You are on page 1of 8

Revisca do Instituto Geologico, Sao Pau lo, 21 (112), 71-78, 2000.

SPACES OCCUPIED BY THE EXPANSION OF DRY CLIMATES IN SOUTH AMERICA


DURING THE QUATERNARY ICE AGES

Aziz Nacib AB'SABER

(Otiginally published in Paleoclimas 3, 1977, Institute de Geografia - Universidade de Sao Paulo, as


Espar;;os ocupados pela expansCio dos climas secas na America do Sui por ocasiiio dos perfodos
glaciais quaternarios. Translated by P.E. Vanzolini, revised by Thomas R. Fairchild.)

At a time when new outlines are drawn 1t is certain that at the beginning of the
of phytogeographic regions and morphoclimatic typical Quaternary climatic variations there were
domains of South America, one feels called to aggressive changes in morphogenetic processes,
fathom the recent past , in searc h of the broad and radical enough, in some areas, to era<>e
paleogeographic and paleoecologic frameworks the landscapes established during the Late Tertiru-y
that succeeded each other within the continent and to make possible the spread of new types of plant
during the Quaternary. cover and new physiographic and ecologic fabrics.
There are no major problems of paleo- DUling the Quaternary, periods of rhexistasy always
spaces, except perhaps in the platform of southern succeeded periods of biostasy (ERHART, 1955);
Argentina and in the coastal areas subject to glacio- long-lasting morphoclimatic systems alternated with
eustatic movements: the present topographic periods of fast degradation through short-lived,
framework may be taken as adequate, especially morphogenetically very active transitional periods.
on the scale of mapping. In terferences of However. to be frank, we know very little about these
neotectonics, epeirogenesis and continental changes of pace of erosive processes, except that
nexures, although relevant to the study of specific they actually paved the way for broader and more
areas, wi II not be considered in this first approach. general ecological changes of regional scope.
The p resent summary is focu sed o n the The critetia that have permitted this first
ap proximate identification of the paleo-spaces synthe tic approach to the morphoclimatic and
occupied by dry climates in South America during phytogeographic mosaic of the last dry period,
the latest glacial and glacio-eustatic periods of the 18,000-12,000 years BP (Wiirm-Wisconsin)
Quaternary. In other words we aim at belong to the field of multiple and correlated
understanding the paths of penetration of dry observations. (FIGURE 1). Our summary is
climates that were associated with a generalized supported by a superposition of geomorphological,
lowering of temperature, consequent upon sedimentological and ecologic facts. We use
glaciation, glacio-eustatic movements and the information accumulated in the literatdre on
action of cold currents that brought a ridi ty geomorphological features (ped iments, fluvial
northward. We shall not discuss short, later dry terraces, residual bolsones), correlative deposits
phases related to the optimum climaticum, when (detritic formations, terrace deposits, piedmont
there were incidental spots of dry climates, due, deposits, stone lines, paleosols), laterites and
among other causes, to global warming and to sig nificant mini-enclaves (evidence of rocky
events of local topographic compartmentation. aridity, local cactus fields, refuges of flora and
Geomorphological, sedimentological fauna). The largest part of the documentation
and phytogeographic research. performed during related to the last Quaternary dry period (18,000-12,(XX)
the last twenty years [i.e., from 195 I to 1971 ], years BP) was taken from data on the superficial
make it possible to assert that between the Late structure of landscapes, including a review of
PJ iocene and the Pleistocene at the time of the main personal observations made over a period of many
phase of Andean uplift, there were alternations in years.
the distribution of soils and floras over relatively We have paid special attention to the
sho11 periods of geological time. mini-enclaves of residual xerophytic vegetation,
As for the complex mechanisms of that locally resisted the changes to general moister
retraction and expansion of the plant cover over climates that occurred in the last 12.000 years. Up
the fundamental space we have only fragmentary to a certain point, the group of mini-enclaves of
information and indirect inferences. xerophytic vegetation seems to indicate the main

