THE BRITISH FERN

GAZETTE

VOLUME TEN PART FIVE

1972

THE JOURNAL OF THE BRITISH PTERIDOLOGICAL SOCIETY

THE BRITISH FERN GAZETTE

VOL. 1 0 PART 5

Contents

F R A S E R -J E N K I N S , C . R . & C O R LE Y , H . V . Dryopteris caucasica ancestral d ip loid in the Male Fern aggregate

an

R U THE R F O R D , A . & ST I R L I N G , A . M cG . Polypodium australe Fee and the

1972

221

tetraploid hybrid in Scotland 233

B O U H A R M O N T , J. Meiosis in apogamously produced .d ip loid plants of Asplenium septentrionale . 237

W A L K E R , T. G . The anatomy of Maxonia apiifolia: a cl imb ing fern . 241

M I T CHE L L , D . s. The Kariba weed : Salvinia molesta 25 1

P R I C E , M . G . Notes on Phi l ippine ferns 253

LO V I S, J. D . , B R O W N S E Y , P. J ., S L E E P , A . & S H I V A S , M . G . (M R S T R E V O R

W A L K E R ). The origin of Asplenium balearicum . 263

R I C K A R D , M A R T I N H . The d istribution of Woods1a ilvensis and W. alpina in Brita in 269

S L E E P , A N N E & S Y N N O T T , D O N A L . Polystichum X il/yricum: a hybrid new to the British I sles 28 1

R E V I E W S 232, 236, 283

The British Fern Gazette Volume 1 0 Part 4 was publ ished on 1 5 October 1971.

Printed by Econoprint Ltd., Edinburgh.

BR IT. F E R N GAZ. 1 0(5) 1 972

DRYOPTERIS CAUCASICA- AN ANCESTRAL DIPLOID

IN THE MALE FERN AGGREGATE

C. R . F RASE R-JEN KINS* & H .V . CO R L EY * *

ABSTRACT

The combination Dryopteris caucasica is made, based on Aspidium caucasicum A. Braun 1 84 1 , which was a substitute name for F ischer & M eyer's i l l egitimate A. affine of 1 838. D. caucasica is described , and regarded as the other parent, with D. abbreviata, of D. filix-mas. Three hybrids are new l y recognised: D. x euxinensis (caucasica x filix-mas), D. x initialis (abbreviata x caucasica) and D. x mantoniae (abbreviata x filix-mas) .

I NTRODUCT I ON

221

l t has long been known that Dryopteris filix-mas sensu lato includes a number of more or less closely related species in Europe and Western Asia. More recent ly since the work of Professor I . M A N T O N ( 1 950) it has been possible to separate from the group an apogamously reproducing species or aggregate of species various ly described as Dryopteris borreri Newman, D- pseudomas (Wol laston ) H olub and Pouzar, D. mediterranea Fomin and D. paleacea ( M oore) Fom i n etc. The classification of th i s complex group, which could be main ly of hybrid origin, may yet present further problems.

The two remaining taxa are both sexua l ly reproduc ing and are the tetraploid D. filix-mas ( L . ) Schott sensu stricto, and the d ip loid D. abbreviata (DC. ) Newm. ( includ ing D. oreades Fomin and D. coreanomontana Naka i ) . M A N T O N ( I .e. ) by the cytological exami nation of hybrids (artific ia l ly made) between the two and by the subsequent production of an apogamously induced d i ploid of D_ filix-mas, showed that D. filix-mas is an a l lotetraploid species to wh ich D_ abbreviata is part parental . The other ancestral d ip loid i n this complex remained unidentified a lthough many of its morphological features can be deduced from a study of D. filix-mas and D. abbreviata. The d iscovery, by C . R . F .-J. , of a Dryopteris with an unusual morphology in Turkey in 1 968 was therefore of considerable interest. The plant, however, received a severe set-back and was nearly lost, and it was not until 1 970, when another exped ition to Turkey was mounted , that more thorough examination of this and severa l other s imi lar populations could be made.

As a result of further study and botanical expeditions to Turkey and the Caucasus it now appears that plants of th i s nature constitute the m issing ancestral d iploid species in the Dryopteris filix-mas aggregate.

NOMENCLATUR E , TYP I F I CATION A N D DESCR I PT I ON

Dryopteris caucasica is not new to science but has been overlooked by present-day workers. lt was described as Aspidium affine by F i scher & M eyer (in H ohenacker, 1 838) . This name was i l l egitimate having been used previously ( 1 828) by B lume for another entirely d ifferent taxon . In 1 84 1 Alexander s R A u N corrected the situation by citing the Fischer & M eyer description and using a new specific epithet, which is now transferred to the genus Dryopteris by the present authors.

*The University, Leicester; now at Charterhouse School, G odalming, Surrey. ** Pucketty Farm, Faringdon, Berks.

222 BRITISH F E R N GAZETTE : VOLUME 1 0 PA RT 5 ( 1 972)

F I GURE 1 : Distribution of Dryopteris caucasica (cross hatch ing) ; stippled areas = doubtful occurrence.

Dryopteris caucasica (A. Br.) F raser-Jenk ins & Corley comb_ nov.

2

Basionym:

Synonym :

Typification.

Aspidium caucasicum A. Br. , Flora 24:707 (7 Dec 1 84 1 ) nom . nov., type as for A. affine.

A sp idium a ffine Fisch. & Meyer ex Hohenacker, Bull. Soc. Nat. Moscou 1 1 : 240 ( 1 838) ; non B lume nee Wal l . ( 1 828) . Type: Khanlar, Azerbaydzhan S.S. R . , Hohenacker, 1 836; L E ! (See below ) . Dryop teris filix-mas (L.) Schott var. athyriformis Fomin, Vest. Tifl is. Bot. Sada 20:35 ( 1 9 1 1 ) . Syntypes from Prov. Terek and Tifl is (TB I ! ) .

I n material loaned to us from the Botanical I nstitute, Len ingrad , a specimen labelled : "Aspidium affine, Helenendorf, Ace. a D. Hohenacker 1836", in Meyer's handwritin'g appeared to be a good candidate for the holotype. lt is probab ly part of H ohenacker's col lection wh ich was publ i shed under the name of Aspidium filix-mas col lected " I n provincia Karabach et in territorio E l izabethopolensi" (H o H E N AC K E R 1 833, p.260) and

F I G U R E 2: Si lhouette of frond of D. caucasica, X �; CRFJ 2854, Ritsa, Abkhazia, U .S.S.R .

F R AS E R-JE N K I NS & CORLEY : DRYOPTE RIS CAUCAS I CA 223

which Hochenacker mentions in F i scher and Meyer's description publ ished by ,h im in 1 838. H elenendorf is the old name for Khanlar, a v i l lage 8km south of K i rovabad.

This is the only specimen we have been able to d i scover which was definitely known by F i scher and Meyer at the t ime they wrote their description. Not only was it label led py Meyer i n 1 836 'Received from H ohenacker', but it also bears another label in Meyer's handwriting which gives a d iagnostic description s imi lar to that pub l i shed in Hohenacker ( 1 838) . A. E . Bobrov of Leningrad states (pers. com m . ) that there is also a second exact dupl icate of the holotype specimen in herb. LE .

Materia l col lected by Hohenacker i n Herb. Boissier, Geneva, wh ich A.H .G . Alston has label led as 'Type Specimen', has no date or precise local ity, and it is far from certain that F i scher and Meyer used it as a basis for their description.

Description.

Plant of gracefu l habit, very variab le. Rhizome up to 35 X 1 0cm ; apex erect and clothed in lanceolate, concolorous, brown scales; older parts procumbent and usua l ly unbranched . Fronds up to 1 05cm long, eglandu lar i n a l l parts, in an erecto-patent rosette, frost-sensitive and not persistent in winter. Petiole ( 1 /6-) i /4-1 /2 as long as blade, very pale yel lowish green or wh itish when mature, sparsely clothed with concolorous, brown, lanceolate to l inear-l anceolate scales up to 2cm long, wh ich become toothed at the margin towards the attenuate apex ; rhach is s imi lar with scales becoming paler and hair- l ike towards apex. Lamina up to 80 X 35cm, ovate- lanceolate to el l i ptic, occasiona l l y truncate at base, fl at, soft in textu re, usua l ly pale green with the lower surface even paler, b ipinnate. Pinnae up tq 20 X 5cm, lanceolate or frequently e l l iptic with attenuated apex , often asymmetrical , the basiscopic pinnu les being longer, pinnate below, pinnatisect h igher up. Pinnules narrowly lanceolate, truncate at base, apex acute; the lowermost shortly stipitate with winged sta l ks, the basal pair often shorter than the pair above. Median pinnules sessi le, usua l l y much the longest, the basiscopic more so than the acroscopic; margins deeply lobed to almost entire, lobes truncate with 2-5 acute teeth; the lowest basiscopic lobe of the lowest basiscopic pinnule somewhat larger, with up to 1 0 teeth, but not forming a large auricle as in D. abbreviata; teeth very acute and distinct at pinnu le apices. Stomata*

44-50,um. Sari (2-) 4--9 pa i rs on each pinnu le; indusia 1 -2mm d iameter, flat, circular or somewhat extended at one side, very thin and membranous, white, without chloroplasts, eglandular, greatly overlapping the sporangia before shrinking to a sma l l size, edges lacerate. Spores (30.8-) 33.0 (-37 .2) X ( 1 9.9-) 22.6 (-25 .3 ) ,urn**

brown, verrucate with very short ridges 2-4,um tal l , densely scattered over the surface; ripe June to August.

Sexual d ip loid, 2n=82. Habitat in forests and forest clearings with high rainfa l l , usua l ly above 750mm per annum ; from 700 to 2,700m altitude, dy ing out above the forest belt.

Geographical range ( F ig. 1 t: B lack Sea coast of Tu rkey , the Caucasus Mts and Transcaucasia, and the southern Caspian forest area of I ran. lt has not been possible to include here a map of the range of D. abbreviata, but it may be said that the two ferns share a large area where they both occur, and there is every reason to suppose that hybrid ization is sti l l taking place between them.

* Tissue b leached in sod ium hypoch lorite solution and mou nted in water; 20 stomata measu red from med ian pinnule of med ian pinna of specimens from each loca l ity.

* * I nner wal l of spore measured dry; average of 20 from each locality.

224 B R ITISH F E R N GAZETTE: VOLUME 10 PART 5 (1972)

Variation: D. caucasica displays a w i de range of variation, partly as a response to conditions of shelter and moisture. I n Turkev. at least, spec imens growing i n very sheltered and luxu riant conditions were seen to atta in an elegant fol iose form with very w i de pinnae, the m ed ian pinnules greatly devel oped and overlapping those of the next pinna. Such specimens also have very markedly asymmetric pinnae, the basiscopic pinnu les being much the longest. However, in cu ltivation in more open conditions in Brita in these features were lost.

The depth of l ob ing of the pinnu les a lso varies greatly, as does the length of the pinnule teeth , both being less in more exposed habitats.

Other features remain constant a lmost w ithout exception, notably the acuteness of the pinnule teeth, the flatness and th inness of the indusium, and, when fu l ly developed, the smal l size and ornamentation of the spores.

The table on p. 228 emphasizes features of distinction between D. caucasica and the species most l i kely to be confused w ith it. l t is noticeable that, when comparing it with D. abbreviata and D. filix-mas, its morphology is more . or less what wou l d be expected for the other parent of the latter. As the D. villarii/pallida cqmplex has been suggested as containing the other parent of D. filix-mas (w 1 o EN 'et a l . · 1 97 1 ), we include data on the morphology of that group in the table. o,he� herbarium s'pepimeru seen include the fol lowing (col lected by C . R . F .-J. u n less otherwise stated ) :

TUR KEY

Bolu : W si de of Bolu Dagi , above Kayna�l i , 1 8km.west of Bolu, Rhododendron scrub, 900m alt. , 18 Aug. 1 970, 2081, 2082 ( B M and herb. C R FJ ) ; c.5km W of Yeni<;aga, Quercus/Carpinus scrub, 1 3 Sept. 1 970, 2,409· (BM and herb. C R FJ ) . Trabzon : 2km below Ku�ukkonak, Zlgana Pass, above. Trabzon, m i xed Fagus forest by stream, 5 Sept. 1 970, 2270-2275 ( B M , herb. C R FJ and herb. Reichstein ) . G um i.isane : c . 1 0km bel ow top of Zigana Pass, above Trabzon, c learing in Picea forest, 5 Sept. 1 970, 2287a, 2287b, 2288 (BM and herb. C R FJ ) .

U.S.S.R.

Abkhazskaya A.S.S. R . Lashipse-Bzib val ley, 25km above lake R i tsa, acid rock in coniferous/hardwood forest, 24 June 1 97 1 , 28 18-2824 (BM & herb. C R FJ ) ; 1 5km above Lake R itsa, acid rock in m i xed con i ferous wood beside road, 28 June 1 97 1 , 2850 (herb. C R FJ ) ; Bzib val ley, 1 9km below Lake R itsa, l imestone i n l ight coni fer/hardwood forest beside road, 28 June 1 97 1 , 2854 (BM & herb. C R FJ ) ; 20km below Lake R itsa, l imestone in th ick, c l imax con ifer/hardwood forest, 'hkm from road, 28 June 1 97 1 , 2867 (herb . C R FJ ) ; 35km above turni ng into val ley W of R i tsa val ley, below Gora Agepsta, Gagrinskiy Khrebet, non-calcareous rocks i n Abies/Fagus forest, C . 1 800m alt., 29 June 1 97 1 , 2889-2896 ( B M , L E , E , herb. C R FJ and herb. Reichstei n ) . Rossiya S. F .S. R . : Kransnodarskiy J(ray, between Chvezhipse a n d Krasnaya Polyana, N of Adler, Mzymta val l ey in Fagus forest, 27 June 1 97 1 , 2839 (herb. C R F J ) ; Karachayevo-Cherkesskaya A.O., in valley between Teberda river a n d K lukhurskiy Pereva l , below Gora Dombay-UI'gen, Teberda val ley, in con i ferous/hardwood forest, c.2000m alt . , 1 1 July 1 97 1 , 3201-3208 (BM and herb. C R F J ) .

PLATE X X I II. Pinnae o f Dryopteris caucasica (A-D) and D. filix-mas (E-G ) , all natural size and all coli. C R FJ. A: 2850, R itsa, Abkhazia. 8: 2288 , Zigana Pass, Giimu�ane, Turkey. C: 2287, same loc. 0: 2270, same loc. E: 2200, l lgaz D ag, �ankiri, Turkey. F: 1130, G erlos Pass, Tyro l , Austria. G: 989, M t . Harghitei, MurerM agyar, R oumania.

PLATE XXITI

PLATE XXIV

F RASER -J E N K I NS & CORLEY : DRYOPTE R IS CAUCAS ICA 225

Kabardino-Balkarskaya A.S.S. R . : 1 0km below Terskol ( =ltkol ) on Bukhaidze road below Gora .Eibrus, Baksan val ley, Pinus forest, c.2000m alt., 1 0 J u ly 1 97 1 , 3076-3079 ( B M , LE, G , E and herb. C R FJ ) . Azerbaydzhan S.S. R . : i n s i lvaticis ad montem schuscha ditionis E l izabethpol [ Khanlar, Ki rovabad] , Hohenacker, Unio itiner., 1 838 ( B M ) ; prov. Baku, distr. Kuba, prope Ozero Kuba, Do/uchanov & Tzaturov, 27 June 1 930 ( B M ) ; Transcaucasia, Karabach meri d., prope Zamur in s i lvis 1 750- 1 800m alt., Karjagin & Minulin, 3 Ju ly 1 932 ( B M ) .

I R AN

Gorgan : Mohammed Reza Shah reserve, 60-80km E N E of Gonbad-e-Oabus, 6km S of Tangar, in l ight Fraxinus wood, in clefts at foot of N -facing l imestone c l iff, 1 050m alt. , 1 9 July 1 9 7 1 , Edmondson 722 ( L TR and herb. C R FJ ) .

I n addition a very large number o f other specimens, from throughout the Euxine region, were seen at the Botanical Institute Tb i l isi (TB I) .

CYTOLOGY AN D HYBR I DS

Little cytological work has been carried out so far beyond chromosome counts to discover the. level of ploidy . These were prepared from root tips in the standard way, using snai l cytase, and being made from specimens collected wi ld and cu lt ivated at the Un iversity of Leicester Botanic Garden. The fol lowing specimens were found to be diploid with 2n=82 :- C R FJ no.2082, C R FJ no.2409, C R FJ no.2272 and C R FJ no.2889 (see above for local ities ) .

The remarkably sma l l spore size of the other specimens not yet cytol ogically examined, being consistent with that of those mentioned above, points to their also being dip lo id.

Certain specimens intermediate in morphology between Dryopteris caucasica and other species in the D. filix-mas group were found to have abortive spores and, on morphological grounds, were assumed to be hybrids, and descriptions fol low.

Dryopteris X euxinensis F raser-Jenkins & Corley, hybr. nov. [=D. caucasica X filix-mas}

Hybrida, media inter parentes, i nfime pinnulae minores dissectae quam D. caucasica; sporae abortivae.

Holotypus: Turkey, Sinop, between Yenikonak and Gokceagac, south of Ayancik, in Pinus/Fagus forest beside road near top of pass on north side, 1 Sept. 1 970, C. R. Fraser-Jenkins 2234 ( B M ) ; isotypi: i n LE, E and G .

Other specimens seen include the fol lowing: T U R K E v Bolu: W si de of Bolu · Dagi , above Kaynas l i , 1 8km west of Bolu, Rhododendron scrub, 900m alt . , 18 Aug. 1 970, C. R. F-J. 2080* ( B M ) . u . s . s . R . Karachayevo-Cherkesskaya A.O . : Severn iyy Priut, N side of K lu khurskiy Pereval below Gora Dombay-u l 'gen, Teberda valley, Pinus forest, 2300m alt., 1 1 July 1 97 1 , C. R.F-J.3209 ( B M and herb . C R FJ ) .

PLATE X X I V. Pinnae of Dryopteris abbreviata (A-C) and D. pallida [2n] (D-F). A : Tutin 4113, Beattock, Dumfriessh ire, U . K. X 2. 8: CRFJ 2841, R itsa, Abkhazia, U .S.S. R ., X 1. C: CRFJ 1895, Col de Verde, Corsica, F rance, X 1Y:.. 0: CRFJ 2123, Akseki, Antalya, Tu rkey, X 1 Y:.. E: CRFJ 2126, same loc., X 2. F: Reichstein 1544, Astara, R asht, I ran, X 1 (as D. raddeana Fomin).

226 B R I TISH F E R N GAZETTE: VOLUME 10 PART 5 (1972)

3 FIGU RES 3 & 4: Si lhouettes of Dryopteris fronds. 3: D. filix-mas X Y.; CRFJ 3262, Kiyev, Ukraine. 4: D. abbreviata X Y., CRFJ 2897, R i tsa, Abkhazia.

FRASER·JENKINS & CORLEY; DRYOPTERIS CAUCASICA 227

5

FIGURES 5 & 6: Silhouettes of Dryopteris fronds. 5: D. villaril (diploid) X %, CRFJ 1175, Mt. Lessini, Vicenza, N Italy. 6: D. pallida X %, CRFJ 2126, Akseki, Antalya, Turkey.

228

RH I ZOME

FROND$

PETIO L E

LAM IN A

PI N N AE

PI N N UL ES

ULTI M ATE SEGM E N TS

STOMATA LEN GTH

INDUS IA

SPO R ES

PLO I DY (x=41 )

HAB ITAT

B R I T I SI-l F E R N GAZ E TTE: VOLUME 10 PART 5 (1972)

D. CAUCASICA

Unbranched.

Eglandu lar, spreading, frost­sensitive.

1 /21ength of frond, sparsely clothed with wide scales.

Ovate-lanceolate, truncate at base, herbaceous in texture and usual ly paler green than any of the other three species.

F lat, ± horizontal, w idest i n middle.

Apex acute, with very acute teeth ; up to two lowest pairs of each pinna stipitate, basal pai r often shorter than those above; basiscopic longer than acroscopic.

Truncate with several teeth, basal segments of basal p in ­nu les not forming auricl es.

41 -50�

1 -2mm diam ., flat, very th in and membranous, eglandu lar, white at a l l stages, consider­ably overlapping sporangia, · shrinking at maturity; margins lacerate.

31 "-37 J.Lm long, verrucate or shortly ri dged.

Diploid.

Forests in alpine regions.

D. FILIX-MAS

Little branch ed .

Sometimes glandu lar, spread­ing, semi -persistent.

c. 1 /3 l ength of frond, ± densely clothed with wid'e and narrow scales.

Ovate-lanceolate, truncate at base, ± herbaceous.

F lat,± horizontal, w idest at base.

Apex acute, with ± acute teeth ; basal pair of each pinna

± stipitate, usua l ly longer than those above; basiscopic usually equal ly l ong as acro­scopic.

Not truncate with few teeth, basal segments usually form· ing auricles.

48-65J.Lm

0.5-2mm diam., convex, thin, sometimes glandu lar, white when young, brown when o ld, sl ightly overlapping sporangia, shrink ing at maturity; marg i ns entire.

37-45J.Lm long, usual ly verru­cate with sma l l lobes and one or two wings especial ly at end of spore.

Tetraploid.

Forests or open places on rocks lowland or a lpine.

F R AS ER -J E N K I NS & CORLEY : DRYOP TE R IS CAUCAS ICA 229

D. ABBREVIATA

Much branched.

Often glandu lar, upright, frost­sensitive.

c. 1 /4 length of frond, densely clothed with w i de and narrow scales.

Lanceolate, tapering to base, ± coriaceous.

Usua l ly concave, i ncl ined to apex, widest at base.

Apex obtuse or rounded with blunt teeth; basal pair sessile or slightly stipitate, longer than the pair above; b<!siscopic equal ly long as acroscopic.

Not truncate, with few or no teeth, basal ones formi ng auricles.

46-5811fll

0.5-lmm diam., h ighly convex, thick,, glandu lar, green when young, grey-brown when ol d, fitting closely round sporan· gia, scarcely shrinking; margins entire.

36-40J.Lm long, verrucate or lobed to broadly and longitudinal ly w inged.

Diplo id.

Open screes and banks in a lp ine regions.

D. PALLIDAIVILLARII

Much branched.

Very gla11du lar,. spreading or upright, . frost-sensitive.

1 /2 length of frond, densely clothed with w i de scales.

±triangular-lanceolate, truncate at base, ± coriaceotis.

F lat, i nc l ined or horizonta l , w idest at base.

Apex acute, with very acute teeth ; lowest 2-3 pairs of each pinna stipitate, basal pair longer thari the pair above; basiscopic equa l to or s l ightly shorter than acroscopic.

Not truncate.' with several teeth, basal ones not forming pronounced auricles.

Variable depending on ploidy .

c . 1 mm diam. , convex or rarely f li!t, thick, glandu l ar, green when young, grey-brown when old, rarely completely covering sporangia, scarcely shrinking; margins entire or lacerate.

Size variable depending on ploidy, irregu larly and/or shal lowly l obed.

Diploid and tetraploid.

Open screes, banks and in forests in alp ine and lowland regions.

230 B R ITISH F E R N GAZETTE : VOLUME 10 PART 5 (1972)

Kabardino-Balkarskaya A.S.S.R . : 1 0m below Terskol (=ltkol ) on B ukhaidze road below Gora Elbrus, Baksan val ley, Pinus forest, 200m alt., 1 0 July 197 1 , C. R.F-J. 3081, 3082, 3084 (BM and herb. C R FJ ) .

Dryopteris X initialis Fraser-Jenkins & Corley, hybr. nov. [=D. abbreviata X caucasica]

·

Hybri da med ia inter parentes sed dentes acuti , et pi nnulae parvae dissectae; simi l i s ad D. filix-mas ; sporae abortivae.

Holotypus: )J.S.S. R . , Karachyevo-Cherkesskaya A.O. , in val ley between Teberda river and Klukhurskiy Pereva l , below Gora Dombay Ul 'gen, Teberda val ley amongst rocks above conifer/hardwood forest, c.2 1 00m alt . 1 1 July 1 9 7 1 , C. R. Fraser-Jenkins 3210 ( B M) ; isotypi: i n LE.and G .

