known. The framework modes and their sources suggest depo-sition in a foreland basin but do not preclude accumulation in apull-apart basin. Resolution of basin type is hindered becausesequences are fragmentary, folded, and sheared, thus obscur-ing stratigraphic relationships and thicknesses. In addition,basin type has not been clarified by lithofacies analysis becausethe lithofacies imply development of alluvial fans and fan deltasin lacustrine and marine environments (Rees and Rowell inpress) and all such associations are common to both basin types.

This work was supported in part by National Science Founda-tion grant DPP 85-18157 to the University of Nevada Las Vegas.

References

Burgess, C.J., and W. Lammerink. 1979. Geology of the ShackletonLimestone (Cambrian) in the Byrd Glacier area. New Zealand AntarcticRecord, 2, 12-16.

Dickinson, W. R., and C. A. Suczek. 1979. Plate tectonics and sandstonecompositions. American Association of Pet roleu in Geologists Bulletin, 63,2164-2182.

Dickinson, W.R., L.S. Beard, G.R. Brakenridge, J.L. Erjavec, R.C. Fer-guson, K.F. Inman, R.A. Knepp, F.A. Lindberg, and P.T. Ryberg.1983. Provenance of North American Phanerozoic sandstones inrelation to tectonic setting. Geological Society of America Bulletin, 94,222-235.

Rees, M.N., and A.J. Rowell. 1987. The pre-Devonian Paleozoic Doug-las Conglomerate of the Transantarctic Mountains: Origin and depo-

sitional setting. Fifth International Syinposiu in on Antarctic Earth SciencesAbstracts.

Rees, MN., and A.J. Rowell. In press. The pre-Devonian Paleozoicclastics of the central Transantarctic Mountains: Stratigraphy anddepositional settings. Volume of the Fifth International Symposium onAntarctic Earth Sciences. Cambridge, U.K.

Rees, MN., A.J. Rowell, B.R. Pratt, and P. Braddock. 1985. The ByrdGroup of the Holyoake Range, central Transantarctic Mountains.Antarctic Journal of the U.S., 20(5), 3-5.

Rowell, A.J., and M.N. Rees. In press. Setting and significance of theShackleton Limestone, central Transantarctic Mountains. Volume ofthe Fifth International Symposium on Antarctic Earth Sciences.Cambridge, U.K.

Rowell, A.J., M.N. Rees, and P. Braddock. 1986. Pre-Devonian Pal-eozoic rocks of the central Transantarctic Mountains. Antarctic Journalof the U.S., 21(5), 48-50.

Rowell, A.J., M.N. Rees, R.A. Cooper, and B.R. Pratt. In press. EarlyPaleozoic history of the central Transantarctic Mountains: Evidencefrom the Holyoake Range, Antarctica. New Zealand Journal of Geologyand Geophysics.

Skinner, D.N.B. 1964. A summary of the geology of the region betweenByrd and Starshot glaciers, south Victoria Land. In R.J. Adie (Ed.),Antarctic geology. Amsterdam: North Holland.

Skinner, D.N.B. 1965. Petrographic criteria of the rock units betweenthe Byrd and Starshot glaciers, south Victoria Land, Antarctica. NewZealand Journal of Geology and Geophysics. 8, 292-303.

Stump, E., M.R. Sheridan, S.G. Borg, P.H. Lowry, and P.V. Colbert.1979. Geological investigations in the Scott Glacier and Byrd Glacierareas. Antarctic Journal of the U.S., 14(5), 39-40.

Paleoenvironmental studies ofnonmarine diatoms in Quaternary

antarctic sediments

REED P. SCHERER

Department of Geology and Mineralogyand

Byrd Polar Research CenterOhio State University

Columbus, Ohio 43210-1293

Nonmarine sediments from ice-free areas of southern Vic-toria Land contain a largely unexploited paleoclimatic resource.Diatom assemblages within lacustrine sediment sequences arebeing used to evaluate Quaternary paleoenvironments in theMcMurdo Sound region. In the first phase of the study, begunduring the 1986-1987 field season, lacustrine and associateddeltaic sediments were sampled at outcrops in lower TaylorValley, the Cape Chocolate region, Wright Valley, Brown Penin-sula, East Dailey Island, and Ross Island. Results from twosequences near Cape Chocolate and one in lower Taylor Valleyare reported here. These samples were processed for diatomanalysis using a method (Scherer in preparation) which allowsthe establishment of absolute abundance (diatoms per gram ofdry sediment). This relationship allows comparison of diatomproductivity and sediment input.