71
Revista do lnstituto Geologico, Sao Pau lo. 21 (112), 71-78,2000.

NATURAl. DOl11lAINS
OF SOUTH AMERiCA
18.110(} IJ.OOO YEARS 81'
FIRST ESSAY

AZIZ AB'SABER
1977

1 0'1 "
.i . ,
II SEMI-AAID ARE.4S WITH CAAJIJ\GAS AND

I ~ UKE FLORAS (WITH CACTI]


~ -!. I

~ SUS.DESERTIC STEPPE AREAS


L (_:) (SPREAD OF THE MONTE)
- : ; :-;1 STEPPE AND COLD DESERTAREAS
L' ' :J(SPREAD OF THE PAV\GONIANSTEPPES]
n
. -
GREAT AV\OOIA DESERT

c::= : GREAT CERRADO.NUCLEI W!1H


~ CAAJINGA ENClAVES
J , fl
, ', I ARAOCARIA NOC!fl (BRAZJUAN AND ANDEAN]
- -J
I : : ~ ROCKY DESERTS AND RAIN-SHADOW DESERTS
1 ' 1
11!.~,:' COLD TEMPERATE FORESTS

FIGURE l Natural domains of South America in last dry period (18,000 - 13,000 years BP)

72
Revista do lnstituto Geol6gico, Silo Paulo. 21 (In), 71-78, 2000.

axes of the great routes of penetration of the glacial periods, and the other to warmer and
Quaternary semi-aridity in the present space of more humid climates during the interglacials. This
Tropical America. scheme is valid for a large portion of the
The earliest significant discoveries geographic space of South America, apparently
regarding paleoclimates of eastern South America in direct opposition to what was known in Africa
were made by CAILLEUX & TRICART (1957) and about pluvial and interpluvial periods.
by TRlCART (1958). The direct association between The d iscovery of the general dynamics
glacial conditions, low sea levels and the spread of and the basic correlations of glaciation and spread
semi-aridity is attributed to BIGARELLA & of semi-aridity clearly did not exhaust the subj ect
AB'SABER (196 1) within a framework of of Quaternary paleoclimates in Brazil nor in South
correlations nowadays little known. In order to America in general. Much remains to be detailed
construct this scheme, field observations by regarding evolution in time and, especially, in
BIGARELLA & AB'SABER (1961, 1964) and space. Such an interclisciplinary effort is still due.
BTGARELLA, MARQUES FTLTTO & AB'SABER During the period of glaciation and
(1961) on gemorphological features and correlative extended eastern cold currents, within the context
deposits were fundamental. The ped iments and of South America, open formations of diverse
correlative deposits at the foot of the Serra de types predominated over the great forested areas.
Tqueririm at Garuva, Santa Catarina, furnished full Nevertheless, one should avoid a mental picture
confirmation of the role of semi-arid morphoclirnatic of homogeneous undifferentiated dry climates. On
Quaternary processes linked to periods of low sea the contrary, everything leads to the concept of a
level corresponding to glacial periods. complex map, closely linked to the conditions of
Based on early findings ( 1964), some of topographic compartmentation of the Brazilian
us thought that the only explanation for the coastal plateau and of the central South American
expansion of the Pleistocene dry system should depressions, as well as to the numerous small
have been a strong activation and advance of the glacial centers and periglacial areas in the Andes.
cold Atlantic current northward along eastern Caaringas were much more widespread
South America into tropical latitudes. This idea than believed a short time ago (TRICART, 1958).
was suppo1ted by the ensemble of conclitions at They penetrated numerous inner compartments of
Cabo Frio, where cold upwelling has produced a the present Brazilian intertropical plateaus in areas
sub-humid climate with mini-enclaves of now covered by forests or cerrados. Cerrados,
xerophytic vegetation on the continent and nearby cerradoes and associated types of vegetation
islands. It was, however, merely a hypothesis, broadly penetrated eastern and central Amazonia,
difficult to prove. perhaps joining similar areas, now of reduced
In 1970 DAMUTH & FAIRBRIDGE extent, in Roraima, the Guianas and the Orinoco
published a fundamental paper on Quaternary dry llanos.ln the space of Amazonian lowlan(,!s forest
climate deposits on the Brazilian coast, based on refuges remained only in the islands of moisture
a very well-conducted research on arkoses on some exposed slopes of small ridges and
deposited on the Brazilian platform. One of the partially convex hills and perhaps also on the west-
authors, Rhodes Fairbridge, before finishing the southwest arch of the Andean slopes and the
article, came to southern Brazil (where he was northern facade of the Guianas. The existence of
assisted by Bigarella) in order to check in situ the these refuges, now engulfed by the expansion and
validity of the observations accumulated in the coalescence of the Amazonia-Guianan forests, has
Brazilian literature. In the paper by DAMUTH & received confumation from the multidisciplinary
FAIRBRIDGE (1970) there is an integrated collaboration of geographers, zoologists and
paleoclimatic interpretation, to us the first attempt botanists (Haffer, Vanzolini, Williams, Journaux,
at a global explanation of the driving forces behind Prance, Brown, among others).
the aridity penecontemporaneous with glacial On ly the domain of the cerrados on the
periods and low sea levels of South America as a central plateau partially resisted the expansion of
whole. More than that, the authors presented the dry climates; in the peripheral and interplateau
schematic maps of the probable course of the cold depress ion s it gave way to caatingas (in
currents during glacial and interglacial episodes depressions between the chapadoes of the Urucuia
(FIG URE 2). In this way the two schemes of and the central Goias Plateau; depressed areas
climati c predominance were defined, one north of Brasilia and Anapolis; pediplain of
corresponding to drier and colder climates during

