it t2 3

1i t\4 Limb -u

Fig. 2. Temperatures measured in channel 2 (18 to 25 um) of the infrared radiometerin Mariner 7 during near encounter, in a scale proportional to energy. The timeswhen the platform was slewed and the crossing of prominent planetary features areindicated. The extreme planetocentric coordinates of the swath through the polarcap are also shown.

significant response to off-axis sources.For example, the response of channel2 of the radiometer aboard Mariner 7to a point source at a distance of 12degrees from axis was only 0.3 per-cent of that for the same source onaxis. Because of the large solid anglessubtended, however, the extended wingsbeyond a central area 1 degree in diam-eter contributed 27 percent of thetotal response of a measurement of theenergy radiated by an isothermal sourcefilling the entire object space. The cor-rection to the minimum temperaturemeasured in Mariner 7, due to the ex-tended response of channel 2 and basedon a model temperature distributionover the planetary surface, amounts toabout -3°K. At low temperatures, thecorrection for off-axis radiation wasgreater for channel 1 than for channel2. A more refined analysis of the sys-tematic effects, based on data obtainedbefore near encounter and during pas-sage across the limb, is in process. Allsystematic errors that have been thoughtof, however, have the effect that theobserved temperatures are higher thanthe true value.The Mariner flights provided the op-

portunity to measure temperatures withan areal resolution approximately tentimes that obtainable from the earth. Al-though transitions between dark (maria)and light (desert) areas appear well de-fined in the data, there were no sharpchanges in temperature exceeding 1°Kin contiguous fields of view (50 km atclosest approach). The only sharp tem-perature fluctuation which does notseem to be associated with any featurein the "classical" maps of Mars was re-corded in both channels of Mariner 6at longitude 307.00 and latitude -3.5°,identified in Fig. 1. The TV pictures of3 OCTOBER 1969

this area show a complex terrain struc-ture that may be related to this tempera-ture anomaly (5).

Because the track of Mariner 6 wasnearly equatorial (3) and extended wellbeyond the terminator, it is particularlysuited for a determination of the pa-rameters characterizing the gross ther-mophysical properties of the martianground. The cooling curve of a homo-geneous solid with (Kpc)12 = 4 x 10-3cal/cm2 sec% per degree Kelvin, whereK is heat conductivity, p is density, cis specific heat, and albedo of 0.85, fitsthe measurements quite well, althoughobvious departures are noticeable inFig. 1. This result agrees with the gen-eral conclusions of Sinton and Strong

(6). Our results should be consideredtentative, although we believe that amore detailed analysis of the data willnot change the conclusions significantly.A thorough analysis of the systematiceffects present in the data is underway;a correlation with the results of thetelevision and infrared spectrometer ex-periments is planned.

G. NEUGEBAUERG. MUNCH

California Institute of Technology,Pasadena 91109

S. C. CHASE, JR.H. HATZENBELER

Santa Barbara Research Center,Goleta, California 93183

E. MINERD. SCHOFIELD

Jet Propulsion Laboratory,Pasadena, California 91103

References and Notes

1. S. C. Chase, Jr., Appl. Optics 8, 639 (1969).2. R. B. Leighton, N. H. Horowitz, B. C. Mur-

ray, R. P. Sharp, A. G. Herriman, A. T.Young, B. A. Smith, M. E. Davies, C. B.Leovy, Science 165, 787 (1969).

3. , ibid., p. 684.4. R. B. Leighton and B. C. Murray, ibid. 153,

136 (1966).5. We thank R. B. Leighton and the entire

Mariner 69 TV team for allowing us to seetheir pictures before publication and for theirvaluable comments.

6. W. M. Sinton and J. Strong, Astrophys J.131, 459 (1960).

7. Many persons within the Mariner 1969 Projectcontributed to the success of this experiment,but our special thanks are due to H. Schur-meier, J. Stallkamp, and C. Kohlhase for theirefforts on our behalf. We also thank J. Bennettand D. Griffith for help in the acquisition andreduction of the data.

