Document e7k5wJYznvedEpRLzd65yEvqE

morphological pattern as assessed by the p measurements taken, SK 84 is ex cluded with 95 percent probability from membership in any group, including Homo sapiens. The closest approach by lire fossil to a centroid in tire discrim inant space occurs in the case of Pan, ihe sample to which assignment is sub sequently made and in fact the only croup for which membership is not ex cluded with 99 percent probability. (Chi-squares relating the fossil to the chimpanzee centroid are 8.36 and 10.12 with three degrees of freedom, for distal lu-inht measurements including and omitting the beak, respectively.) This docs not, of course, prove that Parauthropus at Swarlkrans possessed a thumb (or hand) precisely like that of a modern chimpanzee. But a degree of similarity is certainly implied, with re spect to measurable morphological struc ture and presumably in mailers of func tion as well. The chimpanzee metacar pal and thumb arc short relative to the other digits, and there is sonic limita tion on movement at the carpometacar pal joint imposed by the surrounding '`cutV" of thenar musculature (<5). Thus, despite the potential for mobility at the A'cll-dcvclopcd saddle joint, the animal s seldom, if ever, observed to oppose lie pulp of the thumb to the pulp sur'ace of one or all of the remaining fin:ers, A true "precision grip" is not perormed, although various imperfect apouches to this grip arc possible (S, 9). 'he degree of thenar muscle develop ment, the morphology of the articular apsulc itself (whether loose or closely onslructed), mid the relative lengths of ic pollex and other digits are unknown Dr Parauthropus, hut in overall length ud articular surface configuration, and i other characteristics emphasized in ic discriminant analysis, the fossil 'Clacarpal approximates the chimpante condition and pray thus have fitted hand with similar functional limita- DI1S. In any case, despite its robust nature id tentative association with specimens Homo on functions 1 and 2, the earlkrans bone is far from fully modll in its morphology. Further study of ch materials, including the judicious plication of multivariate statistics, is cessary in order to detail the locoitor and manipulative capabilities of ranthropus and early Homo. G. P. Rightmire partment oj Anthropology, ite University of New York, ighamton 13901 References and Notes J. K. Broom and J. T.# Robinson, Nature 164, K41 (J949;. * 2. J. R, Napier, Fowl Mammals Ajr. 17, 1 (1959). 3. W. E. l.c Gros Chile. Matt*Apts or Ape Men? (Holt, Kincharl & Winston, New York. 1967). 4. J. R. Napier, Nature 196, 409 (1962). 5. B. Patterson and W. \V. Howells, Science 156, 64 (1967); M. II. Day, Nature 215, 323 (1967); -------- and B. A, Wood, Man 3, 4-40 (1968). 6. W. W. Cooley and V. R. Lobncs. Matrix variate Procedures for the Behavioral Sciences (Wiley, New York, 1962). 7. For technical seasons, the small gibbon and baboon samples were treated as un&rouped subjects not included in the actual compula- lion of ihe discriminant uciyhis, Gibbon and baboon centroids were iluis not obtained, although these individuals can be rcfaiec! to the centroids of the other "original'' groups in \hc analysis. 8. R. H. Joule, in 7Ac Chi/npanzec, /'Av.wWory, Behaviour, Seiotusy and Diseu^et's o/ CAioi- panzecs, G. H. Bourne, Ed. (Kurgcr, Basel, 1970), vol. 2, p. 167. 9. J. K, Napier, I'ioc. ZooJ. Soc. London 134, 647 (i960); Sytnp. Zool. Soc, London 5, 115 (1961); R. H. Tuulc, J. Morphol. 128 309 (1909). 