73
Revista do Instituto Geologico, Sao Paulo, 2 1 ( 1/2), 71-78,2000.

Cuiaba; pediplain of the Upper Araguaia; prairies to the south and southeast), and included
monoclinal intra-chapad6es depressions) (Editor's a few forest refuges.
note: chapadoes, plateaus with complex structure Meanwhile it should be noted that a second
or table-like rolling plateaus). Thus a belt of large area of cerrados existed on the tabulei/Vs (.11.:
caatingas remained intercalated between the main low sedimentary plateaus, tablelands) and low
remnants of the core cerrados and the southern chapadoes of Amazonia, with great gallery forests
a nd southeastern Amazonian belt. Between and multiple sub-xerophytic enclaves (caatingas). It
Central Brazil and central Amazonia there was an is not impossible that tlus Amazonian network of
ensemble of very broad bands, made up initially cerrados occupied a larger area than the cenlr"<tl core
of cerrados, passing into caatingas and returning of summit cerrados. The open formation depressions
to ccrrados of diverse types. lt is possible that at of the Rio Branco were much drier than nowadays,
the time the areas of non-xerophytic open and the cerrados extended through northwestern
formations in Amazonia had greater connections Amazonia to the Orinoco.
and coalescences with the areas of the present It is practically certain that no depression,
cerrado macro-enclaves of Roraima-Orinoco and whether denudational (peripheral or monoclinal
the Guianas than with the remaining nucleus in depression) or rectonic (as in the middle valley of
Central Brazil. lt is not out of the question to the Parafba), failed to suffer the penetration of non-
postulate the presence, at certain times in the Late homogeneous, dry climates, originating from one
Pleistocene, of caatinga enclaves in several sub- of the two main areas of principal expansion of
rocky sectors, otherwise occupied by cerrados and the Pleistocene semi-aridity, coupled with the
cerradoes with elongated gallery forests. Along glacial system: the enlarged northeastern area and
the same lines, but with less certainty, a plausible the northemly extended coastal area.
area of dry enclaves would be the depression of On the east-central part of the continent
the Rio Branco grasslands and the low southern only the domain of the cerrados on the Central
slopes of the mountains on the Brazil-Venezuela Brazi li an p lateau partially resisted the
border. encroachment of dry climates and the retraction
Everything indicates that during the of the main areas of great tropical fo rests. We
Quaternary glacial episodes, the core area of the believe that in many areas, especially along the
cerrados was much smaller than today. Part of it, eastern and no1thern margin of the domain of the
on the northern half of the Brazilian Plateau, must cerrados , cerrados and caatingas may have
have been occupied by caatingas; the southern continued to occupy the same space in Central
edge would have been dominated by steppes, Brazil, in a manner similar to a restricted pattern,
mixed prairies and a less dense nucleus of at present much reduced in area, that occurs in
araucaria (southern Mato Grosso and southern north-central Bahia in the region of Ribeira do
plateau). The pediplain of Cuiaba and the Pombal. Another possibility is a scheme like }hose
interplateau and intermontane depressions ofMato found in the Chapada do Araripe or on the south-
Grosso, Goias, Bahia and Minas Gerais have central part of the Chapada da lbiapaba.
always tended toward climates much drier than It is important to note that at the time of
the present ones . In these areas caatingas maximum spread of the open formation s
predominated over cerrados. On the other hand, associated with dry climates or with sub-humid
in the interior of the inner Sao Paulo plateaus - tropical climates having two -seasons (i.e. at the
with the exception of the interplateau and time of the glacial maximum) there was an
intermontane depressions - cerrados must have inversion of the scheme of core areas surrounded
prevailed over forests. In the depressions caatingas by heterogeneous transition and contact belts.
must have occurred. There prevailed in stead a scheme of vast
In a nutshell, the spatial image that can transitional areas. Forests were reduced to refuges
be visualized, especially with respect to the last quartered in propitious topographic situations. The
dry period of the Quaternary, is that of a macro- refuges were fu ndamentally orographic forests,
enclave of cerrados on the high nucleus of the favorably situated in places where moisture could
Central Brazil chapad6es. This summit enclave collect: "brejos", as they are called in the
remained isolated in Goias and Mato Grosso, Northeast, islands of moisture and enclaves
surrounded by a complex network of landscapes (BIROT 1957). The refuges on the Serra do Mar, from
(caatingas to the north, east and west, steppes and Sao Paulo in the south to Espirito Santo in the north,

74
Revista do Instituto Geol6gico, Sao Paulo, 21 (1/2), 7 1-78, 2000.

TROPICOF
CAPRICORN

CURRENT~
1MNDS
L L
L JUlY~
JAN--+

L
FIGURE 2 . Generalized distribution of ocean currents and wind systems (left) for the present and previous warm
interglacial phases and (right) postulated for Pleistocene glacial phases (after DAMUTH & FAIRBRIDGE, 1970).

