0.85

1 1 1 . 5 4.39

Dec, mm (ins.).

1 4 7 . 3 5.80

1 4 6 . 6 5.77

1 8 4 . 4 7.26

1 3 5 . 4 5.33

1 7 2 . 7 6.80

1 3 8 . 9 5.47

1 3 5 . 4 5.33

1 3 3 . 1 5.24

1 5 7 . 5 6.20

136 .7 5.38

Total. mm (ins.).

7 9 3 . 2 31.23 8 8 4 . 7 34.83 8 9 9 . 4 35.41 8 1 8 . 1 32.21 8 6 6 . 1 34.10 8 3 8 . 2 33.00

8 0 0 . 9 31.53 7 4 8 . 8 29.48

8 3 9 . 7 33.06 8 0 3 . 4 31.63

Discrepancies arising where the addition of the monthly totals does not equal the totals given for the year is due to the fact that a single centre in a district omitted to send in a monthly return, whereas the annual total will include this.

Italic figures indicate rainfall in inches.

32

Comments on Rainfall. The year 1952 has been another year of very

unfavourable weather conditions with its meagre total of 803.4 mm (31.63 ins.) of rainfall. During the five months June to October, only 142.0 mm (5.59 ins.) was recorded, with only 67.1 mm (2.64 ins.) during the very important months of September and October when three or four times this quantity is expected. The very favourable precipitation during the last two months of the year has done much to restore the position, and prospects on the whole are satisfactory. There has been an increase in the total amount of rainfall received in the growing seasons this year over that of last, though this is very slight. On the other hand, there has been quite a falling off of rainfall in the April—September period, principally reflected on the North coast, the South coast actually scoring considerably over 1951. The main area affected by the dry winter period has been Zululand, the means for August and September, for example, being only 5.8 mm (0.23 ins.) and 11.9 mm (0.47 ins.), with nine monthly records showing no rain at all. Mkuzi was the most severely affected centre last year, with only 153.9 mm (6.06 ins,) received during the first ten months of the year. At Pongola 232.2 mm (9.14 ins.) was received over the same period. At Hluhluwe the ten months total was 340.9 mm (13.42 ins.), making the average precipitation for the whole area from Hluhluwe northwards for the ten months January to October inclusive only 242.3 mm (9.54 ins.). Fortunately the output of these areas, which are almost wholly dependent upon irrigation, amounts to only about 5 per cent. of that of the industry.

Temperatures. The mean screen temperature for the year at

the Experiment Station was 20.60C. (69.0°F.), which was 0.20C. (0.3°F.) higher than the average of the past 24 years, and 0.80C. (1.4°F.) higher than the previous year 1951.

The high mean temperature for 1952 resulted from high mean temperatures over the winter months June to August, when the mean was 0.60C. (1.2°F.) higher than the corresponding mean for 24 years.

There were no frosts reported during the winter, the mean monthly temperature ranging from 23.50C. (74.2°F.) in February to 17.4°C. (63.3°F.) in July, both months of the normally highest and normally lowest temperatures.

The highest screen maximum temperature re-corded at the Experiment Station was 37.20C. (99°F.), which is the first time in the past seven years that a screen temperature of 37.8°C. (100°F.) has not been reached or exceeded. The lowest screen minimum was 7.2°C. (45°F.). The mean

grass temperature was 14.30C. (57.8°F.), with an absolute minimum of 3.90C. (39°F.) occurring in both June and July.

The temperatures of the ground at the 4 ft., 2 ft. and 1 foot depths were all slightly higher than the averages over the past 18 years.

Atmospheric Conditions. The mean true atmospheric pressure was 29.77 ins.

showing little variation from last year and the average.

The average humidity for the year was 77.1. per cent. of saturation at 8 a.m. and 65.8 per cent. at 2 p.m., both slightly above average.

The total evaporation from a free water surface was 1153.2 mm (45.50 ins.), very slightly below the 17 years average of 1183.6 mm (46.6 ins.). Evaporation exceeded rainfall by 289.3 mm (11.39 ins.).

The total hours of sunshine for the year were 2326.2, or 53.1 per cent. of the available hours of daylight. This is only slightly below the average over the past 25 years.

Conclusion. Following on the disastrous drought of 1951, the

year 1952 has been hardly less disappointing. Nevertheless, steady though insufficient rains over a large part of the 1951-52 growing season and moderately warm temperatures, have left the crops growing and looking well, even though they may not have made the headway they should have. Further good rains over the closing two months of the year will also have contributed to the better harvest anticipated next season.

Experiment Station, South African Sugar Association

Mount Edgecombe. February, 1953.

Mr. Dymond said that Dr. Beater's paper formed a most valuable addition to the data available to the Sugar Industry. He said that he had in his possession weekly records of rainfall for Empangeni going back many years. In graph form these weekly figures gave the effects of rainfall on such things as rate of growth, sucrose, etc. Total rainfall figures might not mean very much. It was the distribution of the fall that really mattered. With Dr. Beater's agreement, he would pass on the figures and graphs in his possession and he suggested that these graphs might be incorporated in the annual weather survey next year. He felt that they would be useful in gaining a better picture of rainfall distribution.

Mr. Barnes said that rainfall was one of the several meteorological elements that affect cane.

33

Referring to the table of rainfall at the Experiment Station by months, he suggested the inclusion of the number of days on which the precipitation reached or exceeded the minimum which would benefit the crop. An arbitrary figure would be necessary for the effective daily rainfall. Elsewhere half an inch (say 12 mm) was used, but a lower figure might apply in Natal.

He suggested that the use of only one station to represent the whole Sugar Belt might be stretching the figures rather far. Consideration might be given to establishing two, or perhaps three, more major stations. The effect of dew deposition seemed to be overlooked. Records of maximum and minimum humidity might well be included in the annual report. Rainfall and climate affected factory performance as well as field crop results. He could not emphasise too strongly the importance of meteorology in connection with rainfall data, as a whole.

Dr. Beater, in reply, drew Mr. Barnes' attention to the fact that the monthly weather reports of the Experiment Station weather service detailed the number of days rainfall ranged from 0.50—0.99 ins., 1.00—1.99 ins., and 2.00 ins. and over. He agreed that there should be more weather stations, but also

pointed out that the Experiment Station was climatically very centrally situated.

Mr. Pearson referred to irrigation and rainfall. He suggested that some means be found of adding to the data some details of wind velocity which played an important part, especially in spray irrigation.

Dr. Beater, in reply to Mr. Pearson, said that the Experiment Station did possess a wind aenometer at one time and that a paper by himself in the S.A.S.T.A. Proceedings for 1936 gave a fairly detailed analysis of prevailing surface winds.

Mr. Main said that as soil temperatures had so great a bearing on productivity he did not think that rainfall alone was a full story. He asked whether records of soil temperature were kept elsewhere than at Mount Edgecombe.

Dr. Beater replied that there were four other stations where soil temperature records were kept.

Mr. Dymond said that he was sure Dr. Beater appreciated the points that had been raised and that these would presumably be incorporated into the official data if this could be done. He then asked the meeting to accord Dr. Beater a hearty vote of thanks.

CENTRIFUGAL MACHINES By CHS. G. M. PERK.

Introduction.

Those of us who have had the opportunity to compare the performance of electrically and water-driven 30 X 18 ins. centrifugals, both curing the same C massecuite, will have observed that the capacity of the electrical ones was at least 50 per cent, more than that of the water-driven centrifugals. Since we knew that the electrically-driven 30 ins. baskets run at 1460 r.p.m. and the water-driven ones at only 1100-1200 r.p.m. (dependent on the available water pressure at any given moment), the difference in the capacity of the two was understandable. When, however, we had to compare baskets of differ­ent dimensions running at different speeds, the question became more intricate. Nowadays, tables showing the gravity factors adjusted to speed and diameter of the baskets obviate the calculations we had to do in the old days when we wanted to com­pare baskets of different diameters running at differ­ent speeds. Moreover, referring to the gravity factor gives us a better insight into the question, since the gravity factor indicates:

{a) the centrifugal force in pounds, exerted on one pound in weight when placed at the cir-cumference of the spinning basket; or

(b) how many times the centrifugal force at the circumference of the basket is greater than the force of gravity.

The gravity factor can be calculated by the formula:

GRAVITY FACTOR = 14.2 (Basket diam. in ins.) x (r.p.m./1000)2.

Note.—A table adjusting the gravity factors according to diameters and speeds (r.p.m.) of the baskets is shown in the appendix.

The range of gravity or "g" factors used in different industries is considerable, depending on the material handled. For example, when dealing with large sul­phate of ammonia crystals a factor of no more than 80 gives satisfactory dryness; in the sugar industry, however, factors ranging from 400 to 2000 are used.

In the table shown in the appendix those factors which are underlined refer to the conventional belt and water-driven centrifugals with basket diameters of 30, 36 and 42 ins. The table shows that the lowest factor is attained by the 42 ins. basket running at 850 r.p.m., and the highest by the 30 ins. basket running at 1200 r.p.m. However, the electrically driven 30 ins, basket also belongs in the "old-fashioned" category as well as the 42 ins. basket running at 960 r.p.m. when driven by a six-pole

AC motor. The centrifugal force of the 30 ins. E.D. basket running at 1460 r.p.m. is 908 times "g" or 50 per cent. more than that of the other ordinary or standard centrifugals. Since 908 times "g" is the highest value attained by conventional centrifugals we will call all centrifugals exerting higher gravity factors than 1000 "High Gravity Factor Centrifugals" and all centrifugals exerting centrifugal forces be­tween 500 and 1000 times "g" "Medium Gravity Factor Centrifugals." This means that all conven­tional centrifugals are to be classed as medium gravity factor centrifugals, with the exception of the 42 ins. basket when running less than 915 r.p.m. A 42 ins. basket running 800, 850 or even 900 r.p.m. exerting less than 500 times "g" is to be classed as a low gravity factor centrifugal. We give special stress to this question because in Natal belt or water-driven 42 ins. baskets are often used to cure C massecuites, while 30 ins. baskets exerting higher centrifugal force are confined to the lighter jobs.

It will be noticed that the popular name "high speed centrifugals" as a descriptive noun for both kinds of modern centrifugals has not been used up to the present. To be frank it has been purposely avoided because the use of this word has led to much misunderstanding. To prevent confusion it is recom­mended that the words "High Duty Centrifugals" be used to describe machines whose chief character­istic is the great number of cycles performed per hour; and to use the name "High Gravity Factor Centrifugals" when the main characteristic is the high centrifugal force exerted at full speed.

When ordering conventional types of centrifugals the choice between the different types of centrifugal was relatively easy. When electrically driven units were considered and the supply was 50-cycle AC, only two basket sizes came into consideration, viz. the 30 ins. and the 42 ins. basket; the former run-ning 1460 r.p.m. for low grade, the latter running 960 r.p.m. for high grade sugars. When 36 ins. baskets were preferred a frequency changer, increas-ing the number of cycles from 50 to 571/2 per second, was required in order that the 36 ins. baskets should run at 1100 r.p.m. (619 times "g"). When water or belt-driven conventional machines were considered, much the same can be said as has been said about electrically-driven centrifugals.

With modem centrifugals, however, besides having to choose between four types of drive, we have also to consider such requirements as acceleration rate, top speed, charging speed, etc. It is this question, i.e. the specifications in the case of modern centri­fugals, that we want to discuss.

35

Operating Conditions of Modern Centrifugal Machines.

The operations of modern centrifugal machines are characterized by three speeds:

(a) the spinning speed; (b) the charging speed; (c) the discharging speed.

The spinning speed is dictated by the rated speed of the motor in the case of individual electrical drive; by the water pressure and the diameter of the Pelton drive in the case of water-driven centrifugals, etc. To obtain the two other speeds, viz. the charging and discharging speeds, special measures have to be taken. For example, the speed of the basket must be prevented from exceeding the speed limit of 60 r.p.m. when discharging so that the operator handling the mechanical plough will run no risk. During the charging period the basket speed must be kept between certain limits to assure the building-up of a sugar cake of uniform thickness; these limits range from 75—150 r.p.m., for easily purging masse-cuites, to 250-350 r.p.m. for less easily purging ones.

In regard to the discharging speed: a positive discharging speed can be obtained with direct coupled motors by connecting them to a low frequency supply. For example, a eight-pole motor connected to a 5-cycle/sec. AC supply (with a corresponding reduced voltage) will turn the basket when idling at a speed of about 70 r.p.m. which speed will slow down to 50-60 r.p.m. when the plough starts cutting into the sugar cake. It is obvious that centrifugals with different numbers of poles require different frequency changers; consequently a battery of machines with two-speed motors of 725/1460 r.p.m. demands a frequency changer different from that required by motors with a speed range of 465/960 r.p.m.

In the case of water-driven centrifugals a positive discharging speed can be arranged by means of an auxiliary shaft to which the basket spindle is coupled during ploughing operations.

In the case of fluid clutch drive combined with torque control the latter prevents a too high torque when the nose of the plough digs too deep into the sugar cake.

With the new Fluid Duplex Drive Coupling (an English patent) which can operate alternately to provide an acceleration and a deceleration torque, the torque can be manually or automatically con-trolled when ploughing.

With the "Turntork" arrangement small auxiliary electric motors can be connected by means of a dutch and V-belt drive to the basket spindles achieving positive discharging speed (in reverse with

respect to the spinning rotation) in the case of electrically, water, belt and gear-driven centrifugals.

Finally, ploughing speed can also be maintained by inching; inching being necessary in all those cases where no special measures are taken to provide a special ploughing speed or a limited ploughing torque.

Returning to the demand of a limited speed range for charging, before the Second World War a Con­tinental firm manufacturing centrifugals driven by pole-changing motors allowing basket speed of about 290 (or 320) r.p.m. during charging and speeds of 1460 for 960) r.p.m. during spinning. Since a motor connected to a 50-cycle supply requires 20 (or 18) poles in order to run at 290 (or 320) r.p.m. the motor concerned was of too complicated a design for its capacity. Moreover, a pole-changing motor with a speed ratio of 1 : 5 (or 1 : 3) requires more energy when accelerating than one with the more common ratio of 1 : 2 and the regenerative braking is less effective too than in the case of 1 : 2 ratio motor.

Note.—For effective super-synchronous braking from full speed to slow speed, full speed may not exceed twice the speed at low speed windings.

With variable speed motors with sufficiently wide ranges of speeds—which for example can be accom­plished with Schrage type motors—it is possible to maintain a positive charging speed, just as in the case mentioned where two-speed motors are wound for 290 (or 320) and 1460 (or 960) r.p.m. The charg­ing is, however, usually performed during the accel­eration period when the correct moment to open the feed valve is based on the estimated basket speed by visual observation. When badly timed, when the feed valve is clogged or when the massecuite flow is retarded due to too low a level of massecuite in the mixer, the machine has to be coasted to prevent it from exceeding the proper charging speed. It is obvious that coasting will be more frequently neces-sary the less attentive the operators are, and the higher the rate of acceleration is. Since coasting puts a severe strain on the switches it must be prevented as much as possible.

At the end of the spinning period the machines must be rapidly decelerated and even where conven­tional types of centrifugals with straight mechanical braking are concerned, brake linings are subject to great wear, and will contaminate the sugar charge and machine by the dust developed. Moreover, the brake linings have to be renewed rather frequently, causing service interruptions as well as expenses. With modern, heavy duty and high gravity factor machines the strain put on the brakes is far more severe, and though it is conceivable to adhere to straight mechanical braking by using amply dimen­sioned and water-cooled brake linings the application

of electrical braking avoids the drawbacks mentioned. Moreover, when electrical braking is performed by means of super-synchronous braking, part of the kinetic energy stored in the rotating parts will be restored to the AC supply.

The above serves only to provide an insight into the range of operating conditions of modern sugar centrifugals. We will continue by saying something about the required top speed and acceleration rate when handling different massecuites.

The top speed or the gravity factor should not be greater than is necessary to obtain the required result in the most economical manner, because, with higher gravity factors (a) initial capital expenditure is greater, (6) maintenance is more expensive and (c) energy requirements are greater. Since fine-grained massecuites of low purity demand a high centrifugal force for the proper separation of the mother liquor from the crystals, it is the C massecuite which requires in the first place high gravity factor centrifugals and more particularly the foreworkers of the C massecuite. On the other hand less finely grained massecuites and higher purity massecuites which purge more easily, for example A massecuites, do not demand high gravity factor centrifugals, but high duty centrifugals. High duty machines are machines capable of performing a high number of cycles per hour—for example 20 or even 25 and more cycles per hour. A high acceleration rate is essential for such high duty machines in order to reduce the time required for a cycle. However, to obtain a high top speed combined with a short cycle is most difficult, since such a combination demands very powerful electric motors which produce excessive power surges when starting.

The high acceleration rate inherent in the short cycle machines is beneficial for the curing process of A massecuites as well, because the mother liquor will leave the massecuite before the temperature has dropped and before the sugar layer has dried out by the fanning action of the spinning basket. When the machines accelerate slowly, the molasses coating around the crystals cools down and dries out, before the centrifugal force is high enough to spin it off; and in addition the increased viscosity of the coating will result in difficulty in washing of the sugar crystals.

Since (well boiled) A massecuites are easily purged, heavy duty machines with short cycles, high accelera-tion and medium gravity factor are particularly suited for handling A massecuites.

In the above we have discussed some of the speci-fications concerning centrifugals purging A and C massecuites. There are, however, more specifications, since there are more massecuites than only those two mentioned. Very enlightening in this respect is the article published by P. F. Grove, Chief Electrical Engineer of Messrs. John Miles & Partners (London) summarizing an investigation on behalf of Messrs. Tate & Lyle Ltd. of London. This investigation was made to determine the most suitable kind of electric motor for driving sugar centrifugals for a complete range of processes such as would be required for modernization of Plaiston Wharf Refinery (Int. Sug. Journ., vol. 51; 1949; 247). In a scheme which we reproduce below the different and varying require­ments demanded for the handling of each type of massecuite encountered in a refinery are specified. It concerns 58 centrifugals of 40 x 24 ins. required to handle the massecuites of a refinery melting 84 short tons of raw sugar per hour.

Sugar Centrifugal Data—For a Melt of 75 Long Tons Per Hour

Number of machines Wall thickness (inches) Weight of dried sugar (lb.) Full speed (revs. per min.) Gravity factor Charging speed (revs. per min.)

Ploughing speed (revs. per min.) Cycle time (secs.):

Charging . ... Accelerating Spinning Decelerating Ploughing

Possible limits of cycle time (min.)

White

15 6

490 1250 888

75-150

0-75

4 34-40 0-90

30 20

11/2-3

Affination

24 6

490 1250 888

100-250

0-75

10 35-50

60-270 30 20

21/2-6

1st Crop

5 6

500 1500 1278

100-400

0-75

15 50-100 0-240

40 20

3-6

2nd Crop

4 6

500 1500 1278

100-400

0-75

20 60-180 30-720

40 40

5-15

3rd Crop

10 4

425 1700 1642

Standing Charge

0-75

30 60-180

240-1360 50

60-120 10-25

36

37

We want to draw attention in particular to the difference in specifications concerning the charging operations of the white and of the other massecuites. The White centrifugals are charged by lower speeds than the Affination or the first and second crop centrifugals to prevent the sugar cake from piling up at the lower end of the basket. To build up a cake of approximately uniform thickness at the top and at the bottom, it is essential that the basket be run very slowly (at 75-150 r.p.m.) when charging with easily purging massecuites such as refinery white massecuites. The acceleration time of the white centrifugals is, moreover, the shortest of all (35—10 sec), in order not only to attain a short cycle but also to prevent cooling off and drying out of the sugar cake during acceleration. Both requirements, viz. slow charging speed and high acceleration rate are essential for efficient washing operations for a type of massecuite such as the refinery white massecuite.

Not only is a slow acceleration rate (acceleration is from one to three minutes in accordance with the prevailing circumstances) prescribed for the C masse-cuite but also a standing charge, in accordance with the above scheme. A slow acceleration, however, is not only confined to refinery low grade practice; it is as essential for proper purging of C massecuites of sugar factories as well. In the case of C massecuite loreworkers (in addition to slow acceleration) it is recommended that after charging the baskets be run first for some minutes at half speed, before changing over to full speed in order to spin off the bulk of the molasses before the sugar cake packs too tightly. However, such an additional procedure is only possible in the case of two-speed and variable-speed motors.

The Different Means of Driving Sugar Centrifugals.

Since it is the drive which has to fulfil all speed and acceleration requirements (and often the de-celeration requirements as well), the different means of driving will be discussed first: The centrifugals can be driven—

(a) by belts and pulleys from a mutual shaft (belt-driven centrifugals);

(b) by bevel gears and clutches from a mutual shaft running over the top of the centrifugals (gear-driven centrifugals);

(c) by Pelton turbines driven by water under pressure (water-driven centrifugals);

{d) by individual electric motors (electrically-driven centrifugals).

While there are probably still more belt and water-driven centrifugals than electrically-driven machines in the world's sugar industry, most modern installations have a direct electric drive. We will therefore particularly consider (d).

Electrically-driven centrifugals can be sub-divided into two categories: (i) according to the way in which the motor torque is transmitted to the basket; and (ii) according to the type of motor used.

(i) Electrically-driven Centrifuges, subdivided accord­ing to torque transmission. Firstly, the simplest way of transmitting the

torque is by using the basket spindle as rotor shaft; or by having the rotor shaft connected to the basket spindle by means of a flexible coupling.

Secondly, a system which has become obsolete, is the transmission of the torque by means of friction exerted when hard wooden blocks or strips of belting are pressed by centrifugal force against the inside surface of a drum which is fastened to the basket spindle. In this instance the motor reaches full speed one to two seconds after starting and gradually spins the basket with it.

Thirdly, the motor torque can be transmitted by means of a fluid clutch drive. In this case the motor does not stop for every charge, but is always running at full speed; the coupling between motor and basket being made by filling the fluid clutch with, or empty-ing it. of oil. Recently an English-patented Duplex fluid clutch drive came onto the market. This drive can also provide a reverse torque for deceleration.

(ii) Depending on the type of motor used the following distinctions can be made. Firstly, the ordinary DC motor which, however,

has become obsolete because direct current motors are not adapted to use in a sugar factory.

Secondly, the slipring AC motor which was rather popular with some Continental firms as a centrifugal drive before the introduction of the modern centri-fugals but which has at present nearly completely lost its place to the squirrel cage motor.

Thirdly, the squirrel cage motor, the simplicity and mechanical robustness of which gives it an advantage over any slipring, commutator or DC motor for sugar factory conditions. By using a two-speed, rather than a single-speed, winding, the losses inherent in squirrel cage motors are halved, and braking from full to half speed can be done elec­trically with some recovery of energy and a reduction of wear on the mechanical brakes. (Mechanical braking from full speed to rest gives the brake the duty of braking four times more than braking from half speed to rest.)

Fourthly, the AC commutator motor; the rotor-fed as well as the stator-fed—the rotor-fed (brush shift or Schrage type) which has more advantages for centrifugal drive.

Fifthly, the DC motor in combination with the Ward-Leonard system and with the Constant Cur­rent system.

A development to be watched is the application of AC commutator motors to centrifugal drives, since this materially reduces the energy consumption. In the report of the investigation quoted earlier in this paper it was the Schrage type AC commutator motor which was finally recommended because it shows the lowest overall energy consumption. With the exception of the two-speed squirrel cage motor the Schrage type motor also shows the lowest price, since it does not require convertors as do the Ward-Leonard and Constant Current systems. Neither does it require frequency changers for ploughing operations as do the stator-fed AC and the squirrel cage motors. Last but not least the Schrage type motor would enable Tate & Lyle Ltd. to change to one type of centrifugal motor since 58 identical Schrage type motors can meet all the varying re­quirements of the different types of massecuites to be handled.

While AC commutator motors may eventually prove themselves preferable for special occasions, squirrel cage motors will probably continue to be standard driving units for some years, and more particularly changing-pole squirrel cage motors with 1 : 2 speed ratio.

To bear out the reasons for the preference of the two-speed motor over (a) the single-speed motor (b) the belt-driven, (c) the gear-driven, and (d) the water-driven centrifugals, we have to go back to the energy requirements for accelerating and de­celerating centrifugals.

Let us assume that a centrifugal basket is con­nected to a steam engine in such a way that both start accelerating simultaneously and equally, and that we neglect all energy losses due to friction. Only in such a case as this is the energy required to accelerate the basket from rest to "n" revolutions per minute equal to the kinetic energy "Z" stored in the rotating basket at the end of the acceleration period.

If, however, the engine is already running at "n" revolutions, and the basket is subsequently con­nected to the engine by means of a clutch so that the engine gradually spins the basket to its own speed "2Z" will be required to accelerate the basket, since "1Z" in excess will be converted into heat.

The second example represents the case when baskets are (a) belt-driven, (b) gear-driven, (c) water-driven, (d) driven by direct-coupled single-speed motors, or (e) driven by single-speed motors by means of fluid drive or mechanical clutch. In each of these methods of driving the basket, "2Z," is required, to store "1Z" in the form of the kinetic energy of the basket, and the "1Z" which is in excess will be dissipated in the form of heat in the

transmission. This is the reason why modern centri­fugals have water-cooled belt pulleys; and why the oil used in the clutch of fluid drive centrifugals has to be water-cooled. In the case of directly coupled single-speed motors "1Z" will be converted into heat in the rotors, or in the case of slipring motors into heat in the separate, external resistances. In the case of water-driven centrifugals the "1Z" in excess will cause the temperature of the water used to rise. In this connection all cases mentioned show the same efficiency, viz. they all require "2Z" to gain "1Z" in the form of kinetic energy; the "1Z" in excess being converted into heat (and wear).

A centrifugal driven by a two-speed motor, how-ever, requires only "11/2Z" when accelerating in the proper manner, viz. at first connected to the half-speed winding accelerating to half speed; and secondly from half to full speed with the aid of the full speed winding. In this case "1/2Z" only will be wasted by conversion into heat in the rotor. (See Appendix III.)

Not only during the acceleration period does the two-speed motor show a lower energy consumption, but also by means of super-synchronous regenerative braking, part of the kinetic energy can be regained during the braking period. In contrast single-speed motors, belt-driven, water-driven and gear-driven centrifugals lose the whole kinetic energy as it dissi­pates in the form of heat (and wear) of the brake linings, which consequently have to be water-cooled and of ample dimensions.

This is not the case when electrical braking can be applied. It is now intended to discuss this. There are three distinct ways in which electrical braking can be put to use in the case of squirrel cage motors:

(a) DC braking; (b) plugging (or braking by reversal of the

current); and (c) regenerative braking under super-synchronous

conditions.

In the case of DC braking, one phase of the stator winding is fed from a DC supply, thus creating a stationary field which sets up current in the rotor circuit until all kinetic energy has been converted into heat in the rotor.

In the case of plugging, the direction of the, rotating field of the stator is reversed to obtain the braking effect; the braking period being of the same duration as the starting period. Since the heat developed in the rotor during braking will be three times as great as that developed during starting, braking by current reversal can only be used in the case of slipring motors, because these motors have separate, external rotor resistances.

38

Where super-sychronous, regenerative braking with pole-changing motors, having pole numbers in the ratio 1 : 2, is concerned, the full speed running motor is suddenly changed over from the low-pole to the high-pole winding and thus braked electrically to half speed. The direction of the rotating field, however, is not reversed. When the ohmic losses are neglected, one half of the kinetic energy "Z" is recovered in this way; one-quarter of "Z" is dissi-pated as heat in the rotor and one-quarter of "Z" is still available at the end of the regenerative braking period since the centrifugal is still running at half speed.

Braking from half speed to rest can now be accomplished (a) by DC injection, (b) by connecting the high-pole winding to the low frequency supply for ploughing, or (c) by straight mechanical braking.

In the case of the Duplex fluid coupling of English make, which has already been mentioned, the Duplex can also be used for braking, since it embodies two separate oil circuits; one for acceleration and the other for deceleration. The electric or hydraulic motor (after it has been started) can be kept running at full speed all the time the battery of centrifugals is operating, just as in the case of the "fluid drive clutch" of American origin. The acceleration and the braking each occur within a fixed time predeter­mined by the designs of the acceleration and the deceleration circuits of the Duplex respectively. When the scoop controlling the flow of oil to the acceleration circuit is moved to the three-quarter "in" position, a ploughing torque of about 150 lbs. ft. is developed at a speed of 50-70 r.p.m. By operating a diverting valve the braking circuit can be filled with oil and a braking torque will be exerted by the Duplex. When braking from full speed to rest the entire kinetic energy "Z" will be converted into heat in the oil. Tn addition to this, energy has to be supplied to the motor driving the Duplex in order to create a contra torque for the braking. However, wear of brake linings is com­pletely eliminated.

When accelerating by means of a fluid drive most of the energy wasted ("1Z") will be used in raising the temperature of the oil of the fluid drive, but part will be converted into heat in the rotor of the motor.

Basket Dimensions.

The conventional machines used to have baskets of the following sizes: 30 x 18 ins.; 36 x 18 ins.; 42 x 20 ins. and 42 X 24 ins.; (48 x 24 ins. baskets were used occasionally). The modern machines have predominantly 42 X 24 ins. baskets.

America, however, has always specialized in 40 ins. diameter baskets and also adheres to them in the

case of modern machines. Recently, however, an automatic batch type centrifugal has arrived on the market with a 48 ins. diameter basket.

Since a motor connected to a 60-cycle AC supply will run 1.20 times faster than a motor with the same number of poles connected to a 50-cycle AC supply, American centrifugals will exert l,22 or 1.44 times the centrifugal force with baskets of the same diameter as the European centrifugals. Although the American baskets are only 40 ins. instead of 42 ins. the gravity factor will still be 1.37 times greater. Even should the European manufacturers change to 48 ins. diameter baskets, the American 40 ins. baskets driven by 60-cycle AC supply would still be in the lead as the following scheme shows:

40 ins. 42 ins. 48 ins. Gravity factors basket basket basket

Four-pole motor: 60-cycle AC supply 1745 — — 50-cycle AC supply — 1273 1453

Six-pole motor: 60-cycle AC supply 753 — — 50-cycle AC supply — 550 628

We see from this how the 50-cycle AC supply is really a handicap when higher gravity factors are demanded and the gravity factors of our 42 ins. machines running 1460 and 960 r.p.m. are both fairly low for the work for which they are used. The C massecuite machine would be better if it could exert a centrifugal force higher than 1273 times "g." The same can be said about our A massecuite mach-ine which exerts only 550 times "g." Changing over to 48 ins. machines would split in half the difference in "g" factor. Bigger baskets, however, have a lower payload than smaller baskets. In a paper read at the Third Technical Conference of the British Sugar Corporation Limited (1950), J. Broadbent showed that the stored kinetic energy in a 40 ins. basket exerting 885 times "g" is 1,255 lbs. ft. per pound of sugar against 1,690 lbs. ft. in the case of a 48 ins. basket exerting the same "g." The kinetic energies per pound of sugar in these cases are there-fore approximately in the proportion of 3 to 4.

If we should want higher gravity factors than those which can be obtained by the rated speeds of our motors, in my personal opinion the best thing to do is to step up the frequency just as is done in the case of electrically-driven 36 ins. baskets where the frequency is stepped up to 571/2 to obtain the required "g." The same motors could even be used when only the voltage of the supply is stepped up in pro­portion to the step-up in frequency.

39

40

A table is added as an appendix showing the volume and the weight of massecuite with which differently dimensioned baskets can be charged. It is adjusted for the apparent thickness of the layer of massecuite. The term "apparent thickness" is used here, because owing to the fact that immediately a basket is charged with massecuite part of the mother liquor is expelled. It is thus possible to charge a basket with a rim or lip width of only 6 ins., with a layer of massecuite of "apparent thick-ness" of 7 ins.

Since it is routine in come countries to refer to the centrifugal capacities as square feet screen area per ton of cane crushed per hour, the number of square feet screen area of the different baskets is shown as well.

Power Factor and Electrically-Driven Centrifugals.

During the period of acceleration the motors of electrically-driven centrifugals are fully loaded and consequently their power factors will be high, i.e. in the neighbourhood of 0.9. During the time the baskets are running at full speed, however, the motors are only opposed by the friction torque of the centrifugals, which incidentally is chiefly com-posed of the resistance of the basket to air, and since the load at that time is only a fraction of the full load the power factor drops to 0.5 or lower. It is a fact that at the end of the acceleration period the power consumption drops to an even greater extent than the power factor does, but the influence of these low individual power factors will still be noticeable at the AC supply.

In this connection it can be mentioned that auto-matic coasting of the machines immediately when they have achieved their top speed or some minutes before the brakes will be applied, will reduce the influence of the low power factors originating from partly-loaded motors.

When the necessity to improve the low power factor resulting from the centrifugal motors becomes imperative (as a low power factor reduces the work-ing capacity of the generating plant and mains, and lowers the efficiency of the system) nowadays static condensors or capacitors assigned to each individual centrifugal motor are installed to improve the power factor on the spot. Static condensors afford one of the simplest and most efficient solutions of the prob-lem indeed, as they require no attendance and cause very little loss of energy. Moreover, as these con-densors counteract the wattless currents at the spots where they originate, they not only improve the power factor at the generator, but also in the mains, thus reducing the losses due to wattless currents running through the mains at the same time.

Static condensors are therefore to be preferred to special synchronous motors when improving the power factor is concerned. The leading current generated by over-excitation of a synchronous motor is nearly always generated at a place located far from the spot where the low power factor originates; consequently the losses in the mains will not be reduced by synchronous motors.

Continuously-Operating Centrifugal Machines

The centrifugal machine was introduced into the sugar industry more than a century ago and since then its design has been continually improved. It is at present a highly efficient machine, even capable of performing automatically all required functions except charging. However, it still works in batches as it did a hundred years ago, and not continuously.

In the chemical industry continuously operating centrifugals have proved their merits for more than twenty years; the introduction of a continuous centrifugal into the sugar industry, however, was delayed till about three years ago due to the diffi-culties encountered by trying to make the con-tinuous machine adaptable to sugar manufacturing conditions.

Advantages of continuously-operating centrifugals are: (a) saving of labour, (b) saving of maintenance costs, and (c) lower and more uniform power con-sumption than that of the batch centrifugals. It is these advantages which incited designers of centri-fugal machines to try to design continuously-operating machines as early as fifty years ago.

The continuously-operating centrifugals can be divided into two categories, according to the means used to discharge the cured product from the baskets:

(i) continuous machines which discharge the cured product by means of a slowly rotating helice, or screw conveyor;

(ii) continuous machines which periodically push the cured product to the discharge side of the basket by means of a disc.

Both categories have a characteristic in common, in that the basket rotates on a horizontal axis. A horizontal position of the basket lends itself better to the operation of continuous discharge than does a vertical position, such as in the batch type centrifugal.

In the so-called push type machine a hydraulic pusher in the form of a fairly tightly fitting disc, periodically pushes the layer of product forward until it is expelled by centrifugal force at the discharge end.

It is the push type of continuous centrifugal sug­gested by Eckstein in 1908 that has been developed

APPENDIX I GRAVITY FACTORS FOR CENTRIFUGAL MACHINES

N O T E S

* These speeds (960 and 1460 r.p.m.) are the rated speeds of induction motors equipped with six and four poles respectively when connected to a 50-cycle AC supply.

¥ These gravity factors (550, 908 and 1272) are the factors obtained at those rated speeds in the case of baskets of 30 and 42 ins. in diameter.

These gravity factors (431, 564 and 614) are the factors generally obtained with water or belt-driven 30, 36 and 42 ins. baskets of conven­tional design, the 42 ins. basket having the lowest factor. In the case of electrically-driven machines the 30 ins. basket running 1460 r.p.m. attains a factor of 908 or 50 per cent, higher than that of the other machines; the 42 ins. E.D. improves to a factor of 550 when running 960 r.p.m.

In the case of high gravity factor machines the electrically-driven 42 ins. basket runs at 1460 r.p.m. and attains a centrifugal force of 1272 times "g."

42

by Escher Wyss into a practical, usable contin­uous machine for the chemical industry; since 1950 a machine has been developed which can be used for easily purging sugar massecuites and magmas.

The difficulty encountered in making the contin­uous machine adaptable for curing sugar factory products is the relatively high centrifugal force required. We have already mentioned that for drying ammonia sulphate crystals a gravity factor of 80 is sufficient; sugar factory products, however, require "g 's" from 400 to 2000.

Let us assume a continuous centrifugal rotating at such a speed that the basket exerts a centrifugal force of 500 times "g" and let the basket wall be covered by a sugar layer of one inch in thickness. This one inch sugar layer will be pressed against the perforated plate covering the basket wall with a force such as would be exerted if 500 inches of sugar cake were resting upon it. Consequently the disc which has to push the sugar cake along the basket

wall has to exert a force such as would be necessary if 500 inches or 42 feet of sugar cake were resting on the perforated covering. Moreover, the pusher has to move the sugar layer gently in order not to crush (too many) crystals. These were the difficulties resulting from the higher gravity factors required for curing sugar factory products, when it was attempted to extend continuous operation from the chemical industry to the sugar industry.

In the beet sugar factory and refinery in Aarberg in Switzerland there are now however two push-type machines designed by Escher Wyss, which cured the whole raw sugar crop of 1951-52. Aarberg has a daily output of 400 tons raws. The power consump­tion of these continuous centrifugals is only 1½-2½ h.p. hours per ton of sugar discharged, depending on the qualities of massecuites and sugar which are cured. One machine will cure 8-10 tons of B sugar per hour. At the end of 1952 an identical machine came into operation at the beet sugar factory Uelzen in Germany.

Volumes Adjusted

APPENDIX II

in Cu. ft. and Weights in Lbs. of Massecuites (weighing 90 lbs. per cu. ft.) to the APPARENT Thickness of the Layer of Massecuite when Charging

43

APPENDIX III

A.—Single-Speed Motors Accelerating and Decelerating

Diagrams I to IV included concern the accelera­tion and deceleration processes of a centrifugal driven by a single-speed electric motor. Diagram I concerns the energy consumption when a squirrel cage motor directly coupled to the basket spindle, accelerates. Diagram II depicts the same when a slipring motor accelerates. Diagram III represents the energy consumption when a single-speed squirrel cage motor accelerates; the motor transmitting its torque, however, by means of a centrifugal slip coupling. Diagram IV concerns the decelerating period of a single-speed motor.

Diagram I.—If it is assumed that the torque developed by the motor has a constant value which is independent of the speed, the basket speed of the centrifugal will increase lincaly with the time. The power transmitted from the motor in a certain unit of time is proportionate to the torque and the speed (r.p.m.) during the given unit of time. Consequently the power also increases lineally with the time as indicated by the line AB in the diagram. The energy supplied to accelerate the mass of the basket during the entire starting process will be proportionate to the time and the power supplied per unit of time; thus it can be represented by the area of the triangle ABC which also represents "Z," the kinetic energy stored in the basket at the end of the period of time AC. Since the same quantity of energy will be con­verted into heat in the rotor the rectangle ADBC represents the entire quantity of energy to be supplied from outside, since the area of triangle ADB is equal to that of ABC.

Diagram II.—In this instance most of the excess energy (AEB) is transformed into heat in the slip-ring resistance (DEB), while part is converted into heat in the rotor (ADB).

Diagram III.—The rectangle ADBC represents the entire quantity of energy supplied to the motor during the acceleration time AC, of which the area of the triangle ABC depicts the quantity usefully consumed and the triangle ADB the quantity wasted. In this case, however, the wasted energy has to be divided into AEF, the energy converted into heat by slipping of the coupling, and EDBF—the heat in the rotor (in this instance F indicates the point where the coupling stops slipping and the basket continues accelerating—now stiffly coupled to the motor).

Diagram IV.—The content of triangle ABC again represents the stored kinetic energy "Z" in the basket; however, in this case AC indicates the deceleration time. Since the entire kinetic energy has to be dissipated by mechanical braking, ABC also depicts the energy converted into heat in the brake linings.

B.—Two-Speed Motor Accelerating and Decelerating

Diagram V concerns the accelration, and Diagram VI the ceceleration process of a centrifugal driven by a two-speed motor.

Diagram V.—In Diagram V ADEG represents the energy supplied in the time AG when the basket is accelerating from rest to half speed; AEG being the stored kinetic energy, and ADE the wasted energy which heated the rotor of the motor.

The rectangle GFBC represents the energy sup­plied in the time GC when the basket accelerates from half to full speed.

The area of ADEFBC indicates the entire energy supply; ABC again represents the kinetic energy "Z" of the basket at full speed and the area of the two triangles ADE and EFB depict the energy con­verted into heat in the rotor. The diagram shows that in this case the area of the blank quadrangle DHFE or "½-Z" less energy has to be supplied than in the case of the single-speed motor.

Diagram VI.—When declerating from full to half-speed by super-synchronous braking the energy represented by the rectangle ADEF or "½Z" will be converted into electrical energy and returned to the AC supply, while DBE will be lost as heat in the rotor. When braking mechanically from half speed to rest EFC will be converted into heat in the brake linings. Since the latter quantity is only "¼Z" the duty imposed on the brake linings will be only one quarter of that imposed in the case of mechanical braking from full speed to rest, which is equal to "1Z" according to Diagram IV.

Mr. Farquharson complimented Mr. Perk on his paper and said he agreed with most of what he had said but there were one or two points which he would like to amplify. He was glad to see that Mr. Perk had drawn attention to the difference between "High Duty" and "High Gravity Factor Centrifugals" and the desirability of charging the baskets at low speed, but he warned against charging at, or just below, the critical speed of the machine. In his experience, with native labour operating 2 speed AC motor driven machines, he found it impossible to get a consistent charging routine and, after a few days, he would find one basket being charged at a very low speed and another at half speed. Strange to say, there did not appear to be a great deal of difference in the final product.

He agreed with Mr. Perk regarding the severe strain put on the switchgear due to "coasting" and stated that for the same reason he was not in favour of "inching" as a means of obtaining a ploughing speed. With regard to brake wear, he thought it might be of interest to note that at

APPENDIX III

44

45

Maidstone their 40 in. X 24 in. AC 2 speed 730/ 1460 r.p.m. centrifugals, installed in 1946, were still running with their original linings; and at Z.S.M. & P. six similar machines, but with constant current motors, had only required 4 linings alto­gether during the past 10 years.