Diatoms within these sediments are compared with modernfloras in sediment from Lake Vanda in Wright Valley and nu-merous meltwater streams and ponds distributed throughoutWright Valley, lower Taylor Valley, Cape Chocolate, Ross Island,Brown Peninsula, and on the McMurdo Ice Shelf around EastDailey Island. Modern diatom assemblages within these sam-ples, as well as those reported in the literature, provide ananalog for ecologic interpretation of fossil assemblages andtheir sedimentary environments. Older nonmarine diatom flo-ras from upper Pliocene/Pleistocene sediments from DVDP-15and CIROS2* drill cores in McMurdo Sound and Ferrar Fjordare also under study. A paleoecologic model is being con-structed based in part on these results.

Discontinuous outcrops of lacustrine and deltaic sedimentsare scattered along valley walls and dry basins of southernVictoria Land (Debenham 1921; Speden 1960; Péwé 1960; Kel-logg et al. 1980). These generally represent only high stands ofprevious lake levels, and thus do not contain continuous strat-igraphic successions. A high-resolution Late Quaternary pal-eoclimatic signal, based on diatom, sedimentologic and chemi-cal analyses, may be extracted from the near-continuoussediment records underlying certain modern lakes. The onlyavailable record of this kind is the upper 4 meters of DVDP-4A inLake Vanda. Although badly disturbed by drilling, this succes-sion showed evidence of major changes in sedimentation, in-cluding periodic evaporite deposition (Brady 1981). Continuoussedimentary records underlying modern lakes need to be ex-

* "DVDP" denotes Dry Valley Drilling Project and "CIROS" denotesCenozoic Investigations of the Ross Sea.

1987 REVIEW 35

tracted using methods that preserve stratigraphy and sedimen-tary structures.

Kellogg et al. (1980) studied diatom fossils from Late Wiscon-sin nonmarine perched deltas in Taylor Valley and defined sixdiatom assemblages. They used these assemblages to evaluatelake-level history but were unable to draw specific pal-eoecologic conclusions due to a paucity of modern ecologicdata. Kellogg et al. (1980) reported nine diatom taxa as endemicto southern Victoria Land, including six species which theydescribed as new. These findings conflict with the results of arecent algal monograph (Seaburg et al. 1979) which reports nononmarine diatom taxa as endemic to southern Victoria Land.For this study, diatoms were identified using the taxonomicassignments of Kellogg et al. (1980). Some taxonomic revision toaccomodate the above discrepancy will ultimately be necessary.

Certain diatoms are cosmopolitan in antarctic nonmarine aq-ueous environments (e.g., Navicula multicopsis and varieties,Navicula quciternaria, Navicula shackleton i, Nitzschia westii), de-spite widely differing ionic concentrations between variousbodies of water. However, it is the distribution of ecologicallyrestricted taxa that permit assemblage differentiation and pal-eoecologic interpretation. Many extant species range back atleast into the Pliocene (Brady 1981) suggesting that a pal-eoecologic model based on ecologically restricted nonmarinediatoms from modern environments has potential for applica-tion to pre-Quaternary sediments. The following is a descrip-tive analysis and comparison of sediments studied, with pal-eoecological notes.

Salmon Stream, Cape Chocolate. Lacustrine sediments nearSalmon Stream in the Cape Chocolate region were described bySpeden (1960) and discussed by Miagkov, Nedeshava, andRiaboba (1976) and Kellogg et al. (1980). Our recent field studiesbrought to light some new observations. A thick succession offinely laminated, moderately calcareous sediments on the northshore of Salmon Stream, near the snout of Salmon Glacier, hasbeen overridden by subsequent glacial activity. The lacustrinesuccession is overlain by a compact tillite and stratigraphy isdisrupted by numerous small thrust faults. Five samples fromthis sequence were studied. Samples are rich in nonmarinediatoms, averaging 6.5 million valves per gram. Marine diatomfragments are rare in this sequence. The diatom assemblage isdominated (65-75 percent) by a small centric diatom, Melosiracharcotii (figure a—c). This diatom is known from intertidal poolsalong the Antarctic Peninsula and associated islands (Peragallo1921; Frenguelli and Orlando 1958) but has not been previouslyreported from the Ross Sea sector of the antarctic. Specimens ofM. charcotii are common in meltpools on the McMurdo Ice Shelftoday, especially near tidal cracks, but were not found in iso-lated inland waters or under exclusively marine conditions. Thepublished ecology for M. charcotii and the distribution of thisdiatom in the McMurdo Sound region suggests an associationwith tidally influenced ice-shelf conditions and other marine-proximal, brackish water environments. Melosira charcotii is alsocommon in Pleistocene nonmarine intervals recovered fromDVDP-15 and CIROS-2 drill cores in McMurdo Sound. Theseintervals, up to approximately 80 meters thick, contain an ex-clusively nonmarine diatom flora, as reported by Brady (1978,1981, 1982) and Harwood (1986), respectively. The sediments ofSalmon Stream and McMurdo Sound drill cores could reflectnonmarine conditions at the ice edge during advance or retreatof grounded Ross Sea ice in McMurdo Sound.