must have occupied discontinuous areas on the masses of angular and rounded pebbles. Tundras
summits of the scarps most exposed to moisture from and cold deserts must have existed in the place of
the sea; the coastal lowlands were relatively dry. today's semi-desertic steppes. Part of the fjords
Some areas of subtropical and even were filled by the same glaciers that had excavated
tropical plateaus on the south-central half of the them during syncopated glacial episodes.
Brazilian Plateau were certainly drier and slightly Cold currents, much broader, somewhat
cooler. Such combinations of aridity and low displaced from their present axes and climatically
temperatures would have facilitated the northward active, hit in full the coast of southern Brqzil and,
extension of the araucarias, as "bridges" along the to a considerable extent, also those of eastern and
highlands of eastern Brazil and the eastern mid- southeastern Brasil. Only the Serra Geral, in its
slopes of the Andes. Xerophytic, sub-xerophytic and east-west stretch in Rio Grande do Sul, comprised
sub-tropical types of vegetation of the Argentinian a barrier high enough to trap moisture, bringing
north and northwest advanced very deep into the rain to its middle and upper slopes. While the area
interior of central South Ame1ican depressions and of the Argentinian and Uruguayan humid pampas
certain sectors of east-central Brazil. was arid, dry and steppe-like, at several times in
Going from south to north, things the Quaternary, the greatest part of the coxilhas
happened by ever more complex and radical (E.n: low rounded hills typical of the rolling
changes. At the time of lower (and receding) sea topography of southernmost Brazil) of Uruguay
levels, when the southern cone was much less and Rio Grande do Sui was under the influence of
funnel-shaped, its dry areas must have been much dry climates and was partially invaded by
more extensive. The glaciers of southern Chile xerophytic formations with cacti. At the time, there
extended to the present day area of the finger lakes were no subtropical gallery forests in the present
of Argentina. Glacial climates and wide belts of areas of mixed prairies in Rio Grande do Sul.
cold deserts and peri-glacial areas stimulated It may be asserted that the drier and
mechanical morphogenesis in high and middle- colder phases of the Quaternary favored the
high areas, favoring the elaboration of great predominance of Argentinian monte with cacti and

75
Revista do Instituto Geol6gico, Sao Paulo, 21 (112), 7 1-78,2000.

steppes of northern Patagonian type, with a it is still too early, as mentioned previously, to try
decrease in the area of humid prairie landscapes. to reconstruct the mosaic of Quatemary landscapes
Monte landscape was present in Rio Grande do and floras in terms of a reference cartographic
Sui, and Patagonian steppes entered deeply into document. What can be done at present,
the present-day area of humid pampas. Instead of expeditiously and with little risk of distortion, is
a core area of humid pampas there was a series of to sketch a rough outline of the main routes of
refuges and refuge clusters on the humid and sub- penetration of the dry climate open formations into
humid topographically prominent slopes of the the areas that at present constitute great forested
pampas of Argentina, Uruguay and Rio Grande domains. In other words, it is only possible to map
do Sul (Sierras del Tandil and de la Venlana, the areas more habitu ally visited by the