He found the table of figures for a melt of 75 tons an hour most interesting, and thought it would prove very useful. Most centrifugal manufacturers endeavoured to submit accurate duty cycle times for their machines, but with the conditions peculiar to the Natal Sugar Belt this was most difficult. The massecuite varied from pan to pan and crystalliser to crystalliser. At Maidstone, for example, the extreme cases had been times when the C massecuite could be cured with a spinning time as short as 60 seconds, and when spinning for 25 minutes failed to cure it, a sticky substance rather like bird lime being formed on the inside of the sugar wall. With difficult massecuites, spinning at half speed for some time had helped, but the most successful results were obtained by charging the baskets to only half their capacity and running a normal cycle. Bad massecuites did not always plough clean and it was necessary to plough a second time or even run the machines empty and steam or wash the baskets every now and again to keep the screens clean. He did not think a higher "gravity factor" would be any benefit under these circumstances, and mentioned that Z.S.M. & P. had actually reduced the speed of their C masse­cuite foreworkers from 1800 r.p.m. to 1600 r.p.m., i.e. "g" from 1841 to 1454, some years ago. He explained that the spinning speed could be readily altered within specified limits on constant current machines, also the rate of acceleration on the later designs.

He was not happy about Mr. Perk's recommend­ation to step up the frequency to increase "g". The motors, he thought, would stand the higher electrical and mechanical loading, but what of the baskets and spindles? He doubted whether any centrifugal manufacturer would agree to running the centrifugals at higher speeds without some strengthening.

Referring to the different types of motor used, he thought a further sub-division into (a) fixed speed motors and (b) variable speed motors, might be helpful. Squirrel cage and slip ring motors came under (a), constant current and to some extent AC commutator motors under (b). Under (a) the operating speeds and acceleration were fixed by the design of the machine and could not be altered except by altering the frequency and voltage. The horse power required was large to cover the losses as shown in the diagrams, Appendix III, and the power supply must have ample current capacity to cope with the heavy

demands when starting, bearing in mind that the current of a 2-speed motor during this period was approximately four times full load, otherwise the supply voltage would drop badly. He thought the power factor of 0.9 mentioned by Mr. Perk during acceleration rather high and felt that 0.6 would be more correct. The motors under (b) could be set to suit whatever working conditions were required and acted like the steam engine starting from rest, so that the total accelerating power required was "IZ". This was borne out in practice by the fact that the constant current motors at Z.S.M. & P. were only 60 h.p. as against the 2-speed AC motors at Maidstone which were 120 h.p. The baskets were identical and the duty cycle almost the same; further, owing to improved regeneration, the power demand from the supply was low and practically constant, so that the motor-generator set operated at high power factor continuously.

Regarding braking by plugging, Mr. Farquharson said he thought this method was obsolete. Apart from the great heat which had to be dissipated by the rotors or their resistances, there was not only no regeneration whatever but actually heavy power demands from the supply to stop the machines. It certainly was not a method that he would recommend.

Mr. Main expressed his interest in the subject and paid tribute to the work Mr. Perk had done.

Dr. Douwes Dekker asked whether he had under­stood correctly that the spinning time of certain C massecuites at Maidstone was only one minute.

Mr. Farquharson replied that the actual running time of spinning was one minute.

Dr. Douwes Dekker said that the aim of C strikes was to crystallise as much sucrose as possible, i.e. the purity of the final molasses should be as low as possible. To exhaust properly the molasses of C massecuite, it was necessary to concentrate the massecuite to a high density and unfortunately a high brix and a low purity inevitably meant a high viscosity of the final molasses. The plant, and in particular the centrifuges, should be capable of processing a C massecuite containing a final molasses of high viscosity and even with powerful centrifuges a spinning time of 10—15 minutes was not excessive. In Hawaii much longer spinning times were con­sidered normal. The fact that it had been possible at Maidstone to spin C massecuites in one minute indicated that the viscosity of the final molasses had been low, for there was a linear proportionality between the time required to obtain a certain separation effect and the viscosity of the molasses of the cuite. A final molasses of such low viscosity could only mean that the molasses had not been properly exhausted. A spinning time of one minute could be achieved at any factory by raising the

46

density and the purity of the molasses, but this did not prove that the factory was operating successfully.

Mr. Walsh said that Mr. Perk had done his best to present a paper which set forth many of the changes that had recently come into being. He could only wish, however, that Mr. Perk had summarised it or tried to give some recommend­ations. The centrifugal manufacturer was faced with -a wide variety of conditions which varied from factory to factory and from country to country, and therefore could not design a standard machine applicable to all conditions. This was where difficulty was being experienced in developing high gravity factor machines. He hoped Mr. Perk would continue his present investigations and present recommendations which he thought would be of great advantage to manufacturers.

Mr. Rault said it would be very desirable to have a record of the present centrifugal practice of all South African factories. He felt sure that very few of them had enough machines and could afford the time to spin their third massecuite for as long as 15 minutes or over in the centrifugals, and the object of replacing the obsolete machines with high speed ones was to limit the curing time to between 4 and 6 minutes and use half the number of centri­fugals to perform the same duty. He did not agree that very high viscosity and difficulty at curing the last grade was a desirable feature and a criterion of good boiling house work and low molasses loss, enforcing the use of a large number of centrifugals. A massecuite that did not purge its molasses com­pletely in 5 minutes was a sign of a badly cooked one or a bad relationship of crystals to molasses and the triplicating of the number of machines for extracting barely the last 10 per cent of the molasses still left uncured did not seem to be an economical solution of the problem of increased recovery and low molasses loss.

In the examples quoted by Mr. Perk on the result of South African factory work for 1952, he would prefer to cure a last massecuite yielding 40 per cent crystal, throwing out a molasses of 39 purity, rather than a massecuite of lower purity with only 35 per cent crystal and the same molasses purity. The high yielding massecuite would give a cleaner brown sugar, more easily washed to the high polarisation of a Grade 2, or a refinery raw, with less recirculation of viscous low molasses returned to process.

Mr. Perk, replying to the discussion, said in regard to the question of high duty and high gravity factor machines, that both were names commonly used in literature. The use of the name "high speed machines" for 42 ins. machines running 960 r.p.m., but performing 20 charges per hour could only create confusion, since the gravity force exerted by these

machines was just the same as that exerted by the old-fashioned electrically driven 42 ins. machines which could only perform 8 to 10 charges per hour. To be correctly named, the former had to be called "high duty" and not "high speed" machines. Since it was not always mentioned where the high gravity factor machines started and the medium gravity factor machines end, an indication of the border line was necessary. In his paper he had assumed that all machines exerting a centrifugal force higher than the old-fashioned ones, the electrically driven 30 ins. machine included, were to be called high gravity factor machines, in this way choosing 1000 times "g" as the border line between high and medium gravity factor machines.

In regard to the Constant Current Drive, he referred to the original paper by Grove about the investigations made on behalf of Messrs. Tate & Lyle, the result of which investigations showed that the Constant Current Drive was advantageous only in cases of very short cycle, i.e. up to three minutes when compared with the Schrage type motor. The Constant Current DC system with the other DC system, the Ward Leonard system, belonged to the most expensive ones.

In reply to the opinion of Mr. Farquharson, that the speed of centrifugals could not be increased without change of construction, Mr. Perk said that with the consent of the manufacturers an increase of 10 per cent in r.p.m. had sometimes been allowed. In the case of 42 X 24 ins. E.D. machines the speed was increased from 960 to 1056 r.p.m., simultan­eously raising the voltage by 10 per cent. Such an increase raised the gravity factor from 550 to 650 times "g". In general, manufacturers had also consented to increase the speed of the 30 x 18 ins. B.D. from 1200 to 1350 r.p.m. when requested.

In reply to Mr. Walsh, Mr. Perk suggested that the chart concerning performances of centrifugals shown in his paper was virtually the description provided by Tate & Lyle to the manufacturers with regard to the requisite characteristics of the 58 machines required. His paper on centrifugals was written for the purpose of drawing particular attention to the fact that when ordering machines a proper description of the required performance characteristics had to be given to the manufacturers.

In regard to the curing time for C massecuites, he said that the concentration and the purity of the C massecuites must be such that the C massecuite centrifugals were kept busy the whole day. Low grade centrifugals ought to be in operation 24 hours per day, and when they operated only 20 hours per day it implied that a lower purity of final molasses could be achieved with the available equipment. Conversely, the more C massecuite centrifugals were available, the lower the purity of the final molasses could be.

47

Mr. Dymond said he thought this was the first paper presented on centrifugal machines and he hoped that next year it would be possible to follow this up. He hoped within that time other engineers

could provide papers which would carry forward what Mr. Perk was doing. He asked the meeting to accord a vote of thanks to Mr. Perk.

INHERENT STEAM LOSSES FOUND IN WATER TUBE BOILERS By Wm. C. LINDEMANN, B.S.c. (Eng.S.A.), M.I.Mech.E., M.I.Cert.E.

It would be safe to estimate that over 90 per cent. of steam used in the Natal Sugar Mills, is generated, using bagasse fuel, by what are known as Water Tube Steam Raisers of the horizontal land and other types, and they possess peculiar features, which inherently interfere with the efficient production of steam.

The characteristics observed during steaming such kind of boilers on load reveal the following innate properties -—

(1) Irregular water levels in the water glass columns. In the double drum horizontal land type, illustrated by Diagram 1, the water level of the left hand side drum is lower than that of the right side one. In regard to the Marine Cross Drum Type, vide Diagram 2, the water level in the drum, is in reverse order to that of the horizontal double drum land class.

(2) The boiler carries a heavier deposit of scale in the right drum and its complement of cir­culating tubes, and in many cases the colour of the scale in the right side is different from the colour of the scale in the left side. This divides the unit into two distinct halves. In the case of the Marine Cross Type Boiler, the scale deposit increases gradually from about the middle row of headers to those on the left side.

(3) The boiler water under steam pressure pos­sesses variable chemical properties.

(4) Furnace flames do not follow through the flue passes to the base of the stack, in a straight line.

(5) Soot deposits follow a similar course to that of the scale deposit.

(6) The extreme right hand unit of a battery of boilers of the same make and type is usually found to be a slower steamer than the left hand side units. The Marine Type Boilers acting reversely.

(7) Magnetic Fields surround all kinds of boilers of the water tube class. In regard to the loco types, which may be termed fire tube ones the earth's magnetic field exercises an influence, but not so in connection with the water tube steamers. Cases on record in­dicate reversals of magnetism when the flue gases change their direction of flow from flue pass to pass.

It is evident that all the aforegoing characteristics arc governed by a force or combination of physical forces, which unknown to the designers entered into the construction of the boilers under review.

In order to get at the origin of the cause of this mysterious happening, 20 Boiler House Plants, comprising 109 of all kinds of steam boilers were surveyed over a period of about ten years, to wit:—

Fuel Used System of Firing No. of Boilers

Coal ... Hand fired 8 Chain grate stoker 27 Mechanical push bar 9

Waste Heat Natural draught ... 7 Bagasse ... Semi-Cook furnace 37

Step grate 12 Wattle Bark Dutch Over hand-fired 7 Wood ... Flat grate ... ... 2

109 Total No. Boilers surveyed..

Diagrams 1 and 2 diagramatically indicate that the directions of flow of water circulation and the travel of the furnace flame move at right angles to each other, the main principle followed by designers when constructing this type of steam raiser. As will be appreciated later on this survey, following this principle unwittingly allows an error to creep in regard to steam generation, in spite of the apparent advantage which designers feel they are deriving when staggering the tubes coupled with the right angle flow of flue gases.

Looking at the two different designs of the boilers vide Diagrams 1. and 2, from the furnace front, it is evident, that the boiler water circulation is in opposite directions, which gives one the first clue to the cause of this mysterious irregular steaming property.

Chemical Differences of the Boiler Water when under Steam Pressure

In connection with this phase of the survey, it was argued that the water properties inside the boiler would be uniform. According to popular belief it could not be otherwise.

In order to test the position, a boiler unit which had its middle pair of downcoming circulating tubes crossed over, was chosen because it was claimed that this crossing over of the two downcomers between the two drums would eliminate the irregular water levels.

48

MULTI-INCLINED WATER TUBE BOILER B & W LAND TYPE

MULTI-INCLINED WATER TUBE BOILER B & W MARINE TYPE

49

50

Result of the test: a representative one.

Left Drum Right Drum

Alkalinity to Phenolphthalein 2-9 c.cs. N/10 acid 4-8 c.cs. N/10 acid

Alkalinity to Methol Orange 7-0 c.cs. N/10 acid 12-25 c.cs. N/10 acid

This test surprised the party who conducted it. It will be appreciated from this simple test,

crossing over the downcoming circulating tubes between the drums only, has no effect upon the chemical differences of the water inside a water tube boiler.

Two Single Drum Boilers with a Common Brick Setting.

When single drum units having the same area of heating surface, one would normally expect the same strength of Alkalinity figures in each drum. Apparently placing them side by side in a common brick setting, produces a similar effect as is associated with double drum boilers, and this is shown in the following test.

Duration of test six weeks.

No. 1 Boiler the left hand one gave a total Alkalin­ity of 33.4.

No. 2 Boiler the right hand one gave a total Alka­linity of 51.8.

These figures speak for themselves, and it would appear that all the other characteristics mentioned and itemised earlier in this survey are dependent upon a physical force of some sort, either mechanical, electrical or chemical, severally or collectively.

Before proceeding further, it would be as well to quote Faraday's discovery, that electric currents are generated in conductors by moving them in a magnetic field and there is an electromotive force produced in the conductor, in a direction at right angles to the direction of motion, and at right angles also to the direction of the lines of force. Further consider the transverse movements of the sun, earth with its magnetism, they follow the principle as enunciated by Faraday. Similarly it is reasonable to assume that since it was found that Magnetism surrounded all kinds of water tube boilers, one was within one's right to use that famous British scientist's principle, on the grounds of heat and water circulation following a transverse movement, which in turn produced a magnetic effect stronger than that of a earth's field, and this can be verified in some boiler plants where it was found that the compass needle swung around 180 degrees when passing the instrument along the brick setting of the boiler.

Survey No. 10 illustrates this extraordinary finding.

SURVEY No. 10

MAGNETIC FIELD PLAN No. 3

Note—N.S. arrows actual directions of compass needle. The circled N. and S. are the true fields which would attract S. and N. seeking pole respectively. The field is apparently changed by the crossing over of down comers between drums. The principal point is that the furnace discloses a N. field.

Another convincing case is that of the right hand side of two separately independent fired boilers of the single drum land type in a common brick-setting, which is generally known as a compound arrangement. In spite of this particular boiler with a much larger heating surface than its sister boiler, it would not deliver its full duty. On the face of it, it all sounds ridiculous. Every time the fires were cleaned, down would come the steam to a very low figure of 7,000 lbs. per hour as re­gistered by the steam flow meter. Its rated capacity was 10,000 lbs. of steam per hour.

Eventually all the downcoming tubes of the circulating system were crossed over in pairs, vide Diagram 3.

The object in mind was to break up the uniform flow of the liquid flywheels as visualised and set up in the header construction. Similarly if the conductor in an armature were likewise treated no current would flow and the magnetism would to a certain extent exercise no influence. In other

51

INTO HEADER SECTIONS

DIAGRAM 3 Depicting crossing over of down-comer tubes in order to

demagnetise boiler.

words damp as much as possible of the magnetising effect of the two elements represented by the furnace flame and the circulating water, producing a thermal reaction in the steaming of the boiler while under load producing better results.

The result of this practical test or trial was astounding. The boiler delivered up to as much as 13,000 lbs. of steam as measured by the Steam Flow Meter. The European firemen expressed themselves as feeling satisfied, most of their troubles were solved.

The following figures obtained under practical operating conditions confirm the previous trial of the survey.

A double drum boiler of the horizontal land type was selected, because of its right hand position in the range, therefore being a lazy steamer.

Remarks in connection with the following test, from the Engineer:

"There were no alterations to the furnace. I think the figures would have been better had we not had the misfortune of a steam coil parting in one of the Vacuum Pans, and not being spotted in

Monthly Results Evaporation Per Cent. Increase (lbs. per hour

H.S.)

August 1938. . . 3.4 Tubes not crossed 1939.. . 3.46 Tubes crossed 1.8 1940.. . 3.70 Tubes crossed 8.9

September 1938. . . 3.25 Tubes not crossed 1939. . . 3.58 Tubes crossed 8.6 1940 ... '3.70 Tubes crossed 13.8

October 1938. . . 3.14 Tubes not crossed 1939. . . 3.42 Tubes crossed 8.9 1940.. . 3.70 Tubes crossed 17.8

November 1938 ... 3.41 Tubes not crossed 1939. . . 3.44 Tubes crossed .88 1940.. . 3.65 Tubes crossed 7.0

time to prevent the sugar finding its way into the feed water.

On opening up the boilers this off-season we found the steam scrubber or separator in the boiler drums half choked with sugar carbon which naturally retarded the steam discharger. We were not short of steam so did not open the boiler during the crop of 1939."

Obviously the figures speak for themselves and it should be clear at this stage that a great saving of steam is possible in the Sugar Industry. The 1940 figures give an overall average saving of 11.9 per cent, which is a very profitable return on a comparatively small outlay of money.

Therefore this practical test is of great value and serves to show that it is possible to rectify the irregularities in a simple manner without having to invest money in contrivances which usually add to the worries of a maintenance engineer.

It is interesting to record that by crossing over all the tubes in pairs, no undue stresses were revealed when the boilers were opened up for a check inspection.

Possible Savings in the Sugar Industry The average amount of Bagasse available during

the two previous years was approximately 1,890,000 tons.

Assuming a L.C.V. of Bagasse at 3,136 B.T.Us./ LB. and a Boiler efficiency of say 70 per cent, which is low, together with only a saving of 10 per cent, of steam by crossing all the down-comers of Water Tube Boilers in the Industry, it will be found that an overall saving of about 600,000 lbs. of Steam per hour can be expected. Correlate this saving with the amount of Sugar produced, an expected saving of £1 per ton is assured, on the basis of reckoning of £1 per lb. of steam evaported when purchasing boilers to evaporate this amount of steam saved.

52

Likewise an extra 189,000 tons of bagasse could be accumulated for the making of paper or paper products.

Mr. Farquharson said he would hesitate to cross swords with Mr. Lmdemann regarding boilers or their operation but he had to agree with remarks made regarding the shielded magnetic circuit. There was also a further point. Mr. Lindemann had stated: "Similarly if the conductors in an armature were likewise treated" (i.e. crossed) "no current would flow and the magnetism would to a certain extent exercise no influence." This was not the case. Crossing the conductors did not alter the direction in which they cut the lines of force, therefore the electromotive force would be the same as before. As a matter of fact, armatures had been manufactured with conductors twisted in the slots and they worked very well, but mechanical difficulties prevented this arrangement becoming popular. The reason for this type of winding was to improve efficiency by minimising the interaction between conductors and reducing circulating currents in the conductors themselves.

Mr. McCulloch said he wished to congratulate Mr. Lindemann on his paper which contained some novel explanations of peculiar boiler characteristics. It seemed to him that it would be possible to account for the magnetic fields associated with boilers by the elctrical currents which would be generated by thermal E.M.F.'s caused by temperature differences across different parts of the boiler structure. The thermal electrical effect was a well known pheno­menon and occurred when there was a temperature gradient across two metals or metallic parts in contact. Although the magnitude of the in­dividual E.M.F.'s was small, the associated current could be of appreciable magnitude owing to the large mass of metal and the relatively low ohmic resistance of the path through which it circulated.

The improvement in the boiler performance due to transposing the downcomer tubes seemed remark­able, but he was doubtful if the explanation was wholly due to this change. Boiler evaporation was substantially affected by scaled and dirty tube surfaces and he considered that part of the improve­ment mentioned was due to the installation of new tubes.

Mr. Gunn submitted that Mr. Lindemann had delivered a very excellent paper which should give rise to much discussion. He noted that Mr. Linde­mann stated that it was always the left hand boiler of the range which did the most steaming. He also stated that boiler water under steam pressure possessed variable chemical properties. One would accept this second statement and one would be reasonably correct to assume that with boilers on the same range the pressures would be equal and therefore the effect of pressure on all the boilers on range would be similar. Mr. Lindemann somewhat contradicted the first statement by saying that after six weeks testing, the alkalinity of the right hand boiler was higher than that of the left. That would indicate to him that the concentration of the boiler water had increased and that could only be caused by increased evaporation. In other words, it indicated that the right hand boiler had been working harder than the left. He asked whether the example quoted was an exception to his rule.

Dr. Douwes Dekker said that the data given at the top of p. 4 of Mr. Lindemann's paper indicated a greater evaporation when the tubes were crossed. In his conclusions, however, Mr. Lindemann assumed a better boiler efficiency, which was not the same. Were data available which showed that the better evaporation was not due to more fuel being burned?

Dr. Parrish said it was difficult to see how any magnetic force could operate within a boiler, since the inside of a boiler would be magnetically shielded. If, however, it was assumed that magnetism was produced within the boiler by the transverse move­ment of gas and water, what possible effect could this magnetism have on the chemical properties of the boiler water?

Mr. Reynolds said that careful tests carried out by the makers of the boilers to which he referred had revealed no increase in steam raising ability due to the provision of a multiplicity of cross-over downtake tubes. He explained that the crossed tubes were originally developed for the purpose of equalising the water levels in the drums of multiple drum boilers, in which capacity they served their purpose reasonably well.

Mr. Dymond asked that a hearty vote of thanks be given to Mr. Lindemann.

A REVIEW OF BAGASSE FUEL QUALITY TRENDS AND OF

RECENT BOILER EFFICIENCY TESTS

By A. F. MCCULLOCH.

Introduction.

It is now some 15 years since a paper concerned with boiler plant efficiency was submitted to the Association,1 and it is considered that a review of the current position would be of interest. During that period the need for maintaining and improving boiler efficiency has become more urgent in order to reduce process costs, to meet the demand of increased throughput and to overcome the problems created by the bagasse fuel yielded from recently introduced cane varieties being poorer in quality than the bagasse formerly available.

Fuel Quality.

Green bagasse docs not compare favourably as an efficient fuel with the more conventional types such as coal and oil on account of the high moisture content: e.g. during combustion approximately 25 per cent, of the heat liberated is irrecoverable and needed to evaporate moisture. While this limitation may be unimportant to the cane sugar factory using a by-product as fuel, the increasing world demand for cellulose-containing raw materials by the rayon and paper manufacturing industries could well create a price trend that may become attractive enough for some factories to sell surplus bagasse as raw material to the chemical process industries, and ultimately develop to a point where a substantial part of the output would be utilised in this way. This prospect is of significance for the national economy, since at present raw materials are imported and the development of rayon and paper making industries is now proceeding here in Natal. One plant is being constructed at Felixton where it is intended to utilise bagasse in the manu­facture of Kraft papers.

In the past the view has been expressed occasionally that fuel shortages may occur owing to the fibre content of the new cane varieties being less than the old varieties, but there is no evidence to confirm this point of view. Data has been re­produced from S.A.S.T.A. proceedings in Appendix 1, and discloses that the average seasonal fibre content for all factories has not changed significantly during the past five seasons. During this period the proportion of N:Co.3I0 cane crushed increased from .07 per cent, in 1948-49 to 21.12 per cent, in 1951-52, and to 36.81 per cent, in 1952-53: the latter figure is based on incomplete returns and will pro­

bably increase when all the data are available. The highest fibre content of. 16.28 per cent, was attained in 1951-52. Similar data are also shown for Um-folozi and Felixton—-factories that have crushed a large proportion of N:Co.310 cane during the current season. At Umfolozi the fibre content declined from 14.16 per cent, in 1948-49 when .01 per cent, of the throughput was N:Co.31.0 to 12.93 per cent, when 75.4 per cent, of the throughput was N :Co.31(). However, this decline cannot be wholly attributed to cane variety since comparing two seasons 1949-50 and 1950-51 the fibre content increased from 13.64 to 13.8 per cent, with N:Co.310 proportions of 3.7 and 27.75 per cent, respectively. Again there was an increase in fibre content from 12.58 per cent. in 1951-52 with 56.39 per cent. N:Co.310 crushed to 12.93 per cent, in 1952-53 with 75.4 per cent. N:Co.31(). At Felixton the proportion of N:Co.310 crushed rose from .028 per cent, in 1948-49 to 36.71 per cent, in 1952-53, and the corresponding average fibre percentage changed insignificantly from 16.46 to 16.44 per cent. Except for the season 1950-51 when the fibre content declined to 15.86 per cent, with a throughput of 17.15 per cent. N:Co. the fibre data has remained sensibly constant at 16.4—16.7 per cent, irrespective of the pro­portion of N:Co.310.

While there is no evidence then for diminishing fibre content there is evidence to show that the fuel quality of the bagasse has deteriorated during the past five seasons and that this may be influenced by the proportion of N:Co.310 cane. Referring to Appendix 1, the moisture content of the bagasse in 1948-49 was 50.53 per cent, and 51.61 per cent, in 1951-52: the current season's data show a further increase to 52.41 per cent. The change has been continuous and always in step with the proportion of N:Co.3I0 crushed, i.e. increasing when the pro­portion of N:Co.310 increased. The Umfolozi data show an increase from 51.01 per cent, in 1948-49 to 53.77 per cent, in 1952-53. Except for the season 1950-51 this increase has been continuous: an explanation for the exception is considered to be that a new mill tandem was installed during 1950-51. At Felixton the moisture content has increased from 51.12 per cent, in 1948-49 to 53.6 per cent, in 1952-3, except for the season 1949-50, when it was 50.79 per cent. It is probable that the explan­ation is partly due to the proportionally small amount (6.97 per cent.) of N:Co.310 crushed during that season which would not have a marked effect

53

51

on the whole throughput. Further evidence is available from the data in Appendix 2, where the results of tests of 7-8 hours duration on N:Co.310 cane are shown. In an 8 hour test at Natal Estates a moisture content of 53.10 per cent, was attained compared with 51.18 per cent, attained during the associated weekly period: at Chaka's Kraal two 7 hour tests showed increases of 0.24 and 0.77 per cent, respectively in moisture content compared with the associated 24 hour averages. At Felixton two 1 hour tests showed increases in moisture content of 3.34 and 1.32 per cent, respectively.

It may be considered that increased throughput rates and inhibition might have been the dominant cause of the increased moisture content, but the evidence available does not adequately support this view. For instance, at Umfolozi the fibre through­put dropped from 13.25 tons/hr. in 1948-49 to 12.5 tons/hr. in 1949-50, whereas the bagasse moisture content increased from 51.01 to 52.45 per cent, respectively, and imbibition decreased from 229 to 202 per cent, respectively. A similar comparison of 1949-50 with 1950-51 discloses that the moisture content diminished from 52.45 to 50.80 per cent, with throughputs of 12.50 and 13.20 tons/hr. fibre respectively, and with imbibition ratios of 202 and 205 per cent, respectively. The throughput data for 1951-52 and 1952-53 at Umfolozi are not directly comparable with the former data since in 1951-52 two mill tandems were operating and only one in 1952-53. At Felixton the moisture per cent bagasse dropped from 51.12 per cent, in 1948-49 to 50.79 per cent, in 1949-50, although the throughputs were 13.4 and 13.55 tons/hr. respec­tively and imbibition was 229 and 202 per cent, respectively. Similarly, comparing 1951-52 with the current season, throughout dropped from 15.07 to 14.3 tons/hr. while the moisture content of bagasse increased from 53.26 to 53.0 per cent, and imbibition remained sensibly constant at 225 and 226 per cent, respectively. It may be also inferred that moisture content increased with increasing throughput, e.g. by comparing the Felixton data for 1949-50, 1950-51 and 1951-52, but since it has been shown that the moisture content may also decrease under the same conditions it is concluded that the effect of throughput rate is not as signi­ficant here as the apparent effect of the property of N:Co.310 bagasse yielding moisture less freely.

Comparatively small changes of moisture content have a marked effect on the quality of bagasse fuel. In Appendix 3, Fig. I the variation of calorific value with moisture content has been plotted for a bagasse having a nominal analysis of 2.5 per cent, sucrose and 50 per cent, moisture. The variation is linear and shows that for each 1 per cent, change of moisture content the lower calorific value changes by 2.5 per cent., e.g. increasing the moisture content

to 52 per cent, would reduce the quality of the fuel by 5 per cent. Of course such a reduction may be insignificant in a factory where there is an abundant supply of bagasse, but in less favoured factories where no surplus bagasse is available difficulties due to drop in steam pressure and reduction in boiler evaporation rate would occur. Aside from the reduction in calorific value increased moisture content reduces furnace temperature and demands increased firing rate to maintain constant evapor­ation. In turn the effect of increased firing rate increases both fan power requirements and unburnt fuel losses which results in lowering the boiler efficiency.

It is not easy to offer a solution to this problem. The first and most obvious remedy would be to dry the bagasse, but this is impracticable on account of the size of the drying installation and an alternative must be sought in increasing and maintaining high boiler efficiency and in process steam economy.2 ' i

Responsibility for maintaining good operating efficiency lies squarely on the back of the engineer and is attainable provided that the boiler super­visors are carefully selected and properly trained for their work and that adequate control instruments are installed and kept in good working condition.

Efficiency Tests. During 1951 and 1952 boiler efficiency tests were

made at two factories and the results are shown in Appendix 4. The plant at Sezcla consists of six similar boilers manufactured by International Combustion Co. Ltd., and it is believed that these are the first of this firm's construction to be installed in the South African sugar industry. The boilers are of the three drum bent tube construction, some­what similar in arrangement to the Stirling boiler and are equipped with conventional pattern Stewart-Murray furnaces: Usco plate type air heaters are fitted. The boilers are arranged in two batteries of three and from each battery flue gas is discharged to engine driven induced draft fans: individually driven forced draft fans are installed on each boiler. The boilers were installed in pairs at intervals since 1940 and No. 6, the one tested, was installed during 1951 so that the plant may be regarded as a good example of modern practice for the sugar factory. The plant at Illovo considered here consists of three two drum Babcock and Wilcox boilers Nos. 4, 5 and 6 which are of conventional construction and are similar to a number that have worked success­fully for many years in the industry. Boilers 4 and 5 were installed in 1935 and are fitted with Stewart-Murray furnaces; No. 6 which was installed in 1940 is also fitted with Stewart-Murray furnaces and has a much larger combustion chamber than the two former. Two coal fired boilers Nos. 7 and 8 and two multi-tubular boilers Nos. 1 and 2 are also

55

installed. The flue gases are discharged into two Green's economises, Boilers 1-5 discharging into the first and boilers 6-8 into the second: no air heaters are fitted. Draft regulation is attained by means of steam turbine driven forced draft fans fitted on each boiler: an induced draft fan is connected to each economiser outlet.

The tests at Sezela in 1951 yielded efficiencies of 57.5 and 58.6 per cent, respectively and evapor­ation rates less than expected. It was realised that the results had been adversely affected by the removal of a gas baffle from the last boiler pass: this had been done to prevent fine ash deposits accumulating and clinkering on the tubes. Sub­sequently an incorrectly assembled soot blower and tube scale deposits caused by juice leakage were also discovered. These defects were corrected and the tests repeated in 1952 yielded efficiencies of 68.2 and 61.5 per cent, and evaporation rates of 25180 and 25700 lbs./hr. respectively. The highest efficiency, 68.2 per cent., was obtained by restricting the entry of cold air to the furnace through the bagasse feed port, and the least 61.5 per cent. with no precaution taken to restrict this air leakage. During routine operation it would not be practicable to restrict air leakage to the extent attained in the first test, but nevertheless the improvement of 6.7 per cent, corresponding to a bagasse fuel con­sumption of approximately 1650 lbs./hr. emphasizes the effect of the leakage usually tolerated.

In a two-hour overload test intended to establish the maximum continuous rating of the boiler an evaporation of 28,000 lbs./hr. was achieved. The maximum draft pressure drop that could be obtained from the fans during the test was 1.18 in. w.g. If it is assumed that the draft is a function of (total draft pressure drop)i the draft required for a given evaporation may be estimated approximately. This has been done in Appendix 5 where the evapor­ation observed during the tests has been plotted against the corresponding value of (total draft pressure drop)½. By extrapolating the curve to evaporations of 30,000 lbs./hr. and 35,000 lbs./hr. the corresponding draft pressures required for these loads are estimated to be 1.25 and 1.56 in. w.g. respectively, which discloses that the capacity of the fans would have to be increased to attain the higher rating. Air heater performance data are tabulated in Appendix 6 and show that approxi­mately 7.5—8.5 per cent, of the heat liberated was recovered and transferred to the furnace air. The performance of the heater is satisfactory although the temperature of the leaving flue gas is high, 428—474°F. This is caused by the relatively high temperature of the entering flue gas 621—666°F.: under good conditions the latter temperature should be less than this, approximately 550—570°F. to yield a flue gas outlet temperature of approximately

370—380°F. and reduce the sensible heat loss to a minimum. The high temperature of the flue gas also yields a high furnace air temperature 498—528°F. as compared with an expected temper­ature of approximately 400°F. These data disclose that the peak boiler efficiency has not been attained, and that results yielded so far could be improved 2—3 per cent.

During the tests at Illovo the boilers were operated in a routine manner, and no exceptional conditions were imposed. The highest efficiency attained was 68.2 per cent, with an evaporation of 21870 lbs./hr. on No. 6 boiler. A second 8-hour test on the boiler yielded 62.9 per cent, with an evaporation of 25730 lbs./hr. The former result is exceptional and probably due to the relatively steady steam pressure that was maintained during the test and to the lower evaporation. No. 4 boiler yielded an efficiency of 60—60.8 per cent, and No. 5,64.5—65.8 per cent, in successive tests. The lower efficiency on No. 4 is partly due to the higher sensible heat loss and CO loss, and partly to the more pronounced effect of steam pressure drop on draft. Although boilers Nos. 4—5 are similar in construction a total draft pressure drop of 3.12—3.21 in. w.g. is required on No. 4 as compared with 1.87—2 in. w.g. on No. 5 for the same evaporation. Both the I.D. and the F.D. fans are steam driven and consequently steam pressure drop reduces the speed and draft pressure. Since the effect of a given change in speed is greater the larger the draft the effect of steam pressure drop on furnace temperature and efficiency is greater on No. 4 boiler than No. 5. The difference in the performance of the two boilers discloses the importance of regular maintenance of brickwork and air ducts, and the need for a fan drive that is in­dependent of steam pressure changes to sustain peak evaporation rates. The economiser perform­ance is tabulated in Appendix 7. A temperature rise 69—91°F. was attained by the boiler feedwater and 4.1—5.9 per cent, of the heat liberated by the fuel recovered yielding an average improvement in boiler efficiency of 7.6 per cent. The flue gas outlet temperature from the economiser connected to boilers Nos. 4—5, i.e. economiser A is 504— 512°F. and from economiser B 467—483°F. These temperatures are partly determined by the economiser heat transmission area which in this case is only 30 per cent, of the boiler heating surface area. During the test the soot scrapers were not in oper­ation on economiser A which has reduced the heat transmission co-efficient by nearly 28 per cent, in comparison with B, and lowered the efficiency of the boiler by 2—3 per cent.

Flue gas analyses are shown in Appendix 8 and disclose that at Sezela the optimum furnace air supply was established during the test of 19th September, 1952 and yielded a combined sensible

56

heat and CO loss of 17.97 per cent. Under these conditions the CO2 content of the flue gas was 12.9 per cent. Changes in CO2 content above or below this amount increase the flue gas loss: for instance decreasing the CO2 content to 12.4 per cent, changes the loss to 18.42 per cent., an increase of 0.45 per cent. The Illovo results show the need for improving the air supply regulation. On Boilers No. 4—5 the CO2 present is too high owing to in­sufficient air being supplied: the CO present is also too high and a combined heat loss of 23.7—25.2 per cent, is yielded. On the other hand CO2 present on Boiler No. 6 is too small owing to excessive air supply, and causes a larger sensible heat loss 20.6— 21 per cent. These losses could be reduced by regulating the air supply to yield 12—13 per cent. CO2 in the flue gas. The results emphasize the advantage of installing a CO2 recorder instrument to enable the regulation of air supply for improving and maintaining maximum efficiency.

A Heat Balance has been calculated in Appendix 9 to show the distribution of the available heat be­tween that transferred to the water and the various sources of loss. The remaining losses or un­accounted losses consist of radiation, ash, unburnt fuel and leakage. Direct measurement of the latter presents many difficulties: for instance, the unburnt fuel consisting of fine bagasse particles cannot be collected and weighed since it is dis­charged to the atmosphere with the flue gases. Past experience on several tests shows that the undetermined losses may be 10—20 per cent.5

of the lower calorific value of the fuel and that these values may be used for the purpose of comparison. Both 1951 tests at Sezela showed values exceeding 20 per cent, and are high because the absence of the gas baffle would allow larger quantities of fine unburnt bagasse to escape into the flues more easily and avoid ignition. This view is supported by the fact that before the baffle was originally removed clinkering occurred at the last tube pass showing that combustion was proceeding there. In the 1952 tests the remaining losses were 13.38 and 20.85 per cent, respectively. The lower result was attained under conditions where cold air leakage to the furnace was restricted and it is con­sidered that the increased loss in the second test is due mainly to the effect of air leakage lowering the furnace air temperature and causing increased unburnt fuel losses.

Summary.

The prospect of utilising bagasse as a raw material for chemical processing and the disadvantage of burning it as a fuel are discussed. Data is pre­sented to show that the seasonal average fibre content of all cane crushed in Natal since 1948-49 has remained sensibly constant: the fuel quality

of the bagasse has deteriorated continuously owing to increased moisture in bagasse which is shown to be due to crushing increasing proportions of N :Co.310 cane. The effect on steam production is discussed and it is suggested that the improvement and maintenance of boiler efficiency and the economic use of steam is needed to meet the problem. The results of boiler efficiency tests made at two factories are given and the source of the losses examined. Recommendations are made for installing flue gas analysis recorders to assist the boiler operators in working at maximum efficiency.

REFERENCES. 1Hedley, E.P.: Boilers, Furnaces and Boiler Equipment'

Proc. S.A.S.T.A. 11, 6-18. 2Perk Chs., G.M.: Exhaust Steam Production and Consump­

tion, S.A. Sugar Jl, 1953, 37,23. 3Perk Chs., G.M.: Reducing Injection Water Requirements,

S.A. Sugar Jl, 1952, 12, 763. 4Comm. Rept.: Boiler Efficiency and Heat Balance in the Cane

Sugar Factory, Proc. S.A.S.T.A., 1928, 77-84. 5Archief v.d. J. S. I.: 1932, pp. 1341-1448.

Acknowledgments.

This Institute wishes to thank the Managements and Staffs of the factories where the boiler tests were made for the assistance and facilities provided during the work. The Institute also wishes to thank Messrs. Patrick Murray & Co. for their collaboration during the test at Sezela.

Mr. Bentley said that Mr. McCulloch was to be congratulated on presenting his paper on this most interesting subject in so clear a manner; they at Maidstone would be pleased if he could find the time during the coming season to carry out similar tests on some of their boilers.

One point in Mr. McCulloch's paper with which he did not agree and which he felt was not entirely supported by the evidence of last year's results was that there had been a deterioration in the quality of bagasse due to the greater percentage of N:Co. 310 crushed. Mr. McCulloch used the argu­ment that moisture per cent bagasse had increased in line with the increase of N:Co. 310 crushed and produced average industry figures to prove it. He felt that a general decrease in extraction not necessarily due to increased throughput of N :Co. 310 was responsible for the increase moisture per cent, bagasse. Had Mr. McCulloch taken his figures from results of those factories such as Gledhow and Maistone where extraction had not dropped, he would have found that moisture in bagasse was, if anything, lower than in previous seasons in spite of a considerable increase in the percentage N :Co. 310 crushed.

57

He drew attention to Mr. McCulloch's statement that "Responsibility for maintaining good operating efficiency lies squarely on the back of the engineer and is attainable provided that the boiler super­visors are carefully selected and properly trained for their work and that adequate control instruments are installed and kept in good working condition." He felt that this subject deserved emphasis. In

many boiler installations in the industry there was a complete absence of reliable instruments and he felt sure that proper instrumentation intelligently used would result in improved boiler efficiency thereby saving fuel and recovering the cost of the instruments in a comparatively short time.

Mr. Dymond asked that a hearty vote of thanks be accorded Mr. McCulloch.

58

APPENDIX I.

Variation of Fibre % Cane, Moisture % Bagasse and N :Co.310 Cane for the Past Five Seasons.

APPENDIX 2.

Comparative Results Obtained from Milling N :Co.310.

59

60

APPENDIX 4.

Test Data and Thermal Efficiencies of Boilers Tested.

61

APPENDIX 6.

Air Heater Performance.

62

APPENDIX 9.

Summary of Heat Balances.

SCALING OF EVAPORATOR TUBES OF NATAL SUGAR

FACTORIES Report on investigations carried out in 1951 and 1952

By K. DOUWES DEKKER.

Introduction.

We are accustomed to regard scaling of evaporator tubes as a natural phenomenon which is largely outside our control. We know that scaling is heavy in some years and in others considerably less. We have a feeling that scaling tends to be more serious in factory A than in factory B, but quantitative data to confirm this feeling are usually lacking. We are so familiar with the occurrence of the phenomenon that we are even likely sometimes to underestimate the detrimental effects on processing.

These effects, however, are not slight. Not only does scaling force us to clean the evaporators weekly, but it seems also to be one of the causes of the undesirably low brix of syrup which is typical for Natal factories.

Studying scaling in the hope of finding means to reduce the rate at which the scale is deposited, or to alleviate its specific effect on heat transfer, is a rather thankless task, and has to be carried out in the firm belief that all bits of available knowledge, when systematically put together, will ultimately provide a clear picture which will allow us to control the phenomenon at our will. But we should not expect an easy solution of the problem which can be had for the taking.

In our plans for collecting more knowledge, how­ever, we should be guided by a few well recognised principles and we should at all times be able to draw up a picture of the main facts which are known without a doubt, and of the questions which cannot definitely be answered: and in particular we should be able to recognise half-truths, the acceptance of which as full truths, may lead our work in the wrong direction.

I think we are safe in stating that scaling depends on two groups of factors: (a) juice properties, and {b) particulars of the evaporator plant, including the way it is operated. The importance of juice proper­ties is immediately evident when we remember that carbonatation factories suffer far less from scaling than sulphitation and defecation factories, and that the normal rate of scaling of carbonatation factories can be reduced to an insignificant value by the application of ion-exchange. We must conclude that there is a definite correlation between the presence of certain components of the juice and the formation of the scale.

About the effect of particulars of the evaporator plant on the amount of scale deposited we have little definite information. Does an oversized quad scale more heavily than a correctly sized one, and what about the rate of circulation inside the vessels ?