9 September 1969

Laser Beam Directed at the Lunar Retro-Reflector Array:Observations of the First Returns

Abstract. On I August between 10:15 and 12:50 Universal Time, with theLick Observatory 120-inch (304-cm) telescope and a laser operating at 6943angstroms, return signals from an optical retro-reflector array placed on the moonby the Apollo 11 astronauts were successfully detected. After the return signalwas first detected it continued to appear with the expected time delay for theremainder of the night. The observed range is in excellent agreement with thepredicted ephemeris. Transmitting between 7 and 8 joules per pulse, we foundthat each return signal averaged more than one photoelectron. This is in goodagreement with calculations of the expected signal strength.

One of the scientific instrument pack-ages placed on the moon by the Apollo11 astronauts is an array of opticalretro-reflectors whose purpose is to per-mit short-pulse laser ranging from sta-tions on the earth (1). The retro-reflec-tor array, for which one of us (J.E.F.)provided the basic design, consists of100 fused silica corner cubes each 3.8cm in diameter mounted in an aluminum

panel 46 by 46 cm. The reflectors,which have a life expectancy in excessof 10 years, are designed to performunder essentially isothermal conditionsthroughout lunar night and most of thelunar day. The retro-reflector packageon the lunar surface eliminates thestretching in time of the return signalwhich would otherwise be produced bythe curvature and irregularity of the

99

on

June

6, 2

016

http

://sc

ienc

e.sc

ienc

emag

.org

/D

ownl

oade

d fr

om

tx

Channel number

.I-I 0 1 2 3 4 5

Time relative to predicted time (i sec)

Fig. 1. A histogram showing the growthof the retro-reflected signal in channelnumber 6.

moon's surface. As a result point-to-point measurements of the distance cannow be made with an accuracy ap-proaching 15 cm. Over a period ofyears, distance measurements to thisaccuracy will permit precise measure-ments of the lunar orbital motion, lu-nar librations, the lunar radius, fluctu-ations in the earth's rotation rate,Chandler wobble of the earth's axis, in-tercontinental drift rate, and a possiblesecular change of the gravitational con-stant.A number of stations have been es-

tablished in the United States for thepurpose of ranging with lasers to theretro-reflector array. We report here thefirst successful observation of a returnsignal which was made with the 304-cm telescope at Lick Observatory. This

III 21 314151 61 71 a I 91101111 121

5 shots

10 shots

O--

° 15 shots

20 shots

m

telescope was specifically instrumented,and telescope time was made availablefor the purpose of assisting NASA inacquiring reflected signals from thearray.

Lick's early acquisition is expectedto aid other stations in acquiring theretro-reflector signal since it eliminatesthe need for them to perform any ex-tended search in either position orrange. It also showed that the arrayhas successfully survived the returningmodule's blast-off; the fact that the sig-nal strength agrees with estimates basedon the overall efficiency of the systemconfirms both that the condition of thearray is satisfactory and that its opera-tion on the lunar surface is as expected.The ranging system at Lick consisted

of a giant-pulse, high-powered rubylaser operated at the ooude focus whichwas optically coupled through the 304-cm telescope and could be fired at 30-second intervals. The angular diameterof the outgoing beam was approxi-mately 2 seconds of arc and made aspot of light on the moon about 3.2 kmin diameter. The return signal was de-tected by a photomultiplier that wasmounted at the coude focus behind a10 arc-second field stop and a narrow(0.7 A) filter which were used to reducethe background illumination from thesunlit moon. A time-delay generator,initiated by the firing of the laser, wasused to activate the acquiring electron-ics some 21/2 seconds (the earth-moonround-trip time for light) later. The de-lay generator was set for each shot withan ephemeris provided by Dr. J. D.Mulholland.

