equivalent d-c drift is 5 mv/hr (maxi-mum) and the temperature coefficientof d-c output voltage is 1 mv per de-gree Celsius. A negative capacitance cir-cuit is provided which improves theoverall frequency response of the ampli-fier but unavoidably adds noise tothe input signal if large stray capaci-tances are compensated. Signals of 40pv through 7 Mohm (microelectrode insaline) can be faithfully reproduced.

The amplifier has been tested in phys-iological experiments in which micro-electrodes (0.5 to 3 IL) were used torecord resting and action potentials(extracellular) from neurons in frogspinal cord, chicken cerebellum, isolatedmonkey brain (2), and nerve fibers invitro. The IGFET amplifier gave con-

sistently better overall results than a va-

riety of commercial electrometer ampli-fiers (vacuum tube and transistor types)that were used to make the same meas-

urements.Many other applications for this de-

vice and simple associated circuits havealready been found. I believe the un-

usual features of this amplifier make itsuitable for routine use in many typesof research; it seems to answer theneeds of large laboratory instructionclasses where 10 to 20 amplifiers are

needed and the cost of commercial de-vices is prohibitive.

RONALD L. CECHNERCase Institute of Technology,Cleveland, Ohio 44106

References

1. H. Borkan and P. K. Weimer, RCA Rev. 24,153 (1963); H. Fein, Inst. Electron. Elec. Eng.Trans. Blo-Med. Electron. 11, 13 (1964).

2. R. J. White, M. S. Albin, J. Verdura, Science141, 1060 (1963).

25 July 1966

"Neutral Hydrogen Survey ofAndromeda Galaxy": Addendum

Edward Argyle (1) has called our

attention to the fact that van de Hulst,Raimond, and van Woerden (2) postu-lated the existence of a neutral hydro-gen ring in M31 in their article pub-lished in 1957. We regret that we (3)did not point this out in our report.Their inference, however, was based on

peaks found along only a single axisthrough the ring. Roberts' survey (4)was the first to be made, in a nearly

continuous manner over much of thegalaxy with sufficiently high resolution,to demonstrate clearly this ring struc-

1550

ture and the marked deficiency of neu-

tral hydrogen inside. Our survey (3)confirmed this result and was the firstto delineate the ring completely aroundthe nucleus.

W. D. BRUNDAGEJ. D. KRAUS

Ohio State UniversityRadio Observatory, Columbuis

References

1. E Argyle, Dominion Radio Astrophysical Ob-servatory, Penticton, B.C.

2. H. C. van de Hulst, E. Raimond, H. vanWoerden, Bull. Astroni. Inst. Neth. 14, 1 (1957).

3. W. D. Brundage and J. D. Krause, Science 153,411 (1966).

4. M. S. Roberts, Astrophys. J. 144, 639 (1966).

22 AuLgtust 1966

Phyletic Position of Tree Shrews

Although much of the recent evi-dence on the taxonomic position ofthe tree shrew argues strongly againstincluding the Tupaioidea in the Lemuri-formes, the new findings still fail toestablish whether or not the treeshrews are closer phyletically to Pri-mates than to any other extant mam-

malian group. If the Tupaioidea di-verged from the Primates very early inthe evolution of the Primates, the tu-paiids would show hardly any more af-finities to the recent groups in the restof the Primates than to the extant mem-bers of any other mammalian order.The data reviewed by Campbell (1)emphasize the distinctiveness of thetupaiids rather than their relatednessto either primate or insectivore types.

Campbell suggests that the extensivevisual system in tree shrews and pri-mates may have resulted from con-

vergent evolution. It is of interest thatserological studies on primate lens pro-teins reveal pronounced affinitiesamong lorisoid, tarsier, and higher pri-mate lenses. However, these studiesdemonstrate divergence of tree shrewlens proteins from those of primatescomparable to that between lens pro-teins of non-primates and primates(2).The serological data on serum pro-

teins, gathered since the Burg Warten-stein conference on Classification andHuman Evolution, further emphasizethe distinctiveness of the tupaiids. Anti-serums produced in rabbits to hedge-hog serum and to tree shrew serum,

while yielding strong -homologous re-

actions, yield very weak reciprocalcross-reactions and unlike the chicken

antiserums fail to detect any specialcorrespondence between tupaiids anderinaceoids. Indeed, the precipitins toalbumin in the antiserum to tree shrewserum develop larger cross-reactionswith human albumin than with hedge-hog albumin and the other nonprimatealbumins tested (3). Thus the originaldata (4) obtained with chicken andrabbit antiserums to human albumin,and confirmed by Hafleigh and Wil-liams (5) suggesting that Tupaia hasserum albumin more like that of pri-mates than insectivores, is now direct-ly demonstrated.The data of Dr. B. H. Hoyer, cited

by Campbell, on the homologies ofpolynucleotide sequences as judged bycompetition of various primate andnonprimate DNA fragments withthose of humans are compatible withthe possibility that the Tu.paioideabranched off from the base of thePrimates. If they did, they should notbe expected to show much more re-latedness to man than would the non-primate mammals. Thus the value of28 percent competition for tree shrewDNA compared to 20 percent for non-primate mammalian DNA's mightprove to be highly significant.

MORRIS GOODMANWayne State University,School of Medicine, Detroit, Michigan

References and Notes

1. C. B. G. Campbell, Science 153, 436 (1966).2. H. Maisel, in Protides of the Biological Fluids,

H. Peeters, Ed. (Elsevier, Amsterdam, 1965),vol. 12, p. 146; H. Maisel and M. Goodman,Amer. J. Opthalmol. 59, 697 (1965).