To complete the picture in the D. filix-mas group a thi rd hybrid is here described, that between D. filix-mas and D. abbreviata. This is the triploid made artificia l ly by Manton in her classic work quoted above. lt has since been found wi ld (by H.V.C.) in three stations in Wales and one i n the north of England, and by A.C. Jermy i n the north of England. We have also found it growing in a garden i n Oxford where the owners were unaware of what they were cu ltivating. The latter plant and the three from Wales have been cytologica l ly checked by S. Walker of L iverpool University, who found they were triploids with approximately equal numbers of pairs and singles.

In all these plants the spores are abortive except for a few large, spherical spores which, however, must be chromosomal ly unbalanced as they appear to be steri le.

In appearance these hybrids somewhat resemble D. filix-mas, but features such as the pinnule teeth and the i ndusium tend to be i ntermediate between those of the parents. l t is l i ke ly to be found wherever the parents grow together.

Dryopteris X mantoniae F raser-Jenkins & Corley, hybr. nov. [=D. abbreviata X filix-ma6]

Hybrida med ia inter parentes; s im i l i s ad D. filix-mas; pinnulae dentes et indusiae intermed iae inter parentes; sporae abortivae; cytotypus triploideus.

Holotypus: on wall of slate quarry on Moelwyn Mawr, c.480m, Merioneth, vc.48; 1 2 Aug. 1 962, Corley 625 ( BM) ; i n cultivation a t Faringdon. Other specimens cytologica l ly investigated : Caernarvon, vc.49 , Ll iwedd c.650m alt . , 29 Sept. 1 96 1 , Cor/ey 623 (BM and in cu lt. at Faringdon ) ; Tremadoc, c. 1 00m, four plants forming sma l l colony Corley 619-622 (NMW, 619 i n cult. at Faringdon ) . 624 (BM) . Oxfordshi re vc.23, i n private garden at Oxford, Corley 626 (BM ) . Jt has been recorded b y various authors from Europe and recently col lected b y C. R . F-J. i n Spain [Lean prov., Puerto d el Ponton between Cangas de Onis and R iano, 1 1 00m, 14 Ju ly 1 972, 3761 (BM) ; Soria prov., Puerto de Piqueras, Sierra Cebol l era, north of Soria, 28 July 1 972, 3809 (BM)) .

• I nvestigation of this plant showed there to be approx. 41 bivalents and 41 univalents at meiosis.

FRASE R-JEN.K I NS 19 .C.QR L,E¥: DRYOPTE R IS .CAU CAS ICA 231

ACKNOWLEOGEI\IlENTS

The authors wi!ih tg thi!nk the Percy Slad�n Memorial Trulit of the Linnean Society ;�nd the Godm<m Tru!it of the British Museum ( N atural History) for their financial assistance, enabling C.R. F-J. to mount a private botanical expedition to the Caucasus, and to thank the l qtter TriJSt for grants on two previous occasions.

Thanks are als.o due to Professor I . Manton, Dr J,O. Lovis and Dr A. Sleep, of Leeds University for their painstaking instruction and constant encouragement of C. R . F-J. in the techniques of counting ch rom osomes and breeding ferns. They also thank Dr G. Vida of Budapest, Professor T, Reichstein of B <!sel and Dr A. E. B obrov of Len ingrad for supplying much extremely usefu l information and material , and they wish especial ly to thank Mr A.O. Chater of Leicester University and Messrs A.C, Jerm y and J.A. Crabbe of the British Museum (Natural H istory) for encouragement, advi ce and help with obta in ing type specimens and essential information, and in the preparation of this paper,

H .V.C. wishes to thank Dr S. Walker of Liverpool University for a great deal of time and trouple in counting chromosomes an(] in gi ving much invaluable advice and encouragement.

F ina l ly C . R . F-J. wishes to express a gre!lt deal of gratitude to Professor T.G. Tutin of Leicester U niversity, who, as his supervisor, gave h im the chance to implement a programme of doctoral resP.arch and ironed out every d ifficu lty in h is way.

REFERENCES

BRA U N , A. 1 84 1 . Bemerkungen uber d ie Flora van Abyssinien. Flora 24: 70!)-720. HO HENACi<E R , R . f. 1 833. Enumer;�tio plantarum in territorio Elisabethopolensi et in provincia

Karabach sponte nasc!lntium, quas collegit R.Fr.Hotwnacker. /lull, Soc. Nat. Moscou 6: 2 1 0-26 1 .

MANTON, I. 1 950.Problems o f cytology and evolution in the Pteridophyta. Cambridge. Wl DEN , C.J., G. V ID A, J. VON EUW & T. R E I CHSTE I N. 1 97 1 . Die f?hloroglucide van Dryopteris

v i l lari i (Bell.) Woynar u nd anderer Farne cl!lr G attung Dryopteris sowie die mogliche Abstammung van D. filix-mas (L.) Schott. He/vet. Chimica Acta 54: 2824-2850.

232 BRITISH FER N GAZETTE: VOLUME 10 PART 5 (1972)

REVI�W

SYSTEMA TICS OF THE NEO TROPICAL SPECIES OF THEL YPTERIS SUBGENUS CYCLOSORUS by A/an R. Smith. University of California Publications in Botany 59: 1-136, 1971. 170 X 258mm. Price $6.50. {U. C.P. 2-4 Brook St, London W1 Y 1AA}

This is a detai led account of the species assembled by Christensen in Dryopteris subg. Cyclosorus in his monograph of Dryopteris ( 1 9 1 3 ); Dr Smith transfers the subgenus to Thelypteris. His account is based on extensive field work in the south-eastern United States, M exico and Costa R ica, on study of p lants from many sources in cul tivation by cytotaxonomic methods, and of more than 4000 herbarium specimens. I n this process he has gathered together a great wealth of new knowledge on a group of c losely related species which have h itherto been very i mperfectly u nderstood.

The total number of species dea lt with is seventeen, and eight natural hybrids have been found . lt is anomalous that Aspidium gongylodes Schkuhr, type of Cyclosorus, is not closely related to the other species which Christensen associated with it, and in fact Dr Smith 's most important new i nformation concerns these other species. They are found in more or less open or d isturbed vegetation, not in primary forest. They are d i visible into two sub·groups, one (3 spp) having known relationships in the Old World and includ ing a single d ip loid the d i stribution of which extends to Malesia (T.quadrangularis; see Brit. F ern Gaz. 1 0 : 1 89-1 90). the other sub·group ( 1 1 spp) being centred on Mexico and including three tetraploids (probably all of hybrid origin ) and five d iploids (a lso three species not cytologica l ly examined ) . The tetraploid T.normalis is i nterm ediate between the two sub-groups. Some of the tetraploids are as aggressive in cult ivation as in the wild; in the tropical fern-houses at Kew spores of T.patens, T.normalis and T.dentata often find their way into cu l tures of other species wh ich they may suppress.

T.quadrangularis and its a l l ies in the Old World belong to the genus Christella as recogn ized by me (Biumea 1 9 :43; 1 97 1 ). These Old World species appear to be centred on the region of N . E . I ndia, Burma and W.Chlna; they have not yet received a study at a l l comparable i n d etai l with that accorded to the American species as dea lt with by Dr Smith . As in America, the common species which colonize open places are tetraploid (T.dentata has spread very widely, fol l owing the destruction of natural vegetation by man, and is very variable; i n recent years it has i nvaded the Americas and so appears in Dr Smith's treatment). There are in Africa a few species which appear to belong to Dr Smith 's second sub-group, namely T.gueintziana, T.callensii, T.chaseana and T.microbasis. These have not yet been col l ectively d istingu ished from other free-veined Thelypteroid ferns in Africa, nor h ave they been stud ied by tra ined observers in the field . Only when they receive the same kind of study now accorded to the American species by Dr Smith sha l l we have a ful l understandi ng of the status of the sub-group, which I bel ieve should a lso be included in the genus Christella.

Or Smith's excel lent study is a model which I hope wi l l be fol lowed by others. l t indicates the subtle k inds of problems i nvol ved i n dea l ing with a closely-a l l ied group o f species, and also the vast scope o f the whole field o f fern taxonomy. l t further expresses the need for m onograph ic work as a background against wh ich studies of individual species can be judged .

R . E . HOL TTU M

B R IT. F E R N GAZ. 1 0(5) 1972

POLYPODIUM AUSTRALE FEE AND THE

TETRAPLOID HYBRID IN SCOTLAND

A. RUTH E R FO R D * and A.McG. ST I R L I NG * *

233

Unti l recently the occurrence of Polypodium australe F ee in Scotland was not suspected despite the possibi l ity that the reference in John SM I T H 's: Botany of Ayrshire, 1 896, to 'Polypodium vulgare var. serratum Wi l ld . ' for Largs arid West Ki lbride and in the British Association Handbook 1 901 (sT E W A R T p .. 1 09 ) to 'Po/ypodium vulgare var. semilacerum Link. ' for West K i l bride m ight refer to the diploid.

s v N No TT ( 1 970) d rew attention to the occurrence of branched paraphyses among the sporangia of some herbarium specimens of 'Polypodium vulgare' col lected in Scotland in the latter half of the n ineteenth century· and these were subsequently determined as P. australe by R . H . Roberts. These early records have been considered by R O B E R TS & S Y N N O TT ( 1 972 ) .

During late 1 9 7 1 and early 1 972 efforts were made to determi ne the present status of P. australe in Scotland using the early records as a basis. Up to the time of writing the authors have knowledge of eleven local ities for the d ip loid l ocated in eight vice-counties.

Vice-county 72 Dumfries

M o o R E ( 1 859 p.67 t. l l l E) mentions a broad·fronded variety of the Common Polypody havi ng been found at 'The Craigs, near Dumfries' . This record appeared to warrant further investigation and on 28th Ju ly 1 972 a visit was made to Maidenbower Craigs near the town of Dumfries a local i ty which appeared to be the most promising in the area for Polypody species. Polypodium australe proved to be plentifu l here, growing on rocks of Permian Sandstone breccia in company with P. interjectum and P. vulgare. This site piffered in two sign ificant respects from the previously exami ned Scottish sites. l t was fai rly heavi ly shaded by trees, th is appearing to inhib it considerably the abi l i ty of the plants to produce sporing fronds; a lso there were no obvious ind ications that the rock was particu larly basic.

Vice-county 73 Kirkcudbright

The earl iest Scottish specimen of P. australe appears to have been col lected by W.G. Johnstone in September 1 850 as 'P. vulgare var. serratum' from 'Col vend coast, Kirkcudbright', a somewhat i l l-defined localisation. On 20 October 1 97 1 P. australe was found groWing abundantly near the mouth of the Southwick Burn at the bas� of cliffs of Si lurian rock on the Solway shore (Grid . ref: 25/9 1 3 560). l t a lso grows a� an epiphyte on elder and ash . At this site P. interjectum is exceed i ngly luxu riant 'in the shel ter of trees, and with it was detected at least one colony of the' rare 4n hybrid P. X rothmaleri Shivas. This is the first record for the tetraploid hybrid in Scotland and only the second for the British Is les. On 8 January 1 972 a second K irkcudbrightsh i re local ity for the d iploid was found on c l i ffs near Port Ling, some two mi les west of the Southwick local ity, also on Si lu rian strata. In contrast this site is in fu l l exposure though south-facing l i ke the other. H ere a lso a colony of P. X rothma/eri grows with the putati ve parents close by. * Rosslyn Cottage, Church Rd, Rhu, Du nbartonsh ire, G84 BRW ** 1 7, Austen Rd, Glasgow, G 1 3 1 SJ.

234 BR I T ISH F E R N GAZETTE : VOLUME 1 0 PART 5 ( 1 972)

Vice-county 75 Ayr

A. Somerv i l le i n 1 896 col l ected 'Polypodium vu/gare var. acutum' from 'woods, Portincross, Ayrshire'. On 1 7 September 1 97 1 a sma l l colony of P. australe was found with some P. X mantoniae Rothm. (the 5n hybrid ) among a larger popu lation of P. interjectum growing on conglomerate of the Old Red Sandstone, l ocal ly known as 'Pudd ing Rock' , part of the raised beach cl iff about a quarter of a m i le north of Portencross Castle (Grid ref: 26/1 7749 5 ) . On 3 January 1 972 at H eads of Ayr (Grid ref: 26/287 1 87) considerab le outcrops of basic volcanic agglomerate were found to support extensive colon ies of the d iploid which grows at a l l e levations above six feet above h igh-tide mark but is most abundant towards the cl iff top at about 200 feet (60m ) . The favoured habitats are exposed ledges and knol ls , often partia l ly concealed by grass and almost invisible from below. In sparse wood land, main ly ash, at the northern end of the cl iffs in the shelter of the trees, fronds of P. australe· frequently atta ined a length of 30 cm includ ing the stipe, with the width across the broadest part of the frond up to 1 5 cm . At th is loca l ity the d iploid grows epiphytica l ly on ash and hawthorn . A further Ayrsh ire local ity was d iscovered on 9 January 1 972 at Lenda lfoot (Grid . ref : 25/1 27895) where P. australe grows in very sma l l quantity on u ltrabasic rock. This site is on the coast six m i les south-west of G irvan. Although su itab le basic coastal rocks were searched from Kennedy's Pass, just south of G irvan, as far as Bennane Head , north of Ballantrae, a d istance of about six m i les, no other colon ies of the d iploid were fou nd .

Vice-county 83 Midlothian

A sheet in the herbarium of the Royal Botanic Garden, Ed i nburgh bears two sma l l fronds of 'P. vu/gare' col lected on Arth u r's Seat, Edinburgh by J . McNab in 1 855. These superficial ly suggest the d iploid, having narrow, d istinctly serrate pinnae, but are not otherwise determ inable with certainty. A far from thorough search of the south-facing rocks of Arthur's Seat on 7 Apri l 1 972 fai led to revea l any P. australe. Soon after this it came to the notice of the authors that B. Simpson of Seami l l , Ayrsh ire had , in a col l ection of ferns made i n the Edinburgh area, a frond of 'P. vu/gare' from 'south-facing rocks, Arthur's Seat' col lected between Ju ly 1 957 and December 1 958. Th is frond is broad ly ovate in outl ine and has narrow, strongly serrate pinnae. Sori were present and were sufficiently mature for microscopic examination to reveal the presence of branched paraphyses among the sporangia thus confirming the occurrence of the d iplo id in this local ity. This represents the earl iest recent record for P. australe in Scotland as well as being one of the few records for the east side of the country. The hab itat at Arthur's Seat is steep basalt rocks with a strong basic character. The plant was aga in seen in this local ity by one of the authors on 23 May 1 972. i t appears to be confined to one comparatively l imited area of rock.

Vice-county 89 East Perth

We have seen a specimen of P. australe from this vice-county col lected by M rs. J. Castle-Smith in a loca l ity which we are not at l iberty to d isclose. This plant was col lected in 1 969.

Vice-county 98 Argyll

On 5th September 1 972 three colonies of P. australe were found i n close proximity on south-east facing l imestone rocks of the raised beach cl iff just to the south-west of Achnacroish on the is land of Lismore (Grid ref: 1 7/846401 ). This record is of considerable interest, representing not only the fi rst recorded occurrence

RUTH E R FO R D & ST I R L ING: POLYPOD I UM AUSTRALE 235

of the d i ploid on l imestone in Scotland , but also the most northern site yet located in the British Isles and, indeed, in Europe.

Vice-county 100 Clyde Isles

The south coast of the I sle of Arran appeared to be a l i kely locus to search for P. austra/e and during visits on 2nd and 1 3th June 1 972 a few colonies were found on cl iffs in the vicinity of Bennan Head ( Grid ref: 1 6/99-20- ) . Due to the rough nature of the ground in th is area it was not possible to examine more than a short stretch of c l i ff in the time ava i lable. Further search may wel l show the d ip loid to be more widely d istr ibuted on the island .

Vice-county 102 South Ebudes

On 4th June 1 972 during a v isit to I slay a colony of P. australe was d iscovered growing on a quartzite rock outcrop i n a sand d u ne area at Ki l i na l lan in the north of the island (Grid ref; 1 6/3 1 3722) . Quartzite is norma l ly a very poor rock type as regard s the qual ity of the flora which it supports. I n th is instance however there was enrichment of the soi l in the rock crevices by blown sand from the dune area. A sim i l ar example in Wales is mentioned by H u G H E S ( 1 969 ) . Although considerable exposures of l imestone occur on I slay P. australe was not detected on th is formation.

Other records

The old records for vice-county 77 Lanark, and vice-county 85 F i fe (sYN N O TT 1 970; R O B E R TS & SYN N O TT 1 972) sti l l lack recent confirmation, search in the original localities having so far proved negative.

Continued search wi l l undoubted ly lead to the discovery of add itional sites for P. australe in Scotland and it is l i kely to prove widespread though extremely local south of the Central H ighlands.

ACKNOWLEDGEMENTS

Our thanks are due to M r R . H . R oberts for determi nations of Po/ypodium material and for help in other ways and to the Regius Keeper, Royal Botanic Gardens, Ed inburgh, and Mr D.M. Synnott, National Botanic Garden, G lasnevin , Dub l in, for the loan of herbarium specimens.

REFERENCES

H UGHES, W.E . 1 969. The distribution of Polypodium vu lgare L . subspecies serratu l um Arcangel i i n N orth Wales. Nature in Wales 1 1 : 1 94-198.

MOOR E , T. 1 859. The octavo nature-printed British ferns, vol. 1 . London. ROB ERTS, R . H . & D.M. S Y N NOTT 1 972. Polypodium australe Fee in Scotland and north-east

I reland. Watsonia 9: 39-41 . ·

SM ITH , J. [ 1896]. The botany of Ayrshire. Ardrossan. STEWA RT, W. 1 90 1 . Ferns and their a l l ies. In Elliot, Laurie & Murdoch (ecU, Fauna, flora &

geology of the Clyde area. (Brit. Ass. Adv. Se. Handb.) G lasgow. SY NN OTT, D.M. 1 970. Evidence of Polypod ium australe F�e in Scotland. Trans. Proc. Bot. Soc.

Edinb. 40: 623-624.

236 B R IT ISH F E R N GAZETTE: VOLUME 10 PART 5 (1972)

REVIEW

A COMPUTOR-MAPPED FLORA by D.A. Cadbury, J. G. Hawkes and R. C. Readett; v + 768 pp., 12 overlays and geological map in black pocket; 225 x 290mm. Academic Press, London and New York, 31 Dec. 1971. Price £10 ($30.00)

This is a vascu lar plant and bryophyte flora of v.-c.38, Warwickshire with a l l the usual subsidiary chapters on soils, geology, c l imate, h istory and land-use by d istingu ished scholars in their fields. Notes on the vascu lar plants present ( i ncluding 41 pteridophytes) are given in 1 1 0 pp. The bu lk of the work consists of 979 ma"ps ( 2 1 ferns and 1 83 bryophytes). The base-maps are printed in b lue and the symbols have been printed by an incremental plotter with a drawing pen control led by a punched tape produced by a computor (at Birmi ngham University). Maps of the bryophytes and a number of vascu lar plants (including Polypodium vulgare s.s. and P. interjectum and Cystopteris fragilis) have been produced by the more conventiona l l i ne printer. A feature of the m apping has been that records have been made a!'ld plotted on a basis of nine major habitats. Early in the project a 1 -km sq in each tetrad was randomly chosen and recorders were asked to concentrate on these thus e l im inating recorders' bias for the more interesting hab itats. N i ne symbols (at two l i ne thicknesses to ind icate frequency of species) may be plotted in each 1 -km square and maps look con fusing unti l one has learnt to read the symbols. The overlays are of course usefu l , but I wished they had narrower borders to a l low easy marrying by using the spine of the book as a gu ide; as it is it must spread across to the other page.

As far as the pteridophytes are concerned there is ample data for thought. Pteridium is recorded for the heavy calcareous soi ls on S and SE but 41 of the records are from hedgerows and scrub . Another interesting habitat record is in Equisetum telmateia where 44 records are from watersides and marshes. A glance at the map wi l l tel l that most of the other records are from hedgerows, waste places and roadsides and that only in 8 squares are other habitats encountered (wood land 6, and grassland and cu l tivated one each ). lt is interesting to note how many ferns turn up on waste places although this may be due to recorders not correctly noting habitats. The distribution of ferns is governed by soi l rather than cl imate it appears.

One very good aspect of the F lora is the notes (and occasional keys and figures) given on characters found useful when identifying that species in Warwicks. Thus it is rightly noted that in Polypodium, frond shape and sori (shape?) are not satisfactory characters to d istinguish segregates and that the annulus number shou ld be checked. One cou ld add colour of annu lus in relation to sporagium wal l, and shape of rhizome scale.

This book is the resu l t of a wel l thought out project and at £ 1 0 may not be on everyone's bookshelf. But every field botanist shou ld read it and browse the maps and I am sure many wil l be tempted to buy it . .

A.C. J E R MY

BRIT. F E R N GAZ. 10(5) 1 972 237

MEIOSIS IN APOGAMOUSLY PRODUCED DIPLOID PLANTS O F

ASPLENIUM SEPTENTRIONALE

J. BOU HAR MONT*

ABSTRACT

Diploid sporophytes of Asplenium sep tentrionale have been obtained by cultivation of gametophytes from the normal tetraploid cytotype. These sporophytes were sterile and their meiosis was very irregu lar: at metaphase I, only 37% of the chromosomes were paired. The higher prophase pairing confirmed the autoploid nature of the original tetraploid, but the homology of its genomes is relatively low comparatively to the tetraploid A. ruta-muraria and A. trichomanes.

I NTRODUCT I ON

One of the best ways to elucidate the origin of a plant species is the observation of chromosome pairing in occasional haploid ind ividuals. I nduction of haploidy by apogamous development of sporophytes has been reported in several pteridophytes and meiosis observations have been used in two Dryopteris ( M A N T O N & W A L K E R 1 954) and Asplenium trichoman(Js L . ( B O U H A R M O N T 1 972a).

The present paper reports some observations performed on Asplenium septentrionale. That spec ies is a tetrap loid, with 1 44 somatic chromosomes, but a d ip loid cytotype is known from the Caucasus ( L O V I S & R E I C H ST E I N 1 964). I n 1 952, M E v E R reported the development of numerous apogamous sporophytes on A. septentrionale protha l l i cu ltivated on d ry soi l or agar, but in humid atmosphere.

MATE R IAL AN D METHODS

The origina l plants of A.septentrionale ( L .) Hoffm . were col l ected o n sch istose ·rocks i n the Lesse v a l l ey, at C iergnon (Belgium). Apogamous development was ach ieved by sowing the spores and culturing the prothal l i on a synthetic medium accord ing to the techniques prev iously described ( B O U H A R M O N T 1 972a). The base medium was K nudson's solution with the m icronutrients of H e l ier and 0.8% agar. Positive resul ts have been obta ined with different concentrations of sucrose or glucose and different growth substances. The first sporophytes appeared ni ne months after sowing and the transfer of prothal lus fragments onto fresh m ed ium sti l l gives apogamous plantlets after four years. Several hundreds of sporophytes have been raised from the same culture.

FIG U R E 1: S i lhouettes of fertile fronds of Asplenium septentrionale; a: from a normal tetraploid p lant; b & c: from dip loid plants induced by apogamy ( natural size).

*l n stitut Carnoy, U niversite Cathol ique de Louvain ( Belgium)

238 B R I T I S H F E R N GAZETTE: VOLUME 1 0 PART 5 ( 1 972)

When the sporophytes had 2-3 leaves they were placed on s i l l iceous substrate and covered to maintain humid ity . The apogamous sporel ings did not possess roots but they produced adventitious ones after several months on the soi l . Their growth was very slow and the first sporangia appeared two years after planting. Adult diploid plants remained smal ler than their tetraploid counterparts, their leaves were shorter and less d i vided ( F ig. 1 ) and the size of their stomata was reduced . The fronds general ly bore few sporangia.

The meiosis of the d iploid Asplenium septentrionale was stud ied after stain ing by aceta-carmi ne, using the classical squash method.