In connection with the way the quad is operated the most important item is probably the brix to which the juice is concentrated. The higher the brix of the syrup, the heavier the scaling, and it has also been stated that working a quad intermittently, i.e. shutting it down for a few hours when Jess juice is produced, favours scaling.

In 1951 the S.M.R.I. started to collect data on the composition of the scale of some of the mills and some information was also gained on the amount of scale deposited in one week in the evaporator tubes of one mill (TS). The results of these tests were discussed in Communication from the S.M.R.I. No. 14, but they will also be shown here in so far as is necessary for a proper understanding of our con­clusions. In 1952 the investigation was continued and it is envisaged that in coming years more work will be done on the subject.

The Chemical Composition of Natal Scales.

It is more or less ordinary routine for a chemist who wants to study scaling to ask what are the constituents of scale, but he should be quite aware of the limits of the information to be derived from analysing scale samples. Analytical data of scale samples can tell us:

(a) what components of the juice are not, or to a very slight extent only, present in scales, and are therefore probably not instrumental in scale formation;

(b) what are the main constituents of the scale. From this knowledge we may infer what com­ponents of the juice should be suspected of being most active in the scale forming phenomenon. We should, however, be careful, for knowledge of the amount of scale deposited is usually also required to draw definite conclusions;

(c) what chemicals are likely to be effective in cleaning operations.

They can also be used to find the correlation, if any, between the composition of the scale and its specific effect on heat transfer.

63

64

Since it is general experience—which was quanti­tatively confirmed by data found at TS mill in 1951 (Table 1)—that the rate of scaling rapidly increases from the first to the last vessel of the set, it was decided to study first the scaling of the last and penultimate vessels.

TABLE 1. Average Weights in Grams of Dry Scale deposited in Six Days per square foot of H.S. in the Tubes of the Tongaat McNeil and

Hagemann Quadruples. No. 1 No.2 No. 3 No. 4 vessel vessel vessel vessel

McNeil 0.12 1.68 3.68 10.5 Hagemann ... 0.32 2.56 4.24 10.3

Accordingly for 1952 an investigation was planned in which the mills were requested to collect quanti­tatively the scale formed, during one week in the middle of the season, in 15 tubes evenly distributed over the tube plate of No. 3 and No. 4 vessels respectively. The scale was to be dried, weighed and forwarded to the S.M.R.T. for analysis. It was requested that representative composite samples of mixed juice, clarified juice and syrup be collected simultaneously, and also some particulars concerning processing operations of that week.

The mills ZM, FX, AK, MV, TS and SZ responded to our request. Particulars of their evaporators are shown in Table 2.

1 The calandria of the first vessel is divided into two equal sections both of which are heated by exhaust steam. The vapour from the first vessel heats the following vessel, and so on, in the normal way. The brix of the syrup was kept more or less

constant throughout the week and averaged 52.9°. A "Superstat" was in use on the clarified juice between the Bach clarifier and the first vessel and operated on 0.85 amperes. The heating surface of No. 1 vessel is 10,000 sq. ft., of each of the following vessels 6,000 sq. ft. Brix ex first vessel was 17.9°.

2 The evaporator is reported by the factory to consist of two pre-evaporators connected in series to two triple effects in parallel. The heating surfaces of the pre-evaporator are 8,400 and 7,000 sq. ft., of the triple effects 6,500 and 5,900 sq. ft. respectively. In general the brix is kept down at the beginning of the week. Vapour is used for heating purposes. The tubes of No. 2 triple effect are slightly larger than those of No. 1, which was sampled. Average brix of the syrup during the week of the tests was 49.2°.

3 The four identical vessels have D calandria. The quad is working at a vacuum of 24 to 26 inches. The syrup standard density is set at 27° B€ (50°Bx) and the average daily brix values during the week of the tests were:

Monday 52.9° Thursday 55.4° Tuesday 50.9° Friday 53.4° Wednesday 54.4° Saturday 51.5°

4 The Melville Quadruple Effect Evaporator con­sists of four identical vessels. The type of calandria is the side D-downtake. Only exhaust steam is used for heating purposes. It is understood to keep the brix at 52° in the beginning of the week. The average brix of the syrup during the week of the tests was 51.0°.

5 The method of heating of the two quadruple effects is normal, with exhaust steam being supplied to the calandria of No. 1 vessel. It is not possible to give the amount of juice handled by each evaporator separately, but the Hagemann was shut down for 37 hours during the week while the McNeil worked continuously for the whole week, i.e. 142 hours. Average daily brix values during the week of the tests were:

Monday 49.8° Thursday 49.2° Tuesday 50.8° Friday 49.7° Wednesday 48.9° Saturday 49.2°

6 There is a third quadruple effect at SZ similar to No. 2 quad which was not operated during the week of the tests. No. 3 quad was operated con­tinuously, No. 2 quad was reported to be shut down for about 25 per cent, of the time of the first week of the tests.

Brix syrup during the three weeks of the tests was: 54.6°, 54.5° and 55.1° respectively.

No vapour is used for heating purposes. The syrup brix on Saturdays is raised slightly (28° Bé) to improve preservation during week-ends.

66

TABLE 5

TOTAL WEIGHTS, AND WEIGHTS PER SQUARE FOOT H.S. OF SCALE DEPOSITED IN ONE WEEK IN No. 3 and No. 4 VESSELS RESPECTIVELY

Note.—The total weight figures for ZM refer to the last and penultimate vessels ; those for FX to the last and penultimate vessels of No. 1 triple effect ; those for TS to the McNeil quad.

The analytical data of the scale samples are shown in Tables 3 and 4. Data of samples collected in 1951 are also included. At SZ, in connection with a special investigation which will be discussed later on, the 1952 investigation covered a period of three weeks, and samples were collected at each of the three week-ends. They are denoted as (1), (2) and (3) respectively.

The weights of scale deposited.

The weights of the scale deposited in one week (six working days) were determined in 1952 only, by collecting the scale from fifteen tubes which had been properly cleaned the previous week-end. Collecting the samples was carried out by the mills' staff. The total weights were calculated, and also the weights of the scale deposited per square foot of heating surface, for the calculation of which the inner

diameter of the tubes was used. The weights are shown in Table 5.

Composition of mixed and clarified juices.

During the week which preceded the taking of the scale samples both mixed and clarified juice were sampled. The composite samples were disinfected with formalin and forwarded to the S.M.R.I, for analysis.

The factory purity figures for the week are re­ported in Tables 6 and 7 and also the percentages of the inorganic constituents, expressed per 100 brix in order to allow for the dilution due to the addition of formalin.

A sample of syrup was also taken for the deter­mination of the amount of suspended solids. This figure was required as a basis with which to compare the quantity of scale deposited.

TABLE 6

MIXED JUICE. SULPHATED ASH AND ASH COMPONENTS ARE EXPRESSED AS PER CENTS. ON REFRACTOMETER BR X

67

TABLE 7 CLARIFIED JUICE. SULPHATED ASH AND ASH COMPONENTS ARE EXPRESSED AS PER CENTS. ON REFRACTOMETER BRIX, SUSPENDED SOLIDS IN SYRUP AS PER CENTS. ON SYRUP

p Unfortunately the juice samples of the second week of the test were thrown away in error. In Table 7 the syrup figures are substituted for the clarified juice figures.

Discussion.

There is a theory that when a bubble of vapour is formed at the surface of a tube the solids contained in the drop of juice which is evaporated will be deposited on that part of the tube where the bubble is formed. However, since juice will stream past the deposit immediately after the bubble is removed from its point of formation, it is difficult to see why it should not dissolve from the deposit those com­ponents in respect of which the juice is under-saturated. Or in other words only those components in respect of which the juice is supersaturated, or perhaps also, nearly saturated, are likely to remain in the scale.

Whether this theory is quite correct we do not know, but it explains satisfactorily that scaling, in respect of the dissolved solids in juice, is a selective phenomenon, and it is conceivable that the selective action has something to do with the solubility of the constituents. Sucrose and reducing sugars are not present in scale, neither are alkali salts, and even magnesium salts whose percentage in clarified juice is of the same order as the percentage of lime salts, are usually absent, or present to only a slight extent. When studying scaling we seem to be allowed to disregard in the first instance the presence of this type of juice component.

A glance at Table 4 shows that (with one exception —ZM) the most important group of components of No. 4 vessel scales are the lime salts, and that calcium is mainly present as sulphate or sulphite. Lime salts formed more than half of the scales which were analysed, and AK shows the maximum amount of CaS04 plus CaS03, viz. 87.5 per cent.

As to the nature of the lime salts it is interesting to quote a report from Imperial Chemical Industries

Ltd., where it was found by X-ray analysis that part of the calcium of Darnall scale was present as gypsum (CaSO4.2H2O) and part as an unknown substance, maybe a mixed crystal of calcium sulphite and calcium sulphate, possibly hydrated.

At ZM, FX, SZ and RN there is more CaSO4 in the scale, at other mills (EN, DL, MV, TS) sulphite is predominant. The amount of CaO not present as sulphate or sulphite is insignificant. It is impossible to offer at this stage an explanation for the fact that the ratio of sulphate to sulphite varies so much for mills working on similar lines, although we tried to correlate this ration to the SO2 content of clarified juice.

The second most important group of components is the group which in analytical routine is described as SiO2 plus matter insoluble in HCl. The highest figure is shown by SZ in the 1951 scale from No. 3 quad, viz. 29.3 per cent. The high figure for MV (22.9 per cent.) is explained by fairly big sand particles embedded in the scale.

The third group of importance consists of the components described as "Organic matter, water in excess of moisture expelled at 105°C, etc." The figures for this group make it clear that organic matter, although scaling seems to be largely an inorganic phenomenon, cannot completely be dis­regarded.

Sesquioxides and P 2 0 5 are present in such small quantities that their effect on the scaling of No. 4 vessels is very probably insignificant.1

The scale of No. 3 vessels usually contains more Si02 , etc., and organic matter, etc., than the corres­ponding No. 4 vessel scales. Calcium sulphite and sulphate are present to a lesser extent and although

1 Scale of first and second bodies contain relatively more P205.

CaO "bound otherwise" is sometimes higher, total CaO is usually lower.

The percentages of sesquioxides and P 2 0 5 tend to be somewhat higher in No. 3 vessel scale than in No. 4 vessel scale, but are quantitatively of little importance. If No. 4 vessel scale contains relatively more sulphate than sulphite, the same can be ob-served in the No. 3 vessel scale, and vice versa.

In some of the scales Cu was found to be present, in a certain case to some extent. The brass filings which could easily be observed in the samples, show that mechanical scrapers do not always remove scale only.

(a) 1952 Tests at SZ.

In 1951 our attention was drawn to the fact that in particular SZ-scale was rich in Si02 and since Bogtstra1 has indicated that the percentage of the silica present in clarified juice which is deposited in the evaporator set diminishes rapidly at higher P 2 0 5

content of the clarified juice, it was decided to carry out an investigation to test a practical implication of Bogtstra's statement. SZ kindly agreed to double the normal amount of phosphoric acid paste in the second week of a three-weekly period, in the first and third week of which the phosphoric dosage would be normal.

It was hoped by increasing the phosphoric dosage to increase the P 2O 5content of clarified juice by which the amount of Si02 deposited should be reduced. SZ follows the normal Natal clarification method of pre-liming, gassing and tempering, the phosphoric paste being admixed in the tempering tanks.

The weights of chemical agents used in the three weeks of the test were:

In Table 8 arc shown the weights of scale and of silica in scale deposited in the tubes of the No. 3 and No. 4 vessels of both quads. Also the weights of Si02 in clarified juice having passed the quads in each week, and the percentages of these weights deposited in scale.

The main conclusion from Table 8 is that doubling the dose of phosphoric has not reduced the total weight of Si02 , etc., deposited in the scale of No. 3 and No. 4 vessels of both quads. On the contrary, in the second week of the test the weight was highest.

According to Bogtstra the weight should have been reduced if doubling the phosphoric paste dose had resulted in a higher P 20 5 content of the clarified juice. This, however, was hardly the case. P 2 0 5 in clarified juice in the three weeks of the test averaged 0.015, 0.017 and 0.015 per cent. respectively. The increase in the second week is insufficient to antici-pate any notable effect on the rate of scaling and the main result of the test is that it seems not permissible to give a practical interpretation to Bogtstra's con-clusion by inferring that an increased dosage of phosphoric paste will reduce the weight of silica deposited in No. 3 plus No. 4 vessels, expressed as a percentage of the total weight of SiO 2 in clarified juice concentrated in the period under observation.

Bogtstra has not specified the nature of the P 2 0 5

to which he related the SiO2 precipitation. Since P 20 5 may be present in the organic and the inorganic form, it is not impossible that later on an explanation may be found on such differentiation.

68

lime sulphur ... phosphoric

paste ...

1st week

37 16.9

5.1

Tons

2nd week

40 18.0

11.1

3rd week

42 18.9

6

lbs. brix

1st week

23.3 10.6

3.21

per 1,000 lbs. in mixed juice

2nd 3rd week keek.

23.3 23.2 10.5 10.5

6.45 3.32

TABLE 9 RATE OF SCALING EXPRESSED AS LBS. OF SCALE DEPOSITED FOR EACH TON OF WATER EVAPORATED

1 Assuming that each of the effects evaporated half of the water. . 2 Assuming that the McNeil effect evaporated 24/39ths. parts of the water. 3 Sum of the weights of scale deposited in corresponding vessels.

On the whole the data of Table 8 are somewhat irregular. No. 4 vessels scaled more heavily than No. 3 vessels. No. 3 vessels of the large quad scaled considerably less than No. 3 vessel of the small quad. The heavy scaling of the small quad in the first week is probably due to that quad being shut down for 25 per cent. of the time, generally only for a few hours at a time, when the juice could be allowed to stand in the various vessels.

Finally it is interesting to note that the increased phosphoric paste dosage resulted—according to SZ —in a better quality white sugar, which paid for the

extra expenditure due to this increased dosage.

(b) The rate of scaling.

There is some confusion in literature about the correct way to express the rate of scaling. From a practical point of view one would be most interested in the reduction of the heat transfer coefficients or the various vessels per ton of water evaporated per square foot H.S., but failing data on heat transfer, and mainly considering the chemical aspects of the problem, the appropriate figure seems to be the weight of scale deposited in a vessel per ton of water evaporated by the whole set. In cases where factories operate two or more parallel quads, to calculate the rate of scaling the volume of juice which passed each quad should be known, or else the weights of the scale in the corresponding vessels of each quad should be added. Table 9 shows the data concerning the factories which participated in the 1952 test.

The most striking feature of Table 9 is the slight rate of scaling, which perhaps is even more pro-nounced if the rate of scaling is expressed as parts of scale per million parts of water evaporated—see Table 10.

It is interesting to compare the weights of scale deposited in No. 3 and No. 4 vessels with the weights of suspended matter in syrup. To obtain these last figures samples of syrup were centrifuged and the deposits washed with diluted alcohol and dried. Unfortunately it was not possible to determine the percentage directly after the syrup sample had been taken, so some material may have been precipitated during the period before the sample was analysed.

The weight of suspended matter in syrup is of course not necessarily a yardstick for the total amount of matter which became supersaturated and consequently was precipitated during the concen-tration process, for the clarified juice may not have been completely clear. In spite of these limitations, however, suspended matter in syrup may be re-garded as matter which could well have been de-posited on the tube walls during the concentration process. The percentages which were actually deposited are shown in Table 11.

Conclusion from Tables 9 to 11 are:

(i) it is tentatively suggested that normal figures for the rate of scaling of evaporators in Natal sugar factories are:

No. 3 vessel—5-15 parts of scale deposited per million parts of water totally evaporated;

No. 4 vessel—-20-60 parts of scale deposited per million parts of water totally evaporated.

(ii) since the weight of scale deposited in No. 3 plus No. 4 vessels amounts usually to less than 10 per cent. of the amount of suspended solids in the syrup produced in the period the scale was formed, it is not expected that the amount of suspended matter in syrup will materially be increased when—

TABLE 10

RATE OF SCALING EXPRESSED AS PARTS (mg) OF SCALE DEPOSITED FOR EACH MILLION PARTS (kg) OF WATER EVAPORATED

ZM FX AK MV TS SZ(1)

No. 3 vessel No. 4 vessel No. 3 + No. 4 vessel ..

1.6 1.9 3.5

41 97

138

4.5 44 48

19 69 88

17 24 40

2.5 22 25

1.0 36 37

1.5 39 40

SZ(2) SZ (3)

69

ZM FX 1 AK MV TS2 SZ3 (1) SZ (2) SZ (3)

No. 3 vessel No. 4 vessel No. 3 + No. 4 vessel ..

. 0.003

. 0.004

. 0.007

0.083 0.193 0.276

0.009 0.087 0.095

0.038 0.137 0.175

0.033 0.048 0.080

0.005 0.044 0.050

0.002 0.073 0.075

0.003 0.077 0.080

PERCENTAGES OF SUSPENDED MATTER IN SYRUP WHICH WERE DEPOSITED IN No. 3 AND No. 4 VESSELS OF THE EXAMINED EFFECTS

by some at present unknown means—the formation of scale in the evaporator is prevented.

(iii) of particular interest for a better under-standing of the mechanism of scaling would be an explanation of the low rate of scaling of No. 3 vessels at SZ and of No. 3 and No. 4 vessels at ZM, and of the excessive scaling at FX.

It was further felt important to compare the com-position of the suspended matter in syrup with the composition of the scales. An analysis of the sus-pended matter isolated from ZM syrup revealed the following composition:

Ignition loss Ash Pure silica Matter insol. in HC1 ... Calcium oxide Phosphorus pentoxide... Sulphate

59.6 40.4 28.0 29.2

9.6 0.21

present

-100

69.3 72.4 23.8 0.53

present There is more similarity between No. 3 vessel scale of ZM and the suspended matter in syrup than

between this latter component and No. 4 vessel scale which consists of 50-65 per cent. of CaS04 and CaSO3. This fact opens new lines of thought on the formation of sulphite-sulphate scale at higher con-centrations which make further investigations de-sirable.

(c) Rate of scaling and juice composition. Since neither the brix values of clarified juice nor

the brix values of syrup differ substantially amongst the various mills one would expect that a certain correlation would exist between the rate at which lime salts are deposited as scale and the lime salts content of clarified juice.

Table 12 shows mg CaO per litre clarified juice and the rate of formation of lime salts in scale (expressed as mg CaO in scale per kg water evapor-ated).

In the same way Table 13 shows the rate of formation of Si02 in scale as a function of the Si02

content of the clarified juice which is concentrated.

PARTS OF CaO IN SCALE OF No. 3 PLUS No. 4 VESSEL DEPOSITED PER MILLION PARTS OF WATER EVAPORATED, AND mg/1 CaO IN CLARIFIED JUICE

TABLE 13 PARTS OF SiO2 IN SCALE OF No. 3 PLUS No. 4 VESSEL DEPOSITED PER MILLION PARTS OF WATER EVAPORATED, AND mg/l Si02 IN CLARIFIED JUICE

Tons of suspended matter in syrup

Scale in No. 3 plus No. 4 vessel % susp. matter

ZM

21.1

1.6

FX

12.2

11.4

AK

9.79

4.6

MV

4.03

8.5

TS

11.6

5.4

S Z ( 1 )

10.9

2.1

SZ(2)

7.17

4.8

SZ(3)

6.98

5.6

70

TABLE 11

TABLE 12

mg/ l CaO in clarified juice Rate of scaling of CaO in

3rd plus 4th vessel ...

ZM

584

0.79

AK

581

17.5

FX

540

36.5

TS

503

13.1

MV

442

26.0

SZ(1)

348

5.79

SZ (2)

347

8.23

SZ(3)

332

9.84

mg/ l Si02 in clarified juice Rate of scaling of SiO2 in

3rd plus 4th vessel ...

ZM

210

1.00

AK

183

3.60

SZ(3)

174

6.85

MV

173

17.8

FX

170

21.4

TS

168

4.51

SZ(2)

145

1.56

SZ(1)

139

5.30

71

TABLE 14

RATE OF SCALING AND EVAPORATOR PERFORMANCE

SZ(3)

Tons of water evaporated per sq. ft. H.S. per week 0.422

Rate of scaling 5.6

MV

0.410 8.5

SZ(2)

0.400 4.8

The conclusion to be drawn from Tables 12 and

TS

0.398 5.4

S Z ( 1 )

0.392 2.1

AK

0.371 4.6

Conclusions.

FX

0.363 11.4

ZM

0.282 1.6

13 is that, although there obviously must be some relationship between the rate of deposition of CaO and SiO2 in the scales of the 3rd plus 4th vessels (expressed as mg per kg of water evaporated) and the content of CaO and SiO2 respectively in the clarified juice which is evaporated, this correlation seems to be obscured by other factors which to a great extent tend to affect the rate of scaling.

Bogtstra suggested that one such factor was the P 2 0 5 content of the clarified juice. Our data do not confirm this suggestion.

When the analytical data for FX clarified juice are compared with those of the other mills, it is obvious that they do not offer an explanation for the heavy scaling of the FX tubes. This heavy scaling must be due either to a juice property not shown by the normal analytical data, or to a par-ticular feature of the evaporator.

(d) Rate of scaling and the performance of the evaporator.

It has sometimes been suggested that an over-sized quadruple will scale more seriously than a correctly sized one. To test this assumption the total weight of evaporated water was divided by the total heating surface of the evaporator, and the quotient, expressed as tons per square foot, compared with the rate of scaling. Table 14 shows the results.

A definite relationship between the performance of the evaporator and the rate of scaling is not shown in Table 14.

(e) The low rale of scaling at ZM. The third conclusion based on the data of Tables

9 to 11 refers a.o. to the low rate of scaling of No. 3 and No. 4 vessels of ZM. The ZM evaporator scaled considerably less than the other sets and one is tempted to relate this feature to the Superstat being installed in this factory.

Scaling, however, is such an unpredictable phe-nomenon that the low rate of scaling may have been due to some other cause. Unfortunately it was not found possible to carry out a special investigation into the effect of the Superstat. It is, however, intended to study this effect in season 1953.

1. The most striking result of an investigation carried out in 1951 and 1952 into the scaling of Natal evaporators seems to be, in spite of the fact that scaling in Natal is considered to be serious, that the rate of scaling when expressed as parts of scale deposited in No. 3 and No. 4 vessels per million parts of water evaporated by the set, is shown as a small figure. Normal figures for No. 3 vessels were found to be 5-15 parts, for No. 4 vessels 20-60 parts. Only a small percentage of the amount of suspended matter in syrup adheres apparently permanently to the wall of the tubes, forming what is usually called a scale.

2. The main constituents of No. 4 vessel scale are lime salts, generally sulphites and sulphates. Silica is also important, and the third group is denoted as "Organic matter, water not expelled at 105°C, etc." In No. 3 vessel scale the two last groups are relatively more important than in No. 4 vessel scales; lime salts may even be present to only a small extent.

3. An investigation carried out at SZ in which it was tried, following a conclusion mentioned in earlier work of Bogtstra, to reduce the amount of silica deposited by doubling the dosage of phosphoric paste, gave a negative result. Doubling the dosage of phosphoric paste did not increase the amount of P 2 0 5 in clarified juice and did not reduce the quantity of SiO2 deposited.

4. Scaling at ZM factory which has a Superstat installed was significantly less than scahng at other factories, but whether it is permitted to connect these factors still has to be investigated.

5. One cannot but feel when trying to form a comprehensive picture of the scaling phenomenon in Natal that a better insight cannot be gained by studying only the relative rates of scaling of factories as functions of the composition of the clarified juices processed, because the differences in these compositions are relatively small and are probably not the main cause of the wide fluctuations of the rates of scaling which were observed. Studying scaling in a specially con-structed test apparatus where conditions are kept strictly constant may reveal more of the chemical

72

and physical laws to which scaling is subject, but a drawback here is also that an enormous quantity of water has to be evaporated to produce sufficient scale to be weighed and analysed. Nevertheless it is hoped to continue our investigation in this direction.

When doing this, it will probably be found neces-sary to see the scaling phenomenon as a process which is quite distinct from the fact that some of the constituents of the juice become supersaturated and insoluble by the concentration of the juice.

Acknowledgment. The analysis of juice and scale samples was carried

our by Dr. J. R. Parrish and Mr. J. B. Alexander of the staff of the Institute.

The S.M.R.I. wishes to thank the management and staff of the factories which collected samples and forwarded details of processing methods, thus enabling us to carry out this investigation.

Sugar Milling Research Institute, Durban.

Mr. Dymond said that the subject was one of considerable interest and some years ago, as a result of a resolution taken at Congress, it was decided to carry out investigations. Dr. Douwes Dekker was asked to carry these out. He congratulated Dr. Douwes Dekker on his paper and on the results of his investigations.

Dr. Parrish asked about the experiment with phosphoric paste at the Sezela factory. He said that although the amount of paste had been doubled, the amount of lime remained the same as in the first and third week. Did this mean that the juice had finished up in an acid state?

Dr. Douwes Dekker said the juice must have been a bit more acid. Perhaps the effect was not too great as the amount of phosphoric paste used was relatively small.

Mr. Galbrailh, in reply to Dr. Parrish, stated that the pH of the clarified juice and syrup was kept normal throughout the three weeks' test. The actual syrup pH values for each week were 6.79, 6.90 and 6.84 respectively. He queried the lime figures given for Sezela in the paper, as there were no means of obtaining an accurate lime consump-tion figure weekly at present.

Dr. Douwes Dekker said the lime figures had been provided by Sezela. If their accuracy was doubted by Mr. Galbraith a note could be added to his paper to this effect.

Mr. McCulloch said he thought a better insight of scaling in the tube could be obtained if samples were taken along the whole length of the tube. If it was contended that scale was formed as a

result of bubble formation it might be expected that more scale would be formed at the bottom of the tube.

Dr. Douwes Dekker said that the possibility that scale was not homogenous had been considered before. As far as he knew the composition of the scale did not vary with the height, but it had been found that sometimes several layers could he discerned and that these layers were of different composition, in particular if the nature of the juice at the beginning differed from that at the end of the period in which the scale was formed.

Mr. Rault congratulated Dr. Douwes Dekker and his staff on their fact-finding work on the subject of evaporator incrustation in South Africa. Although the rate of evaporation (8.5 to 9 lbs. per sq. ft. per hour) at his factory could be considered high, mainly due to the handling of brilliantly filtered juices from the carbonatation process, the heating surface was not free of scale and had to be cleared every week-end, neither did the soluble lime salts reach the low level found in beet carbonatation, clarified juices or even cane carbonatation juices mentioned by Honing in Spencer Meade's book.

Lately the lime salts had been reduced by the systematic use of soda ash, with remarkable lowering of the ash content of the final refined sugar. Natal Estates laboratory had for years determined as a routine test the amount of lime salts left in the juice going to the evaporator. Tn the past they had used the "soap test", which demands a certain amount of judgment and experience from the analyst. It had now been replaced by the quicker and more accurate versanate titration. He hoped that all factories would now follow suit for next season.

The results of reduced scaling at Z.S.M. were not convincing as this factory showed the lowest evaporation per sq. ft. heating surface, a factor which would tend to throw down less mineral matter. The ultimate object of this research was to find conditions that would lead to our factories running for two to three weeks without having to clean the evaporators. So far he did not know of any factory that was bold enough to try this.

Mr. Ash asked whether the water added to the cane during crushing had anything to do with scale formation. He suggested that some factories obtaining good results might be using sweet water.

Dr. Douwes Dekker said he had tried to answer the question regarding the correlation of water used in the milling processes and the rate of scaling. He was not able to find any correlation. He was not prepared to say whether the quality of inhibition water had anything to do with the rate of scaling.

Dr. Dekker said he wanted to refer to the observ-ation made by Mr. Bentley during Mr. McCulloch's

73

paper when he stressed the difficulties raised with boiler water in Natal. Mr. Bentley had referred to the lack of instrumentation. The N.S.M.A. had appointed an instrument maker who was stationed at the S.M.R.I. and whose sole job was to keep the instruments in proper working condition. Mr. Viljoen would in future be able to look after instru-ments at the mills, so that the need for better instrumentation referred to by Mr. Bentley would fall away.

Mr. Dymond suggested that the higher the amount of lime used in the clarification the higher the amount found in the juice. He asked whether this had any connection with an increased form­ation of scale.

Dr. Douwes Dekker said that the correlation between the amount of lime salts and the rate of scaling, although obviously existing, was—at least in Natal mills—obscured by other factors affecting the rate of scaling. He was, however, not prepared to confirm Mr. Dymond's statement that using more lime inevitably resulted in a higher lime salts content of clarified juice.

Mr. Main asked whether Dr. Douwes Dekker could express an opinion on the relative methods of the sulphur burning processes used in Java compared with the normal Natal processes. He mentioned that if wet air entered the sulphur burners trouble could be experienced with scale, the inference being that dry air should be used.

Dr. Douwes Dekker said that he agreed with Mr. Main that sulphitation as carried out in many countries where not sulphur towers, but sulphit­ation tanks, were used had definite advantages, in particular with respect to a better control of pH during the clarification process. Tank sulphitation required the closed type of sulphur burners and the

air blown into these burners was usually dried with unslaked lime. The efficiency of the drying, how-ever, was doubtful and in many cases no difference in the clarification effect had been observed when the drying chambers were cut out altogether. He did not know of any case where cutting out driers had had any effect on the rate of scaling.

Mr. Galbraith asked whether the conditions in which sulphur was burned had any effect on scaling. There was a variation in opinion on this point, as he had noticed that while some factories made no attempt to cool the gases before entering the tower, others went to a lot of trouble to ensure the lowest temperature possible. He thought there must be some difference caused by the temperature of the gases entering the juices.

Dr. Douwes Dekker said he was not in favour of very hot burner gases being drawn into the sulphur tower.

Mr. Wheeler said he thought he might answer Mr. Main's question. He was used to drying the air before he came to South Africa, He had fitted a drier but had found that it made no difference to scaling. What, he had noticed was that if the baffles in the sulphur tower were heavily coated they had a cleaner evaporator.

Mr. Perk said the juices of Amatikulu were darker than those of Felixton and the view was expressed that this was caused because at Amatikulu they did not cool the gas.

Mr. Dymond said that at Umbogintwini they burnt 50 tons of sulphur a day and cooling it was quite a simple procedure. In making acqueous SO2

it was necessary to cool the gases. He asked the meeting to accord Dr. Douwes Dekker a hearty vote of thanks for his interesting paper.

74

SOME NOTES ON AUTOMATIC CONTROL IN SOUTH AFRICAN SUGAR FACTORIES

By H. ROHLOFF.

A comparison between the plants of some vital South African industries such as Sugar Industry, the Iron and Steel Industry, the Chemical, Pulp and Paper Industries as well as Power Stations will single out the South African Sugar Factories for one very noticeable item: The absence of super-vising and controlling instruments for the manu-facturing process and for a number of ancillary processes. By stating this, I want to exclude the power-generating plant such as the Boiler House and electric generators which are almost auto-matically fitted with a number of essential instru-ments.

The almost complete absence of equipment continually supervising and controlling the pro-cessing of sugar is astounding for several reasons: However complex this process may be, it can be reduced to some comparatively simple measure-ments and manipulations which vary for a number of reasons in degree but not in principle. I shall analyze these measurements and manipulations at a later stage.

In the meantime, let me try to assess the. reasons for the comparative lack of instrumentation:

One school of thought claims: "Why should we instal expensive equipment such as instruments and automatic controllers? We have trained staff who do their supervisory work well, who produce fair results in our factory and whom we cannot replace by instruments or machines in any case."

Although extensive experience from all quarters of the processing industry is available, it is often considered unnecessary, too expensive and incon-venient to exploit the experience of others for their own factory.

A very important factor, especially in the South African Sugar Industry is that sugar has been pro-duced quite profitably without any of the modern means of supervising the production process.

It has frequently been argued that instrument-ation must pay for itself and prove its value not by improving manufacturing methods but by cutting losses. Whereas it can safely be said that losses will be cut by suitable supervision and control, it must, at the same time, be understood that these very losses will become apparent only after instru-ments have been installed and that they can only be eliminated after they have been determined.

I would like to quote Llewellyn Young from an article in the South African Electrical Review: 3:

"Building up an automatic control system involves the selection and arrangement of the proper controllers, relays, regulating units and selector stations, and their application in such a manner that the theoretical and practical re-quirements are properly met. Careful study of the proposed plant and collaboration with its designers are therefore necessary. At this stage, the economic aspect cannot fail to assume great importance. If safety of operation is a consider-ation, as in feed water control, no necessary expense should be spared. If, however, efficiency is the only factor, the extent to which automatic control can be applied is limited by the expected savings and the number of years of improved operation which will ensure repayment of capital, interest, and maintenance costs. The accurate determination of improvement in efficiency is a difficult and costly matter even in an operating station; it becomes almost impossible in the design stage to foretell what results will be obtained and assumptions must of necessity be made, giving rise to more or less intelligent guess-work.

Therefore, although technical appreciation of the advantages of automatic control is simple, factual evidence of improvement in efficiency is scarce owing to the difficulties involved in con-ducting comparative experiments over long periods of time."

A further reason may be found in the fact that plant and machinery are often very old. The opinion is held that they do not lend themselves to modern ways and means of production supervision.

This is partly true but if one considers the large amount of capital spent every year for modifications of and additions to plant, one is inclined to think that ways and means could have been devised in our instrument-minded age to also provide, without extra cost, for such alterations as instrumentation and automatic control may necessitate.

A striking example may prove this statement. By courtesy of Messrs. Illovo Sugar Estates Ltd a series of experiments and investigations were conducted. They were to establish that a certain correlation exists between the position of a slide damper controlling the amount of sulphur dioxide

75

to the sulphitation tower and the resulting pH-value of the juice leaving this tower. The general principle of the experimental arrangement is shown Fig. 2. By gradually opening the damper inch by inch it was expected that a gradual increase of sulphur dioxide to the sulphitation tower would ensue and that, subsequently, the pH-value would be influenced in a way that was predicted by theoretical considerations. A table giving the result of this experiment is shown in Fig. 2.

TABLE I.

It reveals a very startling fact:

When the damper is withdrawn by only one inch of the full pipe diameter of 12 in., i.e. if only l/27th part of the area of the pipe is available for the SO2

flow, the full amount of gas produced in the SO2

generator is already passing into the sulphur tower. A further opening of the damper does not change the pH-value materially. The fluctuations which occur are due to changes in the gas quality which is indicated by the figures in column 4, Table 1. It is quite obvious that nobody would conceive the idea that only about one inch in the almost, closed position of the damper-slide must be used to control the pH-value. Let me assume that somebody installed an automatic regulator, utilising this damper as a control organ. It would prove a com-plete failure, and there is little doubt that the controller would be blamed for this failure. But there is even less doubt, in my mind, that the plant described can be altered with very small means to make a controller a success and a paying pro-position.

From the few figures available from this one experiment, I am inclined to conclude: (0 A 12 in. pipe from the sulphur burner to the

tower with a 12 in. damper and the costly operating gear for such a damper are by far

76

them, as one day an automatic pH-controller will form an integral part of the milk-of-lime valve or of the sulphur-dioxide or carbon-dioxide dampers.

One cannot expect and I do not consider it necessary that every sugar engineer should learn to design and understand fully the makings of an automatic controller. It is quite sufficient for him to accept it as part of his plant which has to fulfil very definite functions. One of them is to measure continuously what would have been measured sporadically before, and the other one is to convert the measuring results into proportional forces. They will operate control organs which, in turn, influence plant conditions in such a way that they conform with the requirements. These two func-tions are in brief the object of automatic controllers, and the fact that controllers and instruments very often occupy prominent positions, in a clean and well illuminated space, stealing the show and diverting attention from the plant, should not confuse this issue.

The fact that an automatic gear can measure and control continuously and tirelessly for 24 hours a day whereas a human operator will soon relax and fall off in the accuracy of his reactions, makes the controller superior to the operator. We must, however, beware of slogans, such as "The Auto-matic Sugar Factory" with pictures of operators in white coats, in factories and boiler houses meticulously clean and spotless. They have rather distorted the issue: the task is not to introduce robots which would run the factory but to instal certain gear which can do automatically and un-failingly certain pre-determined operations which will repeat themselves time and again. The human intelligence is not made superfluous by introducing automatic equipment, but it is given the oppor-tunity to assert itself more efficiently.

How futile it would be to hope that anybody engaged in manually regulating the correct ratio juice : milk-of-lime could detect any such delicate third influence as for instance changes of the pH-value of the initial juice! And how easily this could be done if the fluctuation in quantity would be accounted for by an automatic control, and if the operator would be free to observe and assess the much slower fluctuation of the initial pH-value and allow for it by very slow and minute adjustment in the rate of flow of milk-of-lime! Experience proves that, by eliminating such variables which would "draw" other components in a fixed ratio, much higher degrees of efficiency can be achieved permanently than was ever thought possible for an individual plant.

One consideration that frequently prevented even progressive managements from installing adequate supervising and control equipment is the com­

paratively high initial cost. The price of the individual instrument has become relatively lower over the last 20 years, but very often those alter-ations to plant, necessary to make new instruments and control equipment adaptable, are requiring large sums of money. This fact cannot be disputed away but it should be understood, on the other hand, that any such alterations plus instrumentation are en-hancing the value of the plant much more than the capital expenditure would indicate.

This statement should not lead one to believe that instruments and automatic control equipment can make a new machine out of an old one. It will require very careful investigation into the poss-ibilities of the machine, plant or section of a plant, where such installations are contemplated and any decision to purchase expensive equipment should be subject to a critical analysis. More harm than good has been done by irresponsible instrument manu-facturers and dealers by pretending that instruments can do miracles to plant. It must be perfectly understood that instruments and control equipment are supplementary to the plant, and that they cannot do more than help to exploit the plant to full capacity.

It is difficult for me to describe in detail the instrumentation for a sugar factory working under South African conditions because no exemplary factory exists. It can safely be said, however, that it will not differ widely from those instrument-ations applied elsewhere. Let me summarise the problems that will occur: (]) Two or more components have to be mixed in

such a way that a certain result is obtained. The result in this case may be a pH-value which will have to be kept at a certain figure. What does this entail? (a) It necessitates the measuring of the quan-

tities of the various components which have to be mixed. They can be liquids of higher or lesser viscosity, or gases. The normal way of measuring liquids or gases is by means of orifice or similar primary elements built into the pipes which carry them. The design and material of these orifice plates depend on operating conditions. The rate of flow of the various media is normally controlled by valves, slides or butterfly dampers.

(b) The pH-value and its fluctuations due to corrective movements of valves, slides or butterfly dampers has to be measured. This is usually done by means of a method yielding an electric potential which is indicative for the pH-value. Various media can impede the accuracy of this measuring principle by depositing chemical reagents on the electrodes.

The entailing inaccuracies of controllers connected to such electrodes are not due to the controller but to the electrodes. It will be the object of some detailed research to prevent or overcome such deposits of foreign matter.

(2) Temperatures have to be kept at certain levels. This is done by first measuring the temperature and then by correcting control organs which improve or reduce the effect of heaters or coolers, so maintaining a continuous balance between introduced and consumed calories.

(3) A certain density of a liquid must be achieved by either evaporating water or by adding water. In the first case, the intensity of the evaporator must be regulated by turning on or by shutting off steam valves. In the second case, a smaller or larger amount of water must be added to a liquid, to bring it to a certain Baumé figure. In both cases, the density must be measured by an instrument which is sensitive to specific gravity fluctuations. Its indication must be converted into a proportional force which will operate either steam valve or water tap.

(4) A steam accumulator should take up excess steam quantities in periods of low steam con-sumption and should send it back during peak consumption periods. Excess steam makes it-self felt by high pressure, steam deficiency by low pressure. A pressure gauge and a converter using its indications to operate suitable valves would constitute an automatic regulator for an accumulator.

Without going into details, I would like to demonstrate with a few examples some of our most common regulator problems:

(5) Pre-Liming (Fig. 1) Problem: A certain amount (Y) of milk-of-lime

has to be added to the initial juice to bring it to a desired pH-value.

The quantity (X) of juice varies according to crusher operations ;

The initial pH-value (Z) of the juice varies.

Principle of Solution: The quantity "X" of juice is measured by means of a primary element, * i.e. an orifice plate. The measuring result will be a differential pressure. The quantity of milk-of-lime "Y" is measured by means of another primary element which is so cal-culated that it will give for the quantity of milk-of-lime, normally required, the same differential pressure that "X" will produce.

Both impulses are acting through measuring systems "x" and "y" as thrusts Px and Py

on a converter "C". This converter operates a servo-motor coupled with the milk-of-lime valve "V". The servo-motor is so connected to the converter that an increase, for in-stance, of Px which would indicate an increase of X, would open the valve V and so admit more milk-of-lime.

In case of a change in the initial pH-value of the juice, the pre-determined ratio juice: milk-of-line would not be correct anymore. The pH-meter would indicate a change and would transmit its indication to the ratio-alternator RA. By moving the contact element F up or down, the ratio alternator will establish a new ratio juice : milk-of-lime by shifting the equilibrium of the two thrusts Px and Py to a lesser quantity of milk-of-lime for a lower initial pH-value, and vice versa.

Sugar technologists may perhaps stumble over the fact that the milk-of-lime is injected directly into the juice main without an intermediate pre-liming vessel. This method is not customary in Natal sugar factories. I would like to point out, however, that the method of admitting milk-of-lime directly will most decidedly ensure a very thorough mixing of the two components. This will in all probability reduce the curing time which was hitherto accepted to be anything from 8 to 20 minutes. An experiment may well prove that curing vessels can be considerably reduced in size if the milk-of-lime were not admitted in an un-controlled stream direct into a big batch of juice but if it were apportioned by means of an automatic device as suggested in this paper.

This experiment may also prove that the scaling of evaporators will decrease considerably, as in the case of a sugar factory in Germany where it was stated that "errors in pH-value have not occurred any more and faulty operation of the filter plant has been eliminated. This may be proved by the fact that the evaporator has worked uninterruptedly for 12 weeks . . ."2

(6) Sulphitation (Fig. 2) Problem: Sulphur-Dioxide should be admitted

to a sulphitation tower in such quantities as to give a desired pH-value for the juice at the bottom of the tower.

Principle of Solution: Sulphur-dioxide (1) is drawn into the tower by an induced draft (2). On its way up, it encounters the juice coming from above (3). The rate of ab-sorption determines the pH-value at (4). The impulse from (4) is converted in (5) into a proportional pneumatic pressure which acts on diaphragm (6). The resulting thrust

77

78

causes the servo-motor (8) to open or close the butterfly-damper (7) thus admitting more or less sulphur-dioxide to the tower.