10. I thank Dr. C. K. Brain for permission to examine the Swarlkrans metacarpal and Prof. W. \V. Howclb for the loan of indispensable measuring equipment. Project support uas provided by the Research Foundation of \Ua Stale University of New York. 6 December 1971 B Enrichment of Heavy Metals and Organic Compounds in the Surface Microiayer of Narragansett Hay, Rhode Island Abstract. Concentrations of lead, iron, nickel, copper, fatty acids, hydrocarbons, and chlorinated hydrocarbons are enriched from 1.5 to 50 times in the top 100 to 150 micrometers of Narragansett Bay water relative to the bulk water 20 centi meters below the surface. Trace metal enrichment wui observed in the particulate and organic fractions but not in the inorganic fraction. If these substances arc concentrated in films only a few molecular layers thick on the water surface, the actual enrichment factor in the films may be well over JO'1, resulting in extremely high localized pollutant concentrations in the surface microiayer. From a pollution standpoint, the airsea interface is perhaps one of the most important but most poorly characterized regions of the marine environment. The surface microiayer of the ocean has many unique chemical and physical properties, most of which are little un derstood. The organic and trace metal chemistry of this microlayer has re ceived scant attention. The few studies of this region U) indicate the presence of a variety of surface-active substances (for example, fatly acids and fatly al cohols) at the interface. These materials often form a coherent film which be comes evident as a result of Ihe local ized damping of capillary waves. F.vcn when this film or slick is not visible, the surface microiayer may still be en riched with surface-active compounds. The major source of these compounds in the open ocean is the reservoir of natural marine organic matter in the mixed layer. The process of slick for mation is comp'ex, but surfactants are probably concentrated in the surface microiayer by convection currents, ris ing bubbles, and diffusion. The most surface-active of these materials dis place the Jess active compounds, and eventually the microiayer is enriched with lipoid material , and a coherent slick is formed. Many pollutants may be concentrated and stabilized in 1 his layer after its for- mtition or may in fact be incorporated into it directly. This is especially true for lipophilic pollutants such as chlo rinated hydrocarbons (2) and petroleum hydrocarbons. Trace metals may also be concentrated in the microiayer. Or ganic acids, proteinaceous material, and other surface-active organic substances may provide complexing sites for many heavy metals and thus be responsible for the transportation and concentra tion of these metals at the water sur face. The sources of the various pollutants that can reach and eventually concen trate in the surface microiayer are nu merous, They include atmospheric trans port, rivers, sewage and industrial effluents, dumping, pumping, and spills. Once concentrated, pollutants arc read ily accessible to bacteria and other mi croorganisms as well as phytoplankton and zooplankton at the surface. In this way, pollutants initially in the surface microiayer can enter the food chain and eventually be concentrated in the higher trophic members of the marine com munity. Since the chemical nature of the sur face microiayer is poorly understood, both in terms of its natural composition as well as in terms of its pollution gradients, it would seem appropriate to characterize more fully this region of the marine environment. This report de- APRJL 1972 161 osw 025293 STLCOPCB4009248 r\ l': v' l} b. r ij \ l! f. 1 , 1 Table 1. Concentrations and enrichment factors of organic compounds and metals in surface microbyer samples from Narragansett Bay, Rhode Island, Sample f Sample 2 Substance Concentration (^g/liter) Surface Subsurface Enrichment* factor Concentration (pg/liter) Su rfnee Subsurface Enrichment* factor Fatty acids Hydrocarbons PCB's! Lead Particulate Organic Inorganic Iron particulate Organic Inorganic Copper Particulate Organic Inorganic Nickel Particulate Organic Inorganic 128 26 NAt 4.2 1.0 1.4 1.1 1.0 0.8 1.7 0.3 820 47 3.7 1.5 2.8 0.4 7.2 2.3 5.6 0.5 3.4 0.4 11 3.0 4.9 2.6 U 4.0 36 7 NAt 0.15 0.04 0.24 0.17 0.36 0.06 2.7 0.5 28 4 0.60 0.33 1.4 0.2 0.20 0.09 0.19 0.11 3.3 0.3 0.2 0.1 0.48 0.33 14 1.0 3.6 1.0 NAt 28 10 5.8 6.1 2.7 2.2 0.6 0.2 29 5 6.2 4.3 2.0 0.3 36 18 29 17 1.0 0.2 50 30 10 8.7 0.8 0.3 94 . 