CYTO LOG I CAL OBS E R VATIONS

The observations made during the fi rst meiotic d i vision showed the d iploid number of 72 chromosomes and a very variable pair ing from cell to cel l . Counting was easier at the end of diakinesis or m etaphase. For 21 cel ls observed at these stages, the number of u n i valents varied from 30 to 62 (P late XXV; F ig .2 ) ; the mean numbers of chromo­some associations were 1 3. 1 bivalents and 45.7 u n i valents. I n early d iakinesis it was much more d ifficult to count the chromosomes exactly and to make a d istinction between u n i- and bivalents. Nevertheless the un ivalents were clearly fewer and thus pair ing more complete in prophase than in metaphase. (P late XXV; F ig.2a) shows a cel l with probably 8 u npaired chromosomes.

Anaphase I was a lways very irregu lar, the segregation of chromosomes and their migration towards the poles were not synchronous (Plate XXV D ) . The majority of them were however included in the telophase nuclei and very few were lagging. I n spite of an i rregu lar second anaphase, the fou r nuclei of the tetrad were generally regu lar and of similar size but there were a lso a few m icronuclei and some "tetrads" were very abnormal (P late XXV E ,F ) .

I mmature sporangia contained aborted and shrivel led spores, but the proportion of cel ls of regu lar shape was important. However in mature sporangia, the spores were entirely steri le or nearly so. The l ate spore abortion and steril ity were probably due to a random d istribu.tion of the chromosomes during anaphase and to the production of unbalanced nuclei.

D I SCUSSION

The comparison of these resu lts with the observations previously made in induced diploid p lants of Asplenium trichomanes ( s o u H A R M o N T 1 972a) showed some sim i l arities and differences between these two species. Both induced d iploid forms showed a relatively h igh chromosome pairing at diakinesis and a subsequent desynapsis, but the mean num ber of un ivalents at metaphase was h igher in A.septentrionale (45 .7) than in A.trichomanes ( 1 1 .8 ) . The m igration of the chromosomes at anaphase I and 1 1 was a lso more i rregular and the spore steri l ity h igher in A.septentrionale. The spore tetrads were nevertheless relatively regular. The large number of u n i valents in metaphase probably ind icates a lack of chiasmata between some paired prophase chromosomes but it is also a consequence of the delayed anaphase and of the precocious segregation of some bivalents, frequently found in species hybrids. The relatively regu lar shape of tetrads and you ng spores is due to a mechan ical expu lsion of the un ivalents from the spi ndle to the poles; the result of such a random d istribution of the chromosomes is the complete fai lure of the spores at maturity .

PLATE XXV

•

��'-�"() � ' � · �-- J.

A ... �

•

fl ' 8 •

• • • � • • •

• • , . • � • •

• • •

• •• • •

• •

• • • c •

. .

,.

W"Y•

• • • .. .

; . ·.

i f. . -

• (/6 •

• • I

..

• •

• • • • •

• • · •

' •

• •

• . .... •

• •

\ .,

D

•

F

•

. ,

•

P p • " .

., �

•

BOU H A R M ONT : M E I OSIS I N APOGAM O US ASPL E N I UM S EPTENTR IONALE

c

0

0 0

0

239

F IG U RE 2: Camera lucida drawings showing chromosome pairing in induced d ip loid plants of A. septentrionale X 1 500). a: diakinesis with probably 8 u n ivalents; b: late prophase or metaphase with 1 1 bivalents and 50 univalents; c: 18 bivalents and 36 univalents; d: 9 bivalents and 54 univalents.

PLATE XXV. Meiosis in induced diploid A. septentrionale X 1 500. A: d iakinesis ( F ig. 1 a) ; B;C: late prophase o r metaphase ( F igs.2b,c ) ; 0: very irregular first anaphase; E: spore tetrad with two visible micronuclei; F: very abnormal "tetrad".

240 BR IT ISH F E �N GAZETTE : VOLUME 1 0 PART 5 ( 1 972)

The nature of polyploidy in A.septentrionale has been studied after interspecific crossing. A. X murbeckii, a tetraploid hybrid between A.septentrionale and A.ruta-muraria L. showed 60-62 bivalents ascribed to autosyndesis and Indicating the autopo lyploid Origin of both parents (L O V I S 1 963, 1 964; L O V I S and R E I C H STE I N 1 964). Other observations showed the h igh degree of homology between the genomes of A.ruta-muraria (V I D A 1 970; B O U H A R M O N T 1 972b). Thus it is probable that a h igh proportion of the biva lents found i n A. X murbeckii invol ved chromosomes of· A.ruta-muraria and that the contribution of the septentrionale chromosomes was lower. Our observations did not d isagree with the hypothesis of an autoploid orig in for A.septentrionale, but they i nd icated that the homology between the genomes is relatively low in comparison with tetraploid A.ruta-muraria and A. trichomanes. l t is possible to explain the orig in of the present tetraploid form of A.septentrionale by a hybridization between two relatively evolved d ip loid types; another possibi l ity is a chromosome doubl ing in a single d iploid form at a time sufficiently old to a l low a progressi ve d i fferentiation i n the genome.

R E F E R E NCES BOUHARMONT, J. 1 972a. M eiosis and fertility in apogamously produced diploid plants of

Asplen ium trichomanes. Chromosomes Today 3 ( i n the press). BOUHARMONT, J. 1972b. Origine de la polyploidie chez Asplenium ruta·muraria L. Bull. Jard

Bot. Nat. Belgique ( In the press) . LOV IS, J . D . 1 963. M eiosis in Asplenium X murbeckii from Borrowdale. Brit. Fern Gaz.

9 : 1 1 0- 1 1 3. LOV IS, J .D. 1 964. Autopolyploidy in Asplenium. Nature Lond., 203:324--325. LOVIS, J . D . and R E I CHSTE IN , T. 1 964. A diploid form of Asplenium ruta·muraria. Brit. Fern

Gaz. 9 : 1 41 - 1 46. MANTON, 1. and WAL K E R , s. 1 954. I nd uced apogamy in D ryopteris dilatata ( H offm.) A.Gray

and D.filix-mas (L.) Schott emend. and its significance for the interpretation of the two species. Ann. Bot. 1 8 :377-383.

MEYE R, D.E. 1 952. U ntersuchungen uber Bastard ierung in der G attung Aspl enium. Bibliotheca bo tanica 30, H eft 1 23 : 1 -34.

V I DA, G . 1 970. The nature of polyploidy in Asplenium ruta-muraria L. and A.lepidum C. Presl. Caryologia 23: 525-547.

B R I T. F E R N GAZ. 1 0(5) 1972

THE ANATOMY OF MAXONIA APII FOLIA: A CLIMBING FERN

T.G. WAL K E R *

ABSTRACT

Maxonia apiifolia (Sw.) C. Chr. is an example of an epiphytic fern which starts its life rooted as a normal terrestrial species eventually to become scandent. The fronds produced d iffer in their morphology and anatomy depending from which part of the rhizome they originate. These d ifferences are described and the significance and mechanisms producing the effect are briefly d iscussed .

I NTROD UCT I O N

241

In temperate countries the ferns are almost exclusively terrestrial in hab it, thus contrasting with the situation in the tropics where a relatively high proportion of species and even genera are epiphytic. Both the terrestrial and the epiphytic hab itats have their own characteristic featu res and pecu l iarities which differ marked ly from each other even in the same area. Thus, a terrestrial fern l iving near the base of a tree in a forest may well be inhabiting a very u niform environment in which the l ight intensity is very low , the soil i s constantly moist, the atmosphere sti l l and humid and in wh ich there is l ittle change in temperature. A fern growing epiphytica l ly near the top of the same tree wou ld be in a very d ifferent enviro�ment, in which, for example, water is ava i lable during actual rainfa l l and for only a short time afterwards due to the rapid run-off from the surface of the tree. I n add ition, the exposed position of the plant wi l l be subjected to far greater f luctuations of temperature, humidity and air movement than in the shelter of the forest and the l ight intensity wi l l be increased many times over.

DifferentJerns react to these d ifferent habitats in a wide variety of ways but it is nevertheless true that the physiology, anatomy and morphology of species which are more or less related reflect the environment in wh ich they grow and frequently terrestrial and epiphytic forms d iffer marked ly from one another.

There is a sma l l group of ferns in both the Old and the New World tropics which in a sense bridges the gap between the terrestrial and epiphytic habits - ferns such as Stenoch/aena, Teratophyllum, Lomariopsis, Po/ybotrya , etc. These ferns are a l l scandent and start l i fe rooted in the forest floor. The rhizome grows horizonta l ly, often for a considerab le d istance unti l it makes contact with the base of a tree. The d irection of 'growth then switches from horizontal to vertical and the rhizome � i mbs the tree. As the plant thus moves from one envi ronment to another it encounte�s very different cond itions and various mod ifications in morphology, anatomy and physiology take place - some of wh ich are very striking. Whi lst there are detai led accou nts in the l i terature of certain aspects of some of these ferns, e.g. the heterophyl ly of Teratophyllum and other genera (see H O L TT U M 1 954a & 1 954b ) , few members have been examined in detai l despite some remarkably interesting morphogenetic and physiological prob lems wh ich would wel l repay closer study.

During the course of several visits to Jama ica I had the opportunity to col lect and study in the field one such scandent fern, namely Maxonia apiifolia (Sw . ) C. Chr. Maxonia has a very l imited geographical range and consists of the single species of

*Department of Botany, U n iversity of Newcastle upon Tyne, NE1 7 R U .

242 B R IT ISH F E R N GAZETT E : V.OL U M E 1 0 PART 5 ( 1 972)

7

8

CJ 10

F I G U RES 1 - 1 1 . Morphology o f Maxonia. 1 : si lhouette o f sterile scandent frond, X 1. 2: silhouette of fertile scandent frond, X k· 3: sterile scandent segment, X 3.2. 4: steri�e terrestrial segment, X 3.2. 5: abnormal segment of scandent frond, X 3.2. 6: pickled ferti le scandent segment, X 3.2. 7: dried ferti le scandent segment, X 3.2. 8 : scal.e of scandent rhizome, X 4. 9: scale of terrestrial rh izome. X 4. 1 0: mature Indusium, X 6.5. 1 1 : spore, X c.350.

WA L K E R : A NATOMY O F MAXON IA 243

which the typical form occurs in Cuba and Jamaica wh i lst the variety dualis is restricted to Guatemala.

l t is found in a number of local ities i n Jamaica and my own material was col lected at c.2750 feet (825m ) above sea level near the road to H ol lymount, M ou nt Diabolo in the parish of St. Ann where it is relatively abundant . Here it is not enti rely restricted to the forest but also grows i n small open grassy areas from which the forest has been removed .

MORPH O LOGY AN D ANATOMY

Because the morphology, anatomy and general behaviour of this fern d iffers accordi ng to whether it is rooted in the ground or c l imbing a tree it has usual ly been found necessary in the fol lowing account to g ive two (jescriptions of the same organ. The rhizome

The terrestrial * rhizome creeps just u nder the soi l surface - often for long distances - and is more or less smooth in outl ine, i .e. it lacks the longly decurrent stipe bases which are present in the scandent rh izome giving the latter an irregu lar appearance. Scales are borne which measure approximately 8 mm X 1 mm and are concolorous with irregu larly toothed marg ins ( F ig.9 ) . These scales are somewhat sparsely scattered along the rhizome except at the growi ng point where they are abundant and protect the del icate apex. The rh i zome is oval in transverse section ( F ig. 1 3) and measures some 6 mm across at the widest poi nt. The vascular system is a dictyostele and at any one level there may be seen the smal ler traces of at least two fronds. I n the cortex there are a few scattered groups of cel ls bearing thick opaque deposits of brown/black ph lobaphene on their wal ls. A layer of l ignified cel ls surrounds the rhizome immed iately beneath the epidermis.

The rh izome changes its character suddenly and marked ly on a l tering its d i rection of growth from the horizontal to the vertical and scandent. l t becomes flatter and moulds itself more c losely to the contours of the substrate and the roots penetrate the crevices in the bark with the resu l t that the fern adheres very firm ly to the tree and can only be detached with some difficu lty. Very conspicuous golden scales are present which are s imi lar in morphology to those of the terrestria l rhizome but are very much larger, viz. 16 mm X 1 .5 mm ( F ig.8 ) . They invest the whole of the scandent rhizome in a thick covering, contrasting with the situation in the terrestrial rh i zome where the scales are concentrated at the growi ng point and are sparse elsewhere.

The d iameter of the rh izome increases some three and a half times from 6 to 2 1 m m and the outl ine in cross section from a n oval to a more i rregu lar shape i n which three bulges are usual ly present ( F ig . 1 2 ) . These bulges are caused by the decurrent stipe bases retaining their identity for some d istance even though they are fused to the rh i zome. There are normal ly three groups of frond traces (correspond ing with the three bulges) to be seen at any one level and there · are many more groups of phlobaphene·containing cel l s than in the terrestria l rh izome. The outer cortical layer tends to be more heavi ly l ignified than the correspond ing layer in the terrestrial portion and imparts a tougher, more woody character to the rh izome.

*The word terrestrial is used here for any part w h ich has originated on a grou nd-growi ng p lant, e.g. a terrestrial frond has grown from a soil-rooted rhizome. Simi larly, the word scandent is used for cl imbing parts.

244 B R IT ISH F E R N GAZETTE: VOLUME 1 0 PART 5 ( 1 972)

" '- - - . .

..

• 1.-o-. • . . •

• •

. \ , � ·o .. ... ,' I .t:;;,· ·P.· - - · - - , . . . •

18

•

17

F I G U R ES 1 2- 1 8. A natomy of Ma.i<onia. 1 2 : T.S. scandent rhizome, X 3 . 1 3 : T.S. terrestrial rhizome, X 1 2. 1 4 : T.S. stipe vascu lar strand, X 1 35. 1 5 : lower epidermis of steri le scandent frond, X 1 00. 1 6 : epidermis of pneu mathode, X 1 00. 1 7 : L.S. pneumathode, X 1 00. 1 8 : T.S. pneumathode, X 92.

WAL K E R : ANATOMY OF MAXO N I A 245

The fronds The fronds are produced singly at intervals a long the rhizome and are

quadripinnate. Those produced on the terrestrial rhizome are entirely steri le and although very similar in appearance to the steri le scandent fronds are sma l ler, both in overa l l si�e and in the size of the ind ivid ual segments (compare F igs.3 and 4 ) . I nternal ly, there are other d ifferences.

The scandent rh izome produces steri le fronds abundantly ( F ig. 1 ) and a lso ferti le ones. The latter are produced very sporad ical ly and several visits to the same local ity were necessary before specimens were found at a su itable stage for cytological fixation (W A L K E R 1 966) .

The ferti le fronds d iffer marked ly in appearance from the ster i le ones but basica l ly they are constructed in the same way - the strik ing d ifference being produced by the great reduction in the laminar surface of the ferti le segments ( F ig .2 ) . A ferti le segment wh ich was fixed in the field in formal inacetic·alcohol is i l l ustrated in Fig.6. i t can be seen that the area of the lamina is less than one fifth of that of a comparable steri le segment ( F ig.3) and that the venation has been both reduced and simpl ified . I nstead of there being seven or eight pairs of veins which are themsel ves once or twice branched as in a steri le segment there is only a single vein bearing short arms which terminate beneath sor i . The reduction in laminar surface is further accentuated in herbarium specimens where the soft tissues between the veins shrink to a greater degree than the other tissues thus prod ucing the effect shown in F ig . 7.

Very occasional ly one, or at most a very sma l l number, of segments or pinnu les may become ferti le on what is morphologica l ly a typical steri le scandent frond. One such case is i l l u strated in F ig.5 and rather c losely fits the description of forms given by CH R I ST E N S E N ( 1 9 1 6 ) in wh ich he points out that in the earl ier stages the indusium is el ongate and very sim i lar to that found in Cystopteris or Davallia before it broadens out to assume its final mature shape ( F ig . 1 0 ) .

The sori lack paraphyses and the sporangia are long-stal ked and non-glandu lar, with 1 8 to 23 indurated cel ls in the annu lus . The spores are b i lateral , l ight brown in colour and the surface is thrown up into folds. There is a wavy wing to the perispore, the whole being densely covered w ith sma l l fine spinu les ( F ig . 1 1 ) . Cytological exami nation (w A L K E R 1 966) has shown there to be 41 pa irs of chromosomes at meiosis, which is in agreement with the placing of Maxonia in the Aspidiaceae on morphological grounds.

I nterna l ly , a l l three types of fronds (terrestrial steri le, scandent ster i le and scandent fert i le ) have m uch in com mon and , somewhat surprisingly, where they do d iffer the d istinction is not between the ster i le fronds on the one hand and the morphological ly very d ifferent ferti le frond on the other, but between terrestrial and scandent sterile fronds. I ndeed, the scandent steri le and ferti le fronds are so a l ike in the construction of the stipe, rhach is etc. , that the · i l l u strations are completely interchangeab le and only a single series ( F igs. 1 9-24) has been presented for th is reason.

In a l l cases the stipe and rhach is have a broad band of sclerenchyma immediately below the epiderm is and the cel ls lack intercel l u lar spaces. Thus, not only is this layer mechanical ly strong but it wou ld a lso effectively impede gaseous exchange between the interior and the environment if it were complete. However, the layer is interrupted on either side by the pneumathodes wh ich extend the whole length of the stipe and rhachis ( F igs. 1 9 to 22 and F igs.25 to 28 ) . I n many ferns the pneumathode is visib le as

246 B R IT ISH F E R N GAZETTE: VOLUME 1 0 PART 5 ( 1972)

� tff � w � �IQ

� v;;.·· · · ��� \V � . . . � � fi":: � �

§)) (!) (!5 ,<l) r;!J � Q. (!) . ..

fi.\ � � � � '· ·�·- ®···-.. � � �: :•: e. � · . . • G) • . . . :�·

. .

F I G U RES 1 9-24. A natomy of scandent frond of Maxonia. 1 9 : T.S. base of stipe, X 1 0. 20: T.S. mid stipe, X 1 0. 2 1 : T.S. top of stipe, X 1 0. 22: T.S. m id rhachis, X 1 0. 23: T.S. costa, X 20. 24: T.S. costu le, X 20. I n vascular traces: solid black = xylem ; broken l i ne = endoderm is; continuous l ine = phlobaphene-containing cells.

WAL K E R : ANATOMY O F MAXO N I A 247

a conspicuous yel lowish streak but in Maxonia it is not external ly evident. I nternal ly, however, the pneumathode is fu l ly developed and consists of an area of parenchymatous cel l s which are loosely packed ( F ig. 1 8 ) . I n longitud inal section the cel l s are seen to be several times longer than they are wide and there are abundant interce l lu l ar spaces which communicate transversely with one another and also with the l ongitud inal spaces. Thus the pneumathode provides a means of gaseous exchange between the interior of the organ and the exterior. Coupled with this i s the restriction of the stomata to only that part of the epidermis which overlies the pneumathode. Here the stomata are arranged in four or f ive more or less regu lar rows ( F ig . 1 6 ) instead of being scattered over the surface as is the case on the underside of the lamina (F ig. 1 5) . There is a lso a contrast in the shape of the epidermal cel ls, these having stra ight wal l s on the stipe and undu late wal l s on the lamina (F ig . 1 5 ) .

A typical vascular trace ( F ig. 1 4) has a central band of xylem with a broad patch of ph loem ly ing on either side of it, the whole being surrounded by a pericycle and endodermis. The latter is unth ickened except for the sma l l casparian strip on the rad ia l wal l s. The cel l s of the cortex im med iately abutting onto the endodermis are heavi ly impregnated with ph lobaphene which completely obscures some of the cel ls wh i lst only partia l l y covering others. This pattern of ph lobaphene d i stribution is typical of that found at a l l levels of the stipe and rhach is of the steri le and fert i le scandent fronds and also of the midd le and upper parts of the stipe of the terrestria l ster i le frond. I n the l atter, however, the base o f the stipe deviates from the typical pattern. H ere, the cel l s immed iately external to the endodermis develop ph l obaphene only around the margins (as seen in T.S. ) leaving a clear lumen ( F ig.30). I n my specimens the thickening is translucent and l ight in colour, contrasting marked ly with the dense brown/black of typical phlobaphene. In the rhach is the ph lobaphene is agai n l ight in colour and somewhat translucent but in this case i s confined to a band on the inner tangential wa l l s adjacent to the endoderm is ( F ig.3 1 ). At all other levels ( i nc lud ing the costa ) such cel l s are dense, dark and q u ite typica l .

The frond vascu lar traces are relatively numerous when compared with those of many ferns of s imi lar size in the Aspid iaceae and remain approximately constant in number throughout the stipe, although there are m i nor fluctuations from one level to another and it is evident from serial sections that a certain amount of spl itting and anastomosing of the traces occurs. Normal ly f ive traces supply each costa, the number decreasing to three in the costu les. This holds true for both ster i le and ferti le fronds despite the great reduction of the lami nar surface of the latter.

In add ition to the d ifferences of ph lobaphene development noted above, ,the terrestria l steri le fronds have stipes that are only about half the d iameter of tho� of the scandent types (compare Fig . 1 9 to 2 1 with F ig.25 to 27, noting the change in magn ification ) . This reduction in size i s accompanied by a reduction in the number of traces to approximately twelve, wh ich is about half that found in scandent stipes.

The lamina in a l l cases i s wel l provided with aerating tissue, a lthough there is not a very sharp d istinction between pal isade and spongy mesophyl l . Abundant stomata are present ( F ig. 1 5 ) . confined to the lower surface of the frond.

D I SCUSS I O N

Some at l east of the changes that take place on the changeover from the terrestrial to the scandent cond ition can reasonably be l i nked with a change in the environmental cond itions. Thus, the scales on the terrestria l rhi zome (which is underground and not

248

· ·a· . - -

B R ITISH FERN GAZETTE : VOLUM E 10 PART 5 ( 1 972)

. . ,- . .. ' . . . . . · .. . . ·

�: .. , . ... � .. _ _ ,:

31

� 2�

F I G U R ES 25-3 1 . A natomy of terrestrial frond of Maxonia. 25: base of stipe, X 18 . 26: T.S. mid stipe, X 18. 27: T.S. top of stipe, X 18. 28: T.S. m id rhachis, X 18. 29: T.S. costa, X 18. 30: T.S. endodermis and i nnermost layer of cortex of stipe base, X 2 1 0. 3 1 : T.S. e ndodermis and inner­most layer of cortex of top of stipe, X 2 1 0. I n vascular traces: solid black = xylem; broken l i ne = endoderm is; continuous I ine = ,phlobaphene-containing cells.

WA L K E R : ANATOMY OF MAXON I A 249

exposed to the air) are relatively sma l l and are sparse except around the growing point which is most vu lnerab le to damage as it pushes through the soi l . On the scandent rh izome (wh ich is exposed for its entire l ength) the scales are about twice as large and form a thick protective layer over the whole of the rhi zome and not merely at the growing point. Again, the flatten ing of the u nderside of the scandent rh izome and the partial mould ing of its shape to conform with the i rregu larities of the underlying bark enab les it to adhere more securely to the tree.

Presumably the fact that the ferti le fronds are only produced on the scandent portion is also l inked with an environmental change or stimu lus but as to what this may be is a great deal less obvious. Two possible explanations read i ly come to mind but on closer examination neither seems to be entirely satisfactory when viewed in the l ight of field observations. One of the most obvious factors in the case of a fern wh ich is rooted in the ground in the forest and cl imbs to the top of a tree is the d ifference in l ight intensity experienced . However, the idea of a change in l ight intensity being a determ ining factor in the production of fert i le fronds must be dismissed as it does not agree with field observations. For exam ple, one of the areas on Mt Diabolo has been denuded of trees and Maxonia is abundant here. I n the centre of this area a three foot high fence post has been left stand ing and a plant has cl imbed up it, producing a l l the typical scandent changes and a lso a crop of ferti le fronds. Clearly, in this case there was no sign ificant d ifference in the l ight intensity experienced by the terrestrial and scandent portions of the plant. A sim i lar case was noted in New Brita in in the case of Stenoch/aena sp.