(7) Problem: (Fig. 3). The density of milk-of-lime should be kept at a certain Baume figure which has been found most suitable.

Principle of Solution: Lime is mixed with water to form a liquid which will easily flow but which will have a density considerably higher than that required (1). This liquid will overflow continually into the diluting vessel (2) where the density is measured in (3). The result is transmitted to diaphragm (4) as a pneumatic pressure. The resulting thrust is acting against calibrated spring (5) on hydraulic converter (6). This will operate water valve (7) admitting sufficient water to dilute the effluent from (1) to the desired degree.

Anyone acquainted with measuring principles will notice that I am deliberately simplifying these problems. In their essence, however, these examples are giving a true picture of what automatic con-trolling in a sugar factory, as in any other factory, really means. Any factory manager or sugar engineer who is approaching the problem of con-trolling in this straight forward way, should, after a while, be in a position to find applications for effective controllers. That the final control mech-anism will, then, incorporate several components which will take care of occurrences and inherent conditions of the plant, which to describe would take us too far here, should not divert from the basic issue that automatic controlling means:

(1) Measuring and

(2) Converting measuring results into positioning forces for control organs.

In conclusion, I would say that the striking absence of instruments and controllers in the South African Sugar Industry, compared with their abundance in plants of other industries such as the Iron and Steel Industry, the Chemical Industry, Power" Stations, Gas Works, Pulp and Paper Factories, etc., must lead to assume that cogent reasons have prevented their installation. A great number of discussions with managers, chemists and engineers have revealed the following opinions:

(1) Sugar from South African cane is "difficult" to process, and instruments and controllers are not likely to assist materially.

(2) An attitude of caution has created the feeling that it would be better to await positive results from other sources rather than to obtain them in their own factory.

(3) A traditional attitude that sugar has been produced without up-to-date instrumentation for such a long time.

In the face of the obvious advantages of adequate instrumentation and automatic control equipment as they have manifested themselves elsewhere, it appears that the South African Sugar Industry or one of its appointed organisations should create a Study Group of Sugar Technologists, Plant and Instrument Experts who would make it their task to establish the genuine possibilities and the extent to which the South African Sugar Industry could be modernized and made more efficient without major capital expenditure.

Acknowledgments. I would like to thank those firms and individuals

who have willingly put their experience at my disposal and who greatly assisted in compiling this first approach to an important problem. I would specially like to mention The Illovo Sugar Estates Ltd. who have conducted experiments on my behalf and the Sugar Milling Research Institute which has analysed and criticised my suggestions.

REFERENCES. Automatic Regulators shown in schemes 1, 2 and 3

by ASKANIA. 1 Young, Llewellyn: The South African Electrical Review. 2 Fricke: "Zucker" No. 18, May 1952.

Mr. Dymond said that yesterday they had heard Mr. Barnes talk of the jaggery factories in East Africa. They had now swung to the conception of a factory on modern lines with automatic equipment and regulation. They knew that it was not possible to continue with slap-dash methods, but that some adequate instrumentation had to be introduced in due course.

Mr. Barnes said the author of the paper was to be congratulated on a paper of the greatest interest to every milling officer, whatever his status. One reason for the absence of automatic control in sugar factories has been the availability of labour at reasonable cost. A great deal of the automatic controlling introduced into the Sugar Industry has largely been compelled by shortage of labour. The two instances discussed by Mr. Rohloff were only a few among a great number. Time lost through various causes frequently caused losses in other direction s in the factory. Mr. Barnes described a system of automatic control of the cane carrier in relation to the quantity of feed of cane at any one time, to avoid knife chokes. In the system it had been customary for one man to control the feed by means of visual methods. This worked well early in a shift, but later when the man became

79

tired chokes became frequent. A system of auto-matic control was introduced by which the carrier was stopped when the current was less than that at which the knife motor would be tripped out by a choke. After a small interval the automatic device allowed the carrier to move very slowly and thus fed the cane very slowly for a short time through the knives and so avoided an overload and a choke. This system had saved many hours which would have been lost by manual control. Mr. Barnes said there were many other instances Where auto-matic devices could be installed to provide greater efficiency and avoid loss of time by comparison with manual operations.

Mr. McCulloch stated he would also like to con-gratulate Mr. Rohloff on his paper which would be of importance in future developments in the Sugar Industry. Automatic control had been widely adopted in the chemical industries and in oil refinery practice and perhaps had reached a peak of develop-ment in the latter. It was now possible to operate the refinery process by a handful of men, instead of a staff of 300—400 required for the same operations done manually. Referring to the diagram of Fig. 1 he said that he thought there would be a probability of the pressure pipes connecting the milk of lime pipe "Y" to the primary element "y" be-coming blocked since the contained liquid was sensibly stagnant and protecting devices would have to be fitted. He also referred to the diagram dis-played on the blackboard and asked if it would not be preferable to retain the 12 in. bore SO2 pipe, open the damper a limited extent and instal a small bore pipe in parallel. Control of the SO2 pro-portion would be attained only in the latter pipe.

Mr. Rohloff drew attention to the fact that the diagrams in his paper indicated only the principles of control in the various cases that he had cited. He said he knew quite well that the diagrams were over-simplified and required further elaboration in practice. He said that the question of pre-liming was important as in present methods a long tmie elapsed before there was thorough mixture of lime with the juice, whereas with automatic control this mixing was instantaneous, thus saving again on time and improving on the speed of the process.

Mr. Rault also congratulated Mr. Rohloff on the appropriateness as well as the lucidity of his paper. Mr. Rohloff had indicated in his paper why instru-mentation had not been widely introduced in the Sugar Industry. The automatic control of processes by instrument demanded, as a pre-requisite, that

the indications given by such instruments should be true, otherwise automatic control entirely based on their findings would give a false sense of security. His firm had gone to the trouble of putting expensive pH meters at three different stages of the clarific-ation process, as indicators helping the workmen, but not replacing them. He was much in favour of clarification control at the factory itself, by pH meters. He had even praised the work of these instruments at the Technologists' field day held in Mount Edgecombe two years ago, but he was now sorry to admit that they had been of little use to him this season, on account of the lack of a scientific instrument specialist on the maintenance staff of factories, or even in Durban, as the ordinary mechanic or electrician was not trained to do such intricate repair work. It was not surprising that the owners were now somewhat sceptical on the benefit to be derived from these expensive instru-ments which required additional highly paid staff to keep them in working order.

Mr. Rohloff agreed that an important step had been taken by employing an instrument maker for the Industry. Iscor had installed some 120 auto-matic control instruments which were today con-trolling all types of operations at Iscor automatically. He felt that the Sugar Industry was bigger than the iron and steel industry, but at Iscor there was a large team of instrument makers kept busy all the time and Iscor had found that employing these men paid handsomely in that adequate automatic control over all processes was maintained. He felt that the Sugar Industry could adequately employ more than one man.

Dr. Douwes Dekker said that it was not possible to be dogmatic about the necessity of mixing juice and milk-of-lime in the shortest possible time. One of the chemical reactions when juice was limed was the precipitation of calcium phosphate and it was known that time was required to make this precipitation complete. It was probably of ad-vantage to keep the juice alkaline for at least a few minutes.

Mr. Dymond referred to Mr. Rohloff's suggestion that a sub-committee be appointed to go into the question of instrumentation. Mr. Rohloff had opened up a new vista of mechanisation and had indicated the possibilities for the Industry. He felt that here was an important field of research for the Sugar Milling Research Institute or some other body. He asked that a vote of thanks be accorded to Mr. Rohloff.

80

81

CLARIFICATION AND MISCELLANEOUS DATA By G. C. DYMOND.

Items of general sugar interest are classified under the following headings:

1. Preservation of juices. General. 2. Preservation with SO2. 3. Clarification. The use of aqueous SO2. 4. Aqueous SO 2 data. 5. Clarification. Comparison of four methods. 6. Filter Cake. Drying difficulties. Loss of organic

matter and cane wax on composting. 7. Summary.

1. Preservation of Juices. General. The preservation of juices over shut downs has

always been a subject of economic importance. In order to combat the losses incurred, evaporation to syrup, cleanliness, temperature and pH control, together with the use of preservatives, usually For-maldehyde, are the methods employed with varying degrees of success.

The two causes of loss are chemical and bacterio-logical. The former is caused by the inversion of sucrose in an acid medium at high temperatures. The latter by fermentation at both high (thermophilic bacteria) and low temperatures.

The basis of the problem lies in the normal (some-times abnormal) infection of cane juices with a multitudinous array of micro-ogranisms. These may be checked from multiplying abnormally by scrupu-lous cleanliness, by steaming and washing down the mills and by the use of germicides during shut downs, but under the best conditions they are not the complete answers to a universal problem.

2. Preservation with SO2. A closer investigation into the acid pre-clarification

process shows that SO2 in juices below 30° Cent, and a concentration of 0.6-0.7 gms per litre, or 3.0-3.2 pH, acts as an excellent germicide and preservative.

The basis of the acid pre-clarification process lies in the precipitation of a portion of the non-sugars, principally cane wax, protein and silica at a pH of 3.0-3.2 in the cold.

Ordinary methods of comparison by apparent purities on mixed juice, supernatant liquid and clari-fied juice from the acid pre-clarification process are impaired by the alteration in constituents of the Brix solids. The standard of comparison in these tests is the normal weight pol.

A number of test runs using the normal wt. pol. and the standard method by tables (December and January) gave in every case a lower pol. by the standard method. A further peculiarity showed that the pol. of the supernatant liquid (first precipitation) by both methods is usually 0.1 lower than the mixed

juice. In a few cases it is the same and occasionally higher, as it should be owing to the elimination of non-sugars in the precipitate.

Example:

Pol. by Pol. by Brix normal weight tables

Mixed juice 14.4 12.6 12.46 Supernatant liquid . 14.5 12.5 12.36 Slurry ... — 11.7 —

Experiments were then carried out by weighing the mixed juice, supernatant liquid and slurry. Analyses by the two methods of pol., showed a dis-appearance of 1.4 and 1.2 per cent. pol. respectively. Apart from this unknown loss, which might be due to the occlusion of starch in the slurry, the preser-vation of the juices was excellent. Thus the super-natant liquid at 27.5° Cent, over twenty-four hours showed no further drop by either method of pol. determination. At 20° Cent, there was no loss after three days—the period of the test.

If, therefore, the acid pre-clarification process were introduced into our factories, with some cooling device during the hot weather if found necessary, there would be a reduction of sucrose losses during shut downs and to a smaller degree during general running from the same cause, and further, with cleaner factories, there would be easier production of sugars free from bacterial infection.

3. Clarification. The Use of Aqueous S02.

The advisability of using a solution (6 per cent.) of SO2 lies in the sensitivity of the acid precipitate to agitation. If the floe is repeatedly broken pre-cipitation is seriously retarded. This rules out the use of sulphitation boxes with SO2 injection. Rapid absorption by efficient towers is also impractical for controlling the fine end point. There appear to be no practical difficulties in making a 6 per cent, solution of SO2 at atmospheric temperatures.

By courtesy of Messrs. African Explosives, pH curves using both phosphoric and sulphurous acids were plotted. The following shows the abbreviated average results:

Phosphoric Acid Sulphurous Acid Solution— Solution—• 1 ml H8P04 = 7.78 mgs 6 per cent. SO2

H3P04

ml H3PO4

_ 25 50 75

100

pH

5.06 4.16 3.57 3.11 2.81

Drop

__ 0.90 0.59 0.46 0.30

ml H2S04

_ 3 6 9

12

pH

5.06 4.22 3.67 3.20 2.84

Drop

_ 0.84 0.55 0.47 0.36

Note.—The possible use of phosphoric acid instead of sulphurous acid in the acid pre-clarification pro-cess is ruled out for two reasons: the comparative costs and the non-germicidal effect of phosphoric acid at low pH values.

4. Aqueous SO2 Data.

The desirability of a rapid and accurate mixing of SO 2 with cold mixed juice to a pre-determined pH, as in the acid pre-clarification process, has led to a preliminary investigation into the possibilities of making aqueous solutions of SO2 in practice. I am indebted to Messrs. African Explosives for their assistance in working out the following basic figures:

1 lb. of SO2 occupies . ... = 5.65 cu. ft. 1 lb. of 12% gas occupies . = 47 cu. ft. 1 lb. of SO, in 6% solution = 1 . 6 7 gallons. 30,000 gallons of juice con-

containing 0.6 gms. of SO 2 per li tre. . . = 180 lbs. SO2.

= 306 gallons. Sulphur burnt = 90 lbs.

= 306 gallons. Sulphur burnt = 90 lbs. per week = 6.5 tons. Gas to be cooled 45 x 180 = 8,100 cu. ft. per hr.

Note.—With water at 25° Cent, a 6 per cent, solution can be obtained by percolation through towers packed with Raschig rings. At lower tem-peratures higher concentrations can be achieved. The use of an aqueous solution of SO2 with a low sulphur consumption, as in the acid pre-clarification process, gives a simple and positive control, without the use of expensive pH meters. There is a minimum of floc dispersion. The amount of extra water to be evaporated is less than 1 per cent, on mixed juice.

5. Clarification. Comparison of Four Methods.

A comparison was made between the acid pre-clarification process, the same without separation of the first precipitate, simple defecation and sulpho-defecation as practised at Darnall.

Ten comparative juice samples were treated and analysed individually and the precipitates, together with composite sample of juice ash, kept for final analyses.

No. 1. The Acid Pre-clarification Process.

To the cold mixed juice sufficient phosphoric acid preparation is added to give a total P2O5 content of 0.03 per cent. In these experiments 0.5 gms of phosphoric paste, containing approximately 50 per cent, soluble P2O5, was added per litre of mixed juice. This caused a reduction in pH of approxi-mately 0.2 degrees.

SO2 gas was then passed until a reaction of 3:2 pH was obtained, or 780 mg per litre. The juice is then allowed to stand for thirty minutes, when normally an ochre coloured precipitate settles out. In practice this precipitate is thickened by slow moving stirrers. The volume of precipitate normally occupies 10 per cent, of the total juice volume, With efficient slow stirring this can be reduced to 8 per cent, of the total—an important point.

In these experiments the precipitate was filtered and washed once under vacuum and the residue dried at 80° Cent.

The supernatant liquid was decanted and milk of lime of known solid content was added to give a reaction of 7.5 pH. This tempered juice was then boiled and allowed to settle. The second precipitate was then filtered and washed once under vacuum and the residue dried at 80° Cent.

No. 2. The same without separation of the first precipitate.

The same technique is used with the exception that the precipitates are not separated. Thus phos-phoric acid is added as in No. 1, SO2 gas is passed until a reaction of 3.2 pH is obtained. The juice is then immediately neutralised with milk of lime to a reaction of 7.5 pH. The tempered juice is then raised to the boiling point and allowed to settle.

The precipitate thus obtained was filtered and washed once under vacuum, dried at 80° Cent, and weighed.

No. 3. Simple defecation. A comparison was made with the clarification

process usually employed in other sugar countries. Phosphoric acid, which is comparatively low in Natal, was added as in No. 1 and No. 2. Milk of lime was then run in to the stirred juice until a reaction of 8.0 pH was obtained. The juice was then raised to the boiling point and allowed to settle. The precipitate was filtered.as in No. 1 and No. 2,

No. 4. Sulpho-defecation as practised at Darnall. The procedure is that the juice is heated to 50°

Cent, and milk of lime added to give a reaction of approximately 9.5 pH. The juice is then pumped through a sulphur tower, where in this case an average SO2 content of 1.69 gms per litre, were absorbed. The variations which occur under this system range from 2.14 to 1.40 gms per litre.

The juice is now tempered with milk of lime to a reaction of 8.4 pH, and a weighed quantity of phos-phoric paste added, so that after boiling a reaction of approximately 7.3 pH is obtained. In these experiments the final tempered juice, comparative with the other tests, was taken from the factory and heated in the laboratory, the precipitate being collected as in the previous experiments.

Note.—For reasons already stated, the apparent purity variations are not considered significant, but the clarity of No. 1 and 2 is considerably better than in No. 3 and 4.

Comparisons of Lime aad Sulphur used together with Solids produced.

Note.—No sulphur was used in No. 3, but the standard test with iodine showed an average figure of 0.15 gms per litre in the clarified juice. The amounts of lime and sulphur used in Nos. 1 and 2 were half that in use at Darnall this season. In the 1951 crop there would have been a saving industrially in sulphur alone of 3,400 tons.

Slightly less SO2 was consumed in No. 1 than in No. 2, though both juices were brought to the same pH. This peculiarity persisted into the clarified juice.

The total solids are related to the amounts of sulphur and lime used.

p.t .c, an important consideration. As the second precipitate filters readily, cloth filters could be used, a clear juice ready for the evaporators obtained, together with an appreciable saving in wash water.

As has been noted in past years the wax per cent, first precipitate varies from 25 to 35 per cent. The total weight of wax per cent, solids is the same in Nos. 1 and 4. The high figure in the latter is due to the additional wax in the bagacillo, which varies from 1 to 6 per cent.

The reduction in Nos. 2 and 3 is due to the natural emulsification of cane wax in hot water or juice. Ten per cent, of wax is easily lost in this way. Other figures of interest, not shown in the above table, show that the protein content of the first precipitate varies from 12 to 22 per cent, and that the ash contains from 74 to 86 per cent, of silica. This precipitate is a wax, protein, silica complex, con-taining further still unknown constituents. The rate of precipitation varies with the percentage of silica naturally present. Addition of heavy substances, clay or silica itself, increases the rate of precipitation, but these merely increase the weight of solids with no other benefit accruing. Since no losses of sucrose occur at reasonably low temperatures, time can be employed in thickening, diluting and re-thickening the precipitate, so as to obtain the highest percentage of solids in the least volume. Thereafter several methods can be adopted to utilise this valuable raw material for by-products production.

Silica in Clarified Juice.

Analyses were carried out on the ash contents of both precipitates and ash from the clarified juices. An accurate balance of silica in particular was in-validated by the natural silica content of the lime and phosphoric acid used. This is shown in the following figures expressed as lbs. per ton cane.

Note.—In No. 1 the solids in the two precipitates were 5.78 and 4.70 gms per litre. In practice this would mean that the amount of filter cake solids would be reduced from 13.24 gms per litre (No. 4) to only 4.70 gms. Expressed as lbs. solids per ton cane, the reduction would be from 26.48 to 9.40 lbs.

Note.—No. 1 contains the least amount of silica in the clarified juice, while No. 4 contains more silica in juice and precipitate than existed in the original mixed juice. This is due to the silica im-purities in the increased amount of lime used.

Lime Salts in Clarified Juice.

The amount of lime salts in the clarified juice increases progressively with increasing amount of lime used. Thus:

83

Comparisons of App. Purities and Clarity

Juice Clarified

Average 10 Samples

Apparent purity .. pH

Clarity—Luximeter Kopke ..

Mixed Juice

. 8 5 . 7

. 5 . 1 -

—

1

8 7 . 2 6 . 8

68 50

2

8 6 . 8 6 . 7

66 50

8

8 7 . 4 6 . 7

40 27

4

8 6 . 7 7 . 3

50 31

Analyses of Solids

Ash per cent. solid Organic m a t t e r

per cen t . Gms O.M. pe r

l i t re W a x pe r c en t . . . .

gms wax per l i t re

1

first pt.

1 6 . 6

8 3 . 4

4 . 8 2 2 6 . 4 5

1.53

second ppt.

4 0 . 4

5 9 . 6

2 . 8 0 6 . 6 0

0 . 3 1

Total

2 6 . 3

7 2 . 7

7 . 6 2 1 7 . 5

1.84

2

Total

2 7 . 3

7 2 . 7

8 .02 13 .65

1.50

3

Total

2 8 . 0

7 2 . 0

6 . 3 4 1 8 . 7 0

1 .55

4

Total

4 1 . 7

5 8 . 3

7 . 7 2 1 4 . 0 9

1.86

Lbs. SiO2 p.t.c.

Precipitates . Juice

Total ...

Mixed Juice

— 1.47

1.47

l

1.05 0.17

1.21

2

1.14 0.30

1.44

3

1.32 0.28

1.60

4

1.50 0.32

1.82

84

General.

A general assessment of the four processes show that the acid pre-clarification process is better in some respects to No. 2, and better in all respects over either simple defecation or sulpho-defecation.

The principal benefits are: 1. Germicidal action of SO2 in original juice. 2. Reduction in all chemicals. 3. Reduction in quantity of filter cake. 4. Better filtration. No recirculation of muddy

juice. 5. Reduction in undetermined losses. 6. Better clarification, better wax retention and

easier boiling.

6. Filter Cake Drying Difficulties. Loss of Organic Matter and Cane Wax on Composting.

The drying of filter cake is extremely difficult, as the outer layer forms an insulating cover, whereby both heat and moisture are retained for long periods.

Even the daily turning and aerating of small heaps, the passing of cold or hot air under pressure has little effect. Without such aids to drying there is no loss of moisture within the heap itself.

The following experiment illustrates this. A small heap of fresh filter cake approximately 4 cub. ft. was turned daily for twenty-eight days. A similar quantity was kept in a barrel under anaerobic conditions. The results were as follows: A

Lime, gms.per litre in increasing order

Lime. lbs. CaO per ton cane

CaO per cent. Ash .

3

0.76

0.83

10.78

1

1.52

1.04

11.72

2

1.62

1.11

13.09

4

3.22

1.16

14.30

Days

A

Aerobic Moisture %.. . Temperatures

Centigrade .

Anaerobic Moisture %... Temperatures

Centigrade .

On Stacking

73.5

55

73.5

50

l

72.6

50

45

2

72.2

50

73.7

50

3

68.4

62

76.4

50

4

67.3

62

75.1

55

5

64.3

65

56

6

61.2

70

73.1

55

7

58.2

75

71.0

50

8" rain

-

-

9

70.2

55

70.7

50

20

53.1

65

69.6

48

28

52.2

42

73.2

49

Note.—Under aerobic conditions the temperature rose to a peak 25° Cent, higher than under anaerobic conditions. In the former the loss of organic matter was over 20 per cent., the ash per cent, rose from 21.8 to 48.2 and there was a considerable loss of the wax complex.

Logs of Organic Matter. During the stacking of filter cake, the natural

decomposition may be aerobic or anaerobic or both. Under aerobic conditions, there is a sweet earthy smell, temperatures rise to 75° Cent, and over a period of three months the loss of organic matter by weight ranges from 50 to 55 per cent.

Under anaerobic conditions there is a noxious smell of albuminoid putrefaction and sulphuretted hydrogen. Temperatures never exceed 55° Cent, and the loss in organic matter is negligible or com-paratively low, with no loss of moisture except in the outer layers.

Loss of Wax. The percentage of cane wax in fresh filter cake

ranges from approximately 6 to 11 per cent, depend-ing on the variety of cane, its condition and the efficiency of the rotary filters. This efficiency covers fineness and quantity of bagacillo used and the degree of heat and recirculation, whereby part of the wax becomes emulsified and passes into fabrica-tion. Thereafter much of the wax complex may be lost by decomposition. This depends on the nature of the fermentation (aerobic or anaerobic) and time.

In controlled experiments on the availability of P 2 0 5 in rock phosphates, the weighted controls, which were aerated by turning five times in three months, showed only 1 to 2 per cent, of wax at the completion of the tests.

Experiments were finally carried out to determine the rate of loss. The figures are curiously erratic, due possibly to the varying degrees of temperature and aeration throughout the heaps: B

Note.—Under aerobic conditions a whitish my-celium forms a few inches below the surface. The wax per cent, of this layer was 4.9 per cent, whereas the centre and hottest part showed 7.3 per cent.

7. Summary.

The undetermined loss of sucrose in the Industry is over 8,000 tons per annum. There are a number of possible reasons, but one important cause is the loss by bacterial action during shut downs. This loss may be prevented by using SO 2 as a germicide in the cold mixed juice, as in the acid pre-clarifica-tion process. At 27.5° Cent, no loss of pol. was recorded over twenty-four hours. At 20° Cent, no loss was recorded over three days—in both cases the time of the trial. Slight rises and falls of pol. were recorded in the supernatant liquid immediately on settling, which thereafter remained constant. No very definite reason can be ascribed to this peculiarity as yet.

Clarification.

Four methods of clarification were compared, the acid pre-clarification, the same without separation of the first precipitate, simple defecation and sulpho-defecation.

The acid pre-clarification is assessed the best and the reasons given. The use of aqueous SO2 is des-cribed, this method being essential in the acid pre-clarification process, in order to minimise the breaking of the sensitive floc and for accuracy.

In general the clarity of processes 1 and 2 were superior and the amount of chemicals reduced. In the 1951 crop 4,640 tons of sulphur were burnt. This would have been reduced to 1,240 tons, a saving of 3,400 tons of sulphur alone.

The amount of filter cake (second precipitate) is reduced to approximately one-third—26.5 lbs. solids p.t.c. reduced to 9.4 lbs. This would result in easier filtration, less sucrose loss with less dilution water. This saving in water would then be used for dilution of the first precipitate.

Finally the acid pre-clarification process shows the best wax retention in practice, with the least SiO2

and lime salts in the clarified juice.

Filter Cake. The difficulties of drying filter cake are described.

The rise in temperature under aerobic conditions may be 25° Cent, higher than under anaerobic conditions.

The loss of organic matter under aerobic conditions is high—50 to 55 per cent. Under anaerobic con-ditions this loss is minimised. Among the organic materials lost is "cane wax"—a general term cover-ing a variety of high and low boiling point fractions.

Under aerobic conditions this loss may range from 70 to 90 per cent, of the total according to the time and temperature period of decomposition. Under purely anaerobic conditions this loss is reduced, ranging normally from 10 to 20 per cent, during the first month and progressively increasing with time and slow decomposition.

Dr. Douwes Dekker said that, as in past years, Mr. Dymond had again given a paper dealing with his work on clarification problems. The present paper led to the question how in actual processing, the use of SO2 as suggested by Mr. Dymond could prevent the loss of sucrose in clarified juice during shut downs.

Mr. Dymond said that during shut downs there was a loss in clarified juice. This was veiled because on many occasions lime was added and the pH brought up on the Saturday night. He thought this was wrong because the action still went on. Those who had worked in distilleries or acetic acid plants knew that bacteria would only live in a low concentration of the product they produced. To maintain equilibrium the products were often neutralised, thereby enabling the bacteria to live under normal conditions. He suggested that experi-ments be carried out to test this suggestion which he thought worthy of investigation. Under present conditions it was known that a loss was incurred. He maintained that if the clarified juice was cleared off and the pH maintained at 3.2° in the cold mixed juice then less loss would be incurred. In reply to Dr. Douwes Dekker, Mr. Dymond said that he thought that clarifiers should be cleaned out as far as possible at the weekend.

85

Days

B

Aerobic Wax per cent ...

A naerobic Wax per cent. ...

On Stacking

11.7

11.7

l

12.1

8.8

2

5.9

8.6

3

8.5

9.0

4

4.5

7.0

5

8.0

__

6

7.3

_

7

6.3

8.3

9

5.1

10.0

20

3,5

9.9

28

2.3

10.8

Dr. Douwes Dekker said that he was satisfied that Mr. Dymond had shown that if cold mixed juice was sulphited to 3.2 pH, decomposition of sucrose was negligible within a period of about 48 hours. But he did not see how this fact could be used to prevent decomposition of sucrose in the hot juice in the clarifiers during the week-ends. A statement as made by Mr. Dymond should have been supported by the results of factory tests.

Mr. Dymond admitted that the figures were not final, but work would be carried on to verify whether his theory was correct or not.

Dr. Parrish asked about the silica in clarified juice and referred to the figures given in the tables. He asked whether the figure in Process No. 1 was for the final clarified juice or the juice after the use of SO2. The results obtained at the S.M.R.I. with pure lime were somewhat different to those obtained by Mr. Dymond.

Mr. Dymond said he would very much like to see the S.M.R.I. figures with the use of pure materials. He admitted that he had worked with the ordinary substances in the mill and not with purified chemicals. The results from clarification experiments were never spectacular and further experimentation was necessary.

Dr. Douwes Dekker said that the question of silica removal was now a subject of investigation by the S.M.R.I. They knew already that they would have to discern between "total" SiO2 and "available" SiO2. It had also been found that a considerable part of SiO2 in mixed juice could be removed by filtration only. To obtain a proper

flocculation of the suspended nonsugars was an important part of clarification.

Mr. Dymond said he had already indicated that the SiO2 could be removed from the heated juice by straight filtration and if this were the better way methods could be devised of incorporating it.

Dr. Douwes Dekker asked by what method CaO in juice had been determined, by the soap test or by a more perfected method.

Mr. Dymond replied that it was a purely analytical determination.

Mr. Alexander asked whether Mr. Dymond was certain that because there were fewer micro-organisms in the juices there would be fewer in the final product.

Mr. Dymond replied that it had been suggested that his experiments suppressed the bacteria. This could only be proved in general practice and he had merely set out ideas which could be followed by further research work. Mr. Dymond added that it had been decided to go into a pilot plant at Darnall in connection with his process. They would put in a grid of cooling tubes between the sulphur burner and the sulphur tower to bring the temperature of the gas down. The gas would then be passed through a small tower 7 ft. X 3 ft. X 3 ft. and water be allowed to percolate through. From the bottom would then be drawn off whatever acqueous SO2 was required.

Dr. Douwes Dekker then asked the meeting to convey to Mr. Dymond their appreciation in the usual manner.

A BRIEF REVIEW OF A PORTION OF THE LITERATURE DEALING WITH SULPHUR BURNING AND THE FORMATION

AND DISSOCIATION OF SULPHUR TRIOXIDE R. J. HOGARTH and C. T. JENKIN

In this paper an attempt will be made to sum up the facts and figures contained in a host of papers and books already published on the subject of sulphur burning. For the sake of clarity the two sections:—

I. Theory. II. Practice.

have been chosen to enable the reader to be in a better position to differentiate between theoretical hypotheses and conjectures and actual observations recorded from plant practice. In the theoretical section, especially, all calculations have been made with a view to simulating those conditions pre-vailing in the sugar industry in this country.

It should be explained that the sequence of operations, which may be generally referred to as sulphur burning, has been dealt with mainly from the point of view of standard sulphite pulp and paper mill practice. This angle of approach was chosen because of the quantity and quality of literature that is readily available from this source. However, this should not cloud the issue at all from the sugar technologist's point of view as, basically, the problem of the production of sulphur dioxide is the same in both cases. Like the sugar chemist, the pulp and paper technician is concerned with the production of a burner gas which is rich in SO2 and free from all but the smallest possible amounts of SO3, under the operating conditions prevailing. Only after the sulphur dioxide is produced does the application differ for the two industries. We shall therefore concern ourselves only with those operations that are common to either industry.

The source of information has been indicated when a particular work is referred to, and, where possible, the more modern methods of plant practice or theory have been chosen, although in some cases personal preference or dislikes may have predom-inated and influenced the choice of certain facts and theories.

PART I. THEORY.

When sulphur is heated in the presence of excess air it will, under favourable conditions, ignite and continue burning to form, in the main, sulphur dioxide.

If the reaction:—

S + O 2 S O 2

is considered, it will be seen that one molecule of sulphur and one of oxygen combine to form a molecule of sulphur dioxide. However, it is not necessary to use pure oxygen for the purposes of combustion: air will meet the requirement equally well, and, furthermore, its use is more economical. Basically, air may be considered to consist of 21 per cent, oxygen and 79 per cent, nitrogen by volume. Thus, when one volume of oxygen is used to produce one volume of SO2, there will be 79 21 or 3.76 volumes of nitrogen involved. The final burner gas would then consist of:—

1.00 volume SO 2 3.76 volumes N2

4.76 total volume of burner gas.

It is obvious, then, that the theoretical maximum sulphur dioxide content of the burner gas would be

1/ 4.76 X 100 = 21 per cent, by volume. This

calculation assumes that dry air is used and no SO3 is formed during the actual burning of the sulphur.

Continuing, it is apparent from the above equation that in order to completely oxidise 1 lb. of sulphur, it is necessary to have 1 lb. of oxygen present when 2 lbs. of SO2 will be formed. This amount of oxygen is present in 53.4 cubic feet of air at. S.T.P. Practical considerations, howeyer, demand that the air supplied to the burner is in excess of the theoretical minimum in order to ensure that all the sulphur is completely burned off in the space of time allowed. Because of this, the maximum SO2 content of a burner gas in practice would be less than the theoretical maximum. In fact, with the con-ventional rotary burner the maximum SO2 content is approximately 18 per cent.

Quantity of Burner Gas.

Having made these preliminary calculations, the next point of interest is the variation of the quantity of burner gas produced when the SO2

content is raised from a minimum to a maximum value.

On top of this it would be of assistance to know how moisture in the atmosphere would effect these values.

Table I and II, Graphs A and B, supply the answers here, and all calculations were made according to the methods outlined in Lundberg's book.2

Where moist air was assumed in the theory in the calculations, this was taken as air 74 per cent saturated with water vapour at 66.6°F. This represents roughly the average humidity and temperature taken over a number of years at Mount Edgecombe during the months of May to December.1

Graphs A and B show how the volume of the burner gas produced decreases with increasing SO2 concentration. A comparison of the. results with dry and moist air indicates that the effect of moisture is not very marked upon the final gas coming from the burner.

Theoretical Flame Temperature.

When sulphur is burned in air, the reaction is exothermic and therefore the products are at a-higher temperature than that of the reactants. For every particular concentration of SO2 in a gas, there will be a corresponding flame temperature to be reached in the burner.

Unfortunately, the deduction of these flame temperatures from basic principles is a fairly long and tedious procedure. A method suggested by Lundberg2 was used here and it is reasonably rapid and convenient, although comparison with flame temperatures calculated from basic thermodynamic principles indicates that this rapid method may be slightly inaccurate and especially so at the lower concentrations of SO2. It must be stressed that the temperatures listed in Tables III and IV are approximate only, and should be taken more as an indication of the order of magnitude of temperature and not as accurate flame temperatures. Anyone wishing to derive theoretical flame temperatures may do so by consulting the technical literature dealing with this subject.5 8 9 10

Once again, results have been calculated for dry and moist air, and in order to show how the temperature of the flame may vary due to heat losses by radiation, the two cases:

(a) No Heat Losses

(b) Radiation of 15 per cent, of the total heat input have been considered.

The Tables and Graphs show how dilution of the burner gas with air causes the flame temperature to drop. Moisture again appears to have a negligible effect.

Sulphur Trioxide Formation.

Most of the points of theoretical importance have been listed, and all that remains to complete the picture is a consideration of the factors influencing the formation of SO3. It is known in practice that a small amount of this compound is always formed along with the SO2 in the burner, and therefore a knowledge of the conditions which favour SO3

formation should be useful in that it would then be possible to arrange an atmosphere and conditions in the burner which would keep the amount of SO3

formed at an absolute minimum.

When SO3 is formed in the presence of excess air, the following reaction occurs:

2SO2 | O2 2SO3

This reaction is reversible and therefore it will proceed from left to right until equilibrium is reached. At this point, as much SO3 as is being formed will be dissociating once more into its components. The equilibrium can only be dis-turbed by removing one of the reactants or product of the reaction, or changing the steady state of conditions prevailing, otherwise, if temperature, concentration, etc. are kept steady, then the equilibrium will be stable and constant for those sets of conditions.

Fortunately, this reaction has been studied in some detail by many observers, and it is possible to calculate the amount of SO3 that will be formed under almost any given set of conditions. The subject of catalysis and the effects of catalysts have no direct bearing on the theory of SO3 formation as it is conceded that catalysts do not disturb the equilibrium point, but merely serve to speed up the rate of reaction, thereby ensuring that equili-brium will be reached in a shorter space of time. They also allow lower temperatures to be used when SO3

is being formed. Catalysts are therefore of im-portance in the Contact Process for the production of sulphuric acid.

The theoretical conversion percentages which have been calculated here do not take into account the time factor, that is to say, no allowance has been made for the period that is necessary to ensure that the reaction reaches equilibrium before the gases pass from the burning apparatus. Often it will be found in plant practice that the conversion of SO2 to SO3 as measured, is considerably less than that predicted in theory under the conditions specified. Thus, the conversion figures listed may be taken as the maximum possible, which might never be attained in the plant.

In order to be in a position to predict the amount of conversion of SO2 to SO3 under a given set of

which occurs at various temperatures. This re-action, and not the previous one having double the quantities, has been chosen purely for convenience. The equilibrium constant, K, differs for the two reactions, the former being the square of the latter, but the calculated conversions would be the same in each case.

For the purpose of this paper, the equilibrium constant has been calculated from the average of three authorities, viz.

Fairlie*—

where K is the equilibrium constant and T is the absolute temperature in degrees Kelvin.

The above equations show that. K is dependent upon temperature only. Naturally, the equili-brium constant can be calculated from thermo-dynamic first principles along these lines:

where a is the activity co-efficient for each respective reactant or product, and this equation may be re-written as

, where p is the partial pressure of each substance as indicated.

However, the calculation of K from first principles is somewhat lengthy, but should this expedient be necessary or desired, reference should be made to textbooks on thermodynamics. 5 8 9 10

Knowing K, it is then possible to deduce the expected conversion of SO2 to SO3 under various conditions. However, the conversions calculated in this paper were carried out on the lines suggested by Browning and Kress.7 The equilibrium constant, K, was calculated from an equation where the variables were in terms of concentrations of the various products as follows:

where a = initial concentration of SO 2 in %

where b = initial concentration of 02 in %

where x = Fraction ( % . ^ e r s i o n ) of SO,

K is calculated from assumed values of conversion for a gas of specific S0 2 content, and then the equilibrium temperature for that value of K cal-culated from the Lewis and Randall equation given previously.

Table V and Graph E detail the variation of equilibrium constant, K,with temperature, Tables VI VII and Graphs F and G, record the equilibrium temperature for any assumed degree of conversion with a gas of known composition.

With a gas of known strength, the amount of SO 3 likely to be formed can be reduced by increasing the temperature. Generally, in order to reduce the possibility of trioxide formation, the SO2 content of the gas should be increased, i.e. the amount of excess air used in burning should be kept at a minimum. On top of this, high com-bustion chamber temperatures help to reduce the incidence of SO3 formation. However, theory indicates that it is literally impossible to prevent SO2 undergoing a certain amount of oxidation, albeit a very small amount at temperatures over 1,000°C,

The sugar chemist is attempting to produce a burner gas containing SO2 only, and none of the higher oxide, and in the light of the foregoing theory it would appear that this could best be accomplished by the following:—

(a) Maintaining the quantity of excess air used at an absolute minimum for the plant in question.

(b) Ensuring that the flame temperature in the burner is kept as high as possible and heat losses due to radiation are minimised.

(c) Arranging for the hot burner gases to be cooled as rapidly as possible in order that they need not be held at those temperatures for any length of time, where maximum conversion of SO 2 to SO 3 is likely to occur.

(d) On top of this it would be helpful to draw the gases through the burner as rapidly as possible, with the apparatus in use, to en-deavour to reach a state of affairs where the reaction 2SO2 + O2 2SO3, does not have time to reach equilibrium. However, this rate of flow of the gases in the burner will naturally be dictated largely by the rate at which sulphur can be burned without sublim-ation taking place.

conditions, it is necessary to know the equilibrium constant, K, for the reaction:

89

90

This completes the theoretical study of sulphur burning, and all that remains is to observe what occurs in practice. The theory is in fair agreement with what is experienced in actual pant operation and it is hoped that the connections between theory and practice will be obvious.

PART II. PRACTICE. The most convenient raw material for the pro-

duction of SO2 is sulphur, and although there are many alternative sources available, sulphur still proves the popular choice for a variety of obvious reasons. The sulphur used in the sugar industry in this country is obtained almost solely from the U.S.A., this source being chosen with consideration to price, quality and availability.

In 1949 the production of crude sulphur in America amounted to approximately 4¾ million long tons,14 mined by the Frasch-process and it is interesting to note that the Texas Gulf Sulphur Company was the major producer. During the same period the Union of South Africa is credited with the importation of just over 65,000 long tons of crude sulphur,1* most of which was used in the manufacture of sulphuric acid. It can be seen, therefore, that the sugar industry's requirements of between 4,500 and 5,500 long tons per annum, are quite small when compared with the total imports into the country.

Sulphur is abundantly distributed in nature, and on the Gulf Coast it is usually associated with salt dome intrusions. The element is found to occur in the limestone, gypsum and anhydrite cap rock and a few of these domes contain commercial quantities of sulphur.15 The occurrence of these domes appears to be limited mainly to the Gulf Coast region of the States of Texas and Louisiana. Details of the mining and production of sulphur are contained in a highly informative circular en-titled "Sulphur—General Information" issued by the United States Bureau of Mines15 and anyone interested would be well-advised to procure a copy of this publication.

The element sulphur is non-metallic and exists in a variety of allot ropic forms and therefore may­be said to have a series of properties depending on the form in which it is found. Natural sulphur usually occurs as the more stable rhombic or α-sulphur form. However, it can exist in many forms,6 but they all tend to revert to the rhombic form. A study of the various properties and allotropic forms of sulphur is outside the. scope of this paper and a brief account of those properties of relative importance to sulphur burning only, will be given. Sulphur, S8, is a brittle, yellow ele-ment which is solid at room temperature, is in-soluble in water, but will dissolve in carbon bisul-

phide. The rhombic form melts at 112.8°C. and the melt forms a mobile liquid which becomes more viscous as the temperature is increased, until at 200°C. it becomes so thick it will not flow. However, at 350°C, the melt again becomes mobile. It ignites at about 248°C, finally boiling at 444.6OC.2 The above temperatures should be taken as approximate only as Mcllor8 gives specific figures for the known allotropic forms and any particular allotropic mixture would probably have its own characteristic physical and chemical constants.

A typical analysis of the sulphur supplied to the sugar industry is roughly as follows:—16

Moisture at 105°C. ... 0.10% Total Residue on Burning 0.13% Organic residue 0.03% Mineral residue 0.10% Arsenic as As 20 3 2 p.p.m. Selenium as Se Negligible Acidity as II2SO 4 ... < 0 . 1 % Sulphur (by difference) ... 99.6%

Sulphur as supplied contains very little sulphuric acid, but storing in the presence of air and moisture may cause sulphuric acid to be formed.6 It has been found that the oxidation of sulphur is negligible if moisture is absent.6 As sulphur contains a small amount of hydrocarbons which form water on combustion, this causing mists in the burner, and furthermore, as a finite quantity of moisture may be present originally, it is often considered good practice to melt and strain all sulphur going to the burner. This process ensures that moisture and hydrocarbons are removed before combustion takes place.