19 8.5 1.7 0.45 0.11 1.5 0.2 1.4 0.6 6,1 -t- 1.4 35 Hh 7 5.1 : 2.3 17 -f* 8 1.3 0.4 1.6 1.0 1.5 i: 0.6 13 5 5.0 --H 0.6 21 H- 5 * The emichmem factor is equal U> the surface concentration divided by the subsurface concentration. limited sample. f PCB's esiuesscd ns Aroclor J254. 62 12 5.9 1.2 0.05 1.5 0.4 1.4 0.4 9 0.28 0.10 0.27 0.12 3.7 1.0 5.4 2.0 5.2 3.1 1.6 0.6 8.2 1.2 3.8 0.8 12 1 4.3 i.i 1.3 0.6 1.4 0.7 0.26 0.11 0.11 0.04 1.3 .0,4 5.0 2.7 15 11 1.2 0.6 2.1 0.3 1.8 0.7 16 2 6.2 i: 2.5 2.8 1.1 1.3 0.3 t NA, hydrocarbons not delected because of DSW 025294 scribes our efforts to determine the dis tribution of fatly acids, hydrocarbons, chlorinated hydrocarbons, and trace metals in samples from Narragansett Bay, Rhode Island. Several surface microlayer samples were collected near the mouth of the west passage of Narragansett Bay (41 30'20"N, 7123'30"W; 4134'30"N, 7123'\V). The area is designated class SA (Rhode Island Department of Health) and is free of industrial and municipal effluents and major ship traf fic. Samples for trace metal analysis were collected on a polyethylene screen (75 by 75 cm, 20 mesh) mounted on a Plexiglas frame. After the screen had been rinsed in seawater several times, it was submerged in the water and then passed back up through the water sur face with the screen approximately par allel to the water surface. The frame was allowed to drain for 10 seconds, and then the screen was drained for 60 seconds into a polyethylene container through a polyethylene funnel. This col lector samples the top 100 to 150 /^m of the water surface (3). Polyethylene gloves were worn throughout the collec tion, and great care was taken to ensure that the sample was not contaminated by the collection boat (an aluminum rowboat) or the individuals doing the collecting. Samples for organic analysis were collected on a 16-mesh circular stain less steel screen, 30 cm in diameter, in a manner similar to that used for trace metal sampling. The screen was drained into a Teflon bottle through a glass funnel. Subsurface samples were col lected by submerging a Teflon bottle approximately 20 cm below the water surface and then removing the cap (polyethylene gloves were worn during this collection). Sample collection was limited to periods when wind conditions were less than 12 to 15 knots (22 to 28 km/hour) and wave heights were less than 3 feet (91 cm). The collection time was approximately I Vz to 2 hours per station. Surface microlayer samples for organic analysis were acidified and ex tracted with chloroform after the addi tion of fatty acid and hydrocarbon in ternal standards (4). The chloroform extract was evaporated under reduced pressure, dissolved in a mixture of ben zene and methanol (1 : 1), and divided into two equal portions for lipid and chlorinated hydrocarbon analysis. The sample for lipid analysis was saponified and then methylated to con vert total fatty acids (free and esterified) into methyl esters (5). The methyl esters and hydrocarbons were isolated and purified by preparative thin-layer chro matography. Qualitative and quantita tive analyses were carried out with a gas chromatograph (Hewlett-Packard model 700) equipped with a flame ion ization detector. The samples were ana lyzed on polar (diethylene glycol suc cinate or free fatty acid phase) and nonpolar (Apiezon L) columns, and relative retention times were compared to those of authentic fatty acid and nalkanc hydrocarbon standards. In addi tion, the fatty acids were analyzed on these columns after hydrogenation. Quantitative results were obtained by comparing the areas of naturally occur ring component peaks to the area of the internal standards. The