Sierras de C6rdoba, hills of the Uruguay-Rio predominantly dry (but never homogeneous)
Grande do Sui nucleus of the Brazilian Shield). systems that several times during the Quaternary
Based on knowledge of the superficial structure invaded the space at present dominated by tropical
of the landscape of the plateaus of Lajes and systems. It is practically certain that during most
Vacaria, we have reason Lo think that southern of the glacial and glacio-eustatic phases such
Brazilian domain of the araucaria was less compartmented (complex mosaics) dry belts
compact and continuous, interspersed wi th sub- substituted predominantly humid systems similar
rocky, dry, steppe- like sectors; the domain was to those round nowadays on the continent.
additionally somewhat displaced northward in
elongated belts following sub-humid and humid BIBLIOGRAPHY
ridges and summits (Paranapiacaba, Campos do
Jordao, Bocaina, south-central Minas Gerais, I. Present landscape (overviews)
southern Serra do Espinha~o). The retreat of the
AB'SABER, A.N. l97l. Aorgan izac;:ao natural das
cold currents to their present position, beginning
paisagens inter e subtropicais brasileiras.
some 10,000-8,000 years ago, would have been
In : M.G. Ferri (coord), fiT Simp6sio sobre
accompanied by extensive humidification, with
o Cerrado, Editora Edgard BlUcher, Sao
increased density of the araucaria forest,
Paulo, p. 1-14.
expansion of high prairies in sites of former
steppes, and tropicalization of the plateaus of Sao CABRERA, A.L. & WILLINK, A. 1973.
Paulo and northern Parana, thereby interrupting Biogeografia de America Latina. Prog. Reg.
the connection between the core-area of the Des. Cient. Tecnologico, OEA, Washing-
araucarias and outlying refuges on the highlands ton, p. l -120.
of southeastern Brazil. The remains of araucaria EITEN, G. 1974. An outline of the Vegetation of
at Paranapiacaba, near the city of Sao Paulo and South America. In: Symp. Intern. Primat.
at Bragan~a Paulista, intermixed with tropical Soc., 5, p.529 -545.
forests of the plateau, would constitute the last. HUECK, K. 1966. Die Waelder Stidamerikas.
residual document of the expansion a nd Gustav Fischer Verlag, Stuttgart. [Tradu-
coalescence of the tropical forests in Sao Paulo c;:ao: H. Reichardt, 1972. As florestas da
in a landscape previously similar to that of the America do Sul. Editora da Universidade
Sao Paulo-Parana border (Paranagua Plateau, de Brasilia e Editora Poligono S. A., Sao
Upper Ribeira basin on the Parana side). Paulo, 466 p.].
Many questions remain open on the
HUECK, K. & SEIBERT, P. 1972.
paleoclimatic and paleoecological settings of the
Vegetationskarte von Si.ldamerika. Mapa de
Andean macro-region. Possibly all the equatorial
Vegetaci6n de America del Sur. Gustav
area of the Andes has been subject to allitudinally
Fischer Verlag, Stuttgart.
progressive temperature lowering; the western
lowlands may have been less humid. On the II. Palaeoclimates and palaeoclimatology
Amazonian slopes it is possible to conceive of (general works)
discontinuous forest refuges, with predominance AB'SABER, A.N. 1958. Conhecimentos sobre as
of a scheme still observed in the areas of contact flutuac;:oes climaticas do Quaternario no
between forested hillslopes and the Orinoco Brasil. Not{cia Geomorfol6gica, I (1 ): 24-
savannas in Colombia and Venezuela. 30.
In spite of these preliminary inferences - 1962. Revisao dos conhecimentos sobre o ho-