From these brief observations, it is fairly safe to conclude that sulphur should be stored in a cool, dry place under cover to prevent contamination by moisture, carbonaceous matter, etc. When stored in the open, a certain amount of moisture will be absorbed from the air or rains, as will be carbon and silica which are present in the particles of dust in the air. The first component, moisture, helps the oxidation of sulphur to H 2 SO 4 , while the second, carbon, may form a carbonaceous scum on the surface of the molten sulphur in the burner, or, alternatively, ignition of the carbon will occur, the CO 2 formed causing dilution of the burner gas. The silica increases the ash content of the sulphur and this in itself is a nuisance in that it would involve frequent ash removal from the burner.

Obviously, these three constituents, i.e. H2O,, C and SiO2, would, if present in sufficient quantity, affect the rate of burning of the sulphur in addition to altering the final composition of the burner gas. It is not suggested here that storing sulphur in the open would involve the above contaminants being present in quantities sufficient to affect the efficiency

91

of the burner, but it is possible that in many cases they may reach a proportion where they cause trouble.

Types of Burners. (a) Rotary Burner.

This burner is proably the most commonly used and is known as the Glens Falls Rotary Burner. The makers, Glens Falls Machine Works of America, claim that maximum efficiency can be attained with unskilled labour and that installation and operating costs are low. Furthermore, they emphasise that losses due to sulphur sublimation or formation of SO3 can be eliminated almost entirely.11 With this type of burner it is possible to operate con-tinuously at SO2 concentrations ranging between 5 per cent, and about 18 per cent. In order to prevent fluctuations in gas strength it is desirable to feed the sulphur by mechnical means and not by hand.

The sulphur in the lower part of the horizontal rotating drum forms a molten pool as well as a thin film around the circumference of the drum. This film, in effect, increases the surface area of the sulphur exposed to the incoming air and a further increase in surface area is brought about by the sulphur that drips down from the top of the drum. These factors help to increase the capacity of the burner and ensure that combustion is complete. On top of this, it is claimed that the sulphur film protects the drum metal from deterioration due to hot SO2 gases as the heat conductivity of sulphur is less than that of cork.

A combustion chamber is fitted at the back of the horizontal drum and this serves to complete the combustion of any sublimed sulphur as well as to mix the gases and dilute them to the desired strength. This chamber is therefore constructed so as to have air ports and one or more baffles.

Impurities in the sulphur do not readily affect the operation of these burners because the rotation of the drum agitates the molten sulphur sufficiently to prevent any blanketing films forming on the surface. A further refinement that can be added is a sulphur melter and the heat of radiation from the burner may be utilised to keep the sulphur molten. Steam heating coils in the Melting Tank are only required for starting up the burner after a long shut-down. In order to further improve the efficiency of this machine, the molten sulphur should be strained prior to passing it to the burner. The feeding of molten sulphur ensures that the burner is handling moisture-free material.

Ideal results are obtained with this type of burner by using sulphur of a minimum purity of 99.6 per cent. The molten sulphur in the horizontal drum should be at the highest possible level without allowing any overflow to take place.

The makers suggest the following capacities for a burner 30 in. in diameter and 8 ft. long.

Capacity with | in. water draught is 135 lbs. Sulphur per hour.

Capacity with f—1 in. water draught is 200 lbs. Sulphur per hour.

Capacity with 2 in. water draught is 270 lbs. Sulphur per hour.

A burner of identical diameter, but only 6 ft. long would have corresponding capacities of about 75 per cent, of those given above. As a rough guide, it may be taken that:—

For a ¼ in. draught, the burner consumes approxi-mately 2 lbs. Sulphur/sq. ft./hr.

For a ¾ — 1 in. draught, the burner consumes approxi-mately 3 lbs. Sulphur/sq. ft./hr.

For a 2 in. draught, the burner consumes approxi-mately 4.2 lbs. Sulphur/sq. ft./hr.

Another authority12 considers that a burner 36 in. in diameter and 8 ft. long is capable of handling nearly 8 lbs. of Sulphur/sq. ft./hour. Fairlie4

finds it practical, without resorting to high draught, to run a burner 4 ft. in diameter at a rate of 1 ton of Sulphur per day per foot of length.

The size of the combustion chamber is variable between fairly wide limits and it should be borne in mind that a large chamber permits a higher concentration of SO2 in the burner gas without danger of sulphur sublimation. It is suggested that a chamber spa e of 00 cubic feet per ton of sulphur per 24 hours4 is adequate, while the makers of this type of burner indicate that a minimum space of 6| cubic feet per pound of sulphur per hour per square foot of burner area should suffice for burners up to 30 in. in diameter. These figures should be taken as approximate indications only and no hard and fast rules seem to apply. The size of the combustion chamber should be dictated by practical considerations and previous experience.

Sutermeister13 states that rotary sulphur burners produce a gas of varying composition owing to sudden rushes of cold air through the apparatus, but Fairlie* claims that with a continuous feed, it is possible to maintain the gas at a uniform con­centration of SO2 with only minor fluctuations. He stipulates, however, that this is dependent on the depth of the molten sulphur in the rotating cylinder being kept uniform. Any change in this depth will alter the area of the unsubmerged film-covered surface and this in turn will alter the rate of combustion of sulphur.

(b) The Acme Burner. This type is deigned to give a constant rate of

burning for any given concentration of SO2 in the gas. The makers, Acme Coppersmithing and Machine Company, point out that the rate of com-bustion of sulphur bears a definite relation to the amount of air supplied. The amount of air re-quired for a given concentration of SO2 is given by the expression:—

where A = lbs. of air at 70°F. per lb. of sulphur burned when the burner is operating at a rate of 2| lbs. of sulphur per square foot of burning surface. Should the sulphur burning rate be changed, then the above formula has to be modified slightly.

It is claimed that a tray or rotary burner does not satisfy the requirements of a definite constant burning rate because of the following factors:—

(1) Ordinary sulphur contains organic matter and this accumulates as a carbonaceous film on the surface of the burning sulphur which may eventually extinguish the flame.

(2) Additions of fresh charges of sulphur to the burning surface disrupts the burning rate and normal conditions are only returned to after several hours have elapsed.

With this burner, the makers have overcome the first drawback by operating the burner at a rate greater than 2 lbs. of Sulphur per Square foot of burning surface per hour, finding that this prevents the formation of carbonaceous scum by burning it off with the rest of the gases. The drawbacks of factor (2) were prevented by arranging a special feeding device that fed molten sulphur in a manner that did not disturb the burning surface, but, at the same time, ensured that a constant level of molten sulphur was maintained in the burner.

This burner is operated on compressed air and a special sulphur melter is supplied with the apparatus. Sulphur is fed to the bottom of the burner via a main feed pipe, and a side feeder arm on this main feed allows for positive control of sulphur feed, thereby guaranteeing a constant level of molten sulphur in the burner.

All the data18 given here and the brief description of the principles behind this burner were taken from a pamphlet issued by the manufacturers, viz. Acme Coppersmithing and Machine Company of Oreland, Pennsylvania. A description of this burner may be found in the technical literature.19

(c) Spray Type Burner.

The Research Department of the Texas Gulf Sulphur Company is accredited with the invention

of this burner. The development of apparatus was described in a technical paper issued in 193420

and it was therein explained how the burner was operated. Briefly, molten sulphur is fed to a spray nozzle which injects a fine spray of this material into the burner. At the same time the desired amount of air is forced into the burner chamber and ignition of the sulphur takes place. The burner chamber is fitted with baffles to mix the resultant gases and prevent any unburned sulphur passing through to the absorption apparatus. The burning and combustion chambers are in one unit. Normally, the burner chamber is lined with fire-brick to reduce heat losses and minimise the formation of SO3.

Certain advantages are claimed for this burner and among those listed are:—

(1) Operation is simple and the burner requires little attention once the rate of combustion has been set. The combustion rate can be changed simply by altering the stroke of the sulphur pump and adjusting the compressed air supply accordingly.

(2) Starting up and shutting down operations are comparatively easy, and if the burner is started up when cold, maximum gas concentration can be reached within 2½ hours. In order to shut down it is merely necessary to stop the sulphur metering pump and shut off the compressed air supply.

(3) The burner is extremely flexible insofar as rate of combustion is concerned and there-fore it is possible to vary the capacity of the burner within fairly wide limits.

(4) It will produce a gas with an SO2 content of anything up to 20 per cent, when run con-tinuously.

(5) Sublimation of sulphur should not occur provided proper care is taken, and since there is no burning-down period, as with conventional equipment, the hazards of possible sublim­ation are reduced to an absolute minimum.

(6) Provided the gas concentration is kept high and the burner is lined with refractory brick, little SO3 should be formed at the elevated temperatures (2,400° to 2,700°F.) of operation. During tests20 it was found that the SO3

content of the gas passing to the cooler was only 0.14 per cent, of the sulphur burned.

(7) The maintenance costs are not likely to be high and the initial cost of installation should compare favourably with that of any of the more usual types of burner.

(8) Kress20 found that power costs could be re-duced by 75 per cent, over that of conven-tional burners.

93

Anyone wishing to study in detail the design of a spray type burner is well-advised to read the paper by Conroy and Johnstone21 on this subject. The authors indicate that the internal volume of a rotary burner should be 13/14 cubic feet per ton of Sulphur per day and on top of this a combustion chamber space of 60 cubic feet per ton of Sulphur per day is required, whereas the total volume of a spray type burner is 24 cubic feet per ton of Sulphur per day.

It is felt that the three burners listed cover the types of particular and possible interest to the sugar industry. In order to complete our practical con-siderations of sulphur burning, it is only necessary to study how S 0 3 formation occurs and what can be done to minimise its formation as this information should then make it possible to successfully produce a gas that is rich in S0 2 and, at the same time, free of all but traces of S03 .

The Formation and Dissociation of SO3

The conditions favouring the formation of S0 3

are:— (i) Decomposition of any sulphuric acid that

may be present in the commercial sulphur, (ii) Formation of trioxide during the actual

burning of the sulphur. (iii) Oxidation of the S0 2 present in the burner

gas. The first source accounts for only a relatively

small amount of the S 0 3 normally found in burner gas. Naturally, sulphur will oxidise to a certain extent in the presence of moist air, but the amount of sulphuric acid formed is bound to be extremely small.

Undoubtedly, a certain amount of trioxide will be formed during the burning of the sulphur, but this is difficult to determine accurately. It will depend, to a large extent, upon the proportion of excess air present and upon the temperature reached in the burner. It can be seen therefore that the SO 2 content of the gas should be as high as possible for the type of burner in use.

The third source, i.e. the oxidation of S02 , accounts for the major portion of trioxide found in a burner gas and consequently this aspect of the problem will have to be examined in fairly minute detail. In practice it has been found that oxidation will take place only until a certain ratio of dioxide to trioxide is reached. At this point equilibrium is established between the reactants and products of the reaction and as fast as S 0 3 is being formed, it is again decomposed into S0 2 and oxygen. This equilibrium may be distrubed or changed by altering the concentration of the reactants on either side of the equation:

A rotary steel or cast-iron burner operating continuously has been found to convert between 1 per cent, and 2 per cent, of the S0 2 into SO3

2, while a similar burner lined with refractory brick will produce a gas of much lower trioxide content. Obviously, then, the materials of construction have an effect upon the oxidation of SO 2 and an investig-ation led to the discovery that certain metals or their oxides could increase the rate at which trioxide was formed. These metallic substances are known as catalysts and while they do not take part in the reaction in a quantitative way, that is to say, they are not used up or depleted during the course of oxidation, they most certainly speed up the rate of reaction. These substances, therefore, warrant study as well.

It will probably be advisable to study the various effects of physical and chemical conditions upon trioxide formation separately to ensure that the picture be complete.

Effects of Excess Air.

Cornog21 and co-workers found that when sulphur vapour was burned in the presence of 250 per cent. excess air at 460°C, approximately 3.4 per cent, to 3.8 per cent of the sulphur appeared as the trioxide. However, Browning and Kress21 observed less than 0.02 per cent, conversion under similar conditions in the absence of a flame. The obvious inference is that atomic oxygen in a flame has a considerable influence upon the reaction. The results of their experiments indicated that trioxide formation could be reduced by decreasing the equilibrium conversion of S0 2 to S 0 3 by keeping the quantity of excess air at a minimum in con-junction with high combustion chamber temper-atures. A perusal of the theory will show agreement with this statement.

Obviously, then, the amount of air supplied to the burner should be carefully metered, or, failing that, regular analyses of the gases from the com-bustion chamber should be carried out to ensure that the S0 2 content is being maintained at a maximum.

A rough guide to the correct amount of air re-quired is based on observation of the flame in the burner. If the sulphur burns with a blue flame, conditions are just right. A flame with a brown tinge indicates insufficient air and sublimation of sulphur. However, these observations do not prevent one from operating the burner with an excess quantity of air and one is well-advised to rely more on chemical tests and mechanical aids than on rale-of-thumb methods such as those men-tioned above.

A study of Table VII will show how the quantity of excess air can influence the formation of S03 .

94

Should the burner be operated with insufficient air to complete the combustion of sulphur, then sublimation takes place and slight mists of unburned sulphur may form.

Influence of Temperature.

Theory has already indicated to us that temper-ature exerts a considerable influence upon the amount of trioxide formed. In practice it has been found that the optimum temperature range favour-able to the formation of S 0 3 lies between 540°C and 980°C22. Frohberg13 found that maximum oxidation occurred at 400°-—500CC, while at an approximate temperature of 900°/l,000°C., dissoci-ation took place. It has been shown that at 1,000°C. nearly 50 per cent, of the S0 3 formed is decomposed, this decomposition starting at 700°C. at which temperature the conversion of S0 2 to SO 3 is about 60 per cent.23 Above 1,000°C. the dissociation of the trioxide is more rapid than its formation. It may be generally stated that below 400°C. the rate of formation of S 0 3 is too slow to be regarded as a nuisance, while above 1,000°C, even though S 0 3 is formed rapidly, the rate of decomposition predominates. However, even at 305°C, oxidation of S0 2 has been found to occur, albeit at a very slow rate.6 Thus we have two con-flicting states, viz. where the rate of conversion of SO 2 to SO 3 increases rapidly with increasing temperature, but where the dissociation of S 0 3

overhauls the rate of conversion, and at the higher temperature (above 1,000°C), therefore, the per-centage conversion, as measured, tends to decrease.

The reaction:

takes place nearly to completion at 450°C. in the presence of a catalyst, and it has been shown that, while the rate of the forward reaction is only moderate at 400°C, it increases to 40 times this value at 500°C, the decomposition reaction only becoming perceptible at 550°C and upwards. It may therefore be said that up to a temperature of 450°C, reaction of formation prevails, and only well above this temperature does dissociation come into play.25

Sulphur trioxide is very stable in the absence of contact substances and once formed is difficult to dissociate. Generally, decomposition will not take place completely at temperatures as high as l.l00O / 1,200OC, Decomposition of trioxide already formed cannot be relied upon to keep the loss from this source at a reasonable value at temperatures below 1,00°C, the amonnt of dissociation at this temperature only reducing the loss by an amount of less than 0.5 per cent.

It is well worth remembering that the final equilibrium condition of the reaction depends only upon the temperature and the composition of the gas mixture, and oxidation of S 0 2 will take place until a certain ratio of trioxide to dioxide is reached. High temperatures favour a low ratio and low temperatures a high ratio of S 0 3 to S02.7

Presence of Moisture.

The air drawn into a burner is not normally dried, but to prevent the formation of corrosive sulphuric acid mists, the water content of the air xised should not be more than 5 mgm. per cubic foot of air at S.T.P.24 Under conditions prevailing in Natal, this would involve drying the air with either concentrated sulphuric acid or P 2 0 5 or a similar efficient drying agent.

If the air is thoroughly dried with P 2 0 5 , there is little oxidation of S 0 2 up to a temperature of 450°C. and in general it may be stated that drv air helps to retard oxidation. In the presence of water vapour, oxygen does not combine with S02

at 100CC. but oxidation does occur, even at this low temperature, if particles of liquid water are present.8 Mellor* found that moisture does not appear to affect the oxidation of S02 , but the presence of C02 and nitrogen causes more S03

to be formed.

Moisture has a poisoning effect on some catalysts and Tolley27 has shown this to be true for iron oxide. Upon continued exposure, however, cata-lysis slowly increases until after 40 hours the catalysis with wet gases is half that of dry gases. He found that water vapour has its greatest in-hibiting action at a temperature of 475°C, but even at 635°C, the catalytic activity of iron oxide is reduced to of that with dry gases.

Browning and Kress7 investigated the dew-point of burner-gas mixtures and showed that the cor-rosion to be expected would be greatest at the dew-point of the particular gas because of the combined action of scale and condensed liquid upon the metal base. Above the dew-point, only gases are present and corrosion is much reduced, while at temper-atures below the dew-point, the rate of corrosion is considerably less on account of the slowing up of the chemical reactions at these lower temperatures. Their experiments revealed that the dew-point increased as the S 0 3 content of the gas was raised and for safe operating practice, using moist air, the temperature of the gas in iron should be kept above 200°C. to prevent undue corrosion of the metal. When the gas is required to be cooled below this temperature, it should be transferred to a lead pipe.

95

The catalytic effect of iron compounds depends to a large extent upon their physical and chemical state. Experiments have shown that iron oxides, as such, are not good catalysts, but their activity may be increased after a certain period by the formation of sulphates and other compounds. At elevated temperatures both the dioxide and trioxide react with iron, and this would explain the short life of iron under these conditions of service.

In a rotary sulphur burner, there will be surges of SO 3 formed when starting up or burning down as more iron will be exposed during these periods. Furthermore, the temperature range at which maximum S0 3 formation occurs will be passed through during these stages.2

Reverting to the influence of physical state upon catalytic activity, it is worth noting that ferric oxides may exhibit contact properties, and the surface of the material will exert the least influence when freshly precipitated oxides, which are not yet dried, are used. The activity increases if the oxide has been moderately heated or kept for a long time so as to become dry. Oxides obtained by heating ferric- or ferrosulphate give a much lower contact action than that obtained with an oxide prepared by igniting a precipitated hydroxide or pyiites-cinders.26 An oxide containing 2f per cent. Arsenic as As shows considerably more contact reaction than that of a pure oxide at 700°C, while the addition of copper oxide to an oxide of iron is favourable to the formation of S03.26

Observations have shown that when a burner gas is in contact with an iron pipe, maximum conversion occurs at 750°C, there being a 13 per cent, maximum conversion for a gas with a 10.5 per cent. SO2 concentration. This conversion drops to 10 per cent, when the S0 2 content is raised to 14 per cent, and a further decrease in conversion to 3 per cent, for a gas of 19.5 per cent, content is observed.23

At 1,000°C, the conversion at all concentrations is zero.

Ferric oxide begins to exhibit catalytic activity at 550°C, and this activity increases to a maximum in the temperature range 600°—620°C.6 Increasing the S0 2 content of the gas from 2—12 per cent, does not affect the conversion to any appreciable extent, although with higher concentrations, the yield of S0 3 is lower. As a contact substance, this oxide only becomes really effective at a temper-ature of 600°C. when a percentage conversion of SO 2 to SO 3 of 40 to 50 per cent.4 is attainable. In practice, however, the conversion never exceeds 60—66 per cent.26

Iron oxide formed from pyrites-cinders shows little activity, and this only begins above 650°C, increasing with temperature and reaching a con-

As previously stated, moisture in the air can cause mists to form in the burner gas and these are most difficult to remove. A method which helps in the removal of mists consists of cooling the mixed gas to below 46°C, when the S 0 3 will condense. The gas should now be passed through a filter-box packed with charcoal or iron borings and then through a water-spray scrubbing tower. This method, however, is not always very successful and it is easier to prevent mists than, having once formed, to remove them.

Action of Catalysts. Catalysts do not affect the equilibrium point of

a gas mixture, but merely alter the rate at which equilibrium is approached. The ratio of the original and final substances present in the gas will not be changed upon contact with a catalyst. The contact substance will only ensure that this equilibrium ratio is reached in less time.

A list of various contact substances normally encountered follows, and these, with their effects upon SO 3 formation, have been listed separately for the sake of clarity. Anyone wishing to know more about the effects of various catalysts on S 0 3

formation, should read the monumental work of Browning and Kress,7 as this covers the subject most comprehensively.

A. Iron Compounds.

Iron or its oxides have long been known to exhibit catalytic properties as far as the oxidation of S0 2

is concerned and the Mannheim Process for the manufacture of suphuric acid utilised iron oxide as a catalyst. For reasons of efficiency, this oxide has been replaced by either platinum or vanadium in the Contact Process.

Tolley27 predicted that the first reaction to occur when steel was in contact with sulphur dioxide and oxygen at reasonably high temperatures, would be a combination of oxide and sulphide formation, represented as:—

2Fe + S0 2 2FeO + S Fe + S FeS

or alternatively:— 5Fe + 2S02 2FeS + Fe 3 0 4

He pointed out that as FeO cannot exist below 570°C, the first reaction shown above could only occur above this temperature. The experiments showed that the catalytic activitiy of mild steel increased rapidly during the first few hours of exposure to S0 2 , but after about 10 hours the activity became reasonably constant. It was felt that this initial rapid increase in the rate of oxid-ation of SO g was probably due to the formation of iron oxide.

version of 1 per cent, at 1,000°C. with a gas con-taining 15 per cent. S02 . However, it should be considered worthy of note that if 3 per cent, of S 0 3

is formed in a burner gas during normal burning operations, this figure may, under favourable conditions, be raised to three or four times this value if the gases are passed through red-hot pyrites-cinders.26

Pure iron oxide exhibits maximum contact action at 650°C, the respective conversion figures being 15 and 5½ per cent, for gases of 10.5 and 19.5 per cent. SO2 content. These conversion figures drop to 4¼ and 1¼ per cent, respectively when the temper-ature is raised to 1,000°C, the activity in this case being only above 500°C.7

Ferrous sulphate has a maximum contact action at 650°C. when the conversion of a gas containing 15 per cent. S 0 2 is 12½ per cent., this conversion decreasing to ½ per cent, at 1,000°C. Similar con-version figures for ferric sulphate are 8 per cent, at 650°C. and § per cent, at 1,000°C, indicating that it is less active than the ferrous form.

All the above observations indicate that iron in its many forms exhibits varying degrees of catalytic activity and if the formation of SO., is to be mini-mised then obviously gases of high S0 2 concen-tration must be produced at the highest possible combustion chamber temperature that can be attained in practice. It is of further benefit to note that in many instances, the amount of S 0 3

that will be formed according to theory will not be reached in practice as the gas passes through the combustion apparatus before equilibrium is reached.

Silica and Silicates.

It is generally conceded that vitreous fused silica exerts no catalytic effect on the oxidation of S02.7 The same applies to Dialite brick, and therefore these materials are considered satisfactory for the lining of furnaces, combustion chambers, etc.

The Efficient Operation of a Burner.

In America it is a generally accepted fact that the air to the burner should be carefully metered by mechanical means and furthermore, allowance is made for fluctuations in the burner. These changes in burning rate are catered for by the installation of an automatic recording device which continuously analyses the S0 2 content of the gases from the combustion chamber. A diaphragm operated valve then automatically operates the air inlet and sulphur feed and regulates these, thereby keeping the S02 content of the burner gas constant within fairly narrow limits.

Furthermore, the sulphur feed should be mechnical to enable a constant rate of feed to be maintained. With all these refinements it is possible to arrange for high combustion temper-atures and a burner gas containing the highest concentration of S0 2 possible with the type of apparatus in use. Even better results will be obtained if the sulphur is melted and strained before it is fed to the burner, for reasons already given.

The combustion chamber should be fitted with one or more baffles to ensure thorough mixture of the gases and prevent any possibility of unburned sulphur passing to the absorption plant. Coolers are normally fitted after the combustion chamber, and it is imperative that cooling of the gas be as rapid as possible to minimise S 0 3 formation, Initially the gases would be air- or water-cooled and then passed to either a direct or indirect cooler to bring the temperature down to 200°/300°C. At this stage, if further cooling is attempted, the pipes conducting the gases should be lead-lined. The subject of cooling is outside the scope of this paper, but is well-worth pursuing by reading Lundberg2 and others.

Summary. The practice of sulphur burning, the production

of SO 2 and the formation and dissociation of S0 3

has been outlined and the effects of: Excess Air, Temperature, and Catalysts

upon SO 3 formation detailed. Where possible a comparison with the theory has been made and it has been shown that the following points all help to increase the efficiency of this type of plant:—

(a) Tt is preferable to feed the sulphur con-tinuously by mechnical feeder. The feeding of strained molten sulphur is preferable as this eliminates moisture and deleterious hydro-carbons.

(b) The quantity of excess air should be carefully controlled and the S 0 2 content of the burner-gas kept at a maximum.

(c) The temperature of the burner and com-bustion chamber must be maintained above 1,000°C. if at all possible.

(d) Automatic recording and operating apparatus to maintain an even feed of sulphur and constant S 0 2 content of the gas is helpful.

(e) The gas from the combustion chamber should be filtered and cooled as rapidly as possible to prevent formation of SO 3.

97

Conclusions.

The authors have made every endeavour and taken all possible precautions to ensure that the information given is accurate. However, mistakes may have cropped up and no responsibility can be taken for such errata.

Many of the improvements listed may possibly make the production of SO 2 by these methods rather costly and their inclusion should not be taken as a recommendation, but rather as a guide or example of how efficiency may be improved. These refine-ments are in practice in America in many of the larger paper mills, so obviously they are economical for the production of large quantities of S02 .

It is admitted that it is literally impossible to prevent the formation of S 0 3 entirely, but if pre-cautions, along the lines of those mentioned in this paper, are taken, the quantity of S0 3 can be reduced to such a low level that it no longer constitutes a nuisance.

The types of burners listed were limited to those generally used in the Sugar Industry and those which may be of interest. The spray type burner is proving very popular in America and it produces a gas of low SO 3 content and furthermore it is very flexible in that it may be started or shut down in a very short space of time. Control is easier than with the conventional rotary burner and paper mills in America have found that installation costs are not excessive. Maintenance costs are low and power consumption compares most favourably with other types.

For the sake of those who would like to know more about sulphur burning and its applications, a reading list has been added.

Acknowledgments.

Our sincere appreciation and thanks are duly made to those firms and Institutions in America who corresponded with us, supplied technical articles and generally went out of their way to be helpful. In particular we would like to mention :— The Paper Institute, Texas Gulf Sulphur Company, Acme Coppersmithing and Machine Co., and the Glens Falls Machine Works—all American organis-ations that were extremely helpful and co-operative.

REFERENCES. 1 Beater: The Distribution of Temperature in the Sugar Belt

of Natal and Zululand—S.A.S.T.A. 1949. 2 Lundberg: Acid Making in the Sulphite Pulp Industry. 8 Chemical Control Plant Data: Booklet issued by Chemical

Construction Company. 4 Fairlie: Sulphuric Acid Manufacture. 5 Lewis and Randall: Thermodynamics and the Free Energy

of Chemical Substances. 6Mellor: A Comprehensive Treatise on Inorganic and

Theoretical Chemistry—Vol. X.

7 Browning and Kress: A Study of Some Factors Influencing the Formation and Dissociation of SOs in Burner Gases—Paper Trade Journal, 100 No. 19, 31-43, 1935.

8 Glasstone: Thermodynamics for Chemists. 'Dodge: Chemical Engineering Thermodynamics. 10 Whitney, Elias and May: Chemical Reaction Equilibria—

TAPPI, 34, No. 9, 1951. 11The Glens Galls Rotary Sulphur Burner: Pamphlet from

Glens Falls Machine Works. 12Darrah: The Preparation of SOs—Paper Trade Journal,

pp. 132, Nov. 30, 1950. 13 Sutermeister: Chemistry of Pulp and Paper Making. 14 Josephson and Downey: Sulphur and Pyrites—U.S. Bureau

of Mines Yearbook, 1949. "Ridgway: Sulphur—General Information—U.S. Bureau

of Mines I.C. 6329. 16 Sulphur: Technical Service Note No. 32—African Explosives

and Chemical Industries, Ltd. 17 Furniss: Rogers Manual of Industrial Chemistry. 18 General Description of Acme Burner—Pamphlet from

Acme Coppersmithing and Machine Co. 19 Cain and Chatelain: New Low-Capacity Sulphur Burner-

Chemical and Metallurgical Engineering, 46, Oct. 1939. 20 Kress and Others: Spray Type Sulphur Burner—The Paper

Mill, Oct. 1934. 21 Conroy and Johnstone: Combustion of Sulphur in a Venturi

Spray Burner—Industrial and Engineering Chemistry 41, pp. 2741, Dec. 1949.

22 Barker: Sulphite Acid Preparation—Paper Trade Journal, pp. 136, Nov. 30, 1950.

23 Newell Stephenson: Pulp and Paper Manufacture, Vols. I II and III.

24 Modern Chemical Processes. 25 Riegel: Industrial Chemistry. 26 Lunge: Manufacture of Sulphuric Acid and Alkali, 3rd

Edition. 27Tollev: The Catalytic Oxidation of S0 2 on Metal Surfaces Journal Society for Chemical Industry, 67, pp. 369, Oct. 1948.

READING LIST. Chemical Engineering -53, 225 (1946).

54, 221 (1947). Chemical Engineering Process—46,614 (1950). Chemical Markets—30, 261 (1932). Chemical and Metallurgical Engineering 42, No. 7, 374-7 (1935) Chemical Process Principles—Hougen and Watson. Industrial and Engineering Chemistry—17, 593 (1925).

34, No. 9, 1017 (1942). 35, 522, 541-5 (1943). 41, 2741 (1949). 42, No. 4, 713-8 (1950). 42, 2215-7 (1950)

Industrial Chemistry—Riegel. Journal Am. Chem. Soc—48, No. 11 (1926).

„ Chem. Soc—123, 3203 (1923). Paper Trade Journal—94, No. 15, 39-42 (1932).

99, No. 17, 48-51 (1934). Proc. Roy. Soc. London—Series A, 138, 635 (1932). Pulp and Paper Mag. Canada—31, No. 52, 1390-3 (1931).

33, No. 2, 78-80 (1932). "„ "„ „ „ „ Conv. 37, No. 3, 140-8 (1938).

" „ „ „ 44, No. 4, 320 (1943). "„ „ „ „ 52, 108-111 115 (1951). "„ '„ „ „ „ Conv. 243-8 (1951).

Soc. Chem. Ind. Journal—67, 369-73 401-4 (1948). Textbook of Physical Chemistry—Glasstone. Trans. Am. Inst. Chem. Engs.—27, 264 (1931). U.S. Bureau of Mines—Bulletin 406 (1937).

98

100

101

102:

Mr. Dymond said that they were indebted to the authors for this very important paper. He expressed the view that the standard of papers at the Congress was very high. This paper provided data that had been wanted for many years.

Mr. Main agreed and said that the paper was a most valuable addition to the technical knowledge. He referred to Maxwell's book on sulphur burning which was now out of print. Since then he had not found anything as valuable as the paper presented. He hoped that it might be possible to provide illustrations.

Mr. Hogarth said the point had been discussed and photostat copies of a paper dealing more especially with one type of sulphur burner could be made. He hoped that: these could be fairly generally distributed.

Mr. Rohloff said he thought the three types of sulphur burners could be classed as three separate stations in the development of sulphur burners. Much the same could be said about the development of boilers. He drew a. parallel between the different types of sulphur burners and the different types of boilers. This development seemed to suggest that the trend was towards greater flexibility and he thought it was fit to mention this as it indicated a desire to make sulphur burners for the Sugar Industry more flexible and more applicable to sugar production. Mr. Rohloff said that they had asked the S.M.R.I. about the Acme burner and were told that this had been investigated, but the type of sulphur used in the Industry was not applicable to this burner.

Mr. Hogarth said the sulphur supplied to the Sugar Industry did not vary in composition from the sulphur used overseas and presumably used in Acme burners.

Mr. Perk said that speaking from memory, he had told Mr. Rohloff that the method of storing sulphur in Natal precluded the sulphur being used in the closed type of burner used in Java. In order to operate these burners for months without loss

of capacity or interruption for cleaning, the sulphur had to be kept free from dust and moisture. It was therefore routine at the Java factories to store the sulphur in a separate compartment, and particularly not in the store-room where lime was kept.

Mr. Main expressed the view that sulphur storage conditions did not materially affect the sulphur used in burners in Natal.

Mr. Hogarth said he was inclined to agree with Mr. Main, as the contaminants in the sulphur supplied to the Sugar Industry were very small indeed.

Mr. Rault said the problem in the Sugar Industry was not only to burn the sulphur but also to absorb it. He asked whether there was a more modern, compact and controllable absorbtion system than the "eye sore" and dirty plant called the sulphur tower, commonly used in our factories.

Mr. Hogarth said that the absorption of S 0 2 was a most complicated question, but there was equip­ment available which would assist in the absorption more efficiently than was done at the moment in the Industry. He offered to supply this inform­ation to Mr. Rault.

Mr. Dymond agreed that the problem of hot gases going into the juices required further in­vestigation and he hoped this would be done by the S.M.R.I.

Mr. Barnes referred to the possibility of using S02 instead of raw sulphur in the Sugar Industry, as increasing quantities of S02 were being generated by industry. He thought the possibility of using cold gases should be investigated more thoroughly. He referred to the use of liquid ammonia as a fertiliser and said it was not many years since this had been thought impossible, but today most of the nitrogen used in Louisiana was applied m the form of gas or liquid. He felt the same might apply to gaseous S02 in the Sugar Industry.

Mr. Dymond concluded by calling for a hearty vote of thanks to Mr. Hogarth for his excellent paper.

103

THE LABORATORY DETERMINATION OF MOISTURE PER

CENT BAGASSE USING THE DIETERT MOISTURE TELLER By G. D. THOMPSON.

The mechanism of the drying of solids can be divided into two main stages:

(1) The Constant Rate Period, during which stage water is evaporated from completely wet sur­faces. The drying process here is similar to the evaporation of water from a free liquid surface, and the rate is constant for constant drying conditions.

(2) The Falling Rate Period, during which stage the rate of drying decreases. The water content of the solid at the beginning of this period is known as the "Critical Water Content". When dried for an extended period, the water content of the solid approaches an ultimate value which depends primarily on the relative humidity of the air, and is termed the "Equili­brium Water Content".

The Constant Rate Period.

Since evaporation during this stage takes place at the surface of the solid, the rate of drying is limited by the rate of diffusion of water vapour through the surface air film out into the main body of air. The rate of evaporation is increased when heat necessary for vaporization is added by con­duction from adjacent dry surfaces and by radiation. For the most part, however, the heat is supplied by conduction through the surface film of air. The velocity of the surrounding air affects the thick­ness of the surface air film, and hence the rate of evaporation. The higher the air velocity, the thinner the surface film and the greater the rate of evaporation. In effect, the rate of drying is also proportional to the difference between the humidity of the air fed into the dryer and the saturated hu­midity at the temperature of the surface of the solid. Hence, the higher the humidity of the feed air, the slower the rate of evaporation, and, of course, the higher the equilibrium water content.

The Falling Rate Period.

This stage is generally divisible into two secondary zones:

(I) Zone of Unsaturated Surface drying, which follows immediately after the critical water content. The proportion of wetted surface area progressively decreases, but the mechanism of drying is essentially the same as that for the

Constant Rate Period, and the same factors influence the rate of evaporation.

(2) Zone where Internal Liquid Diffusion controls, which prevails when the resistance to internal liquid diffusion is greater than the surface resist­ance to vapour removal. It should be noted that, when internal liquid diffusion controls, air velocity has no influence on the rate of drying, and air humidity is of importance only in its effects on the equilibrium water content. Radi­ation and conduction of heat from the sur­roundings are of little effect except in raising the temperature of the solid and hence improving the diffusion constant of the liquid through the material. Internal evaporation may take place in pulpy solids, but the rate of diffusion of vapour through the solid and surface air film is the same as the rate of internal liquid diffusion.

Apparatus.

The Dietert Moisture Teller is a dryer of the hot-air type, a fan forcing a flow of air over elect­rically heated coils and thence onto the material to be dried.

Referring to the diagram, weighed bagasse is contained in the removable tray (A). The apparatus is operated by an automatic timing device (B). As soon as the timer is turned past the three minute mark, a motor, located within the body of the apparatus at (C), is automatically started. The motor drives a fan below it, which draws in air through the vents in the top cover (D). Electric heating coils, within the apparatus at (E) heat the air as it flows down over the bagasse in the removable tray. This tray has a fine sieve, 500 mesh, bottom, through which the moisture-laden air escapes at (F).

To insert the tray into its operating position, the frame (G) is pressed down at the point (H), a rubber tip. The tray is then placed centrally on the frame, and the frame allowed to return to the horizontal position under the tension of the spring (K).

Control of the hot air temperature is effected by means of the temperature controller at (L).

When the period of time selected on the atitomatic timer has elapsed, the motor cuts out and the heating coils are switched off automatically. The tray containing the now dry bagasse is removed and weighed.

104

Automatic timers are supplied for either 15, 30 or 60 minutes. A 30 minute timer would probably be best suited to a Sugar Factory Laboratory.

The fine mesh screen forming the bottom of the bagasse tray eventually tends to foul with fine particles of bagasse. This can easily be remedied by cleaning regularly (two or three times a week) with high pressure steam from a hose. The screen can be examined by holding it close to a source of strong light.

Experimental.

Routine laboratory bagasse samples were used for all determinations. These samples were com­posited over periods lasting one hour according to S.A.S.T.A. Recommended Methods of Control.

Effect of Temperature.

In an attempt to determine the effect of temper­ature on the rate of drying, samples were dried for successive three-and-a-half minute periods, until constant weight was obtained, at the following temperatures: 212°F., 221 °F., 230°F., 248°F., 257°F. and 266°F. (Tn practice the hot air temperatures varied in cycles over a range of 12°F., increasing and decreasing as the heating elements alternately heated and cooled. Recorded temperatures re­present the means of the extremes of variation.)

After constant weight was obtained, samples were dried for a further 8 minutes at 266°F., this being considered sufficient to effect complete drying.

TABLE I. Moisture % Bagasse (Wet Basis) at Different Temps.

Drying Rates (Grams moisture per minute) at Different Temps. Drying Time

(31/2 min. Periods) 212°F. 221"F. 230°F. 248°F. 257"!*'. 266°F.

1 9.23 9.37 9.06 10.09 10.14 11.34 2 3.51 3.63 3.43 3.57 3.83 2.77 3 0.89 0.77 1.06 0.77 0.77 0.26 4 0.29 0.14 0.23 0.14 0.09 0.03 5 0.09 0.03 0.06 — 0.03 0.09 6 0.03 0.03 0.06 — — —

As will be observed from these results, at no stage of the drying operation was a constant rate period apparent. This is not altogether surprising since the very heterogeneity of a bagasse sample would result in both constant and falling rate phenomena occurring within the first period of drying. Further­more, the progressive increase of humidity of the hot air as it passed through the bagasse layer, and the cyclic variation of the hot air temperature over the range of approximately 12°F., would preclude the required constant drying conditions.

The high drying rate in the initial stages at all temperatures may be interpreted as evidence of evaporation from totally and partially wetted sur­faces, the rapid fall in drying rate being due to the progressive elimination of totally wetted surfaces. The effect of variation of hot air temperatures during this period appears to have been comparatively small.

During the later stages of drying, however, there was a definite tendency towards complete drying in shorter time at higher temperatures. This may have been the result of higher temperatures causing an improved diffusion constant for the liquid through the solid material, and also simply a higher surface drying-rate on areas not directly exposed to the main air flow.

105

The practical implications are that drying a sample for 20 minutes without interruption at 266°F. should more than suffice to dry any normal bagasse sample completely. Decomposition at such a temperature is not significant since this is actually the temperature recommended by the Committee on Uniformity of the International Society of Sugar Cane Technologists1.

Effect of Particle Size.

The fact that drying rates, as shown in Table II, were not a function of temperature alone, promoted the idea that different fractions of the bagasse might have different drying characteristics. A sample was sieved through a coarse wire sieve with J inch square holes, and the two fractions obtained were then dried as above at 252°F. for successive drying periods until constant weight was obtained.

T A B L E III .

Moisture % Bagasse (Wet Basis) at 252°F. for Different Fractious.

T A B L E IV.

Drying Rate (Grams Moisture per Minute) al 252°F. for Different Fractions.

These results illustrate quite clearly the faster dry­ing of fine particles compared with coarse particles. The fact that the initial drying rates were very similar lends credence to the theory that the pro­longed falling rate period for the coarse particles was due to internal liquid diffusion controlling, for were it only a matter of there being areas of the

coarse particles not directly exposed to the main air flow, the initial drying rate for coarse particles would also have been considerably reduced.

In any event, it becomes apparent that the nature of the bagasse sample, i.e. the relative pro­portions of fine and coarse particles, will materially affect the length of time required for drying. In practice, then, the more thoroughly the structure of the sugar cane is broken down in the milling process, the greater will be the surface area freely exposed to the surrounding air, and hence the greater will be the rate of drying from the surface, and the shorter will be the distances through which internal liquid diffusion will have to operate.

Accuracy.

As a means of investigating both the accuracy of the apparatus and the reproducibility of results, one large sample of bagasse was thoroughly mixed. Six one-hundred-gram samples were then dried for 20 minutes each at 266°F. in the Dietert Moisture Teller. A further six one-hundred-gram samples were dried at about 225°F. in a laboratory oven. This procedure was carried out three times.

TABLE V.

Comparative Results, Moisture Teller (266F. 20 Mins.) vs. Laboratory Oven (225'F. 20 hours).

Test No. 1 2 3

Although a greater number of determinations would have been preferred, the above results are sufficient to corroborate results from overseas where this type of dryer has been used for some years with confidence and success.

The variation of moisture within a sample, despite thorough mixing, may appear significant at first sight. However, a considerably greater number of determinations can be carried out with the Moisture Teller than with the ordinary oven under normal conditions, the effect of which is to give greater significance to the average value for moisture per cent bagasse over any period, and consequently to the calculated value for fibre per cent cane.

106

Summary.

To ascertain the operating conditions of the Dietert Moisture Teller which would permit accurate moisture determinations on bagasse under any circumstances likely to obtain, preliminary experi­ments were carried out to study the mechanism of drying. The results are recorded in full, and indicate that the length of time required to dry a sample is not only dependent upon temperature, but also upon the relative proportions of fine and coarse particles in the sample.

The operating conditions decided upon following these investigations were to dry 100 grams of bagasse for 20 minutes at 266°F.