results obtained on standard mixtures of fatty acids and n-nlkanes were in agreement with the stated composition data with a relative error of less than 5 and 8 percent, re spectively, for major components (> 10 percent of the total mixture). The sample used for chlorinated hy drocarbon analysis was separated into a polychlorinated biphenyl (PCB) and a pesticide fraction on a silicic acid column. The fractions were analyzed on a gas chromatograph (Tracer model 220) equipped with two K,Ni electroncapture detectors and two electrometers. The samples were analyzed on two dif ferent nonpolar columns (OV-17/QF-1 and SE-30/QG-1), and the relative re tention times were compared to those in authentic Aroclor ami pesticide stan dards. Surface microlayer samples for trace metal analysis were filtered through Millipore HA filters (47 mm, 0.45 /.mi) science, VOL. 175 STLCOPCB4009249 , . . ,n in-line dosed filtering system con ' .'^-:cd entirely of polyethylene and " i'wopylene. 'I'lic collected particulate 1trial appeared to lie largely bio- The "particulate" sample and fil' J.','\-cie dissolved in a mixture of per- vgyi: and nitric acid for subsequent ' absorption (AA) analysis. The ' nra'c from !'1C Parl'cu'a,c sample was * graded with chloroform. T he chloro- m, .siting with any interfacial mnte jiji, was filtered through a Milliporc i n/i finer (2d mm, 0.45 pm) to remove initrfacial material. The chloroform oir.iet was conccntiHied in a rotary s rvsporator, and the final few milliliters ` vc:c i.vnsfcrred to the 24-inm filter and evaporated. The "organic" phase on the ' f.f.cr was then processed in the same , , manner as the "particulate" sample for ' AA analysis. After filtration and chloroie.-m extraction, llie remaining aqueous , nor;io;i of the sample was acidified and tnifiered to a pH of 3.0. The solution vet then extracted with a 5 percent (by v.debsolution of dielhyldithiocarbami; arid into methyl isobutyl ketone. This 'inorganic'' pliasc was analyzed directly hr AA. All AA analyses were per.'ornicd with an atomic absorption specmykoiomclcr (Perkin-Elmcr model K'.h w,':h standards prepared in a scav.a:er matrix. We have collected several surface mi . relate; samples in Narragansett Bay and analyzed them for trace metals, lipids, and chlorinated hydrocarbons. Trie results of the analysis of two of these surface slicks and subsurface wa ters aic piesentcd in Table 1. Sample 1 was collected in a heavy frothy surface Pick and showed enrichment factors in t.V surface microlaycr relative to those in the subsurface water ranging from 3 to 50 for fatty acids, TCB's, and "part'cukuc" amt "organic" trace metals. Sample 2, from a less pronounced slick, tmd lower enrichment factors varying from slightly over 1 to 15. With the ex ception of iron in sample I, no enrich ment was found for (race metals in the "inuiganie'' phase for these or other surface microlaycr samples collected in it rag* n\cll Bay. In general, (ho fatty xcids were normal saturated and rnonoinsiUiiraied ncids and ranged from 12 to -b ctiihons in chain length. Only three !l.w':oef.rbons were detected in these w;np!e>. and they were tentatively `-'cmjfied as having carbon numbers of ;i C,..and Cjjj.d relative to n`Ttmes o:i an Apiezon J. column. Tile n,.v chloiinafed hydrocarbons found h Al'Kn. 1S72 were PCB's which were measured as Atoclor 1254. An enrichment factor given in Table 1 for any of these substances in the lop 100 to 150 pm of the water surface suggests a much greater enrichment in the film material itself. If the film layer is monomolccular, as has been sug gested (}), it should have a thickness of about 2 X 10~3 pm. If the film thick ness is estimated conservatively as five molecular layers (10-2 pm) and all the chemical enrichment is in this layer, the actual concentration in the film would be J.5 X 104 times the concentration in the lop 150 pm. In sample I, where