Following the pulse produced by thedelay generator, the output pulses fromthe photomultiplier were channeledsequentially into 12 binary scalers witha dwell time for each scaler channel

i -414

IIooo0 11:00 12:00 13:00

Universal time

Fig. 2. A three-dimensional figure show-ing the time of the run (abscissa), therange window during which the equiDmentwas open for receiving data (ordinate),and the number of counts in each chan-nel (width of the bar). In each run inwhich returns were expected they wereseen in the correct channel. During runnumber 19 (3rd from right-hand end) thetelescope was pointed away from thereflector, and no returns were seen.

that was adjustable from 1/4 to 4 ,tsec.The routing of the pu-lses to the scalerswas such that a pulse arriving within0.1 ,usec of the end of a channel wouldals-o add a count to the following chan-nel. The contents of the scalers thencontained a quantized summary of thedetector output for a short time inter-val centered on the expected arrivaltime of the reflected signal. After eachcycling of the scalers that followed alaser firing, a small on-line computerread their contents and reset them. Thecomputer was used to store the ac-cumulated count for each of the scalers,and provided a printed output and acathode-ray tube display of the data.

Scattered sunlight from the lunar sur-face produced a random backgroundthat slowly filled the 12 time channels.Because the return from the retro-

Table 1. Log of observational data showing acquisition.

N Total counts in channel number No T Channelofd1sRno OfTie

width hannel1 2 3 4 5 6 7 8 9 10 11 12 shots (U.T.) (/sec) chanse)

10 12 8 16 18 12 14 10 17 13 27 12 12 20 1021 2.0 -2011 12 12 12 11 11 6 13 11 14 26 10 14 14 1032 2.0 -2012 Data invalidated by range-gate errors 16 2.0 -1013 13 8 8 12 7 18 11 5 6 7 8 12 13 1104 2.0 -1014 Data invalidated by range-gate errors 6 1.0 -1015* 4 3 3 5 4 17 6 8 10 5 6 8 18 1123 1.0 -516 1 1 2 2 6 3 3 1 2 1 3 2 10 1136 0.5 -117 6 3 4 2 11 9 2 7 2 4 2 5 16 1145 0.5 -118 3 1 3 5 3 19 3 3 4 1 0 4 22 1203 0.5 -119t 3 3 3 10 4 3 5 2 5 5 8 5 22 1219 0.5 -120t 2 1 1 0 3 4 6 2 4 2 2 4 10 1223 0.5 -121 5 2 2 3 2 1 12 11 3 4 5 2 22 1245 0.5 -1* Data from three shots with erroneous range gates deleted from tabulation. f Telescope pointed 16 km south of reflector. : Thin clouds notednear moon.

100 SCIENCE, VOL. 166

0

a>-l

1=

-

E

-2a

---i-- Xr--- 1tv-in

on

June

6, 2

016

http

://sc

ienc

e.sc

ienc

emag

.org

/D

ownl

oade

d fr

om

reflector occurred with a predetermineddelay, the channel corresponding intime to the arrival of the signal ac-cumulalted data at a faster rate than theother 11 channels. Figure 1 illustratesthis point by showing the way the dataaictually aIccu-mulated during one of theruns (number 18). Throughout this re-port original data is given; a number ofsmall corrections (instrumental andatmospheric) would need to be madehefore converting to an atctual range.On the night of 1 August a total of

169 shots were fired following acquisi-tion. Range-gate errors occurred on 27shots, and 22 shots were fired with thetelescope pointed away from the re-llector. For the remlaining 1 20 shotsapproximately 1 00 above-backgroundcounts were received. This represents areturn expectation in excess of 80 per-cent and shows that all parts of theexperiment were operating satisfacto-rily. Assumning a Poisson distr-ibLutionof the recordecl photoelectrons, we findthat this correspondls to an average of1.6 detectable photoelectrons per shot.1-his numlber is a lower limit to the trueaverage since interferenice effects aswell as guildinig errors can be expectedto have reduLced the nuLmber of returnsthat were recorded. The. strength ofthis sienral and the lack of "spill" intoadjacent channels clearly shows thatthe signal did not come from thentatural"l lunar sulrface, the, return

from which wouLld be distribuLted overabout 8 Atsec. The timing of the triggertfrom the delay g,enerator relative toMulholland's ephenmeris AXas changedthree times aind the channel widthswsere decreased from 2 to I and thenl'2 1-.sec. After each change, the signalappeared in the appropriate channel.Table I gives the data from ruLns 1 0through 21, the interval from the firstacquisition to the close of operation.(Nine ruLns containing 162 shots werenmade before acqulisition.) Column 1gives the ruLn number, columns 2throuLgh 13 the number of recordedcouLnts in each channel, colunmn 14 thetotal number of shots fired during therun, column 15 gives the tuniversal timefor the middle of the run, colUmn 16gives the channel width, and column1 7 gives the starting time of the I stchannel with respect to the ephemerispredictions. The channel in which thereturn was expected is printed in bold-face.