Hobbs Glacier, Cape Chocolate. A 3-meter-thick outcrop of finelylaminated, (average of 0.8 millimeters per laminade) highlycalcareous sediments is located adjacent to the snout of Hobbs

a-c: Melosira charcotii Peragello, Salmon Creek, Cape Chocolate.(2000 x) d-f: Chrysomonad cyst, Ochromonas miniscula Conrad(?), Lake Vanda, Wright Valley. (2000 x)

Glacier. Small, discontinuous outcrops of similar material arescattered throughout the vicinity. Advances of the HobbsGlacier since 1903 have apparently destroyed similar deltaicsediments (Denton and Hughes 1981). The sediments collectedfrom the outcrop contain an average of 1.2 million diatoms pergram, based on six samples. The diverse nonmarine as-semblage is dominated by Navicula ccntenta, Navicula gibbula,and Hantzschia amphioxys. Marine fragments are uncommon.Melosira charcotii is absent from this sequence.

Commonwealth Glacier, Taylor Valley. A 4-meter-thick exposureof lacustrine and associated deltaic sequences crops out at thesnout of Commonwealth Glacier. The sequence has evidence ofhigh sedimentation rates and rapid shifts in sedimentary regi-me, including several distinct units and a variety of soft sedi-ment deformational structures. Nonmarine diatoms occur inlow concentrations in this succession. Low nonmarine diatomabundance reflects high sediment input rather than low pro-ductivity. Despite the differing character of sediment units, thediatom assemblages in the samples studied are fairly consistentin absolute abundance and taxonomic compositions. Non-marine diatom abundance in four samples from the lightercolored units averages only 200,000 per gram.

Contrary to the observations of Kellogg et al. (1980), whoreport an absence of Nitzschia westii in Commonwealth Glacierlacustrine sediments, N. westii is common in all samples studiedfrom this outcrop. Reworked nonmarine diatom-rich sedimentclasts from older deposits are present in this succession. Theclasts contain abundant Navicula shackletoni and common Melo-sira charcotii. The occurrence of these clasts and common marinediatom fragments indicates significant sediment transport andredeposition.

Lake Vanda, Wright Valley. Recent sediment from the anoxicbasin of Lake Vanda was evaluated using the same methods asabove. This sample has about 3 million nonmarine diatoms pergram of sediment. Species composition is unlike the diatomassemblages present in outcrop or modern meltpool samples.While several cosmopolitan diatoms are present, most notablyNavicula multicopsis and varieties, the dominant diatom is Pin-nularia cymatopleura. Several forms are present which were notobserved in outcrop samples or in modern shallow meltpools.These included Amphora delicatissima, which has been reportedfrom Lake Miers (Baker 1967).

36 ANTARCTIC JOURNAL

A small (5-8 micron) siliceous cyst formed by a planktonicchrysophyte (Ochromonas niiniscula?) (figure, d-f) makes thegreatest contribution to the siliceous microflora in this sedi-ment. The sample contains more than three times as many ofthese chrysomonad cysts as diatoms, amounting to 10 millioncysts per gram, making them an important ecological marker.The distribution of this algal species has been previously de-scribed as abundant in Lake Vanda but rare in Taylor Valleylakes (Seaburg et al. 1979). The cysts are found in very lowabundance in other nonmarine sediments, despite wide dis-tribution. Chrysomonad cysts have been found forming underwinter ice in Scandinavian lakes (Cronberg 1973), suggestingthat cyst formation provides a survival strategy during winterand perhaps other times of environmental stress. Abundantoccurrence of this fossil in a sediment sequence may suggestconditions similar to Lake Vanda today, i.e., thick, clear ice anddeep, chemically and thermally stratified water. Oxygenatedshallow antarctic lake floors are host to a complex community ofmat-forming filimentous blue-green algae, which accumulatestromatolitic laminations (Parker et al. 1981). Benthic diatomfloras are found in association with these algal mats. By con-trast, the microfossil assemblage in sediments from the deepanoxic basin of Lake Vanda is dominated by planktonic forms,reflecting productivity in the water column under the ice but arelatively inactive benthic community. These assemblage dif-ferences may be useful in reconstructing past lake conditions.