The accuracy of the results obtained by using the Moisture Teller was endorsed by results obtained by drying comparable samples in the Laboratory Oven for 20 hours at 225°F. A difference of 0.1 per cent, moisture between the averages for eighteen samples by the two methods satisfies the require­ments of the Laboratory.

The conclusions reached were that this apparatus forms an invaluable adjunct to laboratory equip­ment, and should improve the accuracy of pub­lished figures for moisture per cent, bagasse and fibre per cent. cane.

1Spencer, G. L. and Meade G. P. Cane Sugar Handbook, p. 564. John Wiley & Son Inc., New York, 1945.

Mr. Dymond congratulated Mr. Thompson on his paper and said he was particularly happy that he had been able to present it at this Congress. For many years past they had tried to get the younger men in the Industry to present papers at Congress and Mr. Thompson's paper was particularly valuable. He welcomed it too because of the method by which the determinations and figures had been concisely set out.

Mr. Bentley described the paper as most interesting but he could not reconcile the opinions given by Mr. Thompson that the moisture teller could give a greater number of determinations than obtained through the ordinary oven. The moisture teller could apparently do two samples an hour, while the oven operated at his factory gave five results in two hours. The other point he wished to make was that the line on which they should proceed was in the determination of moisture within two or three minutes rather than half-an-hour. His com­pany was proceeding with the electronic determin­ation of moisture in bagasse. When results were available he would let members know.

Mr. Thompson, in reply to Mr. Bentley, said that the time taken was 20 minutes and not half-an-hour. For most samples 15 minutes was sufficient, but he

gave 20 minutes to be on the safe side.. The results were actually available within a shorter time than by conventional methods. If this time could be further reduced by other means, so much the better.

Mr. Barnes suggested that the method of electronic determination was worth pursuing. He referred to tests carried out in New Jersey where it was astonishing to see the material drying as it were from the inside outward. A further point was that Dr. Wiggins in the West Indies had been studying moisture determination of bagasse and had used a liquid such as benzine with which the water could be distilled out. This method had been used for many years to test the water in grain. He suggested that more investigation should be conducted.

Mr. Rault said that the vacuum oven in Mr. Bentley's laboratory, although of the same design as the one operating at Mount Edgecombe, appeared to be more efficiently operated. For over 20 years in his laboratory the moisture content of the bagasse from individual mill units had been carried out as a daily routine control and this control was limited by the time taken for a moisture deter­mination. He had found that two hours in the Spencer type oven did not always complete the dessication. There was a risk of showing a moisture content half to one per cent lower than the truth if the time was limited to two hours. In sampling a substance such as bagasse, made up of particles of various sizes, he thought that the average of ten determinations on 100 grams would give a more representative picture of mill conditions than the testing of only one sample of one kilo, however carefully mixed and averaged this sample could be prepared. He thought that the tests carried out by Mr. Thompson proved the Dietert Teller to be a welcome acquisition, increasing the sharpness of control for special investigations, as it could give a moisture test within half an hour, but he would continue with the Spencer oven for the usual routine of hourly moisture determinations.

Mr. Dymond asked how many of these tellers were in use in the Sugar Industry.

He was informed that three were in use in Natal, two in Sugar factories and one at the Masonite factory at Estcourt.

Mr. Galbraith asked Mr. Bentley how long it took to get a result from the oven from the time the sample was placed in the oven, and pointed out that the great advantage of the drier under discussion was the fact that the chemist was able to give the engineer moisture contents within 20 minutes.

Mr. Bentley admitted that the moisture teller did appear to give results sooner when calculated from the time the sample was put into the oven. His factory operated two Spencer ovens and by

107

this means they did minimise the time taken in the tests. He also referred to the cheaper cost of the Spencer ovens as compared with the higher speed machines.

Mr. Dymond said that the results obtained at Darnall with the moisture teller were spectacular when compared with their somewhat antiquated type of oven. With the introduction of a new mill,

where moisture information was required in a short time, he felt that the machine teller would be of advantage. Mr. Dymond asked the meeting to accord a vote of thanks to Mr. Thompson. He hoped that in the future more of the younger men would come forward with the presentation of papers to Congress.

108

THE EFFECT ON THE CITRIC SOLUBILITY OF LANGEBAAN PHOSPHORITE (LANGEBAAN PHOSPHATE ROCK) WHEN

COMPOSTED WITH FILTER CAKE By H. E. KRUMM.

(Works Chemist, Bellville Fertilizer Works, African Metals Corporation Ltd.)

Contents.

(1) Abstract.

(2) Introduction.

(3) Materials.

(4) Experiments and results.

(5) Discussion:

(a) Citric solubility changes during composting.

(b) The effect of readily-soluble calcium com­pounds in filter cake on the citric solubility of Langebaan phosphate rock.

(6) Conclusions.

1.—Abstract.

Dymond's 1951 compost experiments with filter cake and Langebaan phosphate rock were continued in 1952. Composting increased the citric solubility of the phosphoric oxide in the rock, although not to the same extent as in the 1951 experiments. Citric solubility decreased with increasing quantities of phosphate rock. Field trials on sugar cane in Natal have been initiated to determine the avail­ability of the phosphoric oxide mixtures of Lange­baan phosphate rock and filter cake after composting.

The effect of readily-soluble calcium compounds in filter cake on the citric solubility of Langebaan phosphate rock is discussed.

The addition of small percentages of formic acid to the composting mixtures did not increase the solubility of the phosphate rock.

2.—Introduction.

Experiments conducted by G. C. Dymond in 1951, showed that the citric solubility of "Langfos" could be increased by composting with filter cake. (Krumm, 1952a). Other waste products of the Sugar Industry, such as cane trash, did not give the same effect. To determine, therefore, whether the results obtained with filter cake could be re­produced, Dymond decided to carry out a further series of experiments in 1952.

The experimental technique employed in 1951 was not changed.

The 1952 experiment, however, was modified as follows:

(1) Langebaan phosphorite was used instead of "Langfos". Chemically and mineralogically, these two materials are the same, but the phosphorite is lower in phosphoric oxide.

(2) The number of mixtures was increased from two to four, the additional mixtures containing 15 per cent, and 40 per cent, of phosphorite, respectively.

(3) Small percentages of formic acid were added to a separate series of three mixtures. Each contained 20 per cent, of phosphorite.

The results of Dymond's 1952 experiments revealed that the increase in citric solubility of the phosphoric oxide was smaller than that obtained in 1951. This cannot be attributed to the use of Langebaan phosphorite in place of "Langfos", but rather to variations in the composition of the filter cake.

The importance of the results, however, lies in the fact that a change did occur, and that it was due almost entirely to an increase in the citric phosphate solubility of the Langebaan phosphorite. It is probable that this change is of more signi­ficance than the figures tend to show, particularly in view of the limitations of the citric acid method for determining availability. (Krumm, 1952b). A field trial, recently laid out at Kearsney in Natal, using composted mixtures of filter cake and Lange­baan phosphorite, and planted to sugar cane, should help to elucidate this point.

In the formic acid series, the increase in citric solubility of the phosphorite was no larger than that of the control.

In the 1951 experiments, the citric solubility of the "Langfos" "actually found", was lower than the corrected figures. This difference has now been clarified.

3.—Materials. («) Langebaan Phosphorite.

At Langebaan Road, Cape Province, large quantities of phosphate rock occur which cannot normally be marketed as "Langfos", owing to its low phosphoric oxide content. It ranges from 9 to 15 per cent., whereas "Langfos" has been registered at 17 per cent, phosphoric oxide with the Department of Agriculture. It could, however, be classed as a Group II Fertilizer. Dymond used this material, in place of "Langfos", in his 1952 experiments.

(b) Formic Acid.

The citric solubility of phosphoiic oxide in mixtures of compost and phosphate rock, is stated to increase in the presence of stinging nettles. (Kolisko). Stinging nettles contain formic acid. Dymond decided to include formic acid in a series of mixtures to ascertain whether this could be used as a substitute for stinging nettles.

(c) Filter Cake. T A B L E 1.

Composition of the Langebaan phosphorite (A) and filter cake (B) used.

109

In the tables below the figures before and after composting are given. The analysis was carried out on the air-dried samples, but the results of each experiment have been calculated on a water-free basis.

Wagner's method was employed for determining citric soluble. (5 gms. shaken with 195 mls. 2 per cent, citric acid for 30 minutes.)

The concentration of Formic acid used in the above experiment appeared to have had no signi­ficant effect on composting.

5,—Discussion.

(a) Citric solubility changes during composting.

It must be clearly understood that when com­paring the results of citric acid soluble phosphate, before and after composting, two important factors have to be taken into account, namely:

(1) Loss of organic matter during composting. (This is reflected in the "Loss on ignition" figures in the tables.) The percentage of mineral matter is, therefore, increased as the organic matter disappears.

(2) Citric solubility is a function of the percentage of phosphorite in the 5-gm. samples taken for the determination. (Krumm, 1952a). Table 4 reflects this function.

The reason for the differences between the corrected Cit. Sol./Total P 2 0 5 ratios (citric solubility) and those actually found will be discussed below.

The citric solubility of mixture (iv) appears to have decreased. This probably represents no change, as the corrected figures clearly show that citric solubility diminishes with increasing weight of phosphorite. Therefore, since the amount of phosphorite is greater in the 5-gm samples after composting, it would appear that:

(1) The decrease in citric solubility of the phos­phoric oxide in the Langebaan phosphorite in mixture (iv) probably represents no change.

(2) The increase in citric solubility of the phos­phoric oxide in the Langebaan phosphorite in mixtures (i), (ii) and (iii) is probably greater than the analytical figures would appear to show.

(b) The Effect of Readily-Soluble Calcium Compounds in Filter Cake on the Citric Solubility of Langebaan Phosphate Rock.

Differences between the corrected citric solubility figures and the amounts actually found were first observed in Dymond's 1951 experiments. The reasons were not then understood. Work carried out in 1952, however, showed that the citric solubility of Langebaan phosphate rock is depressed by the presence of readily-soluble calcium com­pounds in filter cake. This effect will now be described.

It has been shown (Krumm, 1952b), that:

(1) The addition of calcium ions depresses the citric solubility of the phosphoric oxide in Langebaan phosphate rock.

(2) Citric solubility is a function of the pH.

110

REFERENCES. 1Kolisko, "Agriculture of Tomorrow", John Jenningi, Bruns­

wick Road, Gloucester. 3Krumm, H. E. The effect on the citric solubility of "Langfos"

when composted with waste products of the sugar industry. Proc. 26th Congress S.A.S.T.A., 1952, 100-105.

3Krumm, H. E. The limitations of the Wagner method (citric solubility) when applied to rock phosphate. S.A. Journal of Science, 49, 1953, 197-204.

APPENDIX, by G. C. Dymond.

Two series of composting experiments with low grade rock phosphate have now been completed and the results analysed by Mr. Krumm.

Observations on the behaviour of filter cake under aerobic and anaerobic conditions necessitate the continuance of these experiments before any finality can be reached.

In the experiments reviewed by Mr. Krumm the following figures are of interest:—

The amount of filter cake taken for each experi­ment was 1,800 lb.

During the period of three months the heaps were turned four times. Monthly averages showed a regular trend in the loss of organic matter and slow drop in temperature and pH values.

The range of temperature was from 58°C. to 38°C. and the pH from 8.0 to 6.8.

It was obvious that in general the decomposition was mostly anaerobic, except in. the top layers.

A nauseous smell of purification with sulphuretted hydrogen was very noticeable on each turning.

A further series of experiments on filter cake, given in the paper on Clarification showed that an intensive aerobic treatment gave essentially different results from a similar experiment under strictly anaerobic conditions.

Under the former there was a progressive but slow drop in moisture and organic matter with temper­atures rising to 75°C. The loss of cane wax under these conditions was over 70 per cent.

Under anaerobic conditions, there was no loss in either moisture or organic matter; temperatures never rose above 55°C. and the loss in wax was less than 20 per cent.

Table 6 shows that:

(1) Calcium contained in the filter cake was readily soluble in 2 per cent, citric acid,

(2) The citric solubility of the Langcbaan Phos­phate Rock was inhibited when the con­centration of calcium oxide in solution reached 11.8 per cent, and the pH rose to 2.55—2.60.

It would appear, therefore, that when a mixture of filter cake and Langebaan phosphate rock is shaken with 2 per cent, citric acid, readily-soluble calcium compounds in the filter cake pass rapidly into solution. A relatively high concentration of adventitious calcium ions as this established, and the solubility of the calcium compounds of phosphorus in Langebaan phosphate rock is, therefore, suppressed.

6.—Conclusions.

The citric solubility of Langebaan phosphorite increased when composted with filter cake. The change was small, but was probably greater than the analytical figures would tend to indicate. The Wagner method (crtric solumbility) was used to determine the solubility of the phosphorite before and after composting, but it would appear to have very definite limitations in this respect.

The increase in citric solubility was not as great as in the 1951 experiments. This could be ascribed to variations in the composition of the filter cake.

It is anticipated that experiments will be con­tinued, probably along new lines. It would appear that a more reliable method for determining changes in solubility of the phosphate is indicated.

Acknowledgments.

I have to thank Mr. C. W. Sharp, General Manager, African Metals Corporation, and Mr. C. F. Muller, Works Manager, Bellville Fertilizer Works, for permission to publish this paper.

I l l

112

Unfortunately no pH values were recorded, but as Mr. Krumm points out, "a relatively high con­centration of adventitious calcium ions is established and the solubility of the calcium compounds of phosphorus in Langebaan phosphate rock is, there­fore, suppressed."

It appears probable, therefore, that the high calcium content of filter cake is a major factor influencing the pH values during composting and therefore the increase in availability.

Future work will be conducted along the lines indicated above.

Mr. Dymond said he thought the experiments he had started with Mr. Krumm were worth continuing. When one considered the vast amount of filter cake available and the equally vast amount of low grade phosphate, it might be possible to improve the availability of supplies that would. give a greater amount of P 2 0 5 to planters. There was ample literature to indicate how quickly available P 2 0 5

became fixed in the soil and he felt that this question of fixation might be overcome by the series of experiments he and Mr. Krumm had conducted. Mr. Dymond said they were indebted to the Experi­ment Station for taking up this work and co-operating with the author and himself. He hoped the results would be made available in due course.

Mr. Rault asked whether Mr. Dymond had had occasion to use filter cake from the carbonatation process. This cake contained a certain amount of free alkalinity and its pH was decidedly higher than that of the filter cake of sulphitation factories— namely, about ten.

Mr. Dymond said he had not. He added that he had made the point already that the pH of filter cake was high. He had unfortunately not kept the pH values of the tests on filter cake under aerobic and anaerobic conditions, but he felt this point was well worth investigating.

Mr. Pearson expressed doubt whether any marked yield results would be obtained from the experiments being conducted by the Experiment Station. Various trials had been laid down including the ones with Langebaan phosphate. He gave details of some results obtained on plant cane. The tonnages of cane per acre suggested that the results obtained

from raw rock and Langfos did not measure up ,to the results obtained from super phosphate. He doubted whether any material difference would be found between the results from filter cake and from composted rock phosphate.

Mr. Main asked where Mr. Pearson had got the filter cake for his experiments. His point was that filter cake varied considerably in composition. Filter cake from the carbonatation process at Natal Estates near the Experiment Station would be totally different from that of a sulphitation process mill. Here also there would be different filter cake analysis where raw sugar was being produced with­out phosphates as against that from a sulphitation mill producing White Sugar where there would be, probably, a high P 2 0 5 content.

Mr. Pearson replied that the filter cake came from the Glcdhow-Chaka's Kraal mill and was rather high in pH.

Mr. Dymond said the type of filter cake used must, in his view, affect the yields obtained in the field. The object of this experiment was to take an insoluble material and increase the availability from a non-soluble to a more soluble material surrounded by organic matter.

Mr. Palairet asked whether the composting was done with filter cake only or with other materials.

Mr. Dymond said that concentration had been on the use of filter cake.

Mr. Palairet said that according to recent experi­ments, fixation was prevented and phosphates again rendered available in the soil by the addition of compost. As a result of these experiments he had applied phosphates mixed with compost on a soil rich in iron. Although this was not done under experimental conditions, the increase in the yield under these conditions was tremendous. He felt that press cake alone made too alkaline a compost and if trash was used as well the lower pH would help to remove calcium ions and so increase the availability of the phosphates.

Mr. Dymond said the experiment was still in an early stage and he hoped that results at Kearsney would throw further light on the work being done. He asked for a vote of thanks to Mr. Krumm for his paper.

113

MEALYBUG AND ITS EFFECT ON SUGARCANE By J. DICK.

Almost every field of sugarcane in Natal, as in most other sugar-growing countries, harbours popu­lations of the mealybug, Trionymus sacchari, Ckll. For several years investigations, aimed at determin­ing to what extent these insects affect the develop­ment of the cane, have been in progress at the Experiment Station. Although this work can not be regarded as completed, it is thought desirable at this stage to publish an account of the information so far obtained.

Effect on Germination of Setts.

An experiment described in 19426 showed that the presence of mealybugs on setts of Co.290 led to significantly poorer germination. Subsequent pot tests with Co.281 and N:Co.310 were somewhat in­consistent, the figures in Table I being representative of the type of result obtained. In this experiment setts, with or without mealybugs, were planted in tins, half of each batch being washed in water before planting. Fifteen single-budded setts were planted in each tin, and germinated buds were counted (on 26th September, 1950) five weeks after planting.

Required differences: 9.2 at 19 to 1; 12.3 at 99 to 1.

These figures show a significant depression only for the washed "mealybug" setts in N:Co.310 and for the unwashed "mealybug" setts in Co.281. However, if the results for the two varieties are combined, the totals are: mealybug 140; mealybug washed 147; uninfested 163; uninfested washed 162. It is doubtful whether statistical analysis of the figures, in this form, is justified but calculation shows required differences of 14.4 at 19 to 1 and 19.4 at 99 to 1, indicating significant depression in germina­tion in both mealybug and washed mealybug setts f6r the two. varieties together.

Residual Effect on Germination.

The mealybugs secrete a covering of wax, and also produce a syrupy "honeydew" on which a black, sooty mould usually develops. Cane which has been, infested with mealybugs can thus almost always be identified by the presence of wax and sooty mould, even when the insects have disappeared. In an attempt at gaining information on any possible residual effect of mealybug infestation, an experi­ment was carried out in which cane of the varieties N:Co.310, Co.281 and Co.301, with and without these symptoms, was planted in tins. Six replications were planted for each treatment, and each tin re­ceived twelve single-budded setts. Germinated buds were counted (on 23rd June, 1951) six weeks after planting; the results are given in Table II.

TABLE II.

Number of buds germinating.

C — clean setts.

XM — setts showing symptoms of previous. mealy­bug infestation.

31 9 23

Required differences: 8.5 at 19 to l; 11.5 at 99 to 1.

These figures indicate a significant residual effect in all three varieties, the effect on. Co.281 being, in this instance, less significant than that on the other two varieties.

During investigation of a case of bad germination, in N:Co.310 at Compensation, it was thought that mealybug infestation might have been a factor to be considered. The field from which plant material had been derived was inspected, and most of the cane showed symptoms of having been infested, although the insects had disappeared. Gane with and without these symptoms was collected from this field and planted in tins, twelve single-budded cut­tings being placed in each tin. The germinated buds

were counted (on 14th May, 1951) five weeks after planting; the figures obtained are given in Table III.

TABLE III.

Number of buds germinating.

N:Co.310 With mealybug Without mealybug

symptoms symptoms

8 9 7 7 7 8 7 7 6 8 8 6

43 45

The greatest relative difference in numbers of shoots between "clean" and "mealybug" plots was found three months after planting, the results being shown in Table IV.

TABLE IV.

Numbers of shoots—three months after planting.

These figures do not indicate any residual effect in this instance, and poor germination was probably due to some other factor.

Field Experiments.

A field trial was planted at the Experiment Station at the beginning of November, 1950, in 36 plots, each consisting of 4 lines, 4½ feet apart and 16 feet long. A line between plots was left blank, and a break 9 feet wide was left between banks of plots to delay, as far as possible, the spread of mealybugs from one plot to another. The soil in the breaks and in the blank lines was watered with an emulsion of sheep-dip in an attempt at preventing the spread of mealybugs by ants.

Three varieties of cane were planted, namely N:Co.310, Co.281 and Co.301. For each variety, clean setts and setts infested with mealybugs were selected from the same field, an attempt being made at obtaining planting material with as nearly as possible the same degree of infestation for each variety. Only the top sett, about 18 inches long, was taken from each stick. No fungicidal treatment was employed, and the cane was planted end-to-end (single stick) the trash being left on the setts to avoid disturbing the mealybugs. For each variety, six plots were planted with clean and six with infested setts.

Reliable information on the effect of mealybugs on germination was not available from this experi­ment, as it was found difficult to distinguish, in the field, between primary and secondary shoots. Obvious differences in the total numbers of shoots produced in the "clean" and "mealybug" plots were, however, noticed and counts were therefore made, at monthly intervals, of the total numbers of shoots which developed. These counts were continued until, about six months after planting, counting became impracticable.

Required differences: 429 at 19 to 1; 580 at 99 to 1.

At this stage, some cane of each of the three varieties, which had been planted at the same time and under the same conditions as the experiment, was dug up and examined. In the plots grown from infested setts, the insects were found to have dis­appeared from the setts themselves, but young mealybugs, presumably of a new generation, were found clustered at the bases of the primary shoots. It was thought that the presence there of these insects might have been responsible for the poorer development of secondary shoots. Plants from "clean" setts were still free of mealybug.

For comparison, with Table IV, the figures in Table V show the numbers of shoots counted five months after planting.

TABLE V.

Numbers of shoots—five months after planting.

Required differences: 587 at 19 to 1; 794 at 99 to 1.

114

It will be noticed that, although the total difference between "mealybug" and "clean" plots has been maintained, the relative difference has decreased with further development of shoots in both series. It will also be noticed that the differences for Co.301 were considerably higher than for the other two varieties. Statistical analysis of the figures showed that this difference in response between the varieties was significant (at 19 to 1).

Apart from these differences in numbers of shoots, there were obvious differences in growth, the plants grown from clean setts being considerably better developed than those from infested setts. In this respect also, Co.301 was more severely affected than the other two varieties. Although no numerical data are available to demonstrate this difference in growth the difference, for Co.301, is illustrated in the photograph, Fig. 1, .

Figure 1. Co.301, six months old; that on the right grown from setts infested with, mealybugs, that on the left from clean setts.

In spite of attempts at preventing the spread- of mealybugs between plots, all the cane in this field became slightly infested at the age of about eleven months. Apparent differences in growth, however, persisted for some time after this.

The crop from this experiment was harvested in July, 1952. The cane from each plot was weighed, the sticks of millable cane were counted, samples

from each plot were analysed by the staff of the Chemical Department, and the weight of sucrose per plot was calculated. The information is shown in Tables VI, VII and VIII. The weights of cane and sucrose are given in pounds per plot. Calculation of the results as tons per acre is not justified, since each plot was surrounded by breaks, and the yield was considerably higher than would have been expected if the cane had been continuous.

115

TABLE VI.

Weight of cane in pounds per plot. N:Co.310 Co.281 Co.301

Mealybug Clean Mealybug Clean Mealybug Clean

409 188 958

Required differences; 907 at 19 to 1; 1227 at 99 to 1.

As far as weight of cane per plot was concerned, there was thus a significant difference between "clean" and "mealybug" plots only for Co.301.

Required differences: 128 at 19 to 1; 170 at 99 to 1.

For sucrose yield there was thus no significant difference between "clean" and "mealybug" plots, although the figures for Co.301 approached sig­nificance.

There was thus no significant difference in num­bers of sticks of millable cane between the "clean" and the "mealybug" plots. Comparison of the above figures with Table V shows that, for the "mealybug" plots, the numbers of sticks of millable cane at the time of cutting correspond reasonably closely to the numbers of shoots present at five months after planting, whereas, for the "clean" plots, the numbers of sticks of millable cane are considerably lower than the numbers of shoots present at five months. In other words, in the "clean" plots a considerable proportion of the shoots failed to reach maturity. It is considered possible that, in this instance, the depressing action of the mealybugs kept the numbers of shoots within the saturation limit for the field, while the "clean" plots were able to produce shoots in excess of this limit. The possibility still exists that, under different growing conditions, a greater difference between "clean" and "mealybug" plots might be found at the time of harvesting.

Second Field Experiment.

The object of the second field experiment, which was planted in December, 1951, was to determine whether the depressing effect of mealybug infesta­tion on early development would persist after the insects themselves had been removed. The plots were of the same number and size as in the first experiment, and cane of the same varieties was planted. Infested and uninfested setts were selected, the top of each stick, again, being the only portion used. In this experiment, however, the setts were trashed, and both uninfested and infested setts were washed in water to remove all the insects before planting. Counts of the numbers of shoots develop­ing were again made, the numbers present five months after planting being shown in Table IX.

TABLE IX.

Numbers of shoots—five months after planting.

N:Co.310 Co.281 Co.301 Unin- Unin- Unin-

Mealybug fested Mealybug fested Mealybug fested washed washed washed washed washed washed

Required differences: 225 at 19 to 1; 304 at 99 to 1. Required differences: 267 at 19 to 1; 362 at 99 to 1.

116

117

The differences between "uninfested washed" and ''mealybug washed" plots were thus significant for Co.301 and N:Co.310, although not as great as when the cane was not washed. This demonstrates some residual effect from mealybug infestation. However, obvious differences in the size of the cane such as were found, particularly with Co.301, in the first experiment were lacking, and it is not expected that any significant differences will be found when this cane is harvested.

Inoculation Experiments. A sucking insect may influence the growth of a

plant either directly, by the mechanical effects of its puncture and the removal of juice, or indirectly, by the injection of some harmful substance, which may be either a pathogen or a toxin. To discover whether injury caused by the sugarcane mealybug might include injection of harmful substances, several experiments have been carried out involving inocu­lation of cane with the body-fluid of the insects.

In the first experiment, single-budded cuttings of N:Co.310, Co.281 and Co.301 were injected with an extract prepared simply by expressing the contents of the insects through muslin. This extract was not treated in any way, apart from dilution with twice its volume cf distilled water, which was found to be necessary in order to make it sufficiently fluid to pass through a hypodermic needle. Each sett was inoculated with approximately 0.2 cc. of the liquid before planting. A second scries of setts was injected with distilled water to enable any effects of mech­anical injury to be eliminated from the results, and a third scries was left untreated. The setts were planted in tins (on 17th January, 1951) each tin receiving twelve single-budded setts.

After four weeks, the numbers of buds which had germinated were counted; the results are shown in Table X.

TABLE X.

Numbers of buds germinating. C = Untreated setts. W = Setts injected with distilled water. M = Setts injected with mealybug body-fluid.

Required differences: 9.6 at 19 to 1; 12.8 at 99 to 1.

These figures appear to indicate a significant de­pressing effect produced by inoculation with the body-fluid of mealybugs in all three varieties, The effect on Co.30i is apparently more severe than on the other two varieties.

As this preliminary trial had indicated significant results, an experiment was carried out to test the effect of inoculation of setts with mealybug body-fluid treated in various ways, in an attempt at dis­covering whether the effect was due to a pathogen or a toxin. The variety Co.301 was used, and one series of setts was inoculated with fluid prepared from mealybugs which had been surface-sterilized with formalin to destroy any organisms which they might be carrying externally. Other series were inoculated with untreated fluid or with fluid which had been autoclaved or filtered through a bacterio­logical filter. Untreated setts and setts injected with distilled water were also planted. Six tins, each with twelve single-budded setts, were planted for each treatment. Counts of germinated buds were made (on 2nd May, 1951) four weeks after planting; the figures obtained are shown in Table XT.

TABLE XI.

Numbers of buds germinating—Co.301.

C = Control (untreated). W = Injected with distilled water. M = Inoculated with untreated mealybug fluid. A = Inoculated with autoclaved mealybug fluid. F = Inoculated with filtered mealybug fluid. S = Inoculated with fluid from surface-sterilized

mealybugs.

As no valid conclusions could be drawn from this set of figures, the experiment was repeated during the following September but, in this instance also, no significant results were obtained.

Discussion.

The ability of sucking insects to produce disease symptoms in plants, without the presence of a pathogen, has been recognised in a number of instances. Thus, Williams7 and Withycombe8 have concluded that froghopper blight of sugarcane is caused entirely by the feeding of Tomaspis saccharina Dist. and not by fungi, bacteria or viruses transmitted

by this insect. Similarly, mealybug wilt of pine­apples has been studied, particularly by Carter,1-5

and has been shown to be due to the feeding of Pseudococcus brevipes Ckll. and not to a pathogen transmitted by this insect, although some of the symptoms of mealybug attack on pineapples were thought4 to be due to toxins secreted by symbiotic bacteria within the body of the mealybug.

Carter3 has proposed the term "toxicogenic" for insects which are able to cause diseases by the injec­tion of toxic substances. Among such toxicogenic species, not all individuals may be capable of pro­ducing toxic symptoms; moreover, individuals may be able to produce such symptoms only at certain times. Individuals of toxicogenic species, when they are able to produce toxic symptoms, are, therefore, distinguished at "toxiniferous."

As far as the sugarcane mealybug is concerned, evidence so far available is too slight for the con­clusion to be drawn that this species is toxicogenic. However, the depressing effect on early growth caused by infestation of the setts appears to have been proved, and the persistence of at least part of this effect, when the insects have been removed, suggests that the direct effect of feeding is not the only factor involved. If the result of the first inocu­lation experiment is valid, the hypothesis can be suggested that the insect is toxicogenic. The negative results in other experiments would then be due to its not always being toxiniferous. Further investiga­tions on these lines are contemplated.

Acknowledgments. My thanks are due to the Field Staff of the

Experiment Station, particularly to Mr. F. Almond for supervising the planting and harvesting of field experiments; also to the Staff of the Chemical Department for analysis of cane samples.

Summary. The results are given of a number of experiments

on the effects of infestation by the sugarcane mealy­bug. It is shown that the presence of mealybugs on setts at the time of planting has a detrimental effect on early growth. This is indicated by the numbers of shoots developed as well as by their size. Some part of this effect may persist, even if the insects are removed before planting, which suggests that the mechanical effects of feeding are not the only factors involved. Experiments on the inoculation of cane setts with body-fluid extracted from mealybugs have given inconsistent results.

REFERENCES 1 Carter, W. (1933): The Pineapple Mealybug, Pseudococcus

brevipes, and Wilt of Pineapples. Phytopath. 23, 207-242. 2 Carter, W. (1933): The Spotting of Pineapple Leaves caused

by Pseudococcus brevipes, the Pineapple Mealy Bug. Phytopath 3, 243-259.

3 Carter, W. (1936): The Toxicogenic and Toxiniferous Insect. Science 83, 522.

4 Carter, W. (1936): The Symbionts of Psevdococcus brevipes in Relation to a phytotoxic Secretion of the Insect. Phytopath 16, 176-183.

5 Carter, W. (1937): The Toxic Dose of Mealy-bug Wilt of Pineapple. Phytopath. 27, 971-981.

6 Dick, J. (1942): Some Experiments on the Sugarcane Mealybug. Proc. S.A. Sug. Tech. Assoc. 16, 55.

7 Williams, C. B. (1921): Froghopper Blight of Sugarcane in Trinidad. Mem. Dep. Agric. Trin. & Tob. 1, 1-170.

8 Withycombe, C. L. (1926): Studies on the Aetiology of Sugar­cane Froghopper Blight in Trinidad. Ann, appl. Biol. 13, 64-108.

Experiment Station. South African Sugar Association,

Mount Edgecombe.

Mr. Dymond congratulated Dr. Dick on his paper. He said that he did not think anybody could say that there had not been variety in the papers presented to Congress. He felt that nobody could call himself a sugar man until he had taken on the chairmanship of the Association and had had to concentrate on the various types of papers that were presented.

Dr. McMartin said that this experiment of Dr. Dick's provided an example of an insect which was taken very much for granted as having no appreciable effect on growing cane. Agriculturally it had been ignored, although some effect had been previously found on milling. The figures showed that it might be possible to introduce a variety of cane which could be affected by the mealybug. He asked whether the incidence of mealybug was greater or less today than it was in the days of Uba.

Mr. Dymond said he felt that the incidence was very much lower. It was necessary today to look for mealybug, whereas in the days of Uba they could be found almost anywhere. In his view, they formed one of the greatest difficulties in clarifying Uba juice.

Dr. McMartin said that in the South African Sugar Journal of many years ago there was reference to a "gummy disease" in the Chaka's Kraal district. He wondered whether this was due to the effect of mealybug, or whether it was the disease known as gummosis. This disease existed in Mauritius from which many canes had come to South Africa and he felt it might have been introduced. Many of the older varieties were reported to have become gummy and he wondered whether this might have been due to the greater incidence of mealybug.

Mr. Dymond said that with the older varieties of cane after rain there was a swelling and a gummy material formed on the cane, which was due to mealybugs.

Dr. Dodds said that he noticed Dr. Dick had recorded the presence of mealybug clustering round

118

119

the base of young shoots. He thought this was unusual, as in the majority of cases the mealybugs were found clustered near the top of the shoot. He referred to Louisiana experience where mealy­bugs of a different species from ours, however, were found near the base of the shoot, which was con­trary to Natal findings. It was found in Louisiana that the principal detrimental effect of mealybug on the cane was to depress the purity of the juice at the mill, probably because due to the short growing season in Louisiana the juice purity was already uneconomically low. The only serious effect of mealybugs, as far as Natal was concerned, was apparently in the filtration of the juices. Otherwise there appeared to be no significant injury caused by mealybug in its present numbers.

Dr. Dick said the mealybug found in Louisiana was different from that found in South Africa. The existence of mealybugs at the base of primary shoots in the instance quoted was due to the fact

that the growth was comparatively small and the cane itself was not more than 18 in. high.

Mr. Rault confirmed what Mr. Dymond had said about the incidence of mealybug being greater in the days of Uba. He could remember that in the early twenties the manufacturing processes such as nitration and crystallisation were adversely affected by higher viscosity than at present, due to a large proportion of sticky substances in the nature of waxes and gums.' Mr. Dymond at that time had suggested the mealybug as the agent responsible for such viscous substances. Clarification processes were at present more thorough and more chemicals were also used. Wax determination on the filter cake of all South African factories had given him, at that time, a much higher percentage than at present.

Mr. Dymond asked that a vote of thanks be accorded to Dr. Dick for his most interesting paper.

120

Introduction.

The history of mosaic in Natal dates back to the year 1922 when Dr. H. H. Storey was appointed to the Durban Herbarium. At that time Uba was being grown commercially on a large scale, but the thick tropical canes, mainly varieties of Saccharum officinarum, e.g. Port Mackay, Rose Bamboo, Green Natal, etc., were also being grown to a limited extent, mainly by the Indians who were growing them for eating purposes. It was in the highly susceptible S. officinarum varieties that Dr. Storey first found mosaic in Natal. Uba in Natal is immune to mosaic but in the U.S.A. it has been found to be susceptible; this is not due to a change in the susceptibility of the plant itself but to the fact that different strains of the mosaic virus are present in different countries.

In 1922, then, Dr. Storey was confronted with an industry which was growing a large number of varieties varying in susceptibility to mosaic from very susceptible to immune. The obvious solution of this problem of mosaic was to grow only immune Uba which at that time was practically the only variety grown. By allowing only Uba to be grown it was hoped that mosaic would be eradicated from Natal. It was known that mosaic was present in a number of the grasses but it was thought that this would not play an important part in the spread of mosaic. However, in 1927 a law was passed prohibiting the growth of any varieties of cane other than Uba.

Introduction of Mosaic.

Sugar cane mosaic is not endemic to South Africa and it is thought that the disease was introduced from the Argentine in 1914, in a consignment of Java canes. At that time, there was a quarantine station in Durban, but the leaf symptoms of mosaic on some of these varieties, notably P.O.J.213, are very indistinct and it is probable that the disease slipped through the quarantine station. This is borne out by the fact that all the canes grown from this original consignment were infected with mosaic and the original seed probably came from diseased stock.

Period 1927-43.

The period 1927-43 was one of comparative freedom from mosaic in commercial plantations and the only disease of note was streak in Uba. Mosaic was still present but it was under control and looked as if it might almost be a disease of the past. How­ever, the yields of Uba were considerably reduced

by the high incidence of streak and a number of varieties were imported to possibly replace Uba. Co.281 was one of these varieties and its release for commercial planting was withheld for some time because it was found that it could be infected with mosaic. It was, however, eventually released and proved an extremely good cane and was cultivated for some ten years before any mosaic was found. In 1943 mosaic was first noticed in Co.28l at the Experiment Station.

Period 1943-52.

After the discovery of mosaic in Co.281 a survey of the Sugar Industry was made. Mosaic was found to be most prevalent in the Umzinto area and some fields were 60-70 per cent, infected, mainly in Co.281. The North Coast was not so heavily infected and only isolated cases of mosaic were found. This outbreak of mosaic was probably brought about by the presence of mosaic in some of the grasses and it was accentuated by the interplanting of maize with cane, which action considerably increased the number of vectors of the disease. (The vector of sugar cane mosaic, Aphis maidis, colonises on the maize plant.)

The position today is that mosaic is still being found on both the North and South Coasts, but generally the high incidence of mosaic in the Umzinto area is no longer present. Along the North Coast mosaic is still being found but in the majority of cases it is confined to a few isolated stools.

The writer has recently seen a couple of outbreaks of mosaic where the incidence of the disease has been exceptionally high. An investigation showed that the cane had been planted adjacent, to a maize patch. The maize plants were infected with mosaic and had been covered with aphids. There were a number of species of grass present, mainly a species of Panicum, which was heavily infected with mosaic. The conditions here were ideal for the rapid spread of mosaic, and resulted in the plot of cane becoming heavily infected with mosaic. Another plot of the same variety, growing about a mile away and in an area where no mealies were growing, was found to be free of mosaic. Some years ago a paper was presented to a meeting of this Association on the influence of maize on the spread of mosaic in sugar cane. It was definitely shown that the presence of maize in the proximity of growing cane considerably increased the spread of mosaic. Apart from any agricultural consider­ation the practice of interplanting maize and cane is to be condemned.

RESULTS OF MOSAIC TOLERANCE TRIALS By N. C. KING.

Experimental Work.

Recently, two mosaic experiments have been planted with the object of obtaining some inform­ation on the loss in yield due to mosaic. At the same time the secondary spread was investigated but it soon became apparent that the secondary spread (i.e. by Aphids) was so rapid that the original idea of comparing yields from healthy and mosaic canes had to be abandoned and yields from primary and secondary infections were compared.

Influence on Yield on Mosaic.

Experiment I was planted in January, 1949 and since then has been harvested three times. Table I gives the yields for the experiment.

with mosaic but from Table II it is seen that there is no difference in weight between the diseased and healthy sticks. These results show that N:Co.310 is susceptible but not tolerant to mosaic, while the varieties N:Co.291, Co.301 and Co.281 although the plots were almost entirely infected with mosaic the average weight of the sticks from healthy and diseased material was almost the same. These varieties then can be classified as being susceptible but tolerant. From observations it is apparent that the low yield given by Co.281 is not due to mosaic but to some other cause which at present is being investigated.

Experiment II was planted in January, 1950 and has been harvested twice. Table I (a) gives the yields for this experiment.

An analysis of the yields in Table I showed that there was no difference in yield in the varieties Co.281, Co.30I and N:Co.291, between cane which had been planted diseased and cane which had been planted healthy. In N:Co.310 there was a signifi­cantly lower yield from cane which had been planted diseased as compared with cane which had been planted healthy.

From Table II we find that the diseased sticks of N:Co.310 are significantly lighter than those which remained healthy throughout the experiment. From Table I it can be seen that the varieties Co.281, Co.301 and N:Co.291 were almost entirely infected

PHOTOGRAPH I. This Shows the Effect of Mosaic on the Sticks of N: Co.349.

121

TABLE I (a). YIELDS, IN TONS CANE PER ACRE, HARVESTED FROM EXPERIMENT II.

From Table I (a) the statistical analysis reveals that in the variety N:Co.339 there is no difference in yield between cane which was planted diseased and cane which was planted healthy. The varieties Co.331, N:Co.349 and N:Co.310 showed a signifi­cantly lower yield from cane planted from diseased setts when compared with yields obtained from healthy setts.

From Table II (a) we find that in the varieties Co.331 and N:Co.339 there is no difference in weight between diseased and healthy canes. The diseased sticks of N:Co.310 and N:Co.349 were significantly lighter than the healthy canes. These figures from Tables I (a) and II {a) indicate that N:Co.339 is susceptible but tolerant while N:Co.349, N:Co.310 and Co. 331 are susceptible but not tolerant to mosaic. The loss in yield in N :Co.310 and N :Co.349 was due to thinner and lighter sticks while in Co.331 the loss was due to there being fewer sticks in the diseased plots.

PHOTOGRAPH II. N: Co.310 19-months-old Plant Cane showing two lines on left

Mosaic infected, two lines on right healthy.

122

Photograph 1 shows what effect mosaic has on the cane itself. The mosaic infected sticks are much thinner, internodes shorter and more concave than the healthy cane. There were also mosaic stem markings on the rind which can be seen on the right hand mosaic stick.

Photographs 2 and 3 illustrate what effect mosaic has on the growth of cane in the varieties N :Co.310 and N :Co.349 respectively. Note the severe stunting in both cases, due to mosaic.

Infection and Recovery.

The series of graphs of Experiment I illustrate to what extent each variety has been infected with mosaic. In the series "planted diseased" it will

be noticed that very little healthy cane was pro­duced, with the exception of Co.301 (graph 8), which shows a certain amount of recovery from the disease but later these shoots were infected and eventually the whole plot was totally diseased. In the "planted healthy" series all the setts germin­ated healthy but in the varieties N;Co.291, Co.281 and Co.301 the shoots were rapidly infected and as shown in the graphs 3, 5 and 7 the plots were easily infected and even after seven months only about half of the plot was infected with mosaic. The spread of mosaic, although the chances of acquiring the disease were the same, was much slower than in the other varieties and can be con­sidered to be only moderately susceptible to the disease.

P H O T O G R A P H III .

N: Co.349 19-months-old Plant Cane showing two stools on left Mosaic infected, two stools on right healthy.

123

These graphs snow the number of cane stools infected with mosaic during the first seven months of Experiment I. (The broken line Indicates the rate of and total germination)

These graphs show the number of cane stools infected with mosiac during the first seven months of Experiment I I , (The broken line indicates the rate of and total germination)

PLANTED DISEASED PLANTED HEALTHY

N:Co.349 N:Co.349

126

Graphs showing relationship between recovery and age of cane.