76
Revista do Tnstituto Geol6gico, Sao Paulo, 21 (1/2), 7 1-78, 2000.

rizonte sub-superficial de cascalhos FAIRBRIDGE, R. W. 1962. World sea level and


inhumados do Brasil Oriental. Universi- climatic changes. Quaternaria, 6: I 11-134.
dade do Paran<i, Institute de Geologia, 31 GATES, W. L. I 976. Modeling the Ice-age
p. (Boletim da Universidade do Parana, climate. Science, 191 (4232):1138-1144.
Geografia Ffsica, 2).
HAFFER, J. 1969. Speciation in Amazonian forest
BIGARELLA, J.J. 1964. Variary6es climaticas no birds. Science, 165 (3889):131-137.
Quatemario c suas implicary6es no reves-
JOURNAUX, A. 1975. Recherches
timento florfstico do Parana. Boletim
geomorphologiques en Amazonie
Paranaense de Geografia, (10115): 211-
bresiliennc. Cacn: Centre de
231.
Geomorphologic de Caen- C.N.R.S., p.9-
- ; AB'SABER, A.N. 1961. Quadro provi- 15 (Bul letin 20).
s6rio dos fatos sedimentol6gicos,
LIMA, D. A. 1964. Contribuiryao a dinamica da
morfocli maticos e paleoclimaticos na
flora do Brasil. Recife: Institute 9e Cien-
Serra do Mar paranaense e
cias da Terra, Uni versidade do Recife, Ar-
cata rin ense . I n: J.J.Bi ga re ll a,
quivos, (2) : 15-20.
P.L.Marqucs F ilho & Ab'Saber, Ocor-
re ncia de pedimentos remanescentes MAACK, R. 1947 - Breves notlcias sobre a geo-
n as fraldas da Serra do lquererim logia dos Estad os do Parana c Santa
(Garuva, SC), Boletim Paranaense de Catari na. Arquivos de Biologia e
Geografia, (4/5): 91. Tecnologia, 2: 99-200.
- ; - - 1964. Palaeogcographische und MARTONNE, E. DE. J 935. Problemes des
Palaeoklimatische Aspekte des regions arides sud-americaines. Annates de
Kainozoikum in Si.idbrasilien . Geographie, Paris, 44(247): 1-27.
Zeitschrift fiir Geomorphologie, N. F. 8 MERCER, J. H. 1975. Glacial history of Southern
(3): 286-312. South America since mid-Pliocene time.
- ; BECKER, R. (eds). 1975. International In: International Symposium on the
Symposium on the Quaternary. In: Boletim Quaternary. Boletim Paranaense de
Paranaense de Geociencias, (33): 370 p. Geociencias, (33): 57.
-;MARQUES FILHO, P.L.; AB'SABER, A.N. MULL ER, P. 1973. The dispersal centres of
1961. Ocorrencia de pedimentos remanes- terrestrial vertebrates in the Neotropical
centes nas fraldas da Serra do Iquererim Realm. End. W.Junk B. V. Publishers, The