Figure 2 shows a plot of the datataken from Table 1. RuLns 1 2 and 1 4have not been plotted in Fig. 2 becauseerrors in setting the delay genierator3 OCTOBER 1969

0

a)CA--0z

0).-I

,Ea)

a)

10

9

8

7

6:5 t

44

3 -

WiW0- -2-5 -4 -3 -2 -1I

III1 2 3 4 5

Observed time relative to computed time (asec)

Fig. 3. The nulmber of coincidences in3 AuLgLst.

invalidated the timinig. During run 19the tclescope was not pointed at the re-flector and no returns were seen. Be-caLuse of a fortuitousi splitting of the re-turn between two channels on two ofthe runs, an effective timing precisionof 0.1 /isec was achieved. This is equiv-alent to a range c-ror of about 15 m.Clearly seen in Fig. 2 is an atppatrentdritft in the timie the retuLrns were de-tected relative to Mulholland's predic-tions. Although this drift was puzzlingat first. we soon re-al ized that it wascaused by the fact that the 304-cim tele-scope is located approximately 500 meast of the location given in The Amner-icani Eplhe eris anid Nalutical Alltnanlacfor Lick Observatory. A curve showingthis correction to the original ephemiierisis given in Fig. 2. Translation of thiscurve along the ordinate is allowable,and the aimount gives the difference be-tween the observed range and the pre-dicted range. It may be seen that theobservations agree well with the pre-dicted cuLrve Lising the correct coordi-nates.Two nights after the initial acquisi-

tion, a laser with a high repetition rateand a digital time interval measuringsystem was used to record the range tothe retro-reflector. The laser system oper-ated once every 3 seconds with a pulseduration of 60 nsec, full width at halfmalximum intensity. The detection sys-tem consisted of two photomultipliersarranged to detect photoelectron coin-cidenices within a reSsolution of 60 n.sec.The range meaisurement was made witha standard inter-val counter with a reso-lutioni of I nsec, started by a samiipleof the transmitted energy and stoppedby the Oultputt of the coincidence detec-tion system. A range-gating pulse was

100-nsec intervals observed on the night of

derived from a progriammed delay gen-erator.

Over a period of 1 hour iind 45 min-utes on the mor-ning of 3 August thelaser was fired 1230 times. The totalnumber of m1easured range times dulingthis period was 98. The low number ofcoincidences can be partially attributedto ai possible slight misalignment in thedetector optical system and a malfunc-tion of the delay pulse generator. Fur-thermore, since returns were not im-mediately recognized, an angle search,was instituted during which the signalimage was not centered in the coinci-dence area for relatively long times.The nullmber of observed coincidences

was consistent with that expected fromlunar surfalce reflections and r'andomnoise sources. However, retro-reflectorreturns should occur within 100 nsec ofthe true range. The observed nmeasure-ments were therefore comppared to thepredicted r'ange ephemeris and thenlinearly corrected because of the rangeparallax mentioned previously. Thetechnique involved taking the initialrange residuails and fitting to a first-order polynomial. The residuals fallingwithin a certcain rainge with respect tothe polynomial were then used in re-defining the polynomial until a signifi-cant number of points occurred withinthe expected precision of the system.After two repetitions, I I points wereleft with a root-mean-sqquare of 45nsec ahout the new polynomial. Thecoeflicients of this polynomial agreewith expectect devialtions from predictedrange due to the displacement of sitelocation.