H o L T T U M ( 1 954a ) considers that in the case of Stenochlaena the stimu lus to become ferti le is provided by a period of d ry weather. This idea of dry conditions provid ing the stimulus can be extended to explain why ferti le fronds are normal ly only produced on the scandent portions of the plant and not on the terrestrial parts in that a priori one wou ld expect the former to be subjected to drier conditions than the latter at a l l times. The original idea and its extension are more d ifficu lt to prove or disprove by field observations but aga in the balance of the evidence is against them . Thus, a number of species of scandent and c l imbing ferns in Trinidad (where there are wel l -marked wet and dry seasons) do not become fertile in the dry season or at the beginning of the wet season but only when the rainy period is well advanced. I ndeed , for many of them the d ry season represents an interruption or resting period and it wou ld appear that d ryness does not provide the necessary stimulus . Furthermore, in Jamaica in some local ities Maxonia grows i n very wel l drained soi l overlying porous l imestone which dries out during occasional d roughts. Under these cond itions, moreover, the. terrestrial portion is not stim u lated to become ferti le.

Such observations suggest that the terrestrial and scarident phases represen1 two different physiological states of the plant (perhaps somewhat analogous to juveni le and adu l t forms of some organ isms) one of which is normal ly incapable of producing the reorgan ization of growth patterns etc. which the formation of fert i le fronds demands whi l st the other one is. This is consistent with the very striking and rapid changes which occu r in Maxonia and other scandent ferns on passing from the terrestrial to the scandent phase and impl ies the existence of a drastic reorganization of the morphogenetic system which is coincident with the change of physiological state. The mechanism by which th is reorganization is accomplished cou ld conceivably be provided by the changeover from a d iageotropic to a negatively geotropic manner of growth of the rh izome which may affect the d istribution of hormones within the rh izome and its apex . lt is wel l known that some plant hormones such as I AA may

250 BR I TISH F E R N GAZETTE : VOLU M E 1 0 PART 5 ( 1972)

show marked changes in d istribution with in a tissue when the latter is subjected to a geotropic stimulation . These changes have consequent effects upon the patterns of growth .

The identi fication of the stimu lus which triggers off tf:Je a l teration in the geotropic response of the rh izome, together with an analysis of the morphogenetic patterns in the terrestrial and the scandent phases represent prob�ems wh ich would form a fascinating and valuable field of study.

R E F E R ENCES

CHR ISTENSEN , c . 1 9 1 6 . Maxonia, a new genus of ferns. Smithson. mise. Co/lns. 66: 1 -4. HOL TTUM, R . E . 1 954a. Plant L ife in Malaya. London. H O L TTUM, R .E . 1 954b. Flora of Malaya vol.2 Ferns. S i ngapore. WA L K E R , T.G. 1 966. A cytotaxonomic survey of the pteridophytes of Jamaica. Trans. Roy. Soc.

Edin. 66: 1 69-237.

REVIEW

BOTANY OF THE PHIL IPPINE L OWER PLA NTS VOL UME 1 by P.M. Zamora. Pp. 147, 2 12 x 272mm, mimeographed by the author at Oiliman, Quezon City, Philippines, 191 1.

Dr P.M. Zamora has privately produced this i nvaluable book as a precursor to a more permanent and revised printed ed ition. He is to be congratulated. Whi lst a imed at h is students and col leagues i n the U niversity of the Phi l ippines, i t is the k ind of manual that any Un iversity laboratory and l ibrary could wel l possess. The b u l k of the work deals with Lycopsida and Sphenopsida; F i l icopsida is, I assume, the subject of later volumes. F u l l descriptions are given on developmental morphology, anatomy, biosysternatics and ecology; the b ib l iographies are extensive. The most stimu lating part of the work is the frequent mention ing of u nsolved problems yet to be tackled.

I shou ld l i ke to draw attention to a worthwh i l e periodical TH E PH I L I PP I N E BIOTA produced by the Biology Teachers' A$sociation o f the Phi l ippines (C/o University of the Ph i l ippines Col lege of Agriculture, Laguna) . This over the past few years has contained several papers by Dr Zamora and others on the morphology and biology of ferns. These publ ications are reflecting an upsurge of fern stud ies by Zamora and his col leagues M ichael P r ice and Juan Pancho, and their students which Is resu lting in new knowledge from an area of great richness. Their efforts to evaluate and encourage others to join with them to evaluate, the fern flora of these islands before agricu l ture and de-afforestation have taken all are greatly appreciated.

A.C. JERMY

BR IT . F E R N GAZ. 1 0(5) 1972

THE KARIBA WEED: SALVINIA MOLESTA

D.S. M ITCH E L L *

ABSTRACT

The rampant weed on Lake Kariba and elsewhere, h itherto called Salvinia auriculata Aubl . , is now regarded as a separate species, probably of hybrid origin, and is here given the name Salvinia molesta D.S. M itch ell sp. nov.

251

In May 1 959 sma l l patches of a Salvinia wh ich in a l l respects appeared to be S. auriculata Aubl . (s c H E L P E 1 970 et a l . ) were reported on Kariba Lake. The construction of Kariba Dam had brought about a lacustrine habitat with an increasing area of water avai lable for colonisation and the ecology of the plant which soon assumed the characteristics of a weed have been reported by s c H E L P E ( 1 961 ) . As more was learnt of its b iology, morphology and taxonomy (see M I T C H E L L 1 970) it was real ised that it was not the same as S. auriculata Aubl . and it is now proposed that it be regarded as a d istinct taxon, S. molesta. The purpose of this short note is to va l idate this name.

Salvinia molesta D.S. M itche l l species nova.

S. auriculatae, S. bilobae et S. herzogii affinis sed morphologia axis fert i l is d iffert.

Partes vegetativae variabi les ut in S. herzogii. Axis ferti l is ad organa submersa affixus, praecipue plantis aggregatis nascens; ramificatio scorpiodidea. Sporocarpia usque ad 55, ovoidea, apicu lata, d iametro 1 mm vel minus. Sporangia matura, plerumque, vacua. Sporae, ubi adsunt, persaepe deformes, variab i les. Massul ae cavae in microsporangiis interdum adsunt. Macrosporocarpia duo (raro tres) , proximal ia peduncu lata. M icrosporocarpia, usque ad 53, sess i l ia ve l subsessi l i a . Omnia sporocarpia sed macrosporocarpia densius trichomatibus praedita.

Type: Rhodesia, Lake Kariba, Ruz iruhuru R iver I nlet, Mitche/1 1330 (B; BM ; BO; BOL; BR ; CAL; CH R ; EA; G E ; G H ; K; L I SB; LP; M ; MO; NSW; P; PDA; P R E ; RB; SR GH, holotype; US; WAG; Z . ) Other specimens seen: BOTSWANA : Kasane, Chobe R iver, G. E. G1bbs Russe/1 1 341 (SRG H ) . KENYA: Kita le, Barnely H 1 70/57 (BM ) ; Lake Naivasha, Mitche/1 1 1 87 (SRG H ) ;

Nairobi, Ba l ly 1 1 390 ( K) . RHOD E S I A : Chirundu, Zambez i R iver, Bates sn (SR G H 97 1 92 ) ; Brewer 1 82 (�AH ) ;

Lake Kariba, Mitche/1 580 (SR G H ) , 668 (K) , 1 099 ( G E ) , 1 1 93-1 1 96 ( LP ) , ' 1 1 97-1 208 (BM ) , 1 209- 1 2 1 8 ( L P ) ; Umta l i , Chase .46 1 5, 573 1 (SRG H ) ; Victoria Fal ls, Wild 485 1 (SR G H ) .

S. A F R ICA: Knysna, Gouna Forest Reserve, Mitche/1 1 90 (BOL, S R G H ) . ZAM BI A : Lake Kariba, Phipps 28 1 0 (SR G H ) ; Ndola, Town Engineer s n (SRGH

134857); Kazungula, Zambezi R iver, Exe/1, Mendom;a & Wild 1 457 (SR G H ) ; Katam bora, Zambezi R iver, West 3050 (SR G H ) .

CEY LON : Negombo, serious weed in waterway, J. E. Senaratna s.n. ( K) ; very common about Colombo, Sledge s.n. (BM ) .

I N DON ESIA: W . Java, between Tj ibodas and Tj ipanas, Jermy 72 1 3 .

* Botany Department, University of Rhodesia, Salisbury.

252 B R I TISH F E R N GAZETTE: VOLUME 10 PART 5 ( 1 972)

BRAZ I L : Rio de Janeiro Botanic Garden, Kuhlman 06 1 24 ( RB) . QU EENSLA N D : Anakie rlway. dept. dam. Stanton s.n. ( K) . WESTE R N AUSTRAL I A : Harvey, Marchant 253.

The Salvinia au ricu lata group ( incl . S. auriculata Aub l . , S. biloba Radd i , S. hertzogii de la Sota and S. molesta may be d istinguished from other species ofSalvinia, initia l ly by the confirmation of the hairs on the apices of the papi l lae on the u pper surface of the leaves. Here there are four (rarely two in depauperate specimens) uniseriate hairs united at their d istal ends. S. molesta can be d istingu ished from the other three species by the presence of long, straight, secund chains of sessi le to subsessi le male sporocarps, 1 mm or less in d iameter, containing mostly empty sporangia.

S. mo/esta probably has a widespread d istribution in the tropics where it general ly behaves as an aggressive weed on stand ing waters. l t is also probable that, identified as S. auricu/ata , it has been the subject of several previous studies, for exam ple by i< o PP ( 1 936) and a o N N E TT ( 1 955 ) . One specimen has been exam ined in South America. This was a pressed specimen in the herbarium of the R io de Janeiro Jard im Botanico that had been col l ected in 1 94 1 from the Botanic Garden (Kuh/man 06124) . At the same time·, specimens of S. biloba and S. auriculata were col lected from the Garden and th is raises the possib i l ity that S. molesta is a hybrid of horticu l tural origin. Though this is ind icated by the plant's robustness (hybrid vigour) and its steri l ity, there i s l ittle indication of intermed iate characters between S. biloba and S. auricu/ata . However, in some respects, the plant is intermed iate between S. biloba and S. herzogii. Cytotaxonomic and chemotaxonomic studies of members of the genus may provide evidence which could support, or refute th is theory of hybrid origin. In the meanwh ile, consideration must be given to the possib i l ity that th is species may not occur natural ly in S. America, though it is clearly of S. American origin.

Specimens of S. molesta from Lake Kariba and I nd ia ( identified as S. auricu/ata ) have been the subject of a cytogenetical study by L O Y A L & G R E W A L ( 1 966) . They have shown that a h igh proportion of the sporangia abort at the archesporia! stage and, when meiosis does take place, the process i s h ighly anomalous . They cite the num ber of cl>lromosomes to be 45 and suggest that the species is a pentaploid .

Du ring the past year, S. mo/esta, S. auricu/ata and S. herzogii have been grown in cu lture u nder simi lar cond itions. Al l species produced sporocarps wh ich were morphological l y consistent in each case and the species cou ld be regularly distingu ir.hed from one another.

R E F E R ENCES

AUB LET, J .B .C. F . 1 775. Histoire des plantes de la Guiane .franrraise, 2: 969-970. London & Paris. BON N ETT, A .L .M . 1 955. Contribut ion a ! ' etude d es H ydropteridees: recherches sur Salv i nia

auricu late A ubi. Annls. Sci. nat. (bot.) 16: 529-600. DE LA SOTA, E . R . 1 962. Contribucion al conocimiento d e las Salviniaceae neotropicalea. 1 1 .

Salv in ia auricu lata A ublet, I l l . Salv in ia herzogi i nov. sp. Darwiniana 1 2 : 499-520. KOPP, J. 1 936. Uber die kulturbedingungen und die systematischen Merkmale der Salvinia A rten.

I naugural D i ssertation, Munster. LOYA L, o.s. & G R EWAL, R . K . 1 966. Cytological study on steril ity in Salv inia auricu lata Aubl .

with a bearing on its reproductive mechanism. Cytologia 31 : 330-338. MI TCH E L L, D.S. 1970. Auteco/ogical studies of Salvinia auriculata A ubi. Ph. D. Thesis, U n iversity

of London ( u npublished). SCH E LPE , E .A.C.L .E . 1 96 1 . The ecology of Salvinia auricu lata and associated vegetation on Kariba

Lake. J. South A fr. Bot. 27: 1 8 1 - 1 8 7 . SCH E LPE , E .A.C.L.E . 1 970. ,I n E xel l , A.W. and Launert, E . ( ed . ) F!. Zambesiaca, Pteridophyta,

London.

B R I T. F E R N GAZ. 1 0(5) 1 972

NOTES ON PHILIPPINE FERNS

M .G . P R I CE *

CONTENTS

1 . A second Ph i l ippine Dipteris, D. /obbiana. 253

2. The type locality of Ctenitis riza/ensis . 254

3. F irst report of Acrosorus from Luzon . 255

4. T w o a n t i c i pated T h e l y p teroid ferns (T. singalanensis and Pseudophegopteris aurita). 255

5. Confirmation of Vaginularia paradoxa . 256

6. The identity of Trichomanes millefolium Presl (T. maximum 8 1 . versus T. obtusum (Cope!.) Morton) . 257

7. A new species and a new combination i n Diplazium . 258

8. Two rare ferns from Mt. Maki l ing - new to Luzon (Trichomanes rupicola and Tectaria beccariana ) . . 260

9. An earlier name for Pteris asperula J. Smith ex Christ 261 1 0. The genus Lagunaea and Tectaria kehdingiana comb. nov. 262

1 . A SECOND P H I L I P P I N E D I PT E R I S

253

The special sign ificance of the genus Dipteris is that it combines characters of prim itive gleichen iaceou s ferns and advanced polypodiaceous ferns and a lso that it i s very sim i lar to severa l ancient fossil genera. The strange m i ng l ing of unevenly evolved characters added to its status as an isolated rel ic are strong reasons for placing Dipteris in its own fam i ly , as various authors have done. Resu lts of a comparative study by W I LS O N ( 1 959) attesting to the un ique sporangia! structure make that d isposition imperative.

The most abundant and widely d istributed species is D. conjugata and unti l recently it was the only member of the genus known from the Phi l ippi nes. This note is to report the find ing of a second species, on the island of Negros, which is situated near the center of the archipelago. Dipteris lobbiana (Hooker) Moore, I ndex : 80 ( 1 85 7 ) ; H olttum , Gdns. Bu l l . Straits

Sett l . 4 :57, t. ( 1 927) ; Ferns of Malaya 1 33, t. 54 ( 1 955 ) . Po/ypodium /obbianum Hook . , Hooker's J . Bot. 5: 309, t. 1 1 , ( 1 853) non H ook.

( 1 862) . Exsiccatum: Negros Oriental , Amlan Mts, Pasalan Fa l l s, on stones along river bank, 500 m, J. V. Pancho 5005. D. /obbiana is known in the Phi l ippines by only this single col lection. lt has previously been recorded from Sumatra, the M alay Peninsu la, Kal imantan, Sarawak, Sabah, Sulawesi and Amboina, and. has of late been reported by V A N S T E E N I S ( 1 97 1 ) from East New Gu inea.

Holttum ( 1 927) provides a photograph of th is species in its typical floodswept habitat a long with a d iscussion of its ecology and relationships. For its genus, D. /obbiana exh ibits the greatest number of characters considered prim itive, but some of these are possibly later adaptations to its rigorous environment, and derived rather than archaic. Our plant has segments only 4-5 mm broad , at the lower l im it of the range of variation (5- 1 0 m m ) ind icated by Holttum (I.e. ) , but I can detect no other d isparity with h is description.

* Dept. of Botany, U niversity of the Phi l ippines at Los Banos, Col lege, Laguna, Ph i l ippi nes.

Note: Collections cited in th is paper are deposited in the H erbarium, University of the Phi l ippines at Los Banos (CA H P ) un less ind icated oth

'erwise. Dupl icates are being d istributed.

254 B R IT ISH F E R N GAZETT E : VOLUME 1 0 PAR T 5 ( 1 972)

2. THE TYPE LOCA L I TY O F CTE N ITIS R I ZALENSIS

Ctenitis rizalensis was described by H. Christ i n 1 906 and then again i n 1 907. I n the latter, he cited three col lections, the first two being from M i ndanao, the m ajor southern island of the Phi l ipp ines, and the l ast from Luzon, the major northern is land. None of these three was designated as type nor was any reference made to the 1 906 publ ication.

Copeland ( 1 91 0) . whi le report ing on another col lection of this species from Mt Apo, M i ndanao. the locality of the fi rst co l lection cited by Christ in 1 907, stated that Mt Apo was the source of the type. I n 1 929, he a ltered th is opinion and elected San Ramon, Zamboanga, a lso in M indanao, from where had come the second specimen of Christ ( 1 907 ) , wh ich was actua l ly the earl iest col lection. By 1 960, Copeland had become positive that San Ramon was the type local ity. The Luzon plant, wh ich he had not seen , was d ismissed ·as doubtful ; Christ had said it was "a smal ler form" . That the name rizalensis was taken by Christ from the province of R iza l i n Luzon was u nheeded.

The matter i s conclusively c larified by consu lting Christ's original 1 906 publ ication . The solitary co l lection m entioned was from R izal Prov. , Luzon, and it was specified that the type had been deposited in Christ's personal herbarium which is now ·

at Paris. Recent col lections from Luzon serve as confirmation and leave no doubt that the M indanao plant is conspec ific. Ctenitis rizalensis (Christ) Copeland, Gen. F i l . 1 25 ( 1 947) ; F ern F l . Ph i l ip. 2: 290

( 1 960) . Dryopteris rizalensis Christ, B u l l . Herb . Bo iss. Ser.2, 6 : 1 00 1 ( 1 906) ; Phi l ip . J .

Sci. 2C : 2 1 6 ( 1 907) ; Copel . , Leaf l . Ph i l ip . Bot. 3 : 806 ( 1 9 1 0) ; Ph i l ip . J . Sci . 40: 294 ( 1 929) .

Rh izome short, stout, erect, bearing opaque brown acicular paleae 1 -2 c m long and fronds to 1 35 cm long . Stipe s l ightly shorter than lamina, stout at base, densely paleate in lower portion. the paleae gradua l ly modified upward from those of the rh izol')1e to shorter, broader. and c lathrate scales on the rachis. Lamina ovate, comm only 35-50 cm long by 20-33 cm b road and then tripinnatifid at base; base broadly rounded in outl i ne; texture th ick herbaceous; calor beneath pale ol ive-green; eglandular. Stalked p innae 4-7 per side, the lowest only sl ightly expanded basiscopical ly. the basal basicopic p innu le exceed i ng the longest acroscopic pinnu l e by about one-fourth . D issection lax for the genu s, segments becoming 5 m m broad before shal low crenations appear. Al l axes of the lamina bear greyish clathrate scales beneath, mostly oriented longitud inal ly . The main rach i s, however. has numerous such scales on the upper side, oriented transversely and relatively broad, 3-5 mm at base. Sari med ial or i nframed ia l , indusium sma l l , densely setulose. Exsiccata: Luzon, Lagu na Prov . , Mt M ak i l i ng, in stable moist shade, 400-550 m, M. G. Price 683, 932; Quezon Prov. , L lavac, M. G. Price 950; Sorsogon Prov .. Mt B u lusan, M. G. Price 1925.

Endem ic to the Phi l ippines. The largest plants produce magnificent fronds, deeply quadripinnatifid. D ist ingu ished from the related and common Ctenitis setosa (Pres l ) H olttum (C. vi/is sensu Copeland, 1 960) by the paleae of the rachis, the shape of the base of the lamina, and the ampleness a nd texture of the segments.

P R I C E : P H I L I P P I N E F E R N N OTES 255

3. F I RST R EPO RT O F ACROSO R US F RO M LUZON

As a genus, Acrosorus is in a precarious position, for a lthough it probably is a natural group, it may be found not to deserve generic rank after a comprehensive study of the fami ly Gramm itidaceae is made. I n provisionally accepting it here, I fol l ow a conservative course.

The Luzon species, brought to my attent ion by Dr P.M . Zamora, and known only from recent col lections, is : Acrosorus streptophyllus (Baker) Copeland , Phi l ip . J . Sci . 56: 480 ( 1 935) ; H olttum,

Ferns of Malaya 222, t. 1 2 1 ( 1 955) .

Polypodium streptophyllum Bak. , J . Bot., Lond . 1 7 : 42 ( 1 879) ; C . Chr . & Holttum, Gd ns. B u l l . Straits Sett l . 7 : 298 ( 1 934) .

Rh izome short, bearing l ight-brown lanceolate paleae 2 mm l ong and l inear fronds 3-1 1 cm. Stipe to 5 mm long to reduced basal segments. Lamina subcoriaceous, pinnatifid to c. 0.2 mm from the costa with entire segments. Steri le segments to 2 mm long, c. 1 .5 mm broad just above base, varying from triangu lar and blu ntly po inted with large sinuses to oblong and broadly rounded with narrow sinuses. Fert i le segments to 3 mm long, confined to the upper portion of the frond , more stro ngly ascend ing than the steri le, each bearing a single deeply embedded sorus protected by the info ld ing of the apical portions of the segment from both sides which produces a much promi nent ridge on the adaxial side. Veins one per segment, simple in lower segments and once forked near the base in the middle and upper segments; the acroscopic branch of the vein suppl ies a sorus. Ha irs 0,3-0.5 mm long, stiff, mostly persistent, dark redd ish-brown, sol itary or in c lusters of 2-3, borne a long the costa, most numerous abax ia l ly, and a lso a long the margins of the segments. Sporangia not setu lose. Exsiccata: Luzon , Ouezon Prov., L lavac, epiphytic on mossy trunks, scarce and mostly occurring as iso lated ind ividua ls, 250-350 m, P. M. Zamora 1 15, M. G. Price 1501; l sabela Prov . , San Mariano, 500 m, H. Gutierrez PNH 78348.

Our other species, known from M i ndanao, Negros, and M i ndoro, is A. triangularis (Scort. ex Bedd . ) Cope l . , of which synonyms based on P h i l ippine material are A. exaltatus (Copel . ) Copel . , A. merrillii Copel . , and A. symmetricus Copel . lt is possible that this should be combined with A. streptophyllus, as both are variable and the d ifferences, as given by Ho lttum ( 1 955 ) , are mostly quantitative and may be due to age, environment, or ploidy; the latter wou ld then be the correct name for a l l Ph i l ippine specimens. I am unfami l iar with A. sch/echteri (Christ) Christ from New Guinea and A. reineckei (Christ) Cope l . from Polynesia .

4. TWO ANT I C I PATE D T H E LYPTE R O I D FE RNS

If a plant occurs to the south of the Ph i l ipp ines, i r) New Gu inea, Su lawesi , or Borneo, and also to the north, in Taiwan, etc., it m ight wel l be expected to be present here. Thu s Holttum ( 1 955) pred icted for Thelypteris singalanensis : "This species is certain ly found in Sumatra, Malaya and Borneo ; it i s reported also by Ch ing from F ormosa, in which case it probably occurs in some intervening territory." As anticipated, th i s fern has now been found in Luzon, where it is wel l establ i shed . Thelypteris singalanensis (Baker) Ch ing, B u l l . Fan Mem. I nst. Bot. 6: 334 ( 1 936) ;

Holttum, Ferns of Malaya 243, t. 1 38 ( 1 955) .

Nephrodium singalanense Bak. , J . Bot., Lond . 18 : 212 ( 1 880) .

256 B R IT ISH F E R N GAZETT E : VOLUME 1 0 PART 5 ( 1 972)

Exsiccata: Luzon, Laguna Prov., Mt Maki l i ng; in mossy forest near the summit, 1 000- 1 1 00 m, B. F. Hernaez 180, M. G. Price 379, 556, 1252, 1453; Ouezon Prov . , Llavac, P.D. Pages 4.

The Ph i l ippine specimens are s l ightly l arger than those described from M al aya by Holttum ( 1 955). but are not otherwise d ist ingu ishable. The identification has been verified by Or H olttum, who commented on the variabil ity in size and revealed that the species also is known in New G u inea but by another name. Accord i ng to Holttum 's new d ivision of the fami ly Thelypteridaceae (H O L T T U M 1 97 1 ), this species belongs in the genus Metathelypteris.