PLANTED H E A L T H Y PLANTED DISEASED

The series of graphs of Experiment II follow the same pattern as those of Experiment I. In the "planted diseased" series all the setts germinated diseased and virtually no healthy shoots were pro­duced, i.e. there was no recovery. The "planted healthy" series shows the extreme susceptibility of N:Co.349 and N:Co.339. N:Co.310 is very similar to the one in Experiment I (graph 1). Co.331 (graph 13), shows the variety to be moder­ately resistant to mosaic and is very similar to N:Co.310, when secondary infection is compared.

So far it has been seen that there is very little recovery from mosaic in the plant cane crop. Graphs 17 and 18 show that in the varieties N :Co.310 and Co.301 there is considerable recovery in the first and second ratoon crops. The plant cane crop of N:Co.310, planted diseased, gave an in­fection of 85 per cent but in the second ratoon crop this percentage was reduced to 37 per cent. Corres­ponding figures for Co.301 are 80 per cent, and 49 per cent. In the plots which were originally planted healthy, N:Co.310 which in the plant cane crop was 13 per cent, infected was in the second ratoon nearly all healthy. The varieties Co.281 and N:Co.291 showed no tendencies to throw off the disease. This recovery in the ratoon crop is important in that there is less chance of secondary spread taking place as the source of infection is constantly being removed.

Conclusions.

We in Natal are fortunate in that although mosaic is potentially a serious disease it has been kept under control and at present is doing very little

damage to the annual crop. This satisfactory condition can only be maintained by the growing of suitable varieties which are tolerant to mosaic. The ideal position would be to grow only immune varieties but it must be realised that if this was done there would be very few varieties suitable for cultivation. One of the pathologist's problems is to test out new varieties against the more important diseases. One variety may be immune to mosaic but on the other hand extremely susceptible to some other disease. A variety such as this, although excellent from a mosaic point of view, would probably have to be discarded because of the other disease. It is seen, therefore, that a balance has to be found between the various diseases.

It we cannot find immune varieties what then is the solution? As shown earlier in this paper a number of varieties are susceptible but tolerant to mosaic, others are susceptible but not tolerant. In other parts of the world it has been found that the growing of susceptible but tolerant varieties is satisfactory, and there is no reason why the same system should not be adopted here. At present in Natal we are growing both tolerant and non-tolerant varities together and it is the writer's view that if tolerant varieties are to be grown then they must exhibit a certain degree of resistance. It is obvious that it is unwise to grow both tolerant and non-tolerant varieties at the same time.

Summary.

A brief review is given of the introduction of mosaic to Natal and the identification of the disease by Dr. H. H. Storey.

127

The mosaic position in Natal during the period 1927-53 is discussed. The finding of mosaic in Co.281 in 1943 marks the turning point, from a period of freedom from mosaic to a period when the disease is once more present in the cane fields.

Results of tolerance trials are given. The varieties Co.281, Co.301, N:Co.339 and N:Co.291 are very susceptible but tolerant while N:Co.310, N:Co.349 and Co.331 are susceptible but not tolerant. Three photographs illustrate the effect of mosaic on sugar cane.

Graphs have been drawn showing the recovery and infection of the various varieties under trial. In some varieties more recovery took place in the ratoon crops than in the plant cane crop and there are indications that N:Co.310 may completely recover from the disease if allowed to ratoon long enough.

REFERENCES. 1H. H. Storey, 1924: Diseases of Sugarcane of the Mosaic

Type in S.A. Part 1, Journal of Dept. of Agriculture, August, 1824.

2A. McMartin and N. C. King: Some factors influencing the spread of Sugarcane Mosaic in Natal. S.A. Sugar Technologists, 1948.

3E. M. Summers, K.W. Brandes and R. D. Rands: Mosaic of Sugarcane in the U.S.A. with special reference to strains of the virus U.S. Dept. of Agri. Tech. Bull., 955, 1948.

Mr. Adams said he had recently read "Humus, and the Farmer" by Sykes. He suggested that if the fertility of the soil was correct, then everything else was correct. He thought Mr. Dymond had written an article on mosaic in relation to compost. He asked Mr. King whether, in his experiments, he had tried any of the theories propounded by Sykes. In other words, had he planted diseased cane in soil which was 100 per cent healthy and sound.

Mr. Dymond said the subject had been brought up year after year and while he did not agree entirely with Sykes, he felt there was some found­ation for the theory he had propounded of healthy soils producing healthy plants. He was still experi­menting with the theories which Mr. Adams had referred to. His experiment was concerned mainly with streak. He had used Uba cane which was 60 per cent streak infected and in three years had brought the incidence of streak down to nothing. He had not had such good results with a variety of Uba cane that was stunted and diseased but he hoped that it would cure itself in time.

Mr. Barnes observed that mosaic had been a serious problem in most sugar cane countries, and still was important in many. Its menace had been countered by breeding resistant varieties. In the West Indies, only Jamaica of the British territories

experienced difficulty. The principal variety grown there, B34 104, was susceptible but tolerant, and though visible signs of mosaic disappeared with this variety after the early stages of growth, it was regarded as 100 per cent affected. In such cir­cumstances it was impossible to grow a susceptible variety on a commercial scale, whatever its merits in other ways might be.

There was no evidence in Jamaica that the use of farmyard manure" and other organic matter had any effect on the incidence of mosaic disease. In actual fact the epidemic of the 1920's and later was most severe in lands which had been manured with these materials for generations, and where the use of "artificials" was little practised until recent times.

He said that Mr. King had very properly been guarded in his final statement, but the word "dangerous" should replace "unwise."

Mr. King said the reference to the application of manure increasing the spread of mosaic was very interesting, but that in Natal there was not the opportunity of observing this fact as very little farmyard manure was used. The question of curing mosaic with compost had been tried out at the Experiment Station, but with no success.

Mr. Barnes said the question of the recovery of cane from mosaic was one that had confronted him for many years during his work. One very popular variety of cane in the West Indies was B. 34.104 which was accepted as tolerant. It seemed to throw off the symptoms of mosaic and these could not be detected after it had grown for some time. Yet that care was regarded pathologically as one hundred per cent infected. The disease appeared only in the very early stages and then disappeared with growth, which suggested that the variety could throw off the disease. He did not know whether this applied to other varieties.

Dr. McMartin said there were many instances of different varieties throwing off the effects of mosaic, and that the recovery was real, i.e. not merely a masking of symptoms. Some varieties including Uba were able to throw off streak quite readily, irrespective of the treatment accorded to it. The Experiment Station had been able to find no found­ation for the theory that is was possible to treat cane by raising the soil fertility and so enable them to throw off disease. He asked whether the chairman could define what was a healthy soil.

Dr. Dick asked whether once a cane had thrown off mosaic it was afterwards nearer immunity than the original cane.

Mr. King said that he knew of no actual case where mosaic had been thrown off and the plant remained completely immune afterwards. There were indications, however, that some sort of in-

1-28

creased resistance developed, but this could be due either to a change in the host or the virus itself. For example, P.O.J. 213, when it was grown commercially, was nearly completely infected with mosaic, but when healthy sticks were selected and planted in the collection of varieties very little mosaic was found.

Dr. Dodds said that although there had been no major losses from mosaic in this country, other countries had not been so fortunate. He had seen mosaic in many countries and in many different types of soil, but he could not say that he had heard or seen that any attention to the fertility of the soil had any effect on the incidence of mosaic. In Louisiana, mosaic had almost wiped out the sugar industry, but fortunately it had been able to recover and pull through. The relative amount of injury done by mosaic in different countries depended largely on the strains of the disease that happened to be prevalent; different strains of mosaic disease often varied greatly in virulence.

Dr. Dodds asked whether there were any strains newly developed in South Africa which more severely attacked sugar cane.

Mr. King said that about two weeks previously he had found, at the Experiment Station, two distinct patterns of mosaic leaf markings on the same variety. These leaf markings appeared to be the first indications of different strains of mosaic existing in Natal.

Dr. McMartin said that it had been found in Louisiana that if a cane recovered from mosaic no immunity was conferred. It could develop mosaic again. A mild strain might, however, give immunity from a severe strain.

Mr. Dymond said that he would have liked Mr. King's and Dr. Dick's papers to be presented at the Experiment Station, but this was not possible owing to the large number of papers. He felt, however, that the amount of discussion that had taken place clearly showed the interest taken and he asked that a vote of thanks be accorded to Mr. King.

129

By W. F.

This short paper is written at the request of the Agricultural Officer of the Experiment Station, to tie up with other papers being presented on different aspects of Sugar Industry work, and is an attempt to indicate the more immediate directions in which growers might profitably direct their energies with a view to improvement in crops and economy.

The paper concerns itself mostly with the mechanical approach to labour saving, but insofar as they affect the soil and its fertility, mechanical developments must be made in close touch with expert agricultural opinion. There are occasions when the direct apparent advantages of machine use can be offset in time by deleterious effects on the soil; while on the other hand, there are occasions when research indicates possible benefits to be derived from new mechanical techniques as applied to the soil, and it is the object of the several Re­search Organisations in this Industry to maintain that close liaison.

Again, this paper will trespass slightly into the realm of fertilization, but only insofar as fertilizing can be considered a labour saving agent, and it is one of the most important. Proper co-ordination of trash blanketing and fertilization policy is still the most outstanding labour saving "device" that has become available to the Industry in recent years, and is a subject which will repay infinite research. It is not a simple subject—on the con­trary it is highly complex, and introduces a com­pletely new set of factors into agriculture that are not perhaps, at present, fully appreciated. Never­theless, it is safe to say from experience so far gained, that we have in this the means to double the output of sugar per acre in this Industry in a very few years. I make this prediction in full awareness of the probable comments of the cynics.

Below, then, I will review very briefly some of the more promising innovations of method, with short comments where necessary.

Preparation by Chisels. Advantages: Better done in dry conditions, making

possible winter preparation of soils so that they are ready in the early spring. Spring rains not lost as with ploughs, which plough out the moisture. No inversion of soil, with consequent possible upset of bacterial life. No lateral displacement of soil —important on steep slopes. No revolving parts.

Disadvantages: Straw and trash obstruct. Un­suitable for badly drained ground.

i AHEAD C. JEX.

Possibilities: In well maintained soils, such as those trash blanketed for many years, organic content will be high, and friability adequate. These soils will not require extended weathering and exposure by ploughs for preparation of seed bed. Chiselling, supplemented by a rotary hoe, will give instantaneous seed bed, capable of being planted immediately, thereby limiting to a minimum the exposure time to the elements in this severe climate. Soils that are ploughed must have enough time to weather, and recover their balance, before they can be replanted to crops. It is during this period that they are most vulnerable to adverse weather conditions.

Planting. Practical advantages of mechanical planting proved.

Enables planting to commence before spring rains arrive, provided moisture is conserved under trash blanket until time of preparation as described above. Preparation done with chisels does not evaporate moisture, and machine can be used at any time provided sub-surface moisture is adequate. With efficient planting machines now being evolved, depth of cover can be accurately controlled for varying soil types, and varying temperatures. Planting machines must come back in Industry, and will stay. Enormous advantage to be gained by planting at optimum seasons, i.e. Spring, and this is possible with machines.

Weed Control.

Until chemical weed control for this country is better understood, it is essential that advantage be taken of mechanical control as developed in this Industry. Methods compare well with those used in any country, and are especially adapted to suit our more hilly conditions. The speed of mechanical planters must be matched by that of mechanical weeders, otherwise the advantage of the former is lost.

Trash Blanketing.

Trash blanketing has come to stay, and is a labour saving device of the ideal variety. It is simple, cheap, and readily available.

Fertilizing. With trash blanketing, ratoon fertilizing becomes

well worth while, and results do not have to be assessed on a delicate balance—they are obvious to the eye. Fertilizing becomes a "must", and saves its cost many times over in reduced weeding, larger crops, and longer ratoons. The simplest

130

way to apply the fertilizer is to use granulated types, and cheap machines such as the Maxim. These, too, have come to stay.

Loading and Transport.

Labour pressure is only beginning to make itself felt on present methods. For short hauls, expensive lorries are being superseded by more economical tractors and trailers. These go far enough at the moment, in that they economise labour by elimin­ating long carries. The use of two or three detach­able trailers makes possible better tractor efficiency, and as labour pressure increases, self loading trailers will come more into the picture. Two practicable types have been evolved and are on the shelf awaiting demand. The demand is probably not far away now. To get utmost efficiency out of trailers, field lay-out must be considered, and future needs anticipated. Trailers can not cross open drains or obstructions when heavily loaded, nor can they abide bogs. Sub-surface drainage, with tiles, makes it possible to take trailers anywhere, and this should be commenced with any new plantings. Similarly, a trailer loaded to the full capacity of the tractor, can not be towed successfully uphill. Layout must therefore be arranged so that tractors come into the fields at the hill tops— pick up the load on the down slope, and pull away down hill. The hardened roads must be at the bottom of the hills, and if the valley drain is tiled, the old drain can then be used as the road—at the very bottom of the hill. In this way, maximum efficiency can be got from tractor and trailer loadings.

Transloading.

Gantries are proving their worth for transloading into gollovanes, and their use is increasing daily.

Their cost, since they do not need to be mobile, is well within present economics. Machinery 'for transloading trailer loads into S.A.R. wagons is much more expensive, however, and is not economic at anything less than say 100 tons per day, which is beyond the scope of most growers at present. Any reduction in available labour will, however, increase the necessity for conserving the labour at present used for hand loading S.A.R.'s, and the only possible approach that can at present be visualised is the joint purchase by several growers of suitable cranes. There does not seem to be any immediate prospect of producing a mobile crane to handle trailer loads at an economic figure, where only one or two railway wagons are loaded daily.

Harvesting.

This will be the final, and most expensive, item on the list of mechanical developments. Work is progressing steadily towards producing one suitable for our conditions. Mechanical harvesters are available today that involve the burning of the cane, but the economy and general benefit to be derived from trash conservation in present cir­cumstances, make it highly improbable that the use of these would be justified. Mechanical harvesting is still a year or two away for this Industry, and field lay-out will have to be prepared ahead for its use, on the lines indicated above. In the mean­while, general use of the machines mentioned above should relieve labour pressure for a few years yet, and even if the smaller producers can not yet afford the capital cost of equipment, effective use by larger producing units will release labour to the general pool, and so ease the demand on labour for the next few years.

131

SOME OBSERVATIONS I CANE G

By H. L.

I have been asked to write on the above subject and I hope what I have to say will be interesting and at the same time constructive to the growing of sugarcane in this country.

There are three main principles to be carried out and they are:

(1) Trashing or the conservation of trash, mulching, etc.

(2) Fertilizing every acre that is cut each year.

(3) Green manuring.

There has been a lot written and said about the first subject. This practice only started about three years ago and, to say the least of it, it has been revolutionary in agricultural practice in this country. Leaving a blanket of trash and applying the fertilizer on top of the trash was first mooted by the Experi­ment Station and I remember telling the then Director that some of his staff should be in the Lunatic Asylum. However, after seeing a practical demonstration and the consequent result, I was convinced it was the right method and from then we changed the policy of parting trash and culti­vating to the "blanketing of trash". This has saved us enormously in cost, as you can imagine the labour entailed in pulling the trash into alternate lines and the number of men and animals used in cultivating over a large area. The results of this method have been fully justified and in the last three dry seasons we have experienced, the "blan­keting of trash" has kept what little moisture we have had in the ground and not dried up, as when the trash is removed and the ground exposed.

This trash is invaluable to us in this country, with high winds and low rainfall. Enough cannot be said on this subject and perhaps one could go on indefinitely, but I would just like to quote Mr. W. F. C. Jex's article in the October issue of the South African Sugar Journal on trash blanketing and fertilizing which explains the advantages, with which I fully agree.

I might illustrate further the value of trashing when I tell you that my company averaged 34.5 tons of cane per acre on a crop of over 330,000 tons of cane and this after three years of drought or low rainfall—for instance, in 1950 we had only 1| in. of rain in five months between April and September and in 1951 only 2 in. in the same period. In my opinion we would never have averaged this figure if it had not been for this invaluable covering of trash.

N CONNECTION WITH ROWING GARLAND.

Again to illustrate my point of the value of trash, we planted a record area of 4,000 acres this past season and we have never had a better standard of plant—apart from the favourable season this is largely due to trash left on the ratoons as the labour and cultivators which were previously employed on parting trash weeding and cultivating the ratoons is now concentrated on the plant cane with the result that the plants are kept free of weeds and do not get a setback in any way. (I can safely say that due largely to trash mulching our fields today are in a better state of tilth than they were 35 years ago.)

Fertilizing Every Acre that is Cut Each Year.

When I went to the Sugar Conference in Puerto Rico over 20 years ago, the company's fertilizer bill did not reach £4,000 a year, as the policy in those days was to fertilize plant cane only and perhaps a small area of first ratoons. After my visit to Puerto Rico I realized the value of fertilizing and on my return our fertilizer bill jumped to over £20,000 and today it is well over £40,000.

While on this important subject of fertilizing I would like to sound a note of warning in regard to the amount of fertilizer to be applied. In the sugar belt, where we have a limited rainfall and experience severe droughts, over-fertilizing can have a depressing effect, especially in the case of straight out nitrogen such as sulphate of ammonia, in dry areas.

There is an economical limit in the amount of fertilizer to be applied and this cannot be deter­mined solely by soil and leaf analysis. Soil and leaf analysis can be a guide, but must be worked in conjunction with field trials and by practical experience. No man can tell just by soil and leaf analysis, especially in this country where we have prolonged dry spells, exactly how much fertilizer to apply—it can only be a guide. There is a danger that planters will over apply, especially in the dry areas.

I have been in close touch with our Experiment Station since its inception over 24 years ago and they have given me valuable assistance over this period in advising and guiding me in our fertilizer policy.

Green Manuring.

In my opinion after a field has been under a crop of cane for eight years it is essential to give it a rest or fallow and put it under green manure crop.

132

We have practised in the past "short fallowing," i.e. cutting and planting a field the same year with­out a green crop. This may do in a country with a high rainfall such as Hawaii but here, with our limited rainfall, it is important to plough in a green crop which all adds to the conservation of moisture.

Mr. C. H. 0. Pearson, in his papers read at the annual Technologists meeting last year, raised some very contentious subjects and among these was the question of the value of green manuring. Where he maintained that the value of the green manure was lost after a certain period when ploughed in—

this is a new line of thought, but I cannot subscribe to this theory in this country with its low rainfall.

The best green crop to grow is sunnhemp; if that is not obtainable then velvet Somerset runner beans. Various other beans have been tried over a period of years but without success. Lupins have been tried recently, but only with varying success.

There are many other equally important items in the growing of sugarcane on which one could write a book—but for the purpose of this paper I have stuck to the three main principles, which is sufficient to raise an argument.

CANE PRACTICES AND PRODUCTIVITY By C. T. WISE.

I was asked to submit a paper for this year's meeting and this I have done, not in the light of posing as a cane expert, but in the hope that this paper may form the basis of a discussion from which points of mutual benefit may arise.

With the post-war rise in the price of cane-lands, and the almost daily rise in the costs of production, it is exceedingly clear that, at the present price of cane, which has not been increased in proportion to production costs, we must increase the pro­ductivity of our soil if we wish to maintain a reason­able margin of profit.

The S.A. Sugar Association has inaugurated a Fertilizer Advisory Service and when this is really functioning properly, and soil and leaf testing become general throughout the industry, I feel that the overall increase in yields will be considerable. This service is, however, still in its infancy, and without resorting to these technical aids there are certain field practices which, if carried out, can assist us in increasing productivity. I will attempt here to explain some of these practices.

Previously the mills started up in May or June and it became general practice to start the season off by cutting fields which were down to be ploughed out. Rightly so, too, June being, as it is, the worst month in the year for germination. In later years, however, conditions have changed and the emphasis seems to be on early starting, some mills getting off in the first weeks in April.

Owing to habit acquired over many years, a great number of planters and miller-cum-plantcrs still abide by this practice of starting the season on old fields for ploughing out. This I feel is bad practice as one gets into one's better cane in the bad germin­ation (or ratooning) month of June, instead of taking advantage of the month or month and a half of growing weather before winter. On the other hand by cutting good vigorous fields and taking advantage of the remaining growing weather one can, almost without fail, cut that cane again at the end of the following season as age 19—20 months, thus in­creasing productivity by cutting two crops in three years.

I have found too that my vigorous cane gives, over those early months, better sucrose than the old cane.

Last season, for instance, I decided to go against all known practice and cut in April-May a field of plant Co. 301, twenty one months old. In spite of

having had 7 1/2 in. of rain during those two months it yielded 1,024 tons (39 tons per acre) at an average sucrose of 13.58 per cent, as against second and third ratoon fields of N:Co.310 and Co.301, which averaged 12.56 per cent, sucrose.

These fields were cut over the same period as I was operating two separate cutting gangs.

I have also had good results cutting and ploughing one-year old fields in December, when the sucrose is dropping. This is not too late to plant a cover crop of sunnhemp since by planting early the following season one is thus able to re-establish in a very short period and you still have the advantage gained by a nitrogenous cover crop.

Another old and established practice is one which almost reached the proportions of a must and that is that one should complete one's planting by December or at latest mid-January. In the last few years, however, it has been proved that very successful results can be, and have been, achieved, by planting in March and even early April. This practice, I feel, will become increasingly popular, obviating as it does the necessity of short-fallowing, which in my opinion should never be done unless absolutely necessary, and then only if one is able to compost heavily.

By adopting this method, one is able to plough out in June, plant sunnhemp in October (or velvet beans earlier), plough the green crop in in January, and re-establish cane in March. This cane is ready for cutting at the end of the following season and one has again thus obtained the benefit of a nitro­genous cover crop, and only lost a few months in re-establishing, instead of a whole season under the old method with the land lying for months un­protected against the elements.

A further advantage of this cane planted in March is that, if it becomes necessary to leave it over till the next season, it can be left with very little, if any, deterioration.

This late or early planting has slight disadvan­tages, in that one probably requires a little more seed for supplying misses, though this is not always the case. Then again it requires a little more weeding to keep it clean over the winter months. These, however, I feel, are offset by the fact that, by the time your early rains come your field is already established and the cane goes away with a bang.

134

While on the subject of re-establishment of cane, I would like to add one thing further on the times of planting. It is my opinion that canes which are susceptible to arrowing should not be planted too early in the season. By observation over the last four seasons, I have come to the conclusion that the said canes, when planted in August, September, usually arrow twice. By planting in November, December one escapes, to a large extent, the first arrowing, and this cane when cut has very much fewer dead sticks, with the resultant drop in tonnage, than the canes planted early in the season. This is true of cane planted in March, April.

In conclusion just a word on compost. I mentioned earlier that short-fallowing should

be avoided at all costs unless one is able to give the field a heavy application of compost. It is my experience that fields which have been short fallowed never really give of their best, irrespective of one's post-planting treatment. By heavy application of compost I mean a minimum of 20 tons per acre.

I think that most of us are inclined to try and make one's compost, or filter press cake, go too far, instead of doing a lesser acreage well. By this mal­practice we get a depressing effect instead of in­creasing our yields and productivity.

SUGARCANE GROWING IN NATAL

FIELD METHODS AND PRACTICES IN OPERATION IN 1953 By C. H. O. PEARSON.

In 1949 a paper by C. D. Sherrard was read at the Technologists Congress under the title "Sugar­cane Agriculture for Beginners". It is now five years since this paper was compiled and the industry lias adopted certain modifications in its general field practice during this period. A survey of the practices today in the fields, and a study of the progress made during the five years, will not be without its advantages.

Some point in the growth of the cane must be taken for a start of such descriptive analysis and the most convenient is when the land is under an old ratoon crop of cane.

Preparation of Field.

It is the usual practice to cut the fields, due for ploughing out, as soon as the mills open in May. This is more pronounced in the higher altitudes. The field, having been cut during late May, June and early July and the trash blanketed, is left until the first spring rains. It is often only possible to plough after rain has softened the ground and this is certainly the case where the last crop of cane has been burnt standing. It is not the case to the same extent where the trash is left, as the underlying soil is generally moister due to the trash mulch. A wetting of the trash by rain will consolidate this extremely tough material, thus helping when ploughing operations are started.

Even under the most favourable conditions of rain the trash and old stools of cane are not easy to plough completely with one ploughing. It has been the practice for some time to plough the old cane field with disc ploughs, simply bringing up enough soil to harrow and enable a green manure to be sown and germinate. This first ploughing, as a result, often has a ragged, untidy appearance and no alarm should be felt if the field takes on such an appearance.

The period of growth of the green manure will allow sufficient time for trash and cane root to soften and rot, so that when the green manure is ploughed in, about March of the next year, it is possible to make a clear, good ploughing.

Crop Change Over.

Reference has already been made to green manure crops and this practice is extensively carried out using sunnhemp or velvet beans for the purpose.

Sunnhemp, which is sown usually at the rate of 30 to 40 lbs. of seed per acre in December, and velvet beans at the rate of 80 lbs. per acre, grow during the rainy summer period very rapidly and are at their maximum growth about March of the following year. Beans are preferable for earlier sowings, during October and early November, whilst sunnhemp sown in December gives a very rapid cover. January sowings of sunnhemp seem to be preferable if seed is to be collected but despite the high price of sunnhemp seed, the practice of saving seed is not general in Natal, due to the unreliability of the sett and the damage caused by insects and fungus diseases.

Whilst sunnhemp invariably is sown broadcast for a green manure, velvet beans are sometimes drilled in so that the land can be scuffled between the rows until the beans, which are siow to start vigorous growth, have made a complete cover.

It has always been thought that a legume crop would improve the soil and enrich it with nitrogen. This theory has never been proved in the cane belt and, in fact, evidence is being collected which shows that no manurial advantage is gained by the sub­sequent cane crop. This is more pronounced where the more orthodox method of planting cane in August to December is in practice. The green manure having been ploughed in in March, the field is left as a bare fallow for 5 to 9 months and under the high midday temperatures experienced on the coast, even in winter, most of the organic matter and some of the chemicals would appear to be lost through oxidation and bacterial action. The only apparent advantages to be gained by green manuring are: (I) The better tilth created by the ploughing in of vegetable matter; (2) The control of certain insect pests by the period when no cane host is in the soil.

It is felt that, by suitable mechanical methods, a similar tilth might be produced to that created by green manuring, and, if any manurial advantage is gained by the practice, such gains could be rectified more expediently by the use of suitable chemical fertilizers.

Research work is progressing along the line of eliminating, where possible, the need for a green manure, but at the moment, because of insufficient evidence to support any alternative method, the

136

practice of green manuring between crops must of necessity be maintained to allow the old cane roots and trash to rot and to create the seed bed condition necessary for planting.

It may be born in mind that it is considered best at all times to keep the soil under some sort of cover. Under present conditions and planting of cane in March, two to three weeks after ploughing in the green manure would seem the best time to take advantage of whatever manurial residues there may be attributed to a green manure.

To try to eliminate, any period of bare fallow certain planters have been attempting to follow the sunnhemp or velvet bean crop with another crop to fill the period March to September. Lupins have been tried with variable success and it would appear that considerably more research work will need to be carried out on this crop before any degree of certainty of a satisfactory stand can be guaranteed.

Whereas with sunnhemp and velvet beans no bacterial inoculant to start the formation of nitrogen fixing bacterial nodules on the roots is necessary in the coast soils of Natal, such inoculants are very necessary in the case of lupins. The bacteria associated with lupins is of a specific nature for this crop and the method of inoculating the seed needs care, as exposure to light for only a short period is damaging to the life of the bacteria. Lupins seem to be extremely susceptible to attack by eel worm which is present in most of the soils of the cane belt and it has been noted that better results are obtained in the growth of lupins when a sunn­hemp crop, which plant has a repellent effect on eel worm, is grown directly prior to lupins being sown.

Time of Planting. Planting can be carried out when a suitable depth

of soil has been prepared as a seed bed and when this soil contains sufficient moisture to allow germin­ation. This set of conditions usually occurs after the first rains in August or September and again in March. It is considered that the period January and February is not too good for planting, as the open furrow into which the seed material is placed rapidly drys out on the surface and the setts are thus placed in a hot dry medium. At this time of year any cane that germinates makes too much soft growth before the dry winter period sets in and cannot withstand this subsequent drought so well as the smaller, tougher plant that results from a March, April planting. Winter planting during June and July often has dry conditions to contend with in the early stages of growth, but where over­head irrigation is applicable, or the occasional winter rain occurs, winter planting is possible despite the cooler conditions.

It has been found by experimentation and in practice that if cane receives a really good start

there is little difficulty in maintaining good ratoon crops, but where the cane receives a bad start little can be done to effect a substantial recovery in the ratoon crops.

Mention must be made of the planting machine when dealing with the time of planting because if dry conditions are experienced when it is desirable to plant, the opening and immediate shutting action of the furrow by the planting machine and the resultant placing of the sqtt in moist soil conditions considerably helps the germination of the cane. These conditions cannot be achieved by hand planting.

In the higher areas of 2,000 ft. above sea level the period of planting is restricted. It is considered that October is the optimum time for planting, with little latitude into September and November.

Planting.

Having decided on the time to plant and having prepared a good seed bed the actual operation starts with the process of marking and then fur­rowing out. Unless the slope of the field is extremely moderate it is advisable to mark the contour of a similar level and to correspond the lines of cane to these contours. As the gradients become steeper, greater care must be taken in marking out the lines of cane and it is as well to allow a slight slope in one or other direction so that rain water does not accumulate in the line, break through the ridge and thereby start soil erosion. If the slope of 1 in 250-300 is introduced, the water from each furrow can be led away down the slope at a road or other hardened drain.

Where irrigation by furrow methods is con­templated the lines must conform with the require­ments of the supply of water.

The slope and conformation of the field having been accounted for, the land is opened up for planting by a ridging body drawing a furrow 6 to 8 inches below the level surface of the soil and at intervals 4 ft. 6 in. apart. Any fertilizer required is spread along the bottom of the furrow, the cane sett is then put in. position and covered to a depth of 2 inches. It is advantageous to firm the sett in and not enough attention is paid to this process of firming the sett. Air must be reduced to the minimum in the immediate vicinity of the sett to prevent drying out and the moisture in the soil should be brought into close proximity with the sett to encourage germination.

Spacing between Rows and Seed Rate. It is the general practice to plant in rows 4 ft.

6 in. apart. This standard distance may be varied according to the fertility of the soil and the type of cane being planted. In rich flat soil, with a

137

vigorous variety, the rows may be placed 6 ft. apart, but where the soil is poor and the variety upright in growth habit then the inter row may be reduced to 4 ft.

The setts, if prepared before planting, may be placed either as two setts lying alongside one another or with a space between the ends. It was thought that a greater number of sticks per acre were produced when two lines of setts were placed in the furrow. It has been shown by experimental evidence that the same weight of cane is produced from a continuous hne of setts as from a line where setts were placed with 3 ft. between their ends. Counts of sticks in this experiment show that the greater the number of sticks resulting from the continuous planted row were, individually, of a far smaller size that those in the wider spacing treat­ments. It is the practice generally that, when planting a good germinating variety, a gap of 9 to. 12 inches is left between sett ends. Where, however, the germination is not so free, the practice is to place setts slightly overlapping. The amount of cane required to plant an acre when setts are placed double is 4 tons. By spacing 12 inches between sett ends this quantity is reduced to 11/2 to 2 tons per acre. It can be appreciated that when large areas are to be planted a considerable saving in seed material can be effected with the spaced planting.

In connection with space planting, an operation carried out after the germination of the sett may be mentioned here. It has always been the practice to "supply" the area planted with fresh setts wherever germination has failed. It used to be considered necessary to carry out this process where-ever a hoe could be inserted between the growing shoots of cane. Because of the uniform yields recorded from the wider spacings in experimental plots it has been established that no gap in the line of cane under 3 ft. in length need be "supplied".

Selection of Seed Material and Its Preparation for Planting.

Seed material consists of the stick of cane cut into suitable lengths and in such a condition that the buds, occuring at each node, are plump, vigorous and likely to germinate quickly.

To obtain the buds in the best condition young cane is selected. By experimental trial it has been found that the recently produced top of 2 year old cane is equally as good for seed material as cane 12 months old. This fact is important during drought conditions when the growth of cane up to 12 months old is small, as it is possible to use the tops of cane being sent for crushing as seed material.

It is more convenient, however, to use the longer sticks of 12 months old cane and so plant cane or

first ratoon cane is usually used. This cane is cut at ground level and to obtain the speediest germin­ation the trash should be removed and the stick then cut into 15 to 18 inch lengths. The length of 15 to 18 inches is taken so as to ensure that there are at least 3 buds or eyes on each sett. It has been found that where the three bud sett is used the two end buds can be damaged by insect or fungus attack whilst the centre eye germinates satis­factorily.

To prevent the attacks of one fungus disease, pineapple disease, which attacks the setts placed in the ground by entering the cut ends, it is usual to dip the whole sett in a water soluble organic mercurial fungicide. Various proprietary brands of the fungicide are on the market and the solution is made up by mixing one ounce of the powder in a gallon of water. It is useful to bear in mind that a match box full of the powder is roughly one ounce.

If conditions are extremely favourable for germin­ation, namely the soil being nicely moist and the temperatures on the high side, the older method of planting may be adopted. This consists of leaving the trash on the stick, laying the whole stick in the furrow and then cutting it across at suitable intervals. This method speeds up the planting process but retards the appearance of the shoot above ground and in any but the optimum planting conditions exposes the sett to undue risk of attack by insect or fungus. The retarded appearance of the shoot above ground leads to expensive weeding at a later date.

Weeding of Plant Cane.

The eradication of weeds has, due to the practice of trash blanketing the fields, been reduced to those occuring in the plant cane only. By the correct management of the available labour and suitable use of mechanical and horse drawn implements, these weeds should be dealt with as soon as they appear through the surface.

As little hand weeding as possible should be resorted to. The young cane first shows through the surface in the bottom of the planting furrow. In this stage the weeds in the row can be kept in check by the use of a small, very light, semi-circular spiked harrow fitted with a guiding handle and drawn by a mule. It is the object at this stage to preserve the form of the furrow until the cane is somewhat larger.

When the cane is 12 to 18 inches high, a light scuffle can be taken down the ridges to drag out the weed growth from the inter-row that may not, until this stage, have been touched, and to fill in the furrow. By repeated scufflings the furrow is filled in entirely and weed growth is either smothered by

138

the filling in process or pulled out by the tines of the scuffles. Generally one hand weeding at least is necessary in the row of cane and it is felt that with the judicious use of an ormone weed killer the optimum conditions might be achieved when only one hand weeding is required before the cane can furnish sufficient canopy to check any further weed growth.

Cane, once it has germinated and formed its first leaves, is extremely tough and will withstand a considerable degree of rough handling. In some cases the early cultivations are carried out effectively by drawing sets of zigzag harrow or chain harrows over the row and inter-row.

As previously mentioned cane, given a good start, will ratoon satisfactorily and any check such as might be given by competition with weeds in the early stages of the plant cane crop may have a considerable effect on the whole crop. The old saying that the right time to weed is before weeds are visible is especially applicable to the plant cane crop. It is, however, very difficult to compete with the rapid growth of weeds during the wet summer period. This is especially the case in the higher areas where plant cane does not come away as quickly as in the lower areas on the coast.

Harvesting and Trashing.

Harvesting of the cane of necessity corresponds with the opening of the mills. This usually takes place in late May, though in some years, because of economic or supply reasons, an earlier start is made. Cane reaches its maximum sucrose during the period September—October, rising from a low start at the beginning of the season and dropping again after the peak towards the end of the season, about the time of the change of year.

It is usual to start the May cutting by cutting the old ratoon crops that are due to be ploughed out the same year. It has been found that crops cut for ratooning during the period late May to mid July do not come away as well as those cut at a later date. If the season starts before late May then it is preferable to start cutting a ratooning for the late May, mid July period. It is thought, by some that this failure to ratoon during the colder periods of the year is connected with soil fertility and although the fertility has a great bearing on the ratooning qualities of a crop, the temper­ature factor has an over-riding effect as can be seen from the ratoon of June cut crops in the higher altitudes. These, even as first ratoon, are often a complete failure following early July cutting when conditions are cold.

Cutting takes place in most of the Natal belt after 22 to 24 months growth. Certain areas, notably the flats near river mouths where the

water table is high in a rich alluvial soil, can produce high yields in 12 months, but the sucrose content of the cane is low compared with the slower grown hillside cane prevalent in the cane belt. Another short period crop is cut in some areas by selecting the cane Co.301 which can give satisfactory yields in 14—18 months of growth.

At cutting it is nearly a universal practice now to strip the dead leaves, "trash", and cut the tops off the stick of cane and let this crop residue lie as a blanket over the field. In cases where the soil is inclined to be damp it is advisable to part the trash over the line of cane except in the warmest periods of the season, so that the ratooning cane may receive the drying and warming action of the sun.

The layer of trash left after each cutting has to a great extent changed the whole of the farming methods practiced in the cane belt. The extensive weeding of the ratoon cane, which at one time occupied a large labour force for a considerable period of the year, has now been reduced to the attention necessary on plant cane only. The mulch created by the layer of trash is all too evident by inspecting the soil directly underneath. When areas of bare ground are dust dry, areas under trash alongside are moist and friable. These constant dampish soil conditions must have an effect on the growing cane and on the uptake of plant food and fertilizer by the cane.

It has been observed that when trash is carried in to a plant cane crop, the roots will travel freely in the surface soil under the trash layer where they appear as comparatively thickish vigorous feeder roots. The cane growing in ground without a trash layer has roots of a tough, wiry nature which instead of travelling through the surface soil take a down­ward course, penetrating the subsoil. As it is the surface layers of the soil rather than the subsoil that can be most easily affected and altered by cultivation and the additions of fertilizers, it is preferable to keep as many feeder roots as possible in this area. Trash, itself of vegetable origin, will on rotting produce organic matter or humus and this material acts as a sponge absorbing and holding not only water but plant foods as supplied through fertilizers, both of which arc given up to the growing plant as they are required.

It is not unreasonable to suppose, after con­sidering the above facts relating to the practice of trashing, that fertilizer policies relating to cane that has been burnt standing, and thereby removing the bulk of the trash, have little or no relation to the fertilizer requirements of a crop that from the plant cane harvest onwards is under a constant blanket of trash

Trash management has been under experimental scrutiny for some years and it has been found that

139

although beneficial responses to trashing may not be recorded after the first crop of trash has been applied, namely at the first ratoon harvest, benefits can usually be recorded at the second ratoon harvest. Increases of up to 10 tons per acre have been recorded in third ratoons in favour of trashing against burning, but it has also been found that on ploughing out the crop after the third ratoon, green manuring, bare fallowing, and then replanting to cane all the advantages gained by trashing had been lost to the new plant cane crop. It would seem that until some new method of changing crops is arrived at no lasting residual can be expected from the trash treatment. Experimental work on this pro­blem is at present in hand but as yet no conclusions are ready for publication.

Fertilizer Practice.

Since the Experiment Station was started in 1925 a considerable volume of work has been carried out on this subject by members of the Experiment Station staff and other workers in the cane belt. Throughout the proceedings of the S.A. Sugar Technologists, papers are to be found relating to phosphate, nitrogen and potash dressings and the methods of application. Early on it was established that, as is the case in most of the soils of South Africa, phosphates were deficient and that cane in the first instance gave the greatest responses when treated with phosphatic fertilizers. Because of the small amount of movement of phosphates in the soil it was established that the best method of applying phosphates was through an initial dressing given at the time of planting in the furrow. Little or no effect from this type of fertilizer was ever recorded when it was applied as a top dressing to ratooning cane until the trash blanket practice was introduced. It seems that if phosphatic fertilizer is broadcast over a layer of trash, probably in the presence of adequate supplies of nitrogenous fertilizer, there is likely to be a recorded increase due to the phosphates from the second ratoon crop onwards. In the case of virgin soil it will be safer to assume that a dressing of from 600 to 800 lbs. of superphosphate will be required. In the older cane lands, where dressings of phosphatic fertilizers and, perhaps filter cake, a by-product of the sugar mill that has a high phosphatic content, have been applied in fair quantities for a number of years, a phosphatic build up may have been established and in this case somewhat lower furrow dressing may be resorted to. As the crop of cane occupies the ground for a period of from 6 to 8 years the residual effect of an initial furrow dressing of 256 •lbs, on a rich soil and 800 lbs. of superphosphate on a poor soil cannot be expected to last over and above the plant and first ratoon crops. It would, therefore, be advisable to consider the incorporation of some phosphates in the top dressings given to the

second ratoon crop. The determination of the state of the soil and the fertilizer requirements of the crop can to a degree be arrived at by the chemical analysis of soil and growing leaf and to help the cane planter a service has been inaugurated with this object in view. Particulars of this service can be obtained on application to the Experiment Station, Mount Edgecombe.

The need for nitrogenous fertilizers is most marked in ratooning cane where dressings of 300 lbs. of sulphate of ammonia should, if possible, be applied to each ratoon crop. The uptake of nitrogenous fertilizers is to a. large extent dependent on the amount of moisture available to the cane and in this respect higher dressings should be given when irrigation is practiced and when the rainfall is above average, and must be applied to obtain a quick breakdown of trash and the best effects to be gained from the trash blanket treatment.

No mention so far has been made of the use of nitrogenous fertilizer on the plant cane crop. It has been considered unnecessary to apply nitrogen to the cane after ploughing in a green manure crop. The amount of nitrogen accumulated by a leguminous crop in 6 to 7 months of growth is hardly likely to affect a cane crop planted 5 to 6 months after the green manure has been ploughed in. If the effect is to be felt the cane should be planted within a few weeks of the green manure being turned in and this is only possible with March planting, when no nitrogen seems to be needed in the form of fertilizer.

Where the long twelve months fallow has been adopted or new ground is broken tip, it has, been noticed that planters using one of the standard mixed fertilizers as a furrow dressing have obtained extremely satisfactory plant cane crops. The small quantity of nitrogen applied when using a mixture seems to give the young plant cane a needed help that sees.it through the rest of its growing period. From this it would appear that a dressing of 100 lbs. of sulphate of ammonia should be added to the phosphatic dressing applied in the furrow and mentioned earlier. The last main plant food, potash, has until a few years ago appeared to have been adequately supplied by the soils of the coastal belt. Whether it is that the constant drain of this chemical from the cane fields in the removal of molasses for industrial purposes is beginning to make itself felt or for some other reason, the fact remains that in certain areas and on certain soil types deficiencies in potash are now occurring. Resort to the Fertilizer Advisory Service is recom­mended to determine the type of dressing of this material to apply to any given crop. No spectacular increases in yield should be looked for from dressings of potash unless the level of potash in the soil has dropped so low that growth cannot

140

be maintained. In this case, large "shock" dressings of 300 to 350 lbs. muriate of potash will have to be applied and followed up by dressings of 100 to 150 lbs. on all subsequent ratoon crops. It is seldom that such drastic treatment is necessary, but it is wise to ensure against such circumstances occurring by giving small dressings as and when advised to do so.