the enrichment of PCB's in the top 100 to 150 pm is 28, the PCB concentration would be ~ 60 parts per million, which would represent an enrichment of 4 X 104 in the film layer itself. The effects of the high concentrations of chlorin ated hydrocarbons, lipids, and heavy metals on the diversity and species composition of the bacleria, phyto plankton, and zooplankton living in the surface microlaycr arc unknown. Pollutants present in the surface microiaycr of the coastal zone may easily be introduced into the atmosphere for subsequent transport to open ocean wa ters. The surface of the ocean is a ma jor source of atmospheric particulate matter. The primary production mech anism for these particles is the breaking of bubbles, which ejects particles both from a central jet and from llie rup tured bubble film cap (6). There is evi- deuce that these particles aie chemically more representative of the surface mi crolaycr than the bulk water underneath (7), and thus this study suggests that the particles may be considerably enriched in pollutants if they arc generated in highly polluted nearshore areas, Roiilrt A. Duct-: James G. Quinn Graduate School oj Oceanography, University of Rhode Island, Kingston Charles E. Oi.ney Department oj Food and Resource Chemistry, University of Rhode Island Stephen R. Piotkowicz Bahhara J, Ray, Terry L. Wade Graduate School oj Oceanography, University of Rhode Island References and Notes 1. W. D. Garrett, Deep-Sea Res. 14, 221 (1967); . P. M. Williams, ibid., p. 791. 2. 1), D. Seba and ii. F. Corcoran, Pesiic. Motiif. /. 3, 190 (1969). 3. W. D, Garrett. Liiuttol. Occauogr. 10 602 (J9G5). 4. The fatty jicid used whs n-hcptadccanolc acid (Ck, no double fronds). The hytirocaibon used was n-docobane (Cej.o). 5. L. D, Metcalfe. A. A. Schmilz, J. R, Pclka, Anal. Chan.. 38. 514 (1966). 6. D, C. Blanchaid, Progr, Qcetmogr, J, 71 (1963). 7. F. MacIntyre end J. W. Winchester, J. Phys, C'ftc/iu 7;.. 2163 (1969): F. Y. B. Seto and R. Ducc, J. Geophys. Res., in press. 8. This work was supported in part by ihe Oceanography Section, National Science Foun dation, tinder NSF yiam GA-20000 and by the OHice of the International Decade of Ocean Exploration, National Science Founds* tion, under NSF grant GX-2KJ4Q. S.R.P. is supported by Environmental Protection Agency iiaininy grant WP 252, 27 December 1971; revised 22 February 1972 n Cloud Seeding Experiments: Lack of Bias in Florida Series Abstract. There has been concern about the possibility oj selection bias in cloud seeding experiments. Covariates and experimental design have been used to obtain an estimate oj this bias. The results indicate that there was no selec tion bias in the Caribbean and Florida series oj cloud seeding experiments. Stigler (1) has made the general point that selection bias may be in troduced when suitable experimental subjects arrive sequentially; lie referred in particular to the cloud seeding experi ments in Florida. We present evidence here that such a bias is not detectable in the experiments over the Caribbean Sea (2) in 1965 and over Florida (2?) in 1968 and 1970. These references should be consulted for a description of the experimental details. In addition, we wish to make a suggestion about what to do in some experimental situations where selection bias lias not been (or cannot be) eliminated by some appro priate design. In planning the Caribbean' cumulus experiments, careful attention was given to the design. This included consulta tions with experts in experimental -Vsign, one of whom, W. J. Youden (now deceased), picpared the randomization scheme. I hose experiments represent the first time that a numerical model of cumulus dynamics was used to tnake predictions of cloud growlh and to provide a measure of "secdability" of selected clouds. The power of the ex periments was greatly increased by in- 163 DSW 025295 i r7*>****o _ _ STLCOPCB4009250