Holttum and R ay ( 1 965) have reported the d iscovery in New Guinea of another thelypteroid fern , Pseudophegopteris aurita, a l ready known from Southern Chi na to I nd ia a nd from Mt K inabalu in Sabah . I ts presence in New G u i nea greatly increased the l ikel ihood that it occurs in the Ph i l ippines, as is now affirmed . Pseudophegopteris aurita (H ooker) Ch ing, Acta Phytotax . Sin . 8: 3 1 4 ( 1 963) ; Ho lttum

& Ray, B lumea 1 3 : 1 3 1 ( 1 965 ) ; Holttum, B lumea 1 7 : 13 ( 1 969). Gymnogramme aurita Hook ., Hooker's l e . P lant. t. 974 ( 1 854) .

Exsiccatum: Luzon, M t San Cristobal , i n mo ist peaty soi l at edge of bog- l ike crater lake, 1 400 m, M. G. Price 836.

Known only by this one col l ection in the Phi l ippines·, which conforms in a l l respects to the description given by Ho lttum ( 1 969) .

5. CON F I R MATION O F VAG I N U LA R I A PARADOXA

I n excluding this species from the Ph i l ippine fern F lora, Copeland , p. 553 ( 1 960), suggested that Hooker ( 1 864) had attributed it to this country through a mis identification. Previously, both Beddome ( 1 883) and Copeland ( 1 905) had accepted Hooker's report, wh ich was based on an unnumbered Cuming col lection said to have come from Luzon . Then , Copeland ( 1 907) told of receiving a specimen col lected by Steere labeled Phi l ippines, conclud ing that "The determination is correct, and this species should probably be added to our known flora". F ina l ly, Christensen ( 1 933) determined Etmer 22124 from Mt Pinatubo, Luzon, to be Monogramma paradoxa; this col lection is represented by two sheets in the Phi l ippine N ational Herbarium. There have been several recent col lections from Mt Banahaw, Luzon, where both Cum ing and Steere are known to have botanized . Vaginularia paradoxa ( Fee) Mettenius, Ann. Lugd . Bat. 4: 1 74 ( 1 869) .

Pleurogramme ? paradoxa F ee, M em. 3 : 38, t. 4 ( 1 852) . Dic/idopteris paradoxa ( Fee) Carr . , Seem . F l . V it. 370 ( 1 873) . Monogramma paradoxa (F ee) Bedd. , Ferns B rit. l nd . Suppf . 24 ( 1 876) ; H andb .

375, t. 2 1 4 ( 1 883) ; Copel . , Polyp. Ph i l ip . 1 06 ( 1 905) ; Ph i l ip. J . Sci . 2 C : 407 ( 1 907 ) ; C . Chr. , Leafl . P h i l ip . Bot . 9 : 3 1 64 ( 1 933) .

Vaginularia junghuhnii Mett., F i l . H art. L ips. 25, t. 27 ( 1 856) . Monogramma junghuhnii (Mett. ) H ook., Sp. F i l . 5 : 1 23, t. 289b ( 1 864) . R h izome creeping, bearing dense lanceolate-acuminate toothed c lathrate paleae

2 mm long, and narrowly l inear fronds to 1 3 cm or more long, lam i na indistinct from stipe. Frond s c. 1 mm apart, glabrous, entire, c. 0.8 mm broad, apex emarginate to blu ntly po inted but seldom intact. Sorus single, narrow, cont inuou s, borne along the

PR I CE : PH I L I PP I N E F E R N NOTES 257

costa between two indusia! strips, extend ing to as close as 1 cm from both the base and the apex. Paraphyses hair-l ike, not broadened at tip, contorted , decid u ous. Exsiccata: Luzon, Mt Banahaw, on tree fern tru nks in sheltered sites, 1 000- 1 300 m, B. F. Hernaez 2007, J. V. Pancho & B. F. Hernaez 3782, M. G. Price 885, 2 1 12.

The recorded range is from Ceylon to Polynesia and Taiwan to Austral ia, but the above description is based so lely u pon Luzon specimens. c H R I S T E N S E N ( 1 943) regards the Polynesian plant as d istinct from that of Malesia.

Vittaria and the dwarf ferns derived from it are in particu larly u rgent need of a modern rev ision. Among the dwarfs, V. lineata var. trichodes Christ, reported by c o P E L A N D ( 1 905) from M indanao, and Monogramma capillaris Copel . , described from Negros, were, through oversight, om itted by C O P E L A N D ( 1 960) . I am not fa m i I iar with either.

6. THE I DENTITY OF T R I CHOMANES MI LLE F O L I UM P R E S L

Although the Trichomanes millefo/ium of Presl is a later homonym and is therefore nomenclatura l ly i l l egitimate, it is usual ly assumed to be a taxonomic synonym of T. maximum B l . Perhaps the only one to have doubted th is was c o P E L A N D ( 1 933) who assigned it to T. grande Copel . with the remark: "Presl ' s T. millefolium was described with 'rhizoma .. . repens . . . ' wh ich if accurate m u st apply not to th is plant but to T. maximum . The only spec imen cited was Cuming 162, of which a l l the specimens I have seen have an erect rh izome and fascicled stipes ." I n 1 958, however, Copeland retu rned T. millefolium Presl to synonymy u nder T. maximum , without comment.

By courtesy of the late C.V . Morton , I have received on loan from the U .S. National H erbarium a sheet of Cuming 162. Three d ifferent spec ies are represented : T. grande Cope l . , T. apiifolium Presl and a th ird , steri le but with indeed a creeping rh izome and bearing Copeland 's u ndated determ ination, "T. maximum B l . ( ? ) ." This last specimen is T. obtusum (Cope l . ) Morton, which I bel ieve m u st be the plant described by Presl .

Or R . E: H olttum has kind ly searched the Kew herbarium for specimens of Cuming 162, find ing th ree sheets, two bearing exclusively T. grande and the th i rd consisting of a m ixtu re of T. grande with what he bel ieves to be T. obtusum lacking the rh izome. Where T. obtusum is found , it is i nvariably in association with the much commoner T. grande, but T. maximum is confined to h igher elevations and was evidently never co l lected by Cuming.

The fo l lowing is a comparative treatment of T. obtusum and T. maximum :

Trichoman�s obtusum (Copeland ) Morton, Cont . U .S . Nat. Herb. 38: 1 88 (1 968) . Macroglena obtusa Copel . , Ph i l ip . J . Sci . 84 : 1 63 ( 1 955 ) . T. millefolium Presl , Hymen . : 4 3 ( 1 843) ; non Desv. ( 1 82 7 ) . I nvo l ucre ( indusium) 1 .0- 1 .2 mm long b y c. 0.5 mm broad ; u lt imate segments

c. 0.8 mm broad in fresh material (dried material is usual ly curled and shrive l led ) ; lamina (webbing) i n axi ls of forks always l ess than 2 mm long; cel l s of ha irs near rhizome apex yel lowish between septae.

More compact, more finely d issected , genera l ly smal ler (fronds to 40 cm long) than T. maximum . All collections known to me have been from between 300 and 900 m. Endemic to the Ph i l ippines.

258 BR IT ISH F E R N GAZETTE : VOLUME 1 0 PART 5 ( 1 972)

Exsiccata: Negros, G. E. Edano PNH 2 1369 ( i sotype, PN H ) , 2 1372; Luzon, probably Albay Prov., Cuming 162 partim; Ouezon Prov., L l avac, M. G. Price 740; L aguna Prov., Mt M ak i l i ng, M. G. Price 1 180, 1312, 1689.

Trichomanes maximum 8 1ume, Enum. P I . _ Jav. 2 : 228 ( 1 828 ) ; Copel . , Ph i l ip . J. Sci. 5 1 : 2 1 7 , t. 38 ( 1 -4) ( 1 933) ; Ho lttum, F l . M alaya 2, Ferns: 1 07, t. 43 ( 1 955 ) . Vandenboschia maxima (8 1 . ) Copel . , Ph i l ip. J . Sci . 67 : 5 4 ( 1 938) ; Fern F l .

Ph i l ip . 1 : 6 5 ( 1 958) . I nvolucre 1 .5-2.5 mm long b y c. 0.8 mm broad ; u ltimate segments mostly c. 1 .0

mm broad in fresh materia l ; lamina in ax i ls of u lt imate and/or penult imate forks often over 2 mm long; cel l s of ha irs near rh izome apex dark brown between septae.

Usual ly larger and more lax than T. obtusum , Ph i l ippine specimens to 60 cm long, Java specimens to 75 cm or more. A l l Ph i l ippine specimens with data have come from over 1 000 m. I have seen a photo of the syntype from Java, Zippe/ s. n. (Morton Neg. No. 2422) . R ange from Thai land to Polynesia.

In sending me a loan of assorted specimens relating to this problem , Mr Morton included a photo (Morton Neg. No. 2423) of the holotype of T. maximum var. minus 81 . , Enum. P I . Jav. 2 : 228 ( 1 828) , wh ich bears Morton's remark that this variety was not mentioned by Copeland in h is revision of Trichomanes. _ The photo shows a com­plete p lant, read i ly identifiable as T. grande Copel .

7. A NEW S,PECIES A N D A NEW COM B INATION I N D I PLAZ I U M

Diplazium egenolfioides M .G . Price, species nova. ( F ig. 1 ) Ex aff initate D. longissimi (Cope l . ) C . Chr. necnon stipite rhach idique paleis

copiosis n itentibus piceo-atris ornato, lamina semel pinnata, pinnis infimis reductis, axibus omnibus n igris; sed tamen paleis denticu latis, frondibus anguste e l l ipticis, p innis majoribus, in lac in ias dentatas un iformiter subtru ncatas incisas, p innis infim is breviter stipitatis, indusi i s pal l ide brunneis d istinguendo. V ide tabu la .

R,h izome short-erect, bearing sh in ing jet b lack l i near- lanceolate toothed paleae to 8 mm long. Fronds to 50 cm long by 1 5 cm broad, very narrowly e l l iptic, pinnate, pinnae to c. 20 per side, the lowest gradual ly reduced . Stipes crowded , to 1 0 cm long below the lowest reduced pinnae, they and the rachises black and very densely paleate, with paleae l i ke those of the rh izome. Lamina dark-green, paler beneath, d rying dark, texture thin herbaceous, glabrous except for very smal l black paleae along p inna-costae beneath. P innae a lternate, to 8 cm long by ' 2 cm broad, the lowest on stalks to 1 mm, the middle and upper sessi le and then adnate, base truncate; reduced lower pinnae sl ightly deflexed , oblong, broadly rounded at the toothed apex; middle and upper pinnae lanceolate, acute, s l ightly falcate; basa l ly each p inna is deeply cut, to over three-fourths of the way to the costa, the incisions becoming progressively sha l lower towards the merely toothed apex. Segments to 6 mm broad, regul ar, broadest on the basiscopic edge of the pinna, sides nearly para l le l , apices subtruncate; veins to 6 mm broad , regular, broadest on the basiscopic edge of the p inna, sides nearly paral le l , apices subtruncate; veins to four pai rs, each terminating in a sharp marginal tooth. Sma l l fronds from young plants are often less than 20 cm long with d im id iate pinnae 3 cm long, cut only about one-fourth of the way to the costa. The basal acroscopic segment is then developed into a smal l auricle. Sori extend i ng from costu le to margin or more commonly occupying only a sma l l media l portion of the vein, usua l ly d iplazioid on ly on the lowest acroscopic vein in each segment . I ndusium 0 .5 mm

PR I C E : P H I L IPP I N E F E RN N OTES 259

broad, pal e brown, margin slightly undu late. Spores pale brown, reniform , perispore forming a broad plane med ian wing. Exsiccata: Luzon, Laguna Prov. , Mt M ak i l i ng, local on steep slopes between peaks 1 and 2, in moist shade, 800 m, M. G. Price 482, 1762 (PNH, holotypus), 2089. L iving fronds simu late the gross appearance of ster i le Egenolfia rhizophyl/a ( Kau lf.) Fee (Bolbitis, H ennipman) .

F I G U R E l : Diplazium egenolfioides M . G . Price A : H abit, medium-sized plant. 8: M iddle pinna from med ium-sized frond. C: Palea from stipe base. D, E, F : Outl i nes of midd l e p innae from large, med i u m and smal l fronds showing change in shape with change in size. G: Transverse section of stipe. Drawn by E .S. Calara from the type collection.

260 B R IT ISH F E R N GAZETT E : VOLUME 1 0 PART 5 ( 1 972)

Diplazium egenolfioides belongs in the group d iscussed by Cope land ( 1 947 ) , comprising D. acrocarpum Ros., D. altum (Copel . ) C . Chr. , D . ful iginosum ( H ook. ) M.G. Price, comb. nov. (basionym : Asplenium� fuliginosum Hook., Sp. F i l . 3 : 1 20; 1 860) , D. longissimum (Cope l . ) C. Chr. , D. merrillii Copel . and D. porphyrorachis (Bak. ) D iels. This new species is the most d issected of the group and I bel ieve it narrows the gap between the group characterized by narrow el l iptic fronds and copious glossy black paleae, and the rest of the genus via especia l ly D. banahaoense (Copel . ) C. Chr. , of which young plants have nearly the same frond form and b lackish cast, but have fewer and brownish paleae.

8. TWO R A R E F E R NS F ROM MT MAK I L I NG NEW TO LUZON

Trichomanes rupicola Raciborski , Pterid . B u it . : 24 ( 1 898 ) ; Copel . , Ph i l ip. J. Sc i . 5 1 : 1 8 1 , t. 1 9 (2) ( 1 933) ; ibid. 73: 466 ( 1 940) , 'rupicolum. '

Crepidomanes rupicolum ( R acib. ) Copel . , Ph i l ip. J . Sci. 67 : 59 ( 1 938 ) ; Fern F l . Ph i l ip . 1 : 73 ( 1 958 ) .

This attractive spec ies was described from Java and later found in M indanao. I n Luzon , its only known local ity i s Mt Maki l i ng, where it i s found on the lower slopes, just above the campus of the Un iversity of the Ph i l ipp ines. it is a tiny fern, with the largest frond s only 2 1 mm long. Exsiccata: Luzon, Laguna Prov., Mt M aki l ing, confined to shaded portions of large bou lders a long streams, forming dense mats above the reach of the f lood, 1 00-200 m , B. F. Hernaez CA HP 13248, N. M. Orlido 240, M. G. Price 1236.

The epithet rupicola , meaning rock-dwel ler, is always used as a nou n in apposit ion, and so must not be decl ined as an adjective (C .V . M orton, pers. com m . ) Tectaria beccariana (Cesati) C . Christensen, l nd . F i l . Suppl. 3 : 1 77 ( 1 934) ; Copel . ,

Fern Fl . Phi l ip. 2 : 308 ( 1 960) . Polypodium beccarianum Ces., R end. Ac. Napol i 1 6 : 30 ( 1 87 7 ) . Aspidium bryantii Copel . , Perk . F ragm . 3 : 1 75 ( 1 905 ) . Tectaria bryantii (Copel . ) Copel . , Phi l ip . J . Sci. 2 C : 4 1 2 ( 1 907) . Aspidium bolsteri Cope l . , Ph i l ip . J. Sci . 1 Supp l . : 252 ( 1 906) . C O P E L A N D ( 1 960) records th is rare species from M indanao and Negros, and

impl ies that it may have been included with other species in Cuming 356, col lected in Panay. There has been a report from Luzon under the name Aspidium bryanti by CH R 1 S T ( 1 906) , contain ing brief information about a specimen col lected by Loher on Mt Mak i l ing in 1 906. However, the fern described by Christ cannot possibly have been correctly named because of the specified indusiate sori and much narrower stipe wing; most probably it was Tectaria grandifolia (Pres l ) Cope l .

Although characterized by ex indusiate sor i , this species is very int imately related to T. vasta of western Malesia, which has a persistent indusium as described by H o L T T U M ( 1 955, p .5 1 2 ) . Occasional sori on our specimens of T. beccariana do have minute vestigia l i ndusia.

This is a fern of lowland rain forest, 250-350 m, the broad expanse of lamina serv ing to compensate for the low l ight intensity at the forest floor and the broadly winged stipes form ing a humu s-co l lecting basket. Exsiccata: Luzon, Laguna P rov., Mt. Mak i l ing, M. G. Price 1425; Ouezon Prov . , L lavac, M. G. Price 158, 9 15.

P R I C E : P H I L I PP I N E F E R N N OTES 2 6 1

9. AN EA R LI E R NAME FOR PTE R I S ASP E R U LA J . SMITH E X C H R I ST

J. Smith ( 1 841 ) publ ished Pteris asperula without a descript ion, basing the name on two Cum ing Luzon col l ections, 253 from S. I locos wh ich he designated as var. caudata and 4 13 from Zambales, designated as var. ecaudata. The species name was later taken up by Christ ( 1 898) , who publ ished a brief d iagnosis, virtua l ly u seless for d istingu ish ing the plant, but sufficient to validate the name. H ieronymus ( 1 9 1 4) fo l lowed with an elaborate description and was credited for the name by both H O L T T U M ( 1 955) and C O P E L A N D ( 1 958) .

Meanwhi le, other names had been offered . Hooker ( 1 858 ) , quoting the same two Cuming numbers as d id Smith, gave the fern varietal rank under P. quadriaurita. Copeland ( 1 906) described a new species, P. caesia, which was soon reduced to P. asperula by H ieronymu s ( 1 9 1 4) . Ch rist ( 1 907) l isted various Phi l ippine spec imens of th is same taxon under P. parviloba, a name typified by a col l ection from the Asian ma in land. Among the specimens enumerated by Christ was one col lected by Loher at Montalban in 1 890, qu ite l i kely the same as the one cited by him as P. asperula in 1 898 .

A l l authors seem to have ignored P. oppositi-pinnata F ee. C H R I S T E N S E N

( 1 905 , p.603 ) , i ncluded it in P. biaurita L . a long with P. quadriaurita and P. asperula ; but in sorting these out, H ieronymu s fai led to accou nt for F ee's name. Judging from the description, supplemented by a p late, I th ink this species.can be noth ing but the P. asperu/a of Smith, and it was in fact based on a Cuming co l l ection from the Ph i l ippines, a lthough without number. Thirteen years before pub l ish ing P. oppositi-pinnata , Fee ( 1 852) had cited Cuming 4 13 as P. asperula , but did not mention no. 253 which also had been quoted by Smith. I suggest that Fee's unnumbered Cuming specimen , the type of h is new species, may wel l have been Cuming 253; in any case, I am confident of its specific identity. Pteris oppositi-pinn�ta Fee, M em. 1 0 : 1 7 , t. 1 3 (3 ) ( 1 865 ) .

P. asperu/a J . Sm ., J. Bot. 3 : 405 ( 1 84 1 ) , nomen nudum ; F ee, M em. 5 : 1 25 ( 1 852) , nomen nudum ; Ch rist, B u l l . Herb. Bo iss. 6: 1 48 ( 1 898) ; H ieron . , H edw. 55: 362 ( 1 9 1 4 ) ; Cope l . , Fern F l . Phi l ip. 1 : 1 4 1 ( 1 958 ) .

P. quadriaurita Retz . var. setigera Hook. , Sp. F i l . 2 : 1 8 1 , t. 1 35a ( 1 858) . P. setigera (Hook. ) N .C. Nair, B u l l . Bot. Surv . I nd ia 1 1 : 1 87 ( 1 97 1 ) , basionym

incorrectly cited . P. caesia Copel . , Phi l ip. J. Sci . 1 Supp l . : 1 56 ( 1 906) ; Fern F l . Ph i l ip. 1 : 1-4--2

( 1 958) . P. parviloba Christ, Phi l ip . J . Sc i . 2C: 1 72 ( 1 907 ) , as to Phi l ippi ne specimens. Di stingu ish ing characters: fronds redd ish when young, maturing to a dark

grey ish-green ; p innae a l l broadest at base (basa l basiscopic segment a lways longest ) , sessile except for the lowest, cut to c. 0.2 mm from the costa into blunt mucronu late segments common ly 1 0- 1 5 mm long by 2 .5-3 mm broad, with caudate apex 1 .5-5 cm long; termi nal pinna longest and broadest, tru ncate at base, not at all decurrent; paleae at stipe-base l i near-acicu late, foscous with a very narrow pale margina l fringe, to 8 mm long; frond everywhere rough , w ith stipe and rach is muricate, costae and costu les sp inu lose, vei ns raised above and segment margins recurved and carti laginous.

Apparently endem ic to the Phi l ippines. The Peninsu la plant named P. asperula by H O L T T U M ( 1 955, p .406) , d iffers in several respects and shou ld not, I bel ieve, be included when th is species is strictly i nterpreted .

262 B R I T I S H F E R N G A Z E TTE : V O L U M E 1 0 P A R T 5 ( 1 97 2 )

By cou rtesy of the D i rector of the Botanic Gardens and the Curator of the Herbarium, S ingapore, I have been able to exam i ne specimens of Cuming 253 and 4 13. They are nearly identical and are · both P. oppositifpinnata. No. 253, var. caudata of J . Smith, has caudae t o nearly 5 c m long. No. 4 13, var. ecaudata , o f which there are two sheets at S I N G , is hardly ecaudate, having caudae u p to 3 cm long.

1 0. THE G E N US LAGUNAEA V I DAL

Lagunaea was described as a new fern genu s by V idal i n 1 886 and has passed unnoticed since, escap ing even C H R I S T E N S E N ( 1 905 ) . V ida l 's name is i l legitimate, the genu s is a taxonomic synonym, and no species name was ever assigned to it, but the matter is of some interest because V idal 's Span ish description is the earl iest record of a very pecu l iar enedem ic Ph i l ippine Tectaria.

Tectaria dolichosora Copeland , Ph i l ip . J . Sc i . 38 : 1 36 ( 1 929 ) ; Fern F l . Ph i l ip . 2 : 3 1 1 ( 1 960) . Lagunaea V ida l , Rev. P lantas Vase. F i l ip inas: 307 ( 1 886) , non de Loureiro

( 1 790) = Po/ygonum L . V idal provided a fine description, considering th is was the only fern he ever

described, com paring the species to Nephrodium melanocaulon (= Tectaria, Copel . ) as did Copeland ( 1 929) . H e correct ly described the sori as termina l on free veins, but these are i ncluded w ithin areolae.

The idiosyncrasy of this spec ies is an elongate indus ium. Although elongate indus ia are not un ique in tectarid ferns, each case has been given a generic name; comparable indusia are a lso features of Tectaria blumeana ( R egal ) Morton, presumably from Java, and Tectaria kehdingiana ( K u h n ex Luerssen) M .G. Price, comb. nov. (basionym : Luerssenia kehdingiana Kuhn ex Luerss., Bot. Centra lb l . 1 1 : 77; 1 882) from Su matra, of which I have seen a spec imen at S ingapore.

ACKNOWLE D G E M E NTS

I am indebted to the fo l lowing for access to fac i l ities and/or assistance with l iterature ·and specimens: P rof. J .V . Pancho, U n iversity of the P h i l ipp ines at Los Baiios (CAH P ) , D r P . M . Zamora, U n iversity o f the P h i l ipp ines, D i l iman, Quezon Ci ty (PU H ) , Dr R . E . Holttum, Kew ( K ) , Mr C . V . Morton, Smithson ian I nstitution (US) , Dr J .V. Santos (PU H ) . M r E. Bardenas, Ph i l ipp ine National B otanic Garden (PN B G ) , and the D irectors and Curators of the Phi l ippine Nat ional H erbarium (PN H ) and the Si ngapore Botanic Gardens (S I N G ) . I wish to thank the editors of the Br itish Fern Gazette for valuable advice.

R E F E R E NCES

CH R I S T E N S E N , C . 1 906 . /ndex Filicum. Copenhagen. C H R I ST E N S E N , C . 1 943. B. P. Bishop Museum Bull. 1 7 7 : 58. CH R IST, H. 1 906. Bull. Herb. Boiss. , ser 2 , 6: 1 002. COP E LA N D, E.B. 1 905 . Polypodiaceae of the Philippine Islands: 1 07 . Mani la . COP E L A N D, E .B. 1 933. Philip. J. Sci. 5 1 : 224. COP E LA N D, E . B . 1 947. Genera Filicum: 1 5 1 . Waltham. COP E L A N D, E . B . 1 958-60. Fern Flora of the Philippines, 3 vols. Mani la . HOL TTU M , R . E . 1 955. Flora of Malaya 2, Ferns. Singapore. H O L TTU M , R . E . 1 97 1 . 81umea 1 9 : 1 7-52. V A N S T E E N IS, C . G . G . J . 1 97 1 . 81umea 19: 1 1 2. Wl LSO N , K . 1 959. Contrib. Gray Herb. 185: 97- 1 2 7 .