Although many of the soils of the cane belt are extremely acid in nature, sugar cane is a plant that can tolerate a wide range of acid conditions. Ex­perimental evidence has never shown any beneficial gains from applications of lime in the field. In the higher areas it has been a fairly common practice to use dressings of Karroo manure in conjunction with a dressing of lime as a manure at planting time. There is no experimental evidence to support this practice.

Irrigation.

A limited area of the cane belt of Natal is at present under irrigation and it is unlikely that more than one-third of the existing area under cane could ever be brought under irrigation. The reasons for this are: (1) Much of the land is of too steep a gradient; (2) The main sources of water, the rivers, are in too deeply eroded valleys and the cost of raising the water to the cane fields would be ex­cessive; and {3) There is insufficient water in the rivers throughout the year. The conservation of water in sufficient quantities to allow for more than one-third of the area to be supplied would have to be undertaken on an industrial or national basis, but despite these drawbacks considerable advan­tages can be gained by supplying water to the cane wherever possible. The insurance against drought and the possibility of growing cane on land in regions where the rainfall is too low to support growth by itself, were the main reasons for the biggest irrig­ation project that is operating today, the Natal Estates scheme, being originally laid down.

It is undoubtedly possible to substantially increase the yields of cane by applying extra water over and above the natural rainfall. It would appear that to certain limits adequate rainfalls during the normal summer period help to keep the cane growing rapidly. It seems that the addition of water during the winter period can maintain growth despite the lower temperatures which operate at this time of year.

These indications come from some preliminary research investigations which also tend to show that the spray methods of applying water are superior to the furrow method of irrigation.

Further researches into these problems are necessary before any satisfactory conclusions can be arrived at.

Choice of Variety.

Co.301.—On all really sandy soils, and parti­cularly the coastal wind-blown sand, Co.301 seems to have been undoubtedly one of the best varieties. On many other heavier soils this variety will also yield well but has a tendency to "lodge", which makes cutting and loading difficult; it is better, then, on the more fertile soils to cut this variety yearly or between the age of 12 and 18 months. It is easy to cut, being soft and yet brittle, and many sticks snap off at ground level during the cutting process. Co.301 forms a complete canopy at an early stage, which saves much weeding in plant cane and ratoons.

It has been found to be very susceptible to "smut" and for this reason in certain areas may be replaced by N:Co.334. It is doubtful if any other variety will yield quite as well as Co.301 does on the sandy soils.

Co.331.—This variety has found a place in the industry where others are not at their best. It has done well at high altitudes and growers in Eshowe, Entumeni, Doornkop and Powerscourt areas favour it in preference to others. It is also doing well in low-lying, poorly-drained spots typical of a large area around Ginginlovu. To a certain extent Co.331. is also being grown on the better class sands which were previously under Co.301.

Co.331 is a vigorous grower, giving a high yield of cane per acre with rather a low sucrose if cut early or late in the season, but mid-season gives reasonable returns. It has an upright habit lacking the canopy, but it is rather tough to cut, and cutters will often complain about this when cutting it for the first time.

It is susceptible to pineapple disease, which is a disease affecting the cane sett, and bad germination with this variety is usually due to this. Setts should always be treated with a fungicide before planting, as this is a fairly sure control.

In years of deficient rainfall this variety often has many dead or half-dead sticks in a stool, parti­cularly when left over as a two-year-old crop.

N: Co.310.—This variety is doing very well on the better class sandy loams and clay loams and alluvial river flats. On the red-brown coastal sands it also appears to give a good yield but is rather slower growing than C.301, and there is some doubt whether it will be suitable for one-year-old cutting on these soils.

N:Co.310 has given the highest sucrose of any cane yet released and has a good growth habit, forming an excellent canopy similar to Co.301. It is soft to cut but the sticks seem to be more solid

141

and denser than other varieties, and it is easy to underestimate yields.

This variety appears to be an extremely good germinator and it should give some of the highest yields of sugar per acre of any of the released varieties.

This cane is most suited to rich, well watered soils. On the poorer dryer soils the yield of the ratoons tends to drop off rather rapidly.

Co.290.—This variety is not widely grown now, since it has been found to be very susceptible to red rot, and under unfavourable conditions is not a good ratooner.

It is, however, a very good yielder as a plant cane crop, and will also ratoon well if cut yearly and preferably during the summer months. This is an important point, because many of the ratooning failures have been caused by cutting in May, June and July.

With Co.290, then, do not cut until the end of August or later, and the crop should not be older than, say, 18 months.

Treated in this way this variety will give excellent yields on the better class sandy soils and on the well-drained heavier soils. It will not stand a tendency to water-logging.

Red rot has only become epidemic in the mist belt — that is, at altitudes of over 1,500 feet— so there seems no real reason why Co.290 should not be a very useful variety in many other places. It is reasonably resistant to smut, and may prove-to be a substitute for Co.301 on the reddish coastal sand.

It is a fast grower, forming a good canopy at an early stage, and it is easy to cut and load.

Canes Recently Released or Due for Release in 1953 #

N :Co.339.—is a variety giving heavy tonnage, whose sucrose is medium to good, usually slightly higher than Co.301. In variety trials it has shown itself to be adaptable to a wide range of soil con­ditions. It is also slow and difficult to germinate and should be given the best possible conditions at planting.

This cane stools well and maintains its ratoons well.

N:Co.293.—in the high altitude districts has shown considerable promise, doing well in the Eshowe and Braemar districts and proving itself best in a variety trial at Eshowe. It has a sucrose content comparable with Co.301 and gives satis­factory tons cane per acre yield. Its canopy is satisfactory and it resists red rot. Cases of mosaic have been found.

For low coastal areas it cannot be recommended as under coastal conditions it flowers excessively even when young and it has been noticed that its ratoons are poor in the lower areas.

It might, under damp or irrigated conditions, be grown near the coast for annual cutting.

N: Co.334.—This cane has a stiff upright habit of growth of cane with a fair to good canopy which in drought conditions tends to close upwards into the line of cane. It 'has a sucrose content slightly higher than Co.301. The cane does well in sand and is not to be discounted in many other types of soil. Growth is often slow in the first year, but the stand in the second year is uniform and the cane is not given to lodging. The cane is resistant to attack by smut and mosaic. The variety is not immune to smut but is the most resistant of the varieties which have been recently released; only a few cases of mosaic have been recorded.

N :Co.292.- This cane is a useful cane which may prove itself as an alternative variety in planters, cropping programmes. It gives a useful though not outstanding yield in tons cane per acre and sucrose content is slightly below that of Co.301. It does not. seem prone to any disease though cases of mosaic and smut have been found. Canopy is fair and although the leaf yellows under drought conditions, recovery is very rapid when rain has fallen.

Conclusion.

It is intended with this paper to start a series that may, as the years proceed and practice in the cane belt is modified or changed, report these at the Congress. In this way it may be possible to create an interest in the field operations which, after all, are the vital factors in the industry, for without efficient field management, not only would the mills lack their raw materials, but it is surely in the fields that any profit or loss in the whole business of sugar manufacture has its origin.

Thanks must be recorded to Mr. C. D. Sherrard for the use made of his paper of 1949, which has up to now been extensively used and it is a certainty that many planters have sought advice from its pages in the last five years.

Experiment Station, South African Sugar Association,

Mount Edgecombe,

Mr. Dymond said that he could not recall a previous Congress at which the standard of papers presented was as high as during the present Congress. Papers on engineering, chemical and other subjects had been excellent and the standard of agricultural papers presented on the agricultural subjects had come equally up to the standard set by the other

142

papers. He said that discussion on all four papers would take place simultaneously so that those present could ask questions covering all the papers presented at the agricultural session.

Mr. Dymond, of Doornkop, said that Mr. Pearson had not mentioned what should be done with the last crop as far as trash was concerned. He asked whether Mr. Pearson recommended burning the trash.

Mr. Pearson said that as far as was known at present the idea was to plough in the trash and get some cover onto the ground as quickly as possible. He referred to the new method of planting by parting trash, over the cane, rotavating, moving the trash and then rotavating in the inter-row and planting there. Another treatment they had tried was to remove the trash, plough out the cane and replace the trash. That trash had now been ploughed in and he expected to find a complete nitrogen starvation.

Mr. Farquharson said both Mr. Jex and Mr. Pearson had mentioned rotary hoes. He had heard that these machines broke down the soil structure by pulverising it and thereby destroyed its fertility. Was this correct?

Mr. Jex said the rotavator was possibly one of the most potentially dangerous instruments put into farmers' hands. It was also one of the most potentially useful. It depended entirely on the way it was used. It did have a definite use in that it shortened the time within which fields could be replanted. Much had still to be done by way of experiment on this subject. The rotovator if used in conjunction with chisels or sub-soilers for creating a seed bed in the top 5—6 in. of the soil was a remarkably useful instrument. Tf correctly used in the preparation of cane lands a rotovator could not harm the soil. Rotary hoes were con­sidered dangerous for continuous cultivation. Where cane was kept under trash for 8 to 10 years experience suggested that the soil character changed and the texture became more friable. Organic matter accumulated so that when the cane was ploughed out you had a friable soil and not the hard pan which you might get under different conditions. But where soil developed a hard pan through bad agriculture, it was unwise to rely on a rotary hoe to restore tilth—it could, temporarily, but this tilth was misleading if proper agriculture was not adopted. In his view, however, it could, and probably would, replace the plough.

Mr. de Lisle mentioned the lapse of time between the ploughing in of the green crop and replanting of cane. During this period the land was exposed to the elements; his practice was to use velvet beans for a green crop, then rotovate the beans roughly when they were in pod and let them lie. The beans

grew again and gave cover until planting time, when they were ploughed in.

Mr. Pearson said he was grateful for this inform­ation as they were still looking for means for keeping the ground under cover the whole time.

Mr. Palairet said he felt the point made by Mr. Jex required elaboration. If organic matter were mixed in by the rotovator there was an improvement in the structure of the soil, which more than overcame the harm done by the use of a rotovator on hard soil. He referred to Mr. Pearson's reference to small irrigation schemes and the use of irrigation in winter. His experience was that with a small irrigation scheme he had water just when he wanted it, and with proper management there need not be any need for water in winter. He thought those planters who had no water during winter should not be deterred from adopting a small irrigation scheme.

Mr. Garland said that if Natal Estates did not have water for irrigating some of its lands the cane would die. The principal advantage of the water was, in his view, during the winter months.

Mr. Pearson said that experiments in irrigating were still somewhat in their infancy. At Chaka's Kraal, however, they had applied 20 in. more than average rainfall during winter months and had got a crop of up to 50 tons per acre in 13 months. Water during winter often added to increased tonnage and was more beneficial than water applied in summer.

Dr. Dodds said it was most heartening to see that three of the papers presented were written by growers. This was another excellent indication of the co-operation given by the farming community which was further attested to by the large attendance. He referred to the help that had always been given to the Experiment Station by Natal Estates.

Mr. Barnes said that the subjects raised were of deep interest, especially when compared with practices in other countries. The presence of drains in the fields did not prevent the use of machines. Mechanical cultivation was practised with the cambered bed system of the West Indies. Tractor haulage of cane was used, the trailers being loaded in the field. Temporary bridges, readily moved from place to place, and the filling of drains by trash at certain points, enabled tractors and carts to move in and out. In Louisiana the bed and furrow method was no deterrent to mechanical reaping and haulage. It should not be assumed that mechanical haulage was impossible where open drains existed. Tile drains were excellent. They should be laid in accordance with a definite plan, and their positions must be accurately known.

Mr. Barnes expressed the view that reaping should to some extent be considered as a cultural

143

operation when a ratoon crop was to follow. With trash blanketing there was a possibility that the cane might be cut too high instead of at or just under the ground surface. If stubble was left it might be necessary to use stubble shavers to ensure a good ratoon crop, as was the case in Tanganyika and elsewhere.

Mr. Starling asked whether Mr. Pearson could explain Mr. Wise's reference to applications of less than 20 tons per acre of compost having a de­pressing effect.

Mr. Pearson referred to the work of Russell who had used radioactive carbon with compost and had followed this through with geiger counters. It was found that the bacteria population grown on the farmyard manure of 5 to 10 tons per acre very quickly absorbed the manure applied and were left in a starving state, so returning to the nitrogen originally in the soil. The net effect was to deplete the soil of its inherent nitrogen content. The con­clusion drawn was that small quantities of compost left the soil in a worse state than previously.

Mr. Adams said that in his knowledge many planters used less than 20 tons of filter cake per acre. Was it really proved that to put 10 tons of filter cake per acre was a definite waste? He asked whether excessive application of filter cake might damage the crop.

Mr. Pearson said that the proportion of water to solid in filter cake was less than in manure. The amount of organic matter going in was lower than for farmyard manure. At Chaka's Kraal huge quantities of filter cake had been piled on the soil before the Experiment Station took over. They could, therefore, get no phosphate reaction, but there was also no evidence to show that excess application of filter cake had any bad effect on the soil.

Dr. Dodds said that as far as he knew where carbonotation filter cake was used in large quantity it was liable to have some slight deleterious effect on the soil possibly because the minor elements had been completely precipitated by the excess of lime. The optimum figures that he knew of were 8 tons of dried material and between 16 to 20 tons of wet material per acre. This applied chiefly to the sulphitation filter cake.

Mr. Jex said that he would not like to accept Mr. Wise's statement without further knowledge, as otherwise it might do great harm. He felt that anything added to the soil must benefit it and the more that could be put in the sooner would an adequate state of fertility be reached. In his experience, where he had applied less than 10 tons per acre of filter cake the result was possibly small, but it was nonetheless there. Twenty tons per acre undoubtedly was more advantageous, but he

felt they could not discount the effect of applying very much smaller tonnages continuously through­out the year.

Mr. Garland said he had had years of experience with filter cake and he could see no harm in applying 20 tons and more per acre. There was a tendency to burning in dry conditions, but in good conditions this application resulted in wonderful crops.

Mr. Patrick Murray said that all bagasse was burned in boilers. The bagasse came from the fields. Had any experiment been conducted in the returning of bagasse to the fields?

Mr. Steward referred to Mr. Wise's paper and the suggestion that planting could be done in unusual months. At Kearsney a very big expansion pro­gramme was initiated and from 1948 onwards, for over three years, they had planted in every month throughout every year. There was no effect on the final crop. Although cane planted in winter did not germinate as well or as quickly as cane planted in spring, when all the canes were a year old there was no material difference in the stand. This sug­gested a solution to the question of what to do with lands between March and August. He said they should be planted; provided the soil was sufficiently moist he felt that planting at any time was quite safe. Planting machines undoubtedly helped in this respect.

As a result of experience they used D mixtures at planting. They did not know exactly why, but it undoubtedly gave the best results. He suggested that planters should consider mixing their own fertiliser as he felt this would, in the end, -be cheaper. He believed very strongly in the use of green manuring; his company always used sunnhemp, or velvet bean on long-fallow lands and beneficial results had been proved.

Mr. Dymond, replying to Mr. Murray, said that bagasse had not been used in the field for obvious reasons. He referred to Dr. Rapson's reference to bagasse used in paper making. This wras far more valuable than it would be if the bagasse were returned to the field. Several references had been made to filter cake, but little had been done either to enrich filter cake or to dewax it. Professor Owen had shown that dewaxed filter cake had a very much more beneficial result on fields. It had been said that if filter cake were applied to the soils in great quantities over a period of years the wax might have a deleterious effect on the soils.

Mr. Pearson said he was grateful for Mr. Steward's remarks on mixtures. He, too, had observed that planting with D or F mixtures appeared to give better results than planting with straight supers. Tests were being carried out to try to find out why this was so. He felt that nitrogen applied in small quantities in the furrow was beneficial to the young

114

cane, but this still had to be proved. He criticised the combination of N. P. and K. in the mixtures put out by the Government in relation to the Sugar Industry and expressed the view that they were not adequate for sugar cane agriculture. Trials showed there had been great difficulty in getting response to phosphate in top dressing, and he felt that the amount of phosphate in mixtures was too high and farmers had to pay for the phosphate to get the nitrogen onto their crops.

Mr. Jex referred to the question of ploughing in the final trash. The basis of the question was the possible effect on the soil. Practice seemed to bear out that trash on top of the soil was advantageous. The demand for nitrogen was at the bottom of the trash layer. By ploughing in the trash the soil was virtually made to eat the whole substance of trash and nitrogen and the soil might be said to suffer from indigestion. A great deal more research had to be done to determine how the trash should be dealt with. Nowhere in nature would they find any such device as a plough to turn the organic matter into the soil. He felt that if there were any doubt then nature should be followed. Where he had disregarded nature and ploughed trash into the ground in the absence of nitrogen, he had actually immobilised his ground for long periods at a time. If trash were accumulated over one crop for 8 to 10 years you would have on the soil a layer of humus. The organic content of the soil would have been improved to a marked extent, and there­after it would be possible to concentrate on the mechanical aspect of providing a good seed bed. The final trash could either be carried off or burned off, but carrying or bulldozing it off were not found satisfactory, because soil was compacted by treading, and experiments carried out so far indicated that burning off the final trash might be the best method provided the under layer was damp when burned off. He asked that this should not be confused with burning of cane every year. In this case the trash was burnt only once in 10 or so years, and immediately afterwards the work of preparing a seed bed was begun. There was no chance then for a hard surface to be formed. It was possible that burning the final trash might be advisable, but far more experiment­ation had to be done.

Mr. Almond said an experiment on the lines outlined by Mr. Jex had been put down and was being carried out. They would be able to compare such points as turning in the last crop of trash, carting it away and burning off the last crop of trash.

Mr. Jex asked what the nitrogen treatment of the field was.

Mr. Almond agreed with Mr. Jex that a great increase in yield per acre was possible, but thought progress should be more rapid. Planters should

be more alive to the value of the Experiment Station and more willing to accept their advice which was based on results of field trials. An example of delayed progress was the case of trash blanketing which has only recently been accepted as a sound agricultural practice but was first sug­gested by the Experiment Station in 1935.

Mr. de Villiers asked Mr. Garland how many tons of filter cake were usually applied per acre.

Mr. Garland replied that approximately 20 tons per acre were applied every time the fields were ploughed out.

Mr. de Villiers pointed out the very high content of calcium carbonate of carbonatation cakes and asked if deficiencies in minor elements had not been observed on Natal Estates, with particular reference to iron and manganese.

Mr. Rault replied that he did not think this subject had been investigated in Natal.

Mr. de Villiers said that a case of iron deficiency on cane had been found at Triangle owing to too much calcium carbonate in the subsoil under shallow soil.

Mr. Palairet said that Mr. Jex had substantially justified a practice he was working. He used trash blanketing on all ratoons, but where a field was to be ploughed out he carted the cane whole to the siding where he topped and trashed it and made compost. This fulfilled most economically the two points made by Mr. Jex; the trash was removed with no tamping of the soil.

Mr. Dymond said that the question raised by Mr. de Villiers was of the utmost importance. He felt the effect of the continuous application of calcium carbonate should be investigated by the Experiment Station.

Mr. Garland said he was very surprised to hear Mr. Jex suggesting that the last crop of trash should be burned. He was against this practice. He added that he was also in favour of green manuring. He felt there was no need for D mixtures if green manuring was followed. He went on to say that Mr. Steward had raised an important point. Natal Estates was, he thought, the first company to follow the principle of winter planting. One third of the area was now done in March and in April. He did not agree that planting should be done throughout the year. January and February and March and June were, in his view, very bad months for planting.

Mr. Steward said he did not advocate planting all the year round as a general practice; he just wanted to point out that if, for various reasons, planting was necessary it could be carried out at any­time, provided there was sufficient moisture in the soil.

145

Mr. Jex said he had been taken to task by Mr. Garland for speaking a little heresy, but sometimes he liked heresy. He took his cue from nature in which fires were not unknown; it might not be a bad thing to have an occasional fire. A fire every year was of course a bad thing and he could never advocate it. But after 8 to 10 years of covering the soil, he was still not convinced that a final burning might not be advantageous. He questioned whether a green manure crop would be necessary after con­tinued trash blanketing. It would certainly be better to grow a green crop when soil conditions were not right, but where the soil was built up and in good condition it might be as good, and more economical, to grow another crop of cane rather than put in green manure crops—and to get the nitrogen out of a bag !

Mr. Pearson thanked those who had taken part in the discussion, particularly from the planters' point of view. It had been most advantageous to the Experiment Station staff. He was also extremely grateful to the three gentlemen who had contributed papers for the discussion. It was not always possible to set out in writing what could be said in discussion. He felt that the interchange of opinion between the Experiment Station staff and the farmers themselves was a very good thing.

Mr. Dymond expressed the meeting's thanks to the Experiment Station staff for making the Station available and providing the entertainment. He thanked all those who were present and who had contributed their part to the discussion, which he felt was of an extremely high standard. Mr.

Dymond said that shortly eleven members of the Industry would be going overseas to the Inter­national Congress in the British West Indies. He hoped that when they came back it might be possible to have one or two days either in Durban or at the Experiment Station in which these dele­gates could pass on their impressions and their opinions gathered at the Congress.

He referred also to the fact that it was hoped that up to six world experts would be invited to South Africa to meet growers and millers and exchange information. This followed on the Association's decision to invite Mr. de Sornay from Mauritius in the previous year and the system of inviting guest speakers to the meetings of Council. Out of these talks had come such valuable papers as those presented by Mr. Jex and Mr. Hogarth. The Association was aiming at improving continually the standard of papers and the level of information disseminated throughout the Industry. He felt that the Congress had been in every way an out­standing success.

Mr. Rault commented on the remarkably friendly spirit of co-operation between millers and planters. He felt that the Technologists' Association had a great deal to do with this. Mr. Dymond in particular was very alive to the need for this co-operation and had done a very great deal to foster it. He, Mr. Dymond, had conducted the meetings during the past four days with the utmost efficiency and good humour and he asked that a most hearty vote of thanks be accorded to Mr. Dymond.

MEMBERSHIP LIST

The following is a list of members of the South African Sugar Technologists' Association as at 24th August, 1953 :

Honorary Members.

BARNES, A. C 22SussexHouse, WinderSt., Durban HEULEY, E. P. ... ... 16 Gillatts Road, Westridge, Durban. BLACKLOCK, L. L 156 Nicolson Road, Durban. LADLAU, G. P Umhlali. BR ANDES, DR. E. W. ...Washington. HEATON NICHOLLS, G. ..". Cowey's Hill. CAMPBELL, W. A The Natal Estates Ltd., Mount MOBERLY, G. S Eshowe.

Edgecombe. ROBBINS, W. T Umhlali. DODDS, H. H 55 Margaret Maytom Avenue, SIMPSON, J. R. "Uplands," Cowey's Hill.

Durban North. JOHNSTONE, P 12 Marine Court, Durban. FOWLIE, P Richmond.

Members,

ADAM, D. I 28 Manhatten Court, Durban. CORDES, H. Canelands. ADAMS, B. N P.O. Box 55, Stanger. CRAWFORD, W. H P.O. Box 108, Eshowe. ADDISON, L. P P.O. Box 24, Empangeni. CRESSWELL, J. S Hulett's S.A. Refineries, Rossburgh. ALEXANDER, J. B Sugar Milling Research Institute, CROOKES, F. N. N Mount Ashley, Merrivale.

Durban. CROOKES, G. V Lynton Hall, Sezela. ARMSTRONG, E. I Canelands. CUNNINGHAM, J. D. ... p'O. Box 2278, Durban. ARMSTRONG, H Z.S.M. & P. Ltd., Empangeni. ARMSTRONG, Ross P.O. Box 41, Verulam. CHALMRRS, A P.O. Box 45, Rossburgh ARMSTRONG, R. G Verulam CHRISTIE, R. P.O. Box 1484, Durban. ASHE, G. G Z.S.M. & P. Ltd., Empangeni. COTTERELL, W. J P.O. Box 5981, Durban.

ANNETT, L. V c/o Reid Bros. Ltd., Durban. CABRERA, JOSE M Mexico.

ABREU, L. J. S. DE P.O. Box 68, Beira, P.E.A. CARRINGTON, W. B. ... Bulkeley Sugar Co., Barbados, ADKINS, J. E Sena Sugar Estates Ltd., Marromeu. B.W.I.

COSTA, J. DE C P.O. Box 68, Beira, P.E.A. CONNELL, COL. J P.O. Box 236, Barbados, B.W.I.

BAILEY, E. D c/o P.O. Box 1541, Durban. BALCOMB, B. E Sir J. L. Hulett & Sons, Felixton. BEATER, B. K Experiment Station, Mt. Edgecombe. DAVIDSEN O Mount Edgecombe. BECHARD, R. M P/B. Gingindhlovu. DENT, C. E. ' !" '... ..'. The Tongaat Sugar Co. Ltd., BENNETT, R. N. E 46 Glen wood Drive, Durban. Maidstone. BENTLEY, J. P, N Tongaat Sugar Co. Ltd., Maidstone. DENT, II. C "Cartrief," Umhlali. BERRY, A. V Gingindhlovu. DEPEKIND, E. T. J P.O. Rivcrview. BIGARA, L. C P/B. Empangeni Rail. DE BROGLIO, A Mount Edgecombe. BOUCH, W. G Mount Edgecombe. D E CHAZAL-DK CHAMAREL, BOULLE, J. E 26 Springfield Crescent, Durban. E. A. R .' P.O. La Mercy. BOURNE, J. II 46 Glen Road, Montclair, Durban. DICK, J Experiment Station, Mt. Edgecombe. BRETT, P. G. C Experiment Station, Mt. Edgecombe. DICK, j. MoD P.O. Box 301, Durban. BROMLEY, C. K Reynolds Bros. Ltd., Esperanza. DOLLENBERG, A. O. H. ... Z.S.M. & P. Ltd., Empangeni Rail. BROWN, J, W P.O. Box 55, Stanger. DOUWES DEKKER, K. ... Sugar Milling Research Institute, BRUNIQUEL, J. M Sir J. L. Hulett & Sons, Darnall. Durban. BUCHANAN, W. K P.O. Box 1209, Durban. DRAEGER, P. L Illovo Sugar Estates Ltd., Illovo. BYARD, W. E P.O. Darnall. DUCHENNE, J. O Durban.

DUPONT, R. M Esperanza. BONFA, N. J. A c/o P.O. Box 914, Durban. Du PRBEZ, M. S P.O. Box 2204, Durban. BRIGGS, T P.O. Box 11, Brighton Beach, Du TOIT, J. L Experiment Station, Mt. Edgecombe.

Durban. DYMOND, G. C Sir J. L. Hulett & Sons, Ltd., BUCK, W. R c/o P.O. Box 1541, Durban. Darnall.

DYMOND, K Doornkop. BOAST, J. T c/o Rhodesia Sugar Refinery,

Bulawayo. DAWSON, C Electricity Supply Com., Congella. BOTHA, J. A. H c/o Triangle Sugar Estates, Fort DEENIK, Z P.O. Box 734, Cape Town.

Victoria, S.R. DRYBURGH, J. H Hotel Osborne, Durban. DUNN, G. A 20 Rosebery Avenue, Durban North. DYSON, W. H. African Explosives & Chemical

CAMDEN-SMITH, D Pongola, E. Transvaal. Industries, Umbogintwini. CAMDEN-SMITH, E Reynolds Bros. Ltd., Sezela. CAMPBELL, B P.O. Box 5, Empangeni. CARTER, R. A Entumeni. DE DEUS, J. L. D. ... P.O. Box 68, Beira, P.E.A. CHEADLE, W P.O. Box 59, Port Shepstone. DE SORNAY, A Reduit, Mauritius. CHIAZZARI, L, F P.O. Box 55, Stanger. DE VILLIERS, O. D'HOTMAN Triangle Sugar Estates, Fort CHRISTIANSON, W. 6. ... The Tongaat Sugar Co. Ltd., Victoria.

Maidstone. DIEPEVEEN, J. M Sena Sugar Estates Ltd., Marromeu. CHRISTIE, J P.O. Box 1541, Durban. DOVE, D. D Triangle Sugar Estates, Fort COIGNET, I, J. P Reynolds Bros. Ltd., Sezela. Victoria. COLEPEPER, J E ... P.O. Box 11, Eshowe. DUVEKAT, CD Marromeu, via Beira, P.E.A.

ELYSEE, A. D EVANS, C. S

FARQUHARSON, J. C.

FELTHAM, O. A. FRANGS, G. B. FROBERVILLE, L. F. DE

FROBERVILLE, P. DE

FULHAM, J. P

FARNELL, R. G. W.

FEUILHERADE, L

FONSECA, L. M. P. DA FRANCES, C. S. . . . .

GALRRAITH, W. G GARLAND, H. T, GIRDLER, J Goss, T GRANT, J. E

GRINDLEY, L. R GUNN, J. R

GARNER, J GOBLE, A. D

GILMOUR, J GUILI.OTKAU, J. E. H. .

HARDY, R. L

HEDGCOCK, E. W HEDGCOCK, G. HELLETT, J. E HENDRY, D. W HILL, M HILL-LEWIS, C HINDSON, E. E HOWES, N. P HULETT, JACK . . . .

HALL, E. HALSE, R. H. . . . . HILLIARD, J. H HOGARTH, R. J HULL, II. S HUMFRYES, A. G HUTCHESON, R. F

HENDRY, A HINDE, G. H. D ITOWDEN, Hugh ... . HUNTER, J. S. . . . . HVIDT, V. J

IRONSIDE, M

JACKSON, L

JELLEY, C

JEX, W. F. C

KENNY, A KING, N. C KlRKWOOD, J. V ,

KRIGE, P. R KRUGER, A. M

1

... Amatikulu.

... P.O. Box 2278, Durban.

... The Tongaat Sugar Co. Ltd., Maidstone.

... P.O. Box 2160, Durban.

... 141 Benson Road, Montclair, Durban.

... Sir J. L. Hulett & Sons Ltd., Darnall.

... 1 Barbon Court, 273 Bartle Road, Durban.

... P.O. Box 55, Stanger

... Farnell Carbons Ltd., Plumstead, London

... Incomati Estates, Xinavane, Lourenco Marques.

,.. P.O. Box 68, Beira, P.E.A. ,.. Kongwa, Tanganyika.

.. Reynolds Bros. Ltd., Sezela.

.. Mount Edgecombe.

.. Z.S.M. & P. Ltd., Empangeni Rail.

.. Tugela Estates, P.O. Havelock.

.. Sir J. L. Hulett & Sons Ltd., Darnall.

.. Z.S.M. & P. Ltd., Empangeni Rail.

.. The Tongaat Sugar Co. Ltd., Maidstone.

.. 677 Currie Road, Durban.

.. "Byways," Cowie's Hill.

.. Miwani Sugar Mills, Kenya Colony.

.. 57 Rue Cuvier, Paris, France.

.. Doornkop Sugar & Food Industries, Doornkop.

.. P.O. Box 59, Port Shepstone.

.. P.O. Box 59, Port Shepstone.

.. Chaka's Kraal.

.. Reynolds Bros. Ltd. Esperanza.

.. Renishaw Estates, Renishaw.

.. Mount Edgecombe.

.. Clifton Estates, Kearsney.

.. Z.S.M. & P. Ltd., Empangeni Kail.

.. Kearsney.

.. P.O. Box 1242, Durban.

.. Stellenbosch.

.. P.O. Box 346, Durban.

.. P.O. Box 1938, Johannesburg.

.. P.O. Box 922, Durban.

.. P.O. Box 1484, Durban.

.. P.O. Box 1541, Durban.

.. Sena Sugar Estates Ltd., Marromeu,

.. P.O. Box 2076, Salisbury, S.R.

.. Glasgow, S.E.

.. Marromeu, P.E.A.

.. Tanganyika Planting Co. Ltd., Moshi

.. P.O. Box 1163, Durban.

.. Sir J. L. Hulett & Sons Ltd., Darnall,

.. Sir J. L. Hulett & Sons Ltd., Darnall.

.. Umvoti Mouth, via Stanger.

.. P.O. Box 179, Durban.

.. Experiment Station^ Mt. Edgecombe.

.. P.O! Box 29, Mtubatuba.

.. P.O. Box 145, Germiston. ,. Eshowe.

.48

KRUMM, H. E P.O. Box 47, Bellville, C.P.

KINLOCH, J. C P/B. J. 3, Bulawayo, S.R. KOVAC, T. Sena Sugar Estates Ltd., Marrom<

LALOUETTE, H Melville Sugar Mill, Stanger. LATHAM, W. T P.O. Nkwaleni. LEE, G The Tongaat Sugar Co. Ltd.,

Maidstone. LEWIS, S. V. R Reynolds Bros. Ltd., Pongola,

Transvaal. LINCOLN, M. A P.O. Darnall. LINTNER, J P.O. Box 2033, Durban. LIVINGSTONE, A, A. ... Reynolds Bros. Ltd. Sezela. LLOYD, A. A S.A.S.A., Durban. LOVE, A Tllovo Sugar Estates Ltd., Tllovo. LOXTON, A. H. Reynolds Bros. Ltd., Sezela,

LINDEMANN, W. C. ... 1.1 Palm Grove, Durban.

MACKESY, \V. Umhlanga Rocks. MAIN, J. W Dudley Road, P.O. Cowey's Hill. MANN, M. C. ROUILLARD

(MRS.) Darnall. MAROT, J 478 Currie Road, Durban. MF.NNIGKE, A. T P.O. Empangeni Station. MILLAR, J. D 41 Eleventh Avenue, Durban. MOREL, P Sir J. L. Hulett & Sons Ltd.,

Darnall, MORILLION, C Z.S.M. & P. Ltd., Empangeni. MORTII-EE, A. \V. S. ... Mtubatuba, Zululand. MUNGLE, J Lauriston, Frasers. MUNRO, D. B. B P.O. Eive.rview, Zululand. MURPHY, E. S. Caister Hotel, Durban. MURRAY, P 11 Holden Lane, Durban. MACBETH, F. B Mount Edgecombe. MACLEAN, A P.O. Box 1.163, Durban. MCCULLOCH, A. F Sugar Milling Research Institute,

Durban. MCKENNA, H. G Gledhow-Chaka's Kraal Sugar G

Chaka's Kraal. MCKINLAY, D Z.S.M. & P. Ltd., Empangeni Rai MCMARTIN, A. Experiment Station, Mount

Edgecombe.

MAY, R Tanganyika. MARTINDAI.E, C P.O. Box 927, Bulawayo. MEUWS, T Sena Sugar Estates Ltd., Marrome MONTEIRO, V Cassequel. MORGAN, A Rhodesia Sugar Refinery, Bulway

S.R. MUIR, R. Brisbane, Australia. MURRAY, C. W Derby, England.

NICKSON, G P.O. Box 536, Durban.

NELSON, W. J. New York.

OBRIEN, K. M P.O. Box 1938, Johannesburg. O'CONNOR, M. P 551 Musgrave Road, Durban. ODENDAAL, G. A Sir J. L. Hulett & Sons Ltd.,

Darnall. O'FARRELL, M Sir J. L. Hulett & Sons Ltd.,

Darnall.

PACKHAM, F P.O. Felixton. PALAIRET, H. E. H P.O. Box 45, Stanger. PARRISH, J. R Sugar Milling Research Institute,

Durban. PASTOR, O New Guelderland. PATERSON, A. The Natal Estates Ltd., Mount

Edgecombe. PEARCE, K. W Illovo. PEARCE, O. W. M Illovo Sugar Flstates Ltd., Illovo. PEARSON, C. H. O Experiment Station, Mount

Edgecombe.

149

PENNY, S. S. K P.O. Box 2204, Durban. SCHMELZ, G P.O. Box 2033, Durban. PERK, C. G. M Sugar Milling Research Institute, SHAW, D. M P.O. Box 2616, Cape Town.

Durban. SIMPSON, V 34 Birkenhead Road, Durban. PHIPSON, E. H Z.S.M. & P. Ltd., Empangeni. SEMPLE, D. MCHARDY ... c/o P.O. Box 301, Durban. POUGNET, J . F P.O. Box 18, Umhlali. POUSSEN, J. B P.O. Box 55, Stanger. SILVA, ING. F. SERNA . . . Mexico. PRINGLE, D. H Estate Climax, Frasers. SIMMONS, R. I Sena Sugar Estates Ltd., Marromeu.

SINCLAIR, K. B Triangle Sugar Estates, Fort PORTEOUS, J "Roslyn," Musgrave Road, Durban. Victoria. PyoTT, J. T P.O. Box 352, Durban. SMITH, R. H. B Sena Sugar Estates Ltd., Marromeu.

PIIGRIM, H. F Barbados, B.W.I. PILOT, H. G. H. L. ... P.O. Box 68, Beira, P.E.A. TAYLOR, MRS. R. G. ... Experiment Station, Mt. Edgecombe. POWE, W. A Cuba. THOMPSON, G. D Reynolds Bros. Ltd., Sezela. PRADO, J. B. DE R. ... Beira, Mocambique. TONNER, D Hulett's S.A. Refinery Ltd.,

Rossburgh. TURNER, L. E Sir J. L. Hulett & Sons,

RAAFF, J.M Mtunzini. Amatikulu. RAULT, J Mount Edgecombe. TYACK, H. A. G Sir J. L. Hulett & Sons Ltd., RENAUD, C. L 446 Musgrave Road, Durban. Darnall. RENNIE, L. I Sir J. L. Hulett & Sons Ltd.,

Darnall. THURNHEER, M Sena Sugar Estates Ltd., Marromeu. RICHARDSON, T. A lllovo Sugar Estates Ltd., Ulovo. RISHWORTH, A. H P.O. Box 301, Durban. ROBERTS, T Waldene Sugar Estates, Chaka's USHER, H. N Sena Sugar Estates Ltd., Marromeu.

Kraal. ROYSTON, J P.O. Box 36. Chaka's Kraal.

VAN DYK, B. C P.O. Box 51, Mtubatuba. RAE, A. A St. Augustine Court, Escombc. VERINDER, II. N P.O. Box 2160, Durban. ROHLOKF, H. 291 Bree Street, Johannesburg.

VASCONCELLOS, R. DE ... Funchal, Madeira. RABE, A. E. P.O. Box 142, Ndola, N. Rhodesia. VAN DILLEWIJN, DR. C. C/O Victorias Milling Co., Baeolod RAE, W. D 22 Ward Road, P.O. Cranborne, City, P.O. Box 171, Occ. Negros,

ROBERTSON, C. L P.O. Box 268, Salisbury. Ross, E. S. ... Kakira Sugar Works Ltd., Uganda.

SANDSTROM, G. R. A. ... Z.S.M. & P. Ltd., Empangeni Rail. WATERS, C. L Sir J. L. Hulett & Sons, SARGENT, N. V Mount Edgecombe. Amatikulu. SAUNDERS, D P.O. Box 2009, Durban. WATERSON, H. D P.O. Frasers. SAUNDERS, G Sir J. L. Hulett & Sons Ltd., WATSON, R. G. T Tongaat. Sugar Co. Ltd., Maidstone.

Felixton. WHEELER, F. D Messrs. Crookes Ltd., Renishaw. Scott, G. B Z.S.M. & P. Ltd., Empangeni Rail. -WHITLEY, W Z.S.M. & P. Ltd., Empangeni. SEILLIER, H. M Darnall Sugar Mill, P.O. Darnall. WICKES, C. F Entumeni, Zululand. SEYMOUR, G. E Sir J. L. Hulett & Sons Ltd., WILMOT, G. L Reynolds Bros. Ltd., Sezela.

Amatikulu. WILSON, G. C. 131 Folkestone Road, Rossburgh. SEYMOUR, R. W Hulett's S.A. Hennery Ltd., WILSON, J. S Z.S.M. & P. Ltd., Empangeni Rail.

Rossburgh. WYATT, W P.O. Doornkop. SHERRARD, C. D Farm 137/Mtubatuba. SHUKER, H. W Sir j. L. Hulett & Sons, Felixton. WEST, J. A P.O. Box 24, St. Michael's-on-Sea. SIMPSON, W. H P.O. Box 14, Empangeni. SPIES, A. T P.O. Box 106, Eshowe. WARREN, G. T Antigua Sugar Factory, Antigua, STEWART, E. A New Guelderland. B.W.I. STEWART, G New Guelderland. WILLIAMS, J. A Sena Sugar Estates Ltd., Marromeu. STEYN, C. L P.O. Box 59, Port Shepstone. SUI.IN, G. C P.O. Box 1501, Durban. SUTHERLAND, J. R. D. ... c/o E. L. Bateman Ltd., Durban. YOUNG-THOMPSON, I. C. P.O. Box 55, Stanger.

INSTRUCTIONS TO AUTHORS.

1. All papers for the Congress must be in the hands of the Technical Secretary fourteen days before the meeting. It is requested that authors wil l endeavour to send their papers earlier than this date so as to facilitate the work of printing.

2. Where possible the manuscript should be typewri t ten; when not possible the paper should be submitted in a form easily read.

3. References at the end of the paper should be arranged as follows : Name and initial(s) of author; year of publication in brackets; exact t i t le of paper; contracted tit le of periodical; volume number; beginning page number of article. Thus:— Brown, S. J. (1933): Study of Streak Disease. Proc. S.A. Sugar Tech. Assoc, 7, 101.

4. All the authors referred to should be arranged in alphabetical order.

5. Illustrations and diagrams accompanying the papers must be carefully drawn on smooth, white drawing paper in Indian ink so as to be suitable for reproduction. Any lettering on margins should be in pencil so as to allow of easy replacement by printer's type. The size of the largest dia­gram, curve, etc., should not be greater than 10 by 15 inches. Curves should be drawn on black graph paper, which can be obtained from the Technical Secretary.

6. Careful correction of typographical and other errors should be carried out by authors; this would save time on the parts of the Editor and Printers.

7. A summary of the main features described in the paper should be given at the end of the paper, immediately before the REFERENCES.

8. Authors may have 25 copies of their papers on application to the Technical Secretary. Any additional copies required by the author wil l be charged for.

9. Papers read at the Congress wil l not necessarily be published in the Proceedings.