(Gamva, SC). Boletim Paranaense de Ge- Hague.
ografia, (4/5): 82-93. PRANCE, G. T. 1973. Phytogeographic support
BIROT, P. 1957. Esquisse rnorphologique de Ia for the theory of Pleistocene forest ~efuges
region littorale de I'Etat de Rio Janeiro. in the Amazon Basin, based on evidence
Annales de Geographie, 66 (353):80-91. from d istribution patterns in Caryocaracea,
Chrysobalanacea, Dichape talacea and
BROWN Jr., K. S. 1976. Geographical patterns
Lecythidaccae. Acta Amazonica, 3 :5-28.
of evolution in Neotropical Lepidoptera.
1. Ent. B., 44: 201-242. TRICART, J. 1958. Division morphoclimatique
du Bresil atlantique central. Revue de
CAILLEUX, A. & TRICART, J. 1957 . Zones
Geomorphologie Dynamique, 9 (1/2): 1-
phytogeographiques et
22.
morphoclimatigues du Quaternaire du
Bresil. C. R. Soc. de Biogeographie, 293: - 1960. As zonas morfoclim!ilicas do Nordes-
7- 13. te Brasileiro. Notfcia Geomorfol6gica,
3(6):17-25.
DAMUTH, 1. E. & FAIRBRIDGE, R. W. 1970.
Equarorial Deep-Sea Arkosic Sands and 1963. Oscillations et modifications de
Ice - Age Aridity in Tropical South caractere de Ia zone aride en Afrique et en
America. Bulletin of the Geological Amerique Iarine lors des periodes
Society of America, 81:189-206. glaciaires des hautes latitudes. In:
UNESCO, Les changements de climat.
ENJALBERT, H. (s.d.). La Pampa argentin et le
Zone Aride, p. 415-419 (Pub!. 20).
Chaco argentin. (Mimeogr., lnedito).

77
Revista do Instituto Geologico, Sao Paulo, 21 (1/2), 7 1-78,2000.

VANZOLINI, P. E. 1970. Zoologia sistematica, (eds.) Tropical forest ecosystem in Africa


geografia e a origem das especies. Institu- and South America: a comparative review.
te de Geografia, Universidade de Sao Pau- Smithsonian Institution Press, Washington,
lo, 56 p. (Serie Teses e Monografias, 3). p. 255-258
1973. Paleoclimates, reliefs, and species VUILLEUMIER, B.S. 1971. Pleistocene changes
multiplication in equatorial forests. In: B.J. in the fauna and flora of South America.
Meggers, E.S. Ayensu, W.D. Duckworth Science, 173:771-780.

Enderego do autor:
Aziz Nacib Ab'Saber - Institute de Estudos Avanc;ados - USP, Ediffcio da Antiga Reiroria, Av. Prof. Luciano
Gualberto, Travessa J, 374, Teneo, Cidade Universitaria, 05508-900 Sao Paulo, SP, Brasil.

78