B R I T. F E R N GAZ. 10 (5) 1 972

THE ORIGIN OF ASPLENIUM BALEARICUM

J. D. LOV I S* , P. J. B ROWNSEY * , A. SLEEP* AND M . G . SH I VAS* * ( M R S T R EVOR WALKE R )

ABSTRACT

The synthesis of two new hybrids, Asplenium haussknechtii x ba/earicum and A. kobayashii x balearicum, is reported. Both combinations were obtained with unprecedented frequencies for hybrid s between u n re lated species. These hybrids show complete fa i lure of chromosome pairing at meiosis, i ndicating that A. balearicum (and also A. haussknechtii. and A. kobayashii) is of al lopolyploid orig in, and thus provid ing the evidence necessary to confirm the hypothesis of SH I VAS ( 1 969), that A. balearicum is an amph id iploid species originating from hybridization between A. obovatum and A. onopteris.

I NT R ODUCTION

263

Three years ago one of us (M:G.S. ) reported the existence of a tetraploid species of Asplenium in the Balearic I slands, d istinct from either A. adiantum-nigrum L. or A. billotii F.W. Schu ltz, wh ich was described as A. balearicum (S H I V A S 1 969 ) . In the same paper, cytogenetic evidence obtained from synthetic hybrids was presented which suggested very strongly that two d iploid species, A. obovatum Viv. and A. onopteris L., were the parents of A. balearicum , a hypothesis which was clearly consistent with the morphological evidence. H owever, one piece of evidence was sti l l needed t o confirm the correctness o f this hypothesis : a demonstration that A. balearicum was in fact of a l lopolyploid origin. l t is the pu rpose o f this paper to report on the cytology of two new synthetic hybrids which do indeed demonstrate that Asplenium balearicum is an a l lopolyploid .

The hybrid ization technique used was that descri�ed in detai l by L o v 1 s ( 1 968) . I n the cytological investigation, developing sporangia were fixed , stained, and preparations made permanent accord ing to M A N T O N ( 1 950, pp.295-6) .

H Y B R I D I ZATION E XP E R I ME NTS

The new hybrids reported here are Asplenium haussknechtii X balearicum and A. kobayashii X balearicum, synthesized in N:>Vember and December, 1 970.

Asplenium haussknechtii Godet & Reut. ( F ig . 1 ) i s a l ittle-known and d im inutive tetraploid species (Brownsey, Lovis & Reichstein, unpubl i shed results ) with a distribution centred on Asia M in or. On ly very recently was it added to the European flora, hav ing been found in Crete (z A F F R A N 1 970 ) . lt is clearly close to A. lepidum C.Presl of central and south-east Europe.

Asplenium kobayashii Tagawa ( F ig .5) is a rare "and inadequately stud ied taxon known only from Japan and Manchuria. Living material has been avai lable to us from only one local ity, and this col lection, wh ich is tetraploid , has recently been shown to be of rather spectacu lar origin, being derived from two very d i fferent d iploid species, Camptosorus sibiricus Rupr. and · Asplenium incisum Thunb. ( Lovis & Sleep, in preparation ) .

* Department o f P lant Sciences, The University, Leeds, LS2 9JT.

* * Department of Botany, The U n iversity, N ewcastle-upon-Tyne, N E 1 7 R U . .

264 B R ITISH F E RN GAZETTE : VOLU M E 10 PART 5 ( 1 972)

1

3 4

F I G U R ES 1 -5: Si lhouettes of fronds, 2/3 natural size, a l l from cultivation at Leeds. 1 : A splenium haussknechtii, TA 1 628. 2: A. haussknechtii X ba/earicum, PJB 1 A . 3 : A. balearicum, O'N ians cu lture. 4 : A. kobavashii X balearicum, J D L 22 1 4A . 5: A. kobayashii, JDL 2206E ex AS/257.

LOV IS ET AL. : O R I G I N OF ASPLEN I UM BALEAR ICUM

ASPLENIUM HAUSSKNECHTI I � X A. BALEARICUM g Hybridization No.

PJB 1

PJB 2

Date

1 1 Nov.

1 1 N ov.

Prothalli

1 0

1 0

20

Percentage success: 65%

Hybrids

7

6

1 3

ASPLEN I UM KOBAYAS H I I Q X A . BALEARICUM 0 Hybridization No. Date Prothalli Hybrids

J D L 2205 1 1 N ov. 1 0 7

J D L 2206 1 1 Nov. 1 0 8

J D L 2207 25 Nov. 1 0 1 0

J D L 2208 25 Nov. 1 0 1 0

J D L 22 1 3 3 1 Dec. 1 0 7

J D L 2 2 1 4 31 Dec. 1 0 1 0

60 52

Percentage success: 87%

CYTO LOGY

Selfs

0

0

0

Selfs

0

1

0

0

2

0

3

265

Al l three species involved in the h ybridization experiments (A. balearicum, A. haussknechtii and A. kobayashii) are tetraploid and show a regu lar meiosis, with 72 bivalents present (PI . XXV I ) .

M eiosis has been stud ied in five examples of Asplenium haussknechtii X balearicum, and in ten examples of A. kobayashii X balearicum. I n both hybrids, a l l of the plants studied have proved to be tetraploid (as was expected ) and show un iformly complete fai lure of chromosome pair ing, 1 44 univalents being present in a l l cases (PI . XXV I ) .

DISCUSS I ON

A. Interpretation of the cytogenetic evidence: the origin of Asplenium balearicum.

The total absence of chromosome pairi ng seen in Asplenium haussknechtii X balearicum and A. kobayashii X balearicum must mean that i n both hybrids none of the chromosome sets possesses any effective d egree of homology with any of the other three genomes present. lt fol lows that the two chromosome sets contributed by the A. balearicum gametes are so d istinct as to be incapable of pairing together, thus ind icating an al lopolyploid origin for A. balearicum. By the same reasoni ng, it also fol lows that A. haussknechtii and A. kobayashii must be of al lopolyploid origin, a fact which in the case of A. haussknechtii has already been establ ished by the study of other synthetic hybrids ( Lovis, Reichstein et al ia, unpub ., B rownsey, unpub. ) , but which had not previously been determi ned for A. kobayashii.

Knowledge that the chromosome sets present i n the gametes of A. balearicum cannot pair together confirms the interpretation given by s H 1 v As ( 1 969 ) of the patterns of chromosome pairing found in the series of synthetic hybrids obtained by her between A. balearicum and each of A. adiantum-nigrum, A. obovatum and A. onopteris respectively.

266 B R ITISH F E R N GAZETTE: VOLUME 1 0 PART 5 ( 1 972)

The relationships involved in a l l of the hybrids under d iscussion here can be symbol i zed as fol l ows, where letters are used for chromosome sets, and the underl in ing bars i nd icate the constitution of b ivalents formed at meiosis :

SPE C I E S :

A. obovatum: A. onopteris:

A. cuneifolium:

HY B R I DS:

A. ba/earicum: Ob Ob On On c c

A. adiantum-nigrum:

A. haussknechtii: A. kobayashii:

A. haussknechtii X balearicum:

A. kobayashii X balearicum:

A. obovatum X balearicum:

A. balearicum X onopteris:

A. ba/earicum X adiantum-nigrum:

Y Z Ob On S I Ob On Ob Ob On Ob On On Ob On On C

Ob Ob On On -- --

On On C C Y Y Z Z s s 1 1

An important consequence of our new evidence is that it f inal ly establ i shes that A. balearicum is d istinct from A. bil/otii F .W. Schu l tz in its origin, since it is a l ready clearly estab l ished that this latter species behaves cytologica l ly as an autotetrap loid (S L E E P 1 966; G I R A R D & L O V I S 1 968; L O V I S & V I D A 1969 ) . Th is is a point Of some sign ificance, because in the fi rst publ ished reports of Shivas' work on A. ba/earicum (quoted in M A N TO N 1 96 1 (p. 1 1 7 ) and M A N TO N & R E I C H S T E I N 1 962) , th is p lant was referred to respectively as A. obovatum or A. bil/otii, names which we now know to have been i ncorrectly applied to her materia l . This error has since been corrected (G 1 R A R o & L o v 1 s 1 969) , but because a certain amount of confusion sti l l persists in the European l iterature (M E v E R 1 968, 1 969) . i t i s desirable t o take th is opportunity of reviewing our knowledge of th is group of species as represented in western Europe. There are four tetraploid species which are sufficiently s imi lar in morphology to produce some taxonomic confusion, at least in atypical states. These, however, are a l l of d istinct origin, as ind icated below : A. adiantum-nigrum A. cunei folium X onopteris (s H 1 v A s 1 969) . A. balearicum -------- A. obovatum X onopteris (s H I V A S 1 969) . A. forisiense A. obovatum X fontanum (s L E E P 1 966 ) . A. billotii A . obovatum sens. lat . (S L E E P 1 966) .

M E V E R ( 1 969) SUggested that the plant studied b y G I R A R D & L O V I S ( 1 969) in Guernsey was indeed an autotetraploid derivative of A. obovatum , but was not A. bil/otii sensu stricto. This latter species he interprets (M E Y E R 1 968) as being the a l lotetraploid derived from A. obovatum X onopteris ! Meyer's 1 969 paper was evidently written whi le those of Sh ivas ( 1 969 ) and Lovis & Vida ( 1 968) were in press, and therefore inevitably in ignorance both of the description of A. ba/earicum as distinct from A. bil/otii (s H 1 v A S 1 969 ) and Lovis & Reichstein's observation (quoted

PLATE X XV I . Permanent acetocarmine preparations of meiosis. A, C, D & E, d iakinesis; B, fi rst metaphase. Magnification X 1 000. A: A. kobayasllii, J D L 2206E ex AS/257, showing 72 bivalents. B: A. kobayashii X balearicum, J D L 2205A, show ing 1 44 u n ivalents. C: A. balearicum, showing ea. 72 bivalents. 0; A. haussknechtii, TR 1 628, showing 72 bivalents. E: A. haussknech rii X baleari-. cum, PJB 2C, showing 1 44 univalents.

PLATE XXVI

A

D

, tl

B • •

41 • - •

, - ,• .,

' I , , '

I I '

#

� •

c E

LOV I S ET AL. : O R I G I N O F ASP L E N I U M BA LEAR I CUM 267

by Lovis & V ida, 1 969, p.67) that the mater ia l of A. billotii from Ronco, Ticino, used in the resynthesis of Asplenium ( X Asplenophyllitis) x microdon was morphologica l ly very close to material from the locus classicus of A. billotii (Wasigenstein, Alsace) , when cultivated side by side. The point at issue here is at least in part a rather unusual type of nomenclatural prob lem. The question of the nature of A. billotii can only be final ly settled beyond all doubt by a demonstration of the cytogenetic behaviou r of material from the type local i ty, and at present we sti l l do not possess any such evidence. Nevertheless, it is worthwhi le record ing here that A. balearicum has now been cu ltivated in Leeds a longside plants of A. billotii originating from the type local ity in Alsace, from Guernsey, and from Ticino. Wh i le the three A. billotii col lections (two of which are known to behave cytogenetica l ly as autoploids) are clearly seen to belong to one species, the A. · ba/earicum cu l ture maintains its morphological distinction from A. bil/otii, as described by Shivas ( 1 969 ) .

B . The synthesis of Asplenium haussknechti i X balearicum and A . kobayashii X balearicum.

The ease with which both of these hybrids were obtained deserves some d iscussion here, for in general , 'wide' hybrid combinations, i . e. hybrids between entirely unrelated species, have in our past experience never been easy to produce. Usua l ly on ly a sma l l or very smal l minority of the 'female' protha l l i are successfu l ly inseminated , and indeed , our experiments have q u ite often been entirely unsuccessfu l . I n contrast, the degree of success recorded here i s qu ite sensationa l , being unprecedented for any prior attempt to obtain hybrids between unrelated spec ies* . I n th is connection , i t i s worthwhi le record ing that the highest percentage success previously ach ieved by any one of us for a cross between LJnrelated parents is 1 9 .8% (A. lepidum X adulterinum M i lde: 96 protha l l i , 1 9 hybrids: Lovis, unpub l . )

I n view o f the entirely d ifferent character o f the two parents, the success obtained (65%) with Asplenium haussknechtii X balearicum was surprising enough, but the resu lt obtained with A. kobayashii X balearicum (87%) is perhaps even more remarkable, particu larly as this level of success was maintained in six attempts over a period of seven w,eeks, and cannot therefore be d ism issed as merely a fortunate accident. This outcome has a certain piquancy, for although this particu lar hybrid combination was obviously very desirable (it was evident that it might at one stroke yield a demonstration of a l lopolyploidy for two species, both currently under investigation ) the in itial attempts were undertaken primarily only as a demonstration of the techn ique of hybrid ization , and in a spirit of cynicism rather than with any serious expectation of success. The cu ltu re of A. balearicum employed was of unpromising appearance, a l ready bearing young sporophytes when first used. The protha l l i used as males were in fact a second crop which had grown up in the shade of the mature garrietophytes. This same cu l tu re of Asplenium balearicum was used throughout the experimental programme, and it manifestly remai ned in excel lent cond ition for at least seven weeks. Th is entire experience seems to show that it can sometimes pay to be optimistic in fern hybrid ization work.

We are not able to explain why th'e two hybrid combi nations described here proved to be so unexpected ly easy to synthesize. At this point we shou ld perhaps

*An even more spectacular percentage success ( 2 1 4% ! ) has al ready been recorded for a hybridiz­ation experiment i nvolving A. balearicum as a male parent (Sh ivas, 1 969, p.78) . 7 prothal / i inseminated in t h e cross. A onopteris <:;1 X A. balearicum d' yielded 1 5 hybrid sporophytes. Of course, in this case the parents were related.

268 BR IT ISH F E R N GAZETTE: VOLUME 1 0 PART 5 ( 1 972)

emphasise that the technique used in these' hybrid i zation experiments in no way differed from that genera l ly used at Leeds (L O V I S 1 968 ) , a method which has remained u nchanged for a number of years now. lt may be that Asplenium balearicum is an unusua l ly amenable parent, and futu re experimentation with th is . spec ies cou ld resolve this poi nt. lt is certa in ly possib le that the particular strain of A . haussknechtii

used ( TR 1628) is pecu l iar ly compatible, s ince it has subsequently been incorporated in some other un l ikely combinations with an exceptional frequency of success (Brownsey , u npub l . ) Four attempts to synthesise hybrids between Asplenium balearicum and other cultures belonging to the A. haussknechtii complex were also made, but without success. lt is however doubtful whether on th is sca le of effort a negative resu lt has real sign ificance.

ACKNOWLE D G E M E NTS

We thank Prof. I . M anton , Prof. T. R eichstein and Prof. H .W. Woolhouse for their encouragement and interest. Prof. R eichste in we thank specifica l ly for the material of Asplenium haussknechtii used in this investigation . We a lso thank the staff of the Experimental Gardens of the Botany section of the P l ant Sciences Department in the Un iversity of Leeds, and especia l ly Mr Peter Lee, M r F rank H orwood and Mr Albert Henderson for their ski lful care and devoted attention in looking after the protha l l ia l cultures and the resu ltant hybrids. One of us, P .J . Brownsey, thanks the Science Research Counci l for a Research Studentsh ip during the tenu re of which his part in th is study was executed .

R E F E R E N C ES

G I R A R D, P.J . & LOVIS, J .D. 1 968. The rediscovery of X Asplenophyl l itis microdon, with a report on its cytogenetics. Brit. Fern Gaz. 10 : 1 -8.

LOV IS, J . D. 1 968. F ern hybrid ists and fern hybrid ising. 1 1 . F ern hyb rid ising at the U n iversity of Leeds. Brit. Fern Gaz. 1 0 : 1 3-20.

LOV IS, J . D . & V I DA, G. 1 969. The resynthesis and cytogenetic i nvestigation of X Asplenophyl l itis m icrodon and X A . jackson i i. Brit. Fern Gaz. 1 0: 53-67.

MANTON, I. 1 950. Problems of cytology and evolution in the Pteridoph yta. Cambridge. MANTO N , I. 1 961 . Evolution in the Pteridophyta. I n A Darwin Centenary, B.S.B.I. Cont. Rep.,

ed. P.J. Wansta l : 1 05-1 20. MANTON , I. & R E I CHSTE I N , T. 1 962. D ipl oides Asplenium obovatum Viv. Bauhinia 2: 79-9 1 . M E Y E R , D .E . 1 968. Uber neue u nd seltene Asplenien Eu ropas. 5 . M ittei lung. Ber. Deutsch. Bot.

Gese/1. 8 1 : 92-1 06. ·

M E Y E R , D .E . 1 969. Uber neue u nd seltene Asp len ien E u ropas. 6. M itte i lung. Ber. Deu tsch. Bot.

Gese/1. 82: 535-55 1 . S H I VAS, M .G . 1 969. A cytotaxononomic study o f the Asp lenium adiantum·nigrum comp lex. Brit.

Fern Gaz. 1 0: 68-80. SLEEP , A. 1 966. Some cytotaxonomic problems in the fern genera Asplenium and Polystichum.

Ph. D. thesis, U n iversity of Leeds. ZAF F RAN, J . 1 970. Les pteridophytes de l ' i l e de C nhe. Israel J. Bot. 1 9 : 236-244.

APPE N D I X Cultures u sed i n hybrid ization experiments.

Asplenium balearicum: Balearic I slands; col i . E .A . O ' N ians, 1 952. A. haussknech tii: Mt Spath i, D ikti Ori, Crete; co l i . J. Zaffran, 19 (Zaffran 1 7 , TR 1 628 ) . A. kobayashii: Side o f Tomari R iver, M t Ohira, H okkaido, Japan; co l i . A nne S leep, 1 968, 1969.

(AS/257, AS/94 1 8 ) .

B R I T. F E R N GAZ. 1 0(5) 1 972

THE DISTRIBUTION OF WOODSIA ILVENSIS

AND W. ALPINA IN BRITAIN

M A R T I N H . R I C KA R D*

269

Both Woodsia ilvensis ( L . ) R . B r. and Woodsia alpina (Bolton) S . F . Gray were much sought after during the middle of the last century (N e w M A N 1 854 and others) , and unfortunately seriously overcol lected . Numerous whole plants were uprooted for herbarium and garden specimens l ead ing to probable extinctions in several of the better known local ities. R ecently new records of these species have not been pub l ished because either the plants have been overlooked or thei r whereabouts have been kept secret for conservation reasons. I bel i eve the former to be the main factor and the aim of this paper is to clarify the past status and d istribution of these species in Britain and at the same time stimulate further search ing to see if any of the apparently extinct records are sti l l extant.

Several of the more common upland ferns have in the past been mistaken for Woodsias notably JA thyrium spp. , Cystopteris fragilis agg. , Dryopteris filix-mas agg. , and in one case even Asplenium billotii. l t is hoped therefore that the fol lowing guide wil l be of some use when identifying p lants i n the fiel d :

F ronds may b e of a n y size from 0 . 5 to 2 0 c m , usual ly 5 t o 1 0 cm . T h e most conspicuous feature of the genus is the cup-shaped i nd usium which is characteristica l ly spl it into narrow segments; d isti nctive vegetative features are few but fortunately both British species are rarely found steri l e in isolation from mature sparing p lants where exami nation of the indusi u m wou ld a l low certain generic indentification. Key to the B rit ish species:

Frond lanceolate, ha i ry and chaffy ( i .e. scaly) beneath, pi lose above; pinnae oblong with usua l ly 7- 1 3 oblong segments; scales frequent on underside of rhach is and costae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ilvensis

Frond oblong or l i near- l anceo late, g l abrous or only sl ightly hairy beneath; pinnae ovate or triangular w ith u sua l ly 3-'-7 obovate segments; scales on rhachis and costae few or absent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . alpina

Neither size of the p lant nor number of pi nnae segments are very constant characters; frond, p innae and segment shapes are more rel iab le criteria. For example there are fronds of W. aipina, of garden orig in , in the Kew herbarium measur ing 17 cm in length with 1 1 or 1 2 segments per p i n nae, whi le some fu l l y mature fronds of W. i/vensis from Moel yr Ogof i n N o rth Wales are only 4.5 cm long and have only 6 segments on the largest p inna.

Forms, intermed iate between the two species in some characters, are therefore qu ite frequent, particu larly from central Scotland, but to date the hybrid (Woodsia X gracilis ( Lawson) Butters.) has never been reported from the B rit ish I sles. l t has been descri bed from North America (cf . T R Y O N 1 948) and Scand inavia ( R O S E N D A H L 1 9 1 5) and may occu r in B ritai n a lthough the chances are smal l as both species rarely grow close enough in the sam e local ity. H owever there are places where both species do occu r together and hybrids m ay yet be fou nd, notably near Moffat where E .J . Bennett in 1 854 col l ected spec imens o f both species ( a l l label led W. ilvensis i n error) .

* 51 R ylstone Way, Saffron Walden, E ssex.

270 B R ITISH F E R N GAZETTE: VOLUME 1 0 PART 5 ( 1 972)

Plants of W. ilvensis sometimes produce fronds l i ke those of W. alpina accordi ng to Wol laston and Backhouse, jun . ( i n N E W M A N 1 854, p.83 ) . These plants may have been hybrids and if so cou ld be partially ferti le (see T R Y O N 1 948, p. 1 6 1 ) and may therefore be able to reproduce themselves or backcross to either parent producing a local popu lation of hybrids. The author wou ld be very interested to hear of any m ixed popu lations of the two species or of local ities where plants of possible hybrid - origin occur.

I n compi l ing the d istribution data on both species the fol lowi ng herbaria have been consu lted . The abbreviations in brackets are those given in Index Herbariorum ( 1 964) . Aberdeen U n iversity (ABO ) ; U n iversity of North Wales, Bangor (UCNW) ; B ritish Museum (Natu ral H istory) ( BM ) ; Cambridge U niversity (CG E ) ; Royal Botanic Gardens, Kew ( K ) ; National Museum of Wales, Cardiff (NMW ) ; Royal Botanic Gardens, Edinburgh ( E ) ; G lasgow U n iversity ( G L ) ; M anchester U niversity (MANC H ) ; Oxford U n iversity (OXF ); and the museums of Carl is le (CL E ) , Dumfries ( D F S ) , Perth (PTH ) and St. Andrews (STA ) .

The ten k i lometer grid reference i s given after each local ity.

1 . WOODS I A I LV E N S I S

WALES

V-c. 48 Merioneth 1 Precipices of Cader l d ris 23/7 1 Specimens seen : - J. Backhouse 1 877 ( E ) ; Mrs E. Jones 1 894 (UCNW ) . The fi rst publ ished record was b y B A C K H O U S E ( 1 877 ) . There are no recent records but the local ity which was above Llyn Aran may sti l l be extant.

V-c. 49 Caernarvon 2 Moel yr Qgof 23/54 Specimens seen :- J.E. Griffiths 1 884 (NMW ) ; J.L. Williams 1 892 (BM ) , 1893 [ as near Beddgelert] ( B M , CG E ) ; H.J. Ridde/sde/1 1 904 (BM ) ; Sine co/1. 1 9 1 3 (OXF ) ; J. R. Lee 1 93 1 ( G L ) ; E. Vache/1 1 935 ( N MW) ; H. F. Doraston 1 942 [ as Moel H ebog] ( K ) . D Y C E ( 1 962) states that on the sheer wal l of a perpend icu lar chimney n ine plants were counted and these were seen by the author in 1 97 1 but the species is very rare in th is site.

3 Moel Lefn 23/54 Specimen seen : - J.A . Whellan 1 942 (NMW ) .

4 Clogwyn y Garnedd 23/65 Specimens seen:- W. Wilson 1 824 [as Snowdon] ( G L ) ; Bowman 1 824 [ as Snowdon] ( E ) . F irst record D . Lhwyd in R A Y ( 1 690) . T h i s was probably a n error for W. alpina as specimens col lected by Lhwyd from this l ocal ity in the herbaria of Oxford U niversity and the Br itish M useum are W. alpina. The f i rst certai n record is therefore based on the above herbarium specimens col lected in 1 824. N E W M A N ( 1 844) i l lustrates fronds from this site wh ich are u ndo�:�bted ly W. ilvensis, but he bel ieved even then that there were no specimens still in existance, indeed there have been no records of W. ilvensis from Snowdon since.