3. Unpublished results.4. M. Goodman, Human Biol. 34, 104 (1962);

, Ann. N.Y. Acad. Sci. 102, 219 (1962).5. A. S. Hafleigh and C. A. Williams, Jr., Sci-

ence 151, 1530 (1966).29 July 1966

Surveyor I Location

By attempting to correlate the posi-tions of summits of lunar hills, situatedbeyond the horizon of Surveyor I, withfeatures given on the AeronauticalChart and Information Center map ofthe area, Jaffe et al. (1) derive a loca-

Table 1. Surveyor landing sites.

Site

SouthDerived from latitude

(deg)

Photo correlation 2.57 0.02Tracking data 2.49

Westlongitude

(deg)

43.34 0.0243.32

SCIENCE, VOL. 153

on Septem

ber 8, 2018

http://science.sciencemag.org/

Dow

nloaded from

Table 2. Heights (above the plain) of hillsfrom shadow measurements. Feature B cannotbe measured because shadow falls on hillsD and d.

Shadow Solar HeightFeature length altitude (in)

(kcm) (deg)

A 5.4 3.25 300a 3.0 3.10 160C 5.0 2.75 230*D 7.5 2.65 330td 6.6 2.60 290tF 4.5 3.00 230

* Underestimate, because end of shadow fallson hill D. t Overestimate, because end ofshadow falls in large, somewhat depressedarea.

Table 3. Heights of hills from Surveyor data.Feature d is invisible to Surveyor.

Featuredistancefrom

Surveyor(km)

A, 23.4a, 24.8B, 31.3C, 29.8D, 30.9F, 30.0

Height

Seen Belowabove horizon Totalhorizon (calc.) (m)(m) (m)

180 130 310< 0 150 150130 240 37040 220 26080 230 31020 220 240

tion (site I) situated well outside the2cr uncertainty ellipse based upon thetracking data. Furthermore, the correla-tion is only partial. By repeating theprocess with a suitable Earth-basedphotograph, I find that only one loca-tion of Surveyor is possible, well withinthe tracking-data ellipse.

Figure 1 depicts the NE portion ofthe large, incomplete ring Flamsteed P(2); it was made from two stackednegatives taken with my laboratory'sNASA-sponsored 61-inch (153-cm) re-

flecting telescope at 0315 hours U.T.,2 April 1966. The lines of latitudeand longitude were carefully transferredfrom (3). The radial lines represent thedirections of horizon features A-F (1,fig. 16), the azimuths having been ad-justed for the computed inclination ofthe lunar surface from the plane per-pendicular to the line of sight. Thesmall dot indicates the location of Sur-veyor for optimum correlation betweenthese lines and the various hills, whilethe ellipse represents the theoreticalhorizon as seen from Surveyor's camera.

Table 1 gives the coordinates of the

Fig. 1. Northeast portion of Flamsteed P, showing Surveyor location derived fromhorizon features; 10 z 30 km.

23 SEPTEMBER 1966

landing site derived from the precedingcorrelation and from the tracking data;the former is approximately 2.4 kmsouth of the latter, well within the 2cruncertainty ellipse.

In order to verify the correctness ofthe correlations, the heights of severalhills in the group were obtained fromshadow measurements made on a printsimilar to Fig. 1 (Table 2). Thesevalues may be compared with thosecalculated from the angular dimensionsgiven in (1, table 2) and the assumedposition of Surveyor (Table 3). Theagreement is remarkably good in viewof the uncertainties of the shadow meas-urements; differences do not exceed 10m except where the shadows are caston rising or falling terrain. Hill d ap-pears larger than D in Fig. 1, but D ishigher since it casts a longer shadowand thus occults d in the Surveyor view.Feature E is not identified; the walls ofthe small crater situated at the locationindicated are well below Surveyor'shorizon, so this feature is presumablya small object situated relatively closeby. The summit of a must be almostexactly at horizon level.

EWEN A. WHITAKERLunar and Planetary Laboratory,University of Arizona, Tucson

References

1. L. D. Jaffe et al., Science 152, 1737 (1966).2. D. W. G. Arthur et al., Commun. Lunar

Planetary Lab. 3, 62 (1965).3. D. W. G. Arthur and E. A. Whitaker, in

Orthographic Atlas of the Moon, G. P. Kuiper,Ed. (Univ. of Arizona Press, Tucson, 1961),sheet E5.

11 July 1966

Toxic Impurities in NalgeneFilter Removed

I should like to comment on the im-purities derived from the Nalgene FilterUnit which Simpson (1) has found tobe inhibitory to Leishmania tarentolae.

There is little likelihood that thereported effect was derived from somematerials extracted from the plasticbody of the Filter Unit. The plasticis the best grade of polystyrene; it is thekind commonly used in manufacturingdisposable syringes and other medicalitems and has been shown repeatedlyto be nontoxic. Nalgene, incidentally, isnot the name of the plastic, but is theregistered trademark of our laboratoryproducts.The inhibitory effect that was indeed

present has been traced to the residueof a sizing agent in some of the polyole-

1551

on Septem

ber 8, 2018

http://science.sciencemag.org/

Dow

nloaded from

Surveyor I LocationEwen A. Whitaker

DOI: 10.1126/science.153.3743.1550-b (3743), 1550-1551.153Science 

ARTICLE TOOLS http://science.sciencemag.org/content/153/3743/1550.3.citation

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceGovernment Works. The title reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S.Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights

(print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for theScience

on Septem

ber 8, 2018

http://science.sciencemag.org/

Dow

nloaded from