Range residuals corrected to thisequation in time were then used to Coii-struct a histogram (Fig. 3) with 1 00

101

m -

on

June

6, 2

016

http

://sc

ienc

e.sc

ienc

emag

.org

/D

ownl

oade

d fr

om

nsec intervals from -5 to +5 ,usec. Nodata lying outside of ± 5 ,tsec wereplotted since no 100 nsec intervals out-side this range contained more than onepoint. The central 100-nsec intervalcontains nine points with an r.m.s. of35 nsec. Because the probability ofnoise or lunar surface returns occurringat these rates within a given 100 nsectime interval is so low, it can be con-cluded with a high degree of confidencethat the observed signals within this in-terval are indeed from the retro-reflec-tor package. The uncertainty of + 7.5m in the range obtained on the nightof 1 August can therefore be improvedto ± 5 m by the results obtained on thenight of 3 August.

JAMES FALLER, IRVIN WINERWesleyan University,Middletown, Connecticut 06457WALTER CARRION, THOMAS S. JOHNSON

PAUL SPADINGoddard Space Flight Center,Greenbelt, Maryland 20771LLOYD ROBINSON, E. JOSEPH WAMPLER

DONALD WIEBERLick Observatory,University of California,Santa Cruz 95060

References and Notes

1. C. 0. Alley, P. L. Bender, R. H. Dicke, J.E. Faller, P. A. Franken, H. H. Plotkin, D. T.Wilkinson, J. Geophys. Res. 70, 2267 (1965);C. 0. Alley and P. L. Bender, in ContinentalDrift, Sectular Motion of the Pole, and Rota-tion of the Earth, International AstrophysicalUnion, W. Markowitz and B. Guinot, Eds.(Reidel, Dordrecht, Holland, 1968); C. 0.Alley, P. L. Bender, D. G. Currie, R. H.Dicke, J. E. Faller, in The Application ofModem Physics to the Earth and PlanetaryInteriors, S. K. Runcorn, Ed. (Wiley-Inter-science. London, 1969)

2. We thank many people for the success of theLick operation. In particular we wish toacknowledge the efforts of N. Anderson ofthe Berkeley Space Science Laboratory; H.Adams, R. Greeby, N. Jern, T. Ricketts, andW. Stine of the Lick Observatory staff; B.Turnrose, S. Moody, T. Giuffrida, D. Plumb,and T. Stebbins of Wesleyan University; andJ. MacFarlane, B. Schaefer, R. Chabot, J. Hitt,and R. Anderson of the Goddard Space FlightCenter. The Lunar Ranging Experiment is theresponsibility of the LURE group whichconsists of C. 0. Alley of the University ofMaryland, P. L. Bender of the University ofColorado and the National Bureau of Stand-ards, R. H. Dicke of Princeton University,J. E. Faller of Wesleyan University, W. M.Kaula of the University of California at LosAngeles, G. J. F. MacDonald of the Univer-sity of California at Santa Barbara, J. D.Mulholland of the Jet Propulsion Laboratory,H. H. Plotkin of the Goddard Space FlightCenter, and D. T. Wilkinson of PrincetonUniversity. Supported from funds to Lickfrom NASA grant NASS-10752 and NSFgrant GP 6310, from funds to Wesleyan fromNASA Headquarters grant NGR-07-006-005,by in-house funds used by Goddard person-nel, and from funds to JPL from NASAcontract NAS-7-100.

2 September 1969

Age of the Bay of Biscay:Evidence from Seismic Profiles and Bottom Samples

Abstract. Paleomagnetic data and marine magnetic surveys suggest that theBay of Biscay was created by rifting due to an anticlockwise rotation of theIberian peninsula. The period during which movement occurred is not knownprecisely, but a rotation, amounting to 22 degrees, appears to have taken place inpost-Eocene time. To provide independent evidence on the age of the rift, bottomsamples from the Biscay abyssal plain have been related to the distribution ofseismic reflectors within the sediments. The investigation shows that a large part ofthe Bay of Biscay was in existence in late Cretaceous times and that the data canbe interpreted in terms of some early Tertiary rotation. However, the amount ofpossible Tertiary opening is appreciably less than the paleomagnetic results sug-gest. In view of the fact that reflectors of Middle-Upper Miocene age can betraced as undisturbed horizons across the bay, all tectonic movements must haveceased by the early Miocene.