·

R I C KAR D : WOODS IA I N B R I TA I N 27 1

5 Above Cwm l dwal l 23/65 Specimens seen:-W. Wilson 1824, 1 826 (K , CG E ) , 1 828 (G L, BM, K) ; W. Borrer 1 837 (CG E ) ; 'J.M. ' 1 852 ( G L ) . I n this vic in ity there were several d ifferent colon ies o f th is fern.

6 Above the pass of L lanberis 23/65 Specimen seen:- Sine colt. 1 853 (K ) . Latimer Clark ( i n N EW M A N 1 854 p.77) qescribes the habitat a s "rocks of pecul iar character, resembl ing l imestone, that crop out on the G lyder, there may have been a dozen sma l l plants scattered over the rock wh ich was barren, arid and exposed, and the plant seemed to be dwind l ing to extinction". The comment "dwind l i ng to extinction" need not be taken too seriously as the d imensions of the plant vary considerably from summer to summer. Nevertheless, the plant does not appear to have been col lected aga in from th is local ity.

TEES DALE

V-c. 65 North-west Yorkshire Specimens seen without exact local ity : - W. W. Reeves 1 883 (BM ) ; J. Backhouse without date (K ) .

7 Near Cauldron Snout in Yorksh ire 35/82 No specimen s. Ref. Potter in Baker ( 1 854) and Moore ( 1 859) . l t has been general ly assumed that Falcon C l i nts on the Durham side of the Tees was meant, but Cronk ley Scar, wh ich is in Yorksh ire and where it i s suggested by Baker that Potter col lected it, is su itable. Other crags i n· the reg ion cou ld produce this fern but as yet Woodsia has never been recorded with ce�tainty from the county. There are also records from " R ichmond " ( B A K E R & N O W E L L 1 854) , and "near York" ( specimen in CGE sin. cal l . 1 867) ; both these records are errors.

V-c. 66 Durham 8 Falcon Cl ints 35/82 Specimens seen : - J. Backhouse 1 8 2 1 ( E ) ; J. King 1 840 (MANC H ) ; G. Gibson 1 840 (BM ) ; R. Bowman 1 842 ( BM ) ; J. Backhouse, jun. 1 844 ( BM ) , 1 88 1 , 1 883 ( E ) ; J. Tateham 1 845 ( BM ) ; A. Crosfield 1 884, 1 892 (BM) ; H. T. Mennelt 1 895 ( BM ) . The earl iest reference i s in W I N C H ( 1 83 1 , p.68) quoting J . Backhouse and S. Ha i lstone. The Woodsia grew on the basaltic rocks both at the base and at the summit. During the n ineteenth century numerous botanists visited this site and by col lecting cau sed wholesale destruction of the plant. At l east one herbarium sheet col lected from here shows eight whole plants mounted as a resu lt of a single visit!

LAKE D ISTR I CT

V-c. 69 Westmorland Specimens seen without exact local ity : F. C!owes 1 854 (K ) ; W. Crouch 1 862 (K ) ; Mrs East without date ( K ) . F i rst u nlocal ised record - H uddart 1 846 (L I N T O N 1 878, p. 1 5 1 ).

9 Helvel lyn 35/3 1 Specimen seen: Sine colt. 1 87 1 [ "from a cu ltivated specimen orig ina l ly col lected on Helve l lyn"] (OX F ) . Ref: s A K E R ( 1 885) o n the authority of H uddart and C l owes.

272 B R ITISH F E R N GAZETTE: VOLUME 1 0 PAR T 5 ( 1 972)

10 H art Crag, Dove Crag, Fairfield, Scandale Fel l and R ed Screes 35/30 & 35/3 1 Specimen seen: Sine coli. 1 862 [ Fairfield ] (OX F ) : Ref : B A K E R ( 1 885) and W I LS O N ( 1 938) quoting J . Coward and I . H uddart 1853-4.

1 1 G risedale 35/31 Ref: W E I SS ( 1 930) .

1 2 I l l Bel l Range 35/40 No specimens seen. R ef : B A K E R ( 1 885) and W I LS O N ( 1 938) both on the authority of Backhouse. l t is surprising that no specimens have been seen from this locality as Backhouse has col lected freely from other sites.

13 Head of R igginda le and Mardale 35/41 No specimens seen. Ref: B A K E R ( 1 885) and W I LS O N ( 1 938) on the authority of J . Coward.

14 Crosby R avensworth Church 35/61 No specimens seen. R ef : N E W M A N ( 1 854) on the authority of T. Thompson, 1 798. The church has now been demolished but the record was considered by w A TS o N ( 1 883) to relate to the h i l l s i n the vicinity. I f this i s the case the altitude range of W. ilvensis may have to be revised as h i l ls in th is area are a l l relatively low (below 1 400 ft) .l t is more l ikely a case of confusion of habitat.

V-c. 70 Cumberland Specimen seen without exact locality: F. C/owes 1 854 ( K ) . Some of the records for H elvel lyn ( locality 9) may b e from Cumberland.

15 Scafel l Range 35/20 Specimens seen : D. Ratcliffe 1 954 -(CLE ) . Ref : L I N T O N ( 1 878 p . 1 5 1 ) and H O D GS O N ( 1 898 p.366) both quoting I . H uddart and F . C lowes ( 1 847 o r 8 ) . Apparently some 1 00 plants were present near Scafell on high slate crags. On the occurrence of W. ilvensis in the Lake District o . A . R A T C L 1 F F E wrote ( 1 960) :- " l t grows i n at least one habitat i n a dry earthy place and on moderat�ly basic rocks with very sparse plant growth otherwise". There were original ly about 1 00 plants with fronds m easuring up to 7 inches in wet summers (see specimen at C L E ) - one plant had a hund red fronds. R atcl iffe considers th is probably the finest colony of this fern in G reat Brita in . This local ity is almost certain ly H uddart and Clowe's site.

16 Bowness 35/2 1 No specimen seen. Ref: L O W E ( 1 859) . This must be considered a doubtfu l record . I have not found it repeated in any subsequent publ ication possibly because it was assumed that the local ity referred to was B owness in Westmorland where there are no suitable h i l ls. However the record may refer to the Bowness by E nnerdale Water in the neighbourhood of which are some fel ls of moderate altitude which could be a habitat for Woodsia, e.g. the crags on the north and south side of Great Borne.

The existence of so many Lake D istrict habitats is rather suspicious especial ly the glib way in which they are enumerated in the F loras e,g, R iggindale etc. (W I LS O N 1 938 p.267 ) a s i f the fern i s common in the region. The present author has visited, although not thoroughly searched, th is and several of the localities mentioned and fou nd no Woodsia. A few plants of Cystopteris fragilis, with which it is occasional ly confused, were however seen. F u rther suspicion is put on some records by their

R ICKARD : WOODS IA I N B R I TA I N 273

omission from the A tlas of the British Flora (P E R R I N G & W A LT E R S 1 962) , e.g. R iggindale and Mardale; also specimens in the B ritish Museum ( Nat. H ist.) herbarium from the Lake D istrict were immature fronds of Dryopteris sp.

If a l l the habitats given for the region are accepted as good no clear pattern emerges, it rather seems that Woodsia is l ikely to grow on any mountain among the crevices of the ·Slate. The habitat described by RA TC L 1 F F E ( 1'960) is not a rich ly covered mountain ledge, but a comparatively u n interesting area and such an apparently unattractive habitat tends to be ignored by botanists thus contributing to the general fai lure to record the plant.

SOUTH SCOTLAND

V-c. 72 Dumfries Specimens seen without exact local ity : E.J. Bennett 1 854 [as R avine near M offat] (BM , CGE ) ; D.A. Ratcliffe 1 954 (CL E ) . The habitats in th is area were described b y M r . Johnstone in N E W M A N ( 1 854) as always occurring on rocks of western exposure in val leys runn ing from north to south.

17 Devi l 's Beef Tub 36/01 No specimen seen. Ref: W. L ittle in N E W M A N ( 1854) "at 1 200 feet".

18 Near Loch Skene 36/ 1 1 Specimens seen : J. Backhouse 1 858, 1 859 ( E ) . Ref: W . L ittle i n N EW M A N ( 1 854) , and M A T H I S O N ( 1 88 1 ) . Accord ing to M athison the fern grows on rocks of a southern exposure at 2000 feet. s A D L E R ( 1 85 7 ) refers to a "Woodsia ravine" this is possibly a ravine about Loch Skerie.

19 Corriefron G len 36/1 1 Specimens seen: P.N. Frazer 1 857 (BM , E, A B O ) ; J. T. Johnstone 1 872 (CG E ) , 1 889 ( B M ) ; Sine col/. 1 909 (OX F ) . Mrs. C A R R U T H E R S ( 1 863) says the Woodsia is to be found "on rocks at the upper end of the glen". M A T H I S O N ( 1 88 1 ) reports it from the waterfal l of Corriefron on rocks of a western exposure. The fronds in the Oxford herbarium claim to be from the sole rema in ing plant!

20 Saddleback (Saddle Yoke on O .S. maps) 36/1 1 Specimens seen :. P. N. Frazer 1 858 (BM , A B D ) . Th is locality is very close to localit ies 1 9 and 2 1 but i s probably d istinct.

21 Blackhope 36/1 1 No herbarium specimens seen. A sma l l popu lation was located on the B .P.S. Excursion, 1 972. R ef : Scott-E i iot ( 1 896) on the authority of P . N . F razer and J .T. John stone. This local ity is the glen. d rained by the B lackhope Burn.

22 Whitecombe 36/1 1 Specimens seen : W Stevens 1 848 ( B M , CG E ) . The habitat was described a s a rocky cleft near the Whitecombe.

23 Hartfel l 36/1 1 Specimen seen: P.N. Frazer 1 857 (ABO) . M A T H I S O N ( 1 88 1 ) describes the locality as "near Hartfel l on rocks of a western exposure" . This probably refers to the same area as the specimen in Aberdeen herbarium.

274 B R I T I S H F E R N G A Z E T T E : V O L U M E 1 0 PART 5 ( 1 97 2 )

There is a specimen in the Edinburgh herbariu m without col l ector's name, from " N ear Beattock" col lected in 1 852. This i s an u n l i kely way to describe any of the M offat stations; nevertheless it is only just over a m i le d istant from Moffat, and probably does refer to one of the preced ing stations.

V-c. 72/79 Dumfries/Selkirk 24 H i l l s on the D umfries/Selkirk boundary No specimen seen . R ef: Johnstone 1 850 in N E W M A N ( 1 854) . These h i l l s are 2 or 3 m i l es distant from local it ies 1 7-23, but the general form of the h i l ls is rather d ifferent as they are rounded and l ess craggy, however the geology is broadly s imi lar and there is every reason to believe that the Woodsia does sti l l exist in this area . N E W M A N ( 1 854) wrote "the chief station i n the U nited K i ngdom for this fern [W ilvensis] appears to be i n me vic in ity of M offat and K i rkpatrick " . H e cont inued that W. Stevens, a correspondent of his, "found th is fern in considerable abu ndance on very steep crumbl ing rocks amongst the h i l l s d ividing the counties of Dumfries and Peebles" , in fact with in four years Newman heard of 4 or 5 local it ies giving the impression of i t being abundant there. This may have been the case but the Woodsia has s ince suffered so heavi ly at the hands of col lectors that it is now extremely rare. s c o T T - E L I O T ( 1 896) states that in 1 856 there were hundreds of p lants on Whitecombe, whi le at the time of his writing only two plants remained . After several exp lorations of this area the present author can only confirm the scarcity of the p lant.

CENTRAL SCOTLAND

V-c. 86 Sti rling or v-c 87 West Perth 25 Vic in ity of Stirl i ng Ref: L O W E ( 1 859) . Th is record has been ignored by subsequent writers, but the Bencleuch range in v-c 87 ( G rid ref. 27/90, a ltitude 2363 feet) cou ld provide a suitable habitat overlooking Sti l"l i ng .

V-c. 88 Mid Perth 26 Ben Lawers 27/64 Specimen 'seen: F.A . Lees 1 872 ( K ) . Refs: J . B ackhouse in L O W E ( 1 876) , Ouckett and Worland (pers. comms. 1 9 7 1 ) . None of the l iterature records were accepted by P E R R 1 N G & w A L T E R s ( 1 962) confusion with W. a!pina bei ng suspected , but the ex i stence of the herbarium specimen at Kew and the recent independent records are d ifficu lt to ignore. As this is such a rare species neither Ouckett nor Worland col l ected voucher specimens of their records. B R o w N ( 1 964) cited a specimen of W. ilvensis from this local ity in the herbarium of the Missou ri Botan ical Gardens, St . Louis .

27 Ben Chonzie 27/73 No specimens seen . R ef: O r Ba lfour in L O W E ( 1 876 ) . This record may be an error for W. a!pina wh ich grows on the mo u ntain .

V-c. 90 Forfar 28 Corrie F iagh 37/27 Specimens seen : Grevil/e 1 824 [as C lava] ( E ) ; J. D. Hooker 1 836 [C lava] ( G L ) ; J. H.

Balfour 1 840, 1 846, 1 847, 1 859 ( E ) ; H. C. Watson 1 844 (BM ) ; Sine col/. 1 846, 1 882 (OX F ) ; Backhouse 1 849 ( E ) , 1 850 ( K , E ) ; J. Ray 1 865 (AB O ) ; W. Evans 1 87 1 , 1 907 ( E ) ; G. C. Druce 1 882 ( B M , A B O ) ; W.A. Shoo/bred 1 900 ( BM ) ; Foggitt 1 9 1 1 (BM ) ;

R I C K A R D : WOODS IA I N B R ITA I N 275

R.S. Adamson 1 9 1 2 [as G len Clova) (BM ) ; M. C.F. Proctor & K.M. Goodway 1 953 (CG E ) ; N.M. Pritchard 1 956 [as C lova) (AB D ) . The plant i s apparently luxuriant but rare here, there being "about a dozen plants tightly wedged side by side in a narrow crevice, ( R A V E N 1 956) . this has been confirmed again in 1 966 ( H ami lton pers. com m . 1 97 1 ) when there were two colonies.

29 Glen l sla 37/1 7 Specimen seen: R.H. Williamson 1 923 (CL E ) . Ref: Dr Syme in W A T S O N ( 1 883) . W ilvensis is not included in the " F lora of Caenlochan", ( R O G E R 1 954) and this is the mbst l ikely area in the glen where the Woodsia m ight grow, but it may h ave been overlooked.

[V-c. 92/93 Aberdeen There is a specimen in the Kew herbarium from "Anchry H ouse" (col i . T. Frazer 1 875) wh ich is mentioned i n B R O W N ( 1 964) . l t is only included here to correct the error as this specimen is of garden origin . )

V-c. 95 Elgin 30 Near Forres Specimen seen : S. F. Gray without data ( K ) . The area does not appear to b e su itable for th is fern, i t i s more l ikely that confusion of labels occurred . However this far north it is possible that a su itably alpine n iche exists at a relatively low altitude.

V-c 103 Mid Ebudes 31 I sle of M u l l ( ?)

WEST SCOTLAND

Ref : D R u c E ( 1 932) . This is an unlocal ised record accepted by Perring & Waiters ( 1 962) on the ba�is of the Comital Flora record. The only suitable island appears to be M u l l but the plant has not been re-recorded in the Br itish Museum (Natural H istory) survey of the flora of M u l l . This is a lmost certai nly an error by Druce.

2 . WOODS I A A L P I NA

WA LES

V-c 49 Caernatvon 1 Moel yr Ogof 23/54 Specimens seen: J. E. Griffiths 1 884 ( N MW) . G R I F F I T H S ( 1 895 p.1 67) quotes F .C . R oper as having found W alpina "near Beddgelert" wh ich may refer to this local ity. P E R R I N G & W A L T E R S ( 1 962) record W. alpina for the 1 0-km square wh ich includes this mountain since 1 930, but this appears to be an error for the specimen of W ilvensis (in NMW) col lected by J .A. Whel lan in 1 942 ( H ami lton pers. comm. ) . l t i s possible however that a specimen of W alpina was col lected at the same time.

2 Clogwyn y Garnedd 23/65 Specimens seen: D. Lhwyd pre 1 690 (OX F , BM) ; C. Parish 1 848 ( E ) ; J. Kirk 1 849 ( K, CGE ) ; J.M. 1 852 ( G L ) ; T. Butler: 1 855 [ as Snowdon) (AB D, MANCH ) ; Miss Ridley 1 942 [as Snowdon] ( BM ) ; E. Roberts 1 948 (NMW) .

276 B R ITISH F E R N GAZETT E : VOLUM E 1 0 PART 5 ( 1 972)

Refs: Numerous from R A v ( 1 690 - col lected by Edward Lhwyd ) to o Y c E ( 1 962) . The plant i s now extremely rare here and i naccessible without ropes.

3 Pass of Llanber i s on the N E side of Snowdon 23/65 Specimens seen: Sine col/. 1 852 ( G L ) ; W. lngham 1 895 (NMW ) .

· Refs: N EW M A N ( 1 854) to Duckett (pers. comm . 1 97_1 ) . There appear to b e several local ities in this area, the Woodsia may sti l l exi st i n a l l of them.

4 Near L lyn y cwn 23/65 Specimen seen: W. Wilson 1 824 ( G L ) . There h as been n o record of this species i n the local ity s ince.

5 Cwm Bochlywd 23/65 No specimen seen. Ref : G R 1 F F I T H ( 1 895) . Th i s record needs confirm ing. ·

6 Carnedd range 23/66 No specimens seen. J. O uckett has reported (pers. comm . 1 97 1 ) f ind ing th is local ity recently. C l imatical ly and edaph ical ly there is no reason why the species should not qe present i n several places in the Carnedds.

LAKE D I ST R I CT

Al l the Lake D istrict records are very doubtfu l , but there is no reason why W. alpina should not grow i n the area.

V-c 69 Westmorland 7 Dove Crag , Fa i rfield 35/31 No specimen seen. Recorded in B A K E R ( 1 885) as "having been seen" there. w 1 LS O N ( 1 938) did not include the record i n h is F lora.

8 Helvel lyn 35/31 Specimen seen: Knowleton without date ( K ) . th is local ity is prefaced on the herbar ium sheet by a question mark . If it was not col lected on H elvel lyn it seems l i kely it came from the v ic in ity and presents the best evidence avai lable for citing W. alpina as a Lake D istrict p lant.

[V-c 70 Cumberland Kinprite

Specimen seen: J. Tay/or 1 860 ( K ) . The local ity i s written " K i nprite Co. Cumberland" and is not included in standard Gazetteers of the B ritish I sles. I suspect th i s specimen to be of foreign (probably the U .S.A . ) or garden orig in . ]

SOUTH E R N SCOTLAN D

V-c 72 Dumfries Specimens seen without exact local ity: E.J. Bennett 1 854 [ R av ine near Moffat] (BM , MANCH ) .

9 Near Loch Skene 36/1 1 Specimen seen : Sine col/. and date (AB O ) . Refs: N E W M A N ( 1 854) and L O W E ( 1 859) . Both of which refer general ly to the "Moffat H i l ls". L ittle, in Newman, expresses doubt over the ex istence of th i s species in the area, but Lowe shows no such hesitation. Subsequent writers, i nc lud i ng

A ICKAAD : WOODS IA IN BR ITA I N 277

sco TT-E L I O T ( 1 896 p206 ) . have omitted thi s recqrd. U nless there has been some confusion of labels in the Aberdeen herbarium there can be l ittle doubt that W. alpina did at one time occur near Loch Skene.

10 Whitecombe 36/ 1 1 Specimen seen: J. Sad/er 1 857 ( D FS ) . There is some confusion over this record a s the locality and col l ector's name have been added at a later date. ·

CENTRAL SCOTLAND

V-c 87 West Perth 1 1 Glen Fal loch 27/32 Specimens seen: E.S. Marsha/1 1 890 (CG E, B M , PTH ) ; F.J. Hanbury 1 890 (BM ) . Ref: W H I T E ( 1 898) o n the authority of S . Grieve i n 1 888. There are several suitable h i l l s on the south side of the glen, but this local ity may be identical with locality 1 3 which i s over the watershed of the R iver Fa l loch i n v-c. 88, but E.S. M arshal ! has chosen to label specimens from both local ities suggesting they are d istinct.

V-c 88 Mid Perth 1 2 Ben L u i ( Laoigh ) in Perthsh i re 27/22 Specimens seen: H.M. Drummond-Hay 1 874 ( BM ) ; F.B. White 1875 (PTH ) ; R. Kidston 1 879 ( B M , E) ; F.M. W. & R. Robinson 1879 '[as Tyndrum, Perth] (GL ) ; Sine coli. 1 9 1 1 ( N MW ) . Ref: Wh ite ( 1 8913) . The specimen i n NMW was col lected a t an altitude o f 2300 ft. Some of the above mentioned specimens may have been col lected in v-c. 98, Main Argy l l , as the vice-county boundary bisects the mountain .

13 Cruach Ardra in 27/42 Specimens seen: E. F. Linton 1 891 [ as Corrie Ardrain] ; E.S. Marsha/1 1 891 [ as Corrie Ardra in] (CGE ) ; A. Somerville 1895 ( E ) .

1 4 Ben More 27/42 Specimens seen: Sine co// 1 905 [at 3480 ft] ( NMW) . If this estimate of a ltitude is correct it constitutes an i ncrease i n the upper l imit of this species so far recorded in the British Isles; the species has not otherwis� been recorded above 3000 feet.

1 5 C reag M hor 27/33 Specimens seen : A. Somerville 1 897 ( E ) ; J. R. Lee 1 929 (G L ) . R ef: W H I T E ( 1 898) .

16 Beinn H easgarnich 27/43 Specimens seen : P. Ewing 1 892 (NMW) ; 1 908 (CG E , K ) ; F. C. Crawford 1 902 ( E ) . l t may have been common here a t one time a s the sheet · i n the Edinburgh herbarium contained 1 2 complete plants.

17 Meall a' Churain 27/43 Specimen seen : P. Ewing 1 872, 1 902 [as Mea l l Chiri rn, K i l ! in] ( G L ) .

1 8 Stuchd a n Lochain 27/44 No specimens seen . R ef: H am i lton (pers. comm. 1 972) on the authority of E .C . Wal lace, 1 938. This is a large mountain and it is possible, but unl ike ly that W. alpina was recorded in 27/5 4.

278 BR ITISH F E R N GAZETTE : VOLUME 10 PART 5 ( 1 972)

Local ities 1 9-22 are close together form ing a chain of h i l ls about four m iles long.

19 Meal ! Ohuin Cro isg 27/53 No specimens seen . Ref: Wi l son 1 827 ( i n N E W M A N 1 854) .

20 Creag n a Cai l l ich 27/53 Specimens seen: J. Hardy 1 828 (MANCH ) ; A. Maingay 1 856 (MANCH ) ; J. Backhouse 1 859 ( E ) ; Sine col/. 1 860 (OXF ) ; 'A. McK. ' 1 860 ( G L ) ; 'W.R.' 1866 (AB O ) ; J. Roy 1866 ( B M ) ; E. F. L inton 1 882, 1 884 (BM ) , 1 883 (CGE ) ; F. C. Crawford 1 900 ( E ) . Ref: F i rst record Wi lson 1 827 ( i n N E W M A N 1 854) .

21 Cam Chreag 27/53 Specimens seen : J. Sad/er 1 875 (ABO, B M ) , 1 866 ( E ) ; Sine col/. 1 883 [as Mea l l nan Tarmachan] (NMW); F. Hanbury 1 888 (CGE, BM) ; E.A . Richards 1 908 [as Mea l l nan Tarmachan] (BM ) . Cam Chreag i s a common name i n Scotland and some of the herbarium specimens so label led cou ld possibly origi nate from the simi larly named mountains at 27/3734 or 27/5348.