These results are very encouraging, but further work isneeded. Detailed ecologic and distributional studies on themodern diatom floras will refine the paleoenvironmentalmodel. To apply this model best, long, undisturbed coresshould be extracted from the bottom of several modern dryvalley lakes. Continuous sedimentary successions and the lackof bioturbation make these records attractive for high resolutionpaleoclimatic studies. The record spanning the last few 1,000years maybe compared with the antarctic ice-core record. Long-er time-scale lacustrine successions may contain trends for com-parison with nearshore antarctic marine core records. Thesesedimentary successions have the potential for providing anexcellent record of terrestrial Quaternary climate change inAntarctica.

This research was supported by National Science Foundationgrant DPP 85-17625 to Peter-Noel Webb. Lake Vanda sedimentmaterial was collected under National Science Foundation grantDPP 85-16465 to William Green and was generously provided forthis study. Peter Webb and Dave Harwood made helpful com-ments regarding the manuscript.

References

Baker, A. N. 1967. Algae from Lake Miers, a solar-heated Antarctic lake.New Zealand Journal of Botany, 5, 453-468.

Brady, H.T. 1977. Freshwater lakes in Pleistocene McMurdo Sound.Antarctic Journal of the U.S., 12(4), 117-118.

Brady, H.T. 1981. The significance of fossil marine and non-marinediatoms in DVDP cores. in L.D. McGinnis (Ed.), Dry Valley DrillingProject. Antarctic Research Series, Vol. 33, Washington, D.C.: AmericanGeophysical Union.

Brady, H.T. 1982. Late Cenozoic history of Taylor and Wright Valleysand McMurdo Sound inferred from diatoms in Dry Valley DrillingProject cores. In C. Craddock (Ed.), Antarctic geoscience. Madison:University of Wisconsin Press.

Cronberg, G. 1973. Development of cysts in Malloinonas eoa examinedby scanning electron microscopy. Hydrohiologia, 43, 29-38.

Debenham, F. 1921. Recent and local deposits of McMurdo region.British Antarctic Expedition "Terra Nova," 1910. Natural History Report,Geology, 1(3), 63-100.

Denton, G. H., and T.J. Hughes. 1981. The last great ice sheets. New York:Wiley.

Frenguelli, J., and H. Orlando. 1958. Diatoms and silicoflagellates of theSouth American sector of the Antarctic. Buenos Aires: Instituto AntarticoArgentino. (In Spanish)

Harwood, D.M. 1986. Diatom hiostratigraphy and paleoecology with aCenozoic history of antarctic ice sheets. (Doctoral dissertation, Columbus:Ohio State University.)

Kellogg, D.E., M. Stuiver, T.B. Kellogg, and G.H. Denton. 1980. Non-marine diatoms from Late Wisconsin perched deltas in Taylor Valley,Antarctica. Palaeogeography, Palaeoclimatology, Palaeoecology, 30,157-189.

Miagkov, SM., G.N. Nedeshava, and E.I. Riabova. 1976. McMurdoSound sea level changes in the last 50,000 years. Antarctic Journal of theU.S., 11(5), 233-235.

Parker, B.C., G.M. Simmons, Jr., F.G. Love, R.A. Wharton, Jr., and K.G.Seaburg. 1981. Modern stromatolites in Antarctic dry valley lakes.Bioscience, 31, 656-661.

Peragallo, M. 1921. Freshwater and saltwater diatoms. In J . Charcot(ed.), Botanique. Paris: Second French Antarctic Expedition1908-1910. (in French)

Péwé, T. 1960. Multiple glaciation in the McMurdo Sound region, Ant-arctica: A progressive report. Journal of Geology, 68, 489-514.

Scherer, R.P. In preparation. An absolute abundance method for quan-titative analysis of diatoms or other fine sand and silt-sized sedimen-tary particles. Micropaleontology.

Seaburg, K.G., B.C. Parker, G.W. Prescott, and L.A. Witford. 1979. Thealgae of Southern Victorialand, Antarctica. Bihliothecza Phycologica,Vol. 34. Valduz: J. Cramer.

Spederi, I. 1960. Post-glacial terraces near Cape Chocolate, McMurdoSound, Antarctica. New Zealand Journal of Geology and Geophysics, 3,203-217.

1987 REVIEW 37