The hypothesis, put forward byWegener (1) and others, that the Bay ofBiscay formed as a result of an anti-clockwise rotation of the Iberian penin-sula relative to the rest of Europe hasrecently received some support frompaleomagnetic studies (2-5) and fromthe fanlike pattern of linear magneticanomalies between the French andSpanish continental slopes (6). Girdler(2) found a consistent difference of 38°(+ 160) in the declinations in rocks ofPermian, Triassic, and early Jurassicage from north and south of the

102

Pyrenees, which can be accounted forby a 400 rotation of Iberia in lateMesozoic and Tertiary time about apoint near 44°N, 1°W. According tobathymetric charts, this same amountof movement is necessary to open upthe Bay of Biscay, if the 500-fathom(900-m) contour is taken as the bound-ary between ocean and continent. Thereis some area of overlap in fitting north-ern Spain against France in the easternpart of the bay, but this is probably dueto outbuilding -of the French continentalslope by thick accumulations of sedi-

ment deposited during the Tertiary (7).The paleomagnetic data of Van Don-

gen (3) and Van der Voo (4) based onmeasurements made in Spain indicatea similar amount of rotation. Theformer, working with rocks from theeastern Pyrenees, proposed a 300 rota-tion after Permian time, whereas thelatter, using observations from theMeseta, advocated a post-Triassicmovement. Watkins and Richardson(5) have examined some late Eocenevolcanic rocks near Lisbon and havesuggested that the Bay of Biscay waspartially open during the Eocene andthat a 22° rotation occurred sometimelater in the Tertiary.

In view of these conclusions, particu-larly those of Watkins and Richardsonwhich imply that the bay is of fairlyrecent origin, it seems appropriate toconsider some results of a study of thestratigraphy of the sediments of theabyssal plain which forms the floor ofmuch of the Bay of Biscay. We haveexamined the relation of some cores anddredge samples to several hundredkilometers of seismic reflection profiles,recorded with an air-gun profiler (8),in order to determine the age of variousparts of the abyssal plain sediments andthus to provide some independentevidence on the age of the Bay of Bis-cay. The positions of the seismic linesare indicated in Fig. 1. Figure 2 showsthe locations and ages of bottom sam-ples, both published (9-11) and unpub-lished (12, 13), from the plain andcontinental slope.The reflection records indicate that

the sediments of the abyssal plain arewell stratified and that they are prob-ably composed of many turbiditelayers interbedded with finer-grainedmaterial, an assumption strongly sup-ported by the core data and by thepronounced leveling effect of sedimenta-tion close to small seamounts (Fig. 3).We have divided the sequence intothree units which can be recognizedover a wide area. The uppermost sec-tion (the upper turbidites) is about 300m thick and is made up of many closelyspaced reflectors. The middle "homo-geneous zone" is approximately 200 mthick and generally lacks strong reflec-tors in the frequency range in whichthe recordings were made (60 to 150cycles per second); this suggests thatthe sediments are more lithologicallyuniform than those above and below.The lowest formation (the lower turbi-dites) is primarily composed of severalclosely spaced strong reflectors.A reflector marking the top of a suc-

cession of stratified sediments clearlySCIENCE. VOL. 166

on

June

6, 2

016

http

://sc

ienc

e.sc

ienc

emag

.org

/D

ownl

oade

d fr

om

10.1126/science.166.3901.99] (3901), 99-102. [doi:166Science 

and Donald Wieber (October 3, 1969) Johnson, Paul Spadin, Lloyd Robinson, E. Joseph Wampler James Faller, Irvin Winer, Walter Carrion, Thomas S.Observations of the First ReturnsLaser Beam Directed at the Lunar Retro-Reflector Array:

 Editor's Summary

   

This copy is for your personal, non-commercial use only.

Article Tools

http://science.sciencemag.org/content/166/3901/99and article tools: Visit the online version of this article to access the personalization

Permissionshttp://www.sciencemag.org/about/permissions.dtlObtain information about reproducing this article:

AAAS. is a registered trademark ofSciencethe Advancement of Science; all rights reserved. The title

York Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association forweek in December, by the American Association for the Advancement of Science, 1200 New

(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the lastScience

on

June

6, 2

016

http

://sc

ienc

e.sc

ienc

emag

.org

/D

ownl

oade

d fr

om