22 Creag an Lochain 27/54 Specimens seen : Sine col/. 1 828 ( E ) ; Graham 1831 ( E ) ; J. Leitch (col i . ) 1869 (CL E ) ; B. White 1 875 (PT H ) ; F.M. Webb 1878 ( E ) ; W. R. & E.F. Linton 1 883 (BM , MAN C H , ABO, E ) ; P. Ewing 1 883 (MANC H ) , 1 897 ( B M ) 1 9 1 0, 19 1 2 (GL) ; W.B. Wareterfa/1 1 885 ( E , MANCH , BM , K ) ; W. G. Craib 1 885 (ABO) ; G. Sim 1 893 (STA) ; W. Young 1 893 (STA ) ; W.C. Warsde/1 1 895 ( K ) ; D. T. Playfair 1 900 ( BM ) ; F. C. Crawford 1 900 [ 1 9 complete p lants mounted ] ( E ) ; A. Somervi/le 1 900 (E ) ; Me. T. Cowan 1 906, 1 907, 1 9 1 0 ( B M ) . 1 909 (MANCH) , 1 9 1 0 (K, E ) ; T.J. Foggitt 1 907, 1 93 1 ( BM ) ; R. & M. Corstorphine 1 907 (Sl:"A ) ; Sine col/. 1 9 1 0 (OX F , N MW ) ; 1 945 (STA ) ; J. R. Lee 1 9 1 5 ( G L ) ; D. M. Henderson 1 953 ( E ) . Probably the best known local ity for the species in Scotland but th is popu l ation cannot however stand such intense col lecti ng in the future.

There are specimens in CGE and BM col lected on Catjagh iamman by J.T. Syme in 1 85 1 , this is possibly with in th is area ( local ities 1 9-22 ) .

23 Ben Lawers 27/64 Specimens seen: J. Dickson 1 79- (BM , E ) ; A rnott 1821 ( G L ) ; J.H. Balfour 1823 ( G L ) , 1 828 ( B M , E ) ; W.J. Hooker 1 824 ( K ) ; J. McNab 1 829 (BM ) ; Greville 1 837 ( BM ) ; W. Wilson 1 836 ( N MW , BM ) ; Graham 1 838 ( G L ) ; F.J. Ivory 1847 ( E ) ; P.N. Frazer 1 856, 1 862 ( E , A B O ) ; Elizabeth Metcalf 1 857 ( K) ; W. Bell 1 860 (AB O ) ; J. Sad/er 1 860 ( E ) ; S. Sim 1 864 (MANCH ) ; J. Roy 1 866 (ABO ) ; F. B. White 1 862, 1 872 (AB O ) ; F.A. Lees 1 872 ( K ) ; P. Ewing 1 892 (G L ) ; C. E. Salmon 1 9 1 3 (BM ) ; W.A. Shoo/bred 1 9 1 3 (NMW) ; 'D.R. ' 1 92 1 ( G L ) . Ref: D I C K S O N ( 1 790) . "Ben Lawers" i s qu ite l ikely a n over general i sation of local ity, no doubt many of these specimens were col lected on Creag an Lochain (22) or An Stu ic (24 ) , however some were probably col l ected on Ben Lawers itself.

24 An Stuic 27/64 Specimens seen : J. Dickson 1 97- ( BM ) ; J.H. Balfour 1 823 (G L ) , 1 828 ( BM ) ; Greville hb. 1 856 ( E ) ; F.J. Hanbury 1 888 (BM ) ; S.M. Mc Vicar 1 891 [as Ben or Mea l l Garbh] (BM ) ; C. E. Salmon 1 9 1 3 ( BM ) ; E.S. Marsha/1 1 9 1 3 (CG E ) . There is a spec imen in H b. R.A. Graham co l lected in 1 9� 1 ( H ami lton pers. com m . 1 972 ) .

R I C KA R D : WOODS IA I N B R ITA I N

25 Meall Ghaordie 27/53 or 54 Specimen seen : E.S. Marsha/1 1 89 1 (CG E ) ; E.F. Linton 1 89 1 ( B M ) . There have been n o recent records_.

26 Ben Chonzie 27/73

279

Specimens seen : J. H. Balfour 1 850 ( E , G L ) , 1 853 ( E ) . 1 854 ( E ) ; Sine col/. 1 850 ( BM ) ; D. Nicho/ 1 856 ( E ) ; J. Bryce 1 859 (CG E ) ; F.B. White 1 86 1 (PTH ) ; Sine col/. 1 867 (OX F ) ; P. N. Frazer 1 867 ( E ) . I n 1 96 1 A . McG. Stir l ing reported W alpina "rare o n d ry rocks at 2000 .feet" (Hami lton pers. comm. 1 97 2 ) .

V-c 8 9 East Perth D R u c E ( 1 932) l i sts an un local i sed record . No specimen has been found to substantiate it.

V-c 90 Forfar 27 Glen l sla 37/1-7 Specimens seen : Greville hb.' 1 832 [as Caen l ochan ] ( E ) ; J. H. Balfour 1 847 ( E ) ; J. Roy 1 869 (BM, MANC H ) ; Sine col!. 1 923 (OX F ) . The most recent records are 1 958 and 1 966 by Dr U . K. Duncan ( Hami lton pers. com m . 1 97 2 ) . Al l these records probably refer t o the cl iffs above the Caenlochan Glen.

28 Glen Cl ova 37/27 Specimens seen : H.C. Watson 1 844 [ as G len F ee] ( B M ) ; Sine col/. 1 845 (OX F ) ; F. Nay/or 1 863 [as Glen Dol l ] ( B M ) ; Gower 1 873 ( E ) . N E W M A N ( 1 854) quotes Balfour, 1 847, from Glen F iadh . The most recent record i n 1 967 from Corrie F iadh b.Y A.C. Jermy (pers. com m . 1 97 2 ) .

V-c 98 Main Argyll 29 Ben Lui ( Laoigh ) 27/22 Specimens seen : see local ity 1 2 . This loca l ity in Cio�han Beinn Laoigh which i s partly i n Perthshire, b u t i t seems l ikely that the fern was found in the Argy l l ha lf (o v c E 1 969 ) . A.McG. Sti r l ing (pers. comm . ) confirms its occurrence in the Argyl l half .

30 Coire Gamn hain 27/23 No specimens seen : Hami lton (pers. comm. 1 97 1 ) has a record of it being found in 1 950.

31 Mea l l nan Tighearn 27/22 Specimen seen : �.S. Marshal/ 1 9 1 0 (CG E ) .

32 B e inn an Dothaidh 27/34 Specimens seen: E.S. Marshal/ 1 893 ( B M , CG E ) ; E.S. Marshal/ & W.A. Shoo/bred 1893 (NMW) . The post- 1 930 record for the square in the A tlas of the British Flora is an error.

V-c 92 South Aberdeen, 94 Banff or 96 Easterness 33 Cairngorms Specimen seen : J.M. Balfour 1 822 ( G L ) . Th is i s a very doubtfu l record but cou ld originate from the Cai rngorms i n any of these vice-counties. By 1 823 (the next year) Balfour knew of the Ben Lawers loca l ities, he may therefore have kept the record d istinct on pu rpose. I n addition, on a single record

280 B R ITISH F E R N GAZETTE : VOLUME 10 PART 5 ( 1 9 7 2 )

card covering parts of v-cs 89, 90 and 92 J .G . R oger recorded W. alpina from "rocks of the Dalradian series" (Hami lton pers. comm. 1' 972) , this most l i ke ly refers to v·c 90 but could be in v-cs 89 or 92. G len Cal later in v-c 92 is of a simi lar geological structure to both G lens ls la and Clova and may house this species.

V-c 1 04 North Ebudes 28 I sle of Skye Specimen seen : E. F. Linton 1 886 ( B M ) . This cou ld b e virtually anywhere o n the upland portion o f the island .

ACKNOWLEDGEMENTS

I wou l d l ike to thank M iss N . H am i lton of the B iological Records Centre for her generous assistance, and a lso the curators of a l l herbaria consu lted , for the loan of specimens.

R E F E RE NCES

BACKH OUSE , J. 1877. Botanical Record Club Report 1876: 227-246. BAKE R , J .G . 1 863. North Yorkshire, London. BAK E R , J .G . 1 885. Flora of the English Lake District. London. BAKER, J.G. and NOWELL, J. 1 854. Supplement to Baines Flora of Yorkshire, London. BROWN, D. F.M. 1964. A monographic study of the fern genus Woodsia. Beih. Nova Hedwigia. 1 6 :

1 - 1 54. CAR R UTH E RS, M RS. 1 863. The ferns of Moffat, l\lloffat. DI CKSON , J. 1 790. Plantarum cryptogamicarum Britanniae 2, London. DR UCE, G.C. 1 932. Comital Flora of the British Isles. A rbroath. DYCE, J.W. 1 962. British Pteridological Society A n nual E xcursion. Brit. Fern Gaz. 9: 87-92. DYCE, J .W. 1 969. B ritish Pteridological Society A n nual Excursion. Brit. Fern Gaz. 1 0: 1 00- 1 04. G R I F F ITHS, J .E . 1 895. Flora of Anglesey and Carnarvonshire, Bangor. HODGSON, w. 1898. Flora of Cumber/and, Carlisle. HYDE , H.A. , WA DE, A.E . and H A A R ISON, S.G. 1 969. Welsh Ferns, 5th ed . Cardiff. LI NTON, W.J. 1 878. Ferns of the English Lake Country, 2nd eel. Windermere. LOWE, E.J. 1 859. Ferns British and Exotic 7. London. LOWE, E.J. 1 876. Our Native Ferns London. MATHISON, R. 1 88 1 . Note on stations of Woodsia ilvensis, Hist. Berwicksh. Nat. Club 9: 542. MOOR E , T.

-1 859. Octavo

- Nature Printed British Ferns London. NEWMAN, E . 1 844. History of British Ferns. London. NEWMAN, E . 1 854. British Ferns: 2, London. PE A R l NG, F .M . & WAL TEAS, S .M. 1 962. A tlas of the British 'Flora. London. RATCLI F F E , D.A. 1 959. The Mountain Plants of the Moffat H i l ls. Trans. Proc. bot. Soc. Edinb.

37: 257-27 1 . RATCL I F F E, D.A. 1 960. The mountain flora of Lakeland. Proc. bot. Soc. Br. Is/. 4: 1 -25. RAV E N , J. & WAL TE AS, S.M. 1956. Mountain Flowers. London. RAY, J. 1 690. Synopsis methodlca stirpium britannicarum. London. ROGE R, J.G. 1 954. The F lora of Caenlochan. Trans. Proc. bot. Soc. Edin. 36: 1 89-194. ROS E N DA H L, H. V. 1 9 1 5. Uber Woodsia alpina u nd eine sud liche B innenlandform derselben sowie

uber Woodsia a lpina X ilvensis nov. hybr. Svensk bot. Tidskr. 9: 4 1 4-420. SADLER, J. 1 857. Narrative of a ramble among the wild flowers of the Moffat hills in August.

Moffat. SCOTT· E L I OT, C. F . 1 896. The Flora of Dumfriesshire. D umfries. TRYON, R .M . 1948. Some Woodsias from the north shore of Lake Superior. Amer. Fern J. 38:

1 59. WATSON, H .C. 1883. Topographical Botany. 2nd ed . : 508, London. WEISS, F .E . 1 930. N. West.Nai. 5 : 15 1 .

-

WI L'soN, A. 1 938. Flora of Westmorland A rbroath. Wl NCH, N.J . 1 831 . Flora of Northumberland and Durham: N ewcastle. WH I TE, F .B .W. 1 896. Flora of Perth. Edinburgh.

PLATE XXVII

frond. 1llnp. &nd dllluet:tl.Ofl .. 1�1 •••cUr ,,.....,r d.l<"Le =v�� �Tt� !��r=�th "'" .. [t is � op1n1on tllat tll 11 1pecima 11 the h.yb["ld �iMi!!lii:hr. � 1y r;1�;.,.. t>et: Ann.o Sh••P• llth Hay , m>.

f>�J_ � • .J,..-1 �� P.lv-c-1 p J.

· ¥·-_1., "'(?.1 (.1 ·�

B R I T. F E R N GAZ. 1 0(5) 1 972 281

POLYSTICHUM X ILLYRICUM :

A HYBRID NEW TO THE BRITISH ISLES

AN N E SLEEP* & DONAL SYN N OTT * *

I n the last volume of the Gazette i t was suggested that the hybrid Po/ystichum X illyricum (= P. acu/eatum ( L. ) Roth X P. lonchitis ( L. ) Roth ) might be :found in the British I sles, and we are pleased to report here the real isation of this pred iction. One sheet of a rather u nusual- looking Po/ystichum came to l ight during a search for Polystichl!m x bicknellii specimens in the herbarium of the National Botanic Gardens, in Dubl in ( D B N ) , -and on further examination th is proved to be a fine specimen of the hybrid P. x ilfyricum.

This specimen was col lected by R . L loyd Praeger in August, 1 932, and it bears th� label "Polystichum acu/eatum, with P. lonchitis, S. side of G lenade, Co. Leitrim". The sheet consists of two very fine mature fronds, both of which are intermediate in morphology between P. - aculeatum and P. lonchitis. The fronds are 55cm long, narrowly lanceolate in shape, a�d rather stiff. Although simply pinnate, as in P. lonchitis, the fronds are deeply •pinnatisect, and the pinnae in the midd le third of the frond are again d ivided into tiny, acute,, obl iquely inserted pinnu les, at least in the proximal part. Dista l ly, the pinnae are entire, a lthough deeply toothed. The proximal acroscopic pinnule tends to be conspicuous, and the pinnae are sick le-shaped, as i n P. lonchitis. The spores are shrivel led and abortive, wh.ich confirms the hybrid nature of

' tpis specim�n'.

· ·

, I The d iscovery 'of th is bybrid in I reland is in itself rather remarkable, for one

parent, Polystichum /ohchitis, is ve�y ri<!rE1 there, being confined to mountain cl iffs in the west. - l t is, however, fa ir ly p'lentiful on the l im!;!stone pl<jteal! of Ben Bu lben, where it occurs at altitudes from 1 ,000 to 2,000 feet (300-600m) . G lenade is a beautiful c l iff-walled val l ey which intersects the Ben Bu lben range. Extend ing in a north-westerly d i rection from the town of Manor Hami lton in County Leitrim, it is about 1 ,000 ft. deep and a mi le wide, with a lake in its southern half. I n places the massive cl iffs are rel ieved by patches of wood land, a lthough for the most part they consist of vertical c l iffs with some free-stand ing crags, deep crevices and chasms and loose and overgrown l imestone scree. The local ity referred to on the label is a lmost certainly a mi le or so of steep c l iff to the north of the lake and on the western side of the glen. These north-east·facing l imestone escarpments harbour a great number of rare · a lpines, includ ing Arabis petraea and Epilobium alsinifolium. Coch/earia alpina, Chrysosplenium oppositifolium, Silene acau/is, Draba incana, Dryas octopetala, Saxifraga oppositifolia, S. aizoides, S. hypnoides, Polystichum. lonchitis and Asple!f.ium viride are among some of the other a lpine species present. On the talus under these cl iffs l imestone scree is m i xed with the remains of a former covering of peat, and ferns are unusual ly abundant and luxuriant, a l l of the fol lowing species being fou nd within a relatively restricted area : Polystichum lonchitis, P. aculeatum, P. X ilfyricum, P. setiferum, Dryopteris filix-mas, D. carthusiana, D. aemu/a, A thyrium filix-femina,

* Department of P lant Sciences, The U n iversity, Leeds LS2 9JT.

* * National Botanic Garden, G lasnevin, D ub l in, E ire.

PLATE XXVI I . Polystichum x illyricum in herb. National Botanic Gardens, G lasnevin. ( R . L. Praeger s. n . )

282 B R ITISH F E R N GAZETTE: V O L U M E 1 0 PART 5 ( 1 972)

Oreopteris limbosperma, Blechnum spicant, Cystopteris fragilis, Asplenium trichomanes, A. viride, A. adiantum-nigrum, 'A. ruta-muraria, A . scolopendrium, A. ceterach, Polypodium vulgare and Hymenophyllum wilsonii.

P. X illyricum is the commonest Polystichum hybrid occurring on the continent, and is often found, sometimes in- profusion, in the mountainous regions of central Europe. lt certainly could occur elsewhere in the British Isles and it is hoped that the publ ication of th is note on our f ind w i l l stimu late pteridologists to search for this hybrid in other local ities. lt should be noted, however, that juveni le fronds of P. aculeatum can be very easi ly m istaken for P. X il/yricum, as can sma l l , stunted specimens of the same species growing in u nfavourable habitats. I f a m icroscope is ava i lable, qu ick scrutiny of the contents of a sporangium is the best method of positive identification.

R E F E R E NCES

P R A E G E R , R.L I . , 1 934. The Botanist in Ireland. Dubl in. SLE EP, A. 1 97 1 . Polystichum hybrids in B ritain. Brit. Fern Gaz. 1 0: 208-9.

REVIEWS

SYSTEMA TICS OF THE ONOCLEOID FERNS by Robert M. L/oyd. University of California Publications in Botany 61 : 1-86, 197 1. 170 X 258mm. Price $3.50. [U. C.P. 2-4 Brook Street, London W1 Y 1AA]

This paper provides a valuable and detai led account of the morphology and taxonomy of the genera Onoclea, Onocleopsis and Matteuccia and the conclusion that they shou ld be reorgan ised as a natural group seems fu l l y justified . They are compared very fu l l y with both the B lechnaceae and the Aspidiaceae ( in the broad sense of Copeland ) and it is considered that they are related to the Aspidiaceae and are not ancestral to the B lechnaceae as postu lated by Bower. This wou ld probably be genera l ly accepted , but some aspects of the discussion, dealt with below, do not seem entirely satisfactory.

The classification on which the argument is based is set out in a figure derived from one publ ished by Wagner in 1 969. The Aspidiaceae and B lechnaceae are placed in a group termed " I nd usiatae" by Lloyd (together with the Dennstaedtiaceae, Lindsaeaceae, Asplen iaceae and Dava l l iaceae; and also the Hymenophyl laceae ) . L loyd states (p.54) that it is not necessary to d iscuss the evolution of the sorus and indusium a lthough, accord ing to Bower's interpretation, wh ich he accepts in fu l l , the presence of a "false" introrse indusium in the B lechnaceae as opposed to a "true" extrorse indusium in the onocleoid ferns and the Aspidiaceae becomes an important distinguish ing character. Since this interpretation was rejected by Christensen in 1 938 and by Holttum in 1 946 (with supporting argument) an adequate d iscussion shou ld have been provided .

By the recognition of a group " l ndusiatae" in a phyl ogenetic scheme it seems to be impl ied that the "true" indusium in these fami l i es has a common origin . The present reviewer wou ld support th is theory. lt may be added here that the ancestral arrangement is apparently seen in Dicksonia and Dennstaedtia where the sorus arises from the marginal meristem and is protected by both dorsal and ventral flaps. I n those genera (the majority ) w ith superfic ial sori the dorsal flap has retreated with the sorus onto the lower surface of the leaf and has become the normal ly extrorse "true" indusium. In the "Gymnogrammoid " ferns of L loyd 's d iagram, on the other hand ,

R E V I EWS 283

there is no trace of a "true" indus ium. Those genera that have marginal sari (e.g.; Pteris, Pel/aea, Adiantum ) usual ly have these protected by the inrol led and more or less mod ified leaf margin wh ich is termed a "false" i ndusium. Accord i ng to Bower the leaf margin in 8/echnum has formed an introrse "false" i ndusium as in the "Gymnogrammoid" ferns. The heterophy l lous species w ith contracted fert i le pinnae (formerly assigned to Lomaria ) are prim itive and the expanded lamina of the fert i le fronds of the homophyl lous species and of other genera (e.g., Woodwarc/ia, Doodia ) is a secondary development. Lloyd accepts this derivation of the indusium in the Blechnaceae and a lso, apparently, this evolutionary sequence, a lthough he does state that heterophyl ly cannot be used to show affinity between the B lechnaceae and the onocleoid ferns. Holttum, on the other hand, suggested that the sorus of 8/echnum had been derived by the fusion of subcostal, v i rtua l ly introrse sari such as can be found in Asplenium, together with the formation of costal areoles, and pointed out the sign ificance of 8/echnum punctulatum var. krebsii in this respect. He also regarded the heterophyl lous cond ition as derived . The introrse indusium of the B lechnaceae wou ld thus be homologous with the "true" indusium of the rest of the " l ndusiatae".

I f we return to the comparison of the three groups, which is fac i l itated by a usefu l tabular summary, it is evident that the B lechnaceae and the onocleoid ferns are marked ly d istinct from each other .even when indusia! characters are excluded. The author's ma in contention is thus wel l estab l ished . However, it may a lso be observed that, accord ing to th is tab le, there are no further major d ifferences between the B lechnaceae and the Aspid iaceae s.l. apart from the development of the gametophyte, so that an estimation of the evolutionary (and taxonomic) d istance between these two groups depends to a significant extent on the interpretation of the indusfal structure. The final comparison to be made is that between the onocleoid ferns and the rest of the Aspidiaceae sensu Copeland and, as L loyd states, this group fits wel l with in the diversity found in this fam i ly. lt is however th is very d iversity which b lurs the d iscussion of the c loser affin ities of the onocleoid ferns. The Asplen iaceae were a lready spl it off from the Aspidiaceae by Copeland but the Thelypteridaceae and the Lomariopsidaceae are equa l ly worthy of recognition, whi le the remain ing genera fa l l into fair ly well-defined groups. Recogn ition o f these groupings - a t whatever level -wou ld have en l ightened the d iscussion and more extended comparative stud ies with in th is a l l iance wou ld be of great morphological and taxonomic va lue.

The suggestion of the hybrid origin of Matteuccia intermedia, a key species in Bower's supposed phylogenetic series, between 'the onocleoid ferns and the B lechnaceae, is of considerable interest.

F RANCES M. JARR ETT

SOUTH A FRICAN POL LEN GRA INS AND SPORES Ed. E. M. Van Zinderen 8akker Part VI THE SOUTH A FRICAN FERN SPORES by Miss W. G. We/man. Pp. ix, 1 10, 1970, 16 x 24.5cm. 8alema: Cape Town.

This vol.ume fol l ows others of the series where pol len of S African plants are described. The arrangement of fami l ies and the 75 genera i s that of E. Schelpe's forthcoming monograph . There is a key to the "sporomorphae" ful ly described later in the text and a tota l of 262 species and varieties are covered, 75 of them clearly i l l ustrated by drawings. The glossary describes the terms used very expl icitly a lthough it is clear that concepts may differ with d ifferent workers. References to other workers' descriptions

284 B R IT ISH F E R N GAZETTE : V O L U M E 1 0 PART 5 ( 1 97 2 )

are given under each species, however, and assuming the other worker's taxonomic concept is the same th is i s most usefu l . In Miss Welman's case, collector, number and herbarium where i t is deposited are quoted for each specimen.

To a fern taxonom ist anywhere this k ind of book is i nvaluable. I f descriptions of the spores of a cosmopol itan taxon is d ifferent to the one the reader knows there may be a regional (and i ntrinsic) variation wh ich is worth investigating (e.g. Anogramma leptophylla ) .

Megaspores of lsoetes stellenbossiensis Duth ie were not avai lable to M iss Welman . Those on the syntype material (shortly to be designated lectotype) i n the British Museum herbarium show a d istinct sporoderm reticu late on both proximal and distal faces. The microspores are, u nder the S . E . M . , verrucate with m i nute pap i l lae both on and between the verrucae.

POL L EN AND SPORES OF CHIL l by Calvin J. Heusser. Pp. xiv, 167, 1911. 23.5 x 31cm, University of Arizona Press, Tucson. Price $15.00.

Much the same may be said of this book as said of the above review, although Dr Heusser includes flowering plants and describes the microspores of 698 species, of which 54 are pteridophyta . Besides main land Chi le, Easter Is land and Juan Fernadez are i ncluded. This work is i l l ustrated by photomicrographs which are beautifu l l y reproduced yet leave much t o b e desi red - a gap that the S . E . M . w i l l eventua l ly fi l l . The study of palynology i s essentia l to fern taxonomy and I welcome th is · book and more I ike it.

A.C. J E R M Y