Full text of DOI:10.1080/00173130150503795

This article was downloaded by: [University of South Carolina ]
On: 26 June 2013, At: 10:22
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office:
Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Grana
Publication details, including instructions for authors and subscription
information:
http://www.tandfonline.com/loi/sgra20
Pollen of the Asteraceae from the Paleocene-
Eocene of South Africa
Michael Zavada & Susan de Villiers
Published online: 05 Nov 2010.
To cite this article: Michael Zavada & Susan de Villiers (2000): Pollen of the Asteraceae from the Paleocene-
Eocene of South Africa, Grana, 39:1, 39-45
To link to this article: http://dx.doi.org/10.1080/00173130150503795
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial
or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the
contents will be complete or accurate or up to date. The accuracy of any instructions, formulae,
and drug doses should be independently verified with primary sources. The publisher shall not
be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or
howsoever caused arising directly or indirectly in connection with or arising out of the use of this
material.

Grana 39: 39±45, 2000
Pollen of the Asteraceae from the Paleocene-Eocene of South
Africa
MICHAEL S. ZAVADA and SUSAN E. DE VILLIERS
Zavada, M. S. & de Villiers, S. E. 2000. Pollen of the Asteraceae from the Paleocene-Eocene of South
Africa. ± Grana 39: 39±45. ISSN 0017-3134.
Two species of dispersed pollen (Tubuliforidites antipodica and T. viteauensis) that have af®nity with the
Asteraceae were recovered from Paleocene-Eocene sediments of South Africa. This is the earliest
unequivocal record of the Asteraceae. The two species were examined with light, scanning, and
transmission electron microscopy. The pollen wall structure of both taxa are of the Anthemoid type, a
pollen wall structural type considered to be primitive among asteraceous pollen. The wall structure type
is indicative of many taxa of the Mutisieae. The appearance of this pollen type in the Paleocene-Eocene
of South Africa supports a early Tertiary, west Gondwana origin of the Asteraceae, and the
evolutionary basal position of the Mutisieae.
Michael S. Zavada, Department of Biology, Providence College, Providence, RI 02918, USA; Susan E.
de Villiers, Bernard Price Institute for Palaeontology, University of the Witwatersrand, Private Bag 3,
WITS 2050, Johannesburg, Gauteng, South Africa.
(
Manuscript accepted 9 June 2000)
The Asteraceae are comprised of about 1,500 genera and Dispersed pollen recovered from the Oligocene and
23,000 species arranged in 3 subfamilies and 17 tribes. The younger sediments is assignable to the Asteraceae based
Asteraceae have a cosmopolitan distribution, and may be on features examined with the light and scanning electron
one of the most conspicuous plant families in the world. The microscopy (Muller 1981, and references therein; Barreda
size and distribution of the family intuitively suggest a long 1993). There are reports of Eocene Asteraceae pollen from
evolutionary history.
Egypt (Kedves 1971) and Afghanistan (Lang & Meon-Vilian
The Asteraceae are separated into three subfamilies, the 1976), which may be the oldest record of pollen of the
Barnadesioideae, the Cichorioideae and the Asteroideae Asteraceae, however, these reports are uncon®rmed by
(
after Bremer 1994). The numerous palynological studies on detailed morphological investigation. There are other
extant asteraceous pollen provide a substantial data base reports of pre-Oligocene dispersed asteraceous pollen
e.g Thanikaimoni 1977) for comparison with fossil (Turner 1977, Kemp & Harris 1975, 1977), however,
(
dispersed pollen. Many taxa in the Asteraceae have Muller (1981) rejects these as being unreliable. Dettmann
distinctive pollen characteristics that can be easily recog- & Jarzen (1988) report the occurrence of the genus
nized in dispersed fossil pollen record using light, scanning Tubuli¯oridites from the Upper Cretaceous of Australia
and transmission electron microscopy. Paleopalynologists and Antarctica, however, these taxa, although morphologi-
traditionally separated the pollen into two groups based on cally assignable to this form genus, lack the characteristics
the gross morphology of the pollen grain, the liguli¯orae- that permit their unequivocal association with the Aster-
type characterized by echinolophate pollen, which is aceae.
generally found in the Cichorieae, and the tubuli¯orae-
The ®rst megafossil evidence for the Asteraceae may be an
type, which is found in many other tribes (Muller 1981). Eocene leaf, Hieracites salyorum from Aix-en-Provence,
Skvarla & Turner (1966) and Skvarla et al. (1977) using France (Saporta 1862), and Collinson et al. (1993) compared
gross morphology of extant pollen recognized three pollen a fruit recovered from the Eocene of Europe with the
types, i.e., psilate (this type can not be easily recognized as Asteraceae, however, other than super®cial appearance,
Asteaceae in the dispersed pollen record), echinate, and there are no other features to suggest af®nity with the
lophate pollen. Based on pollen wall ultrastructure, Skvarla Asteraceae. The Oligocene-Miocene taxon Viguiera cron-
&
Turner (1966) and Skvarla et al. (1977) further re®ned quistii from Montana (Becker 1969) was thought to be the
these pollen types and recognized ®ve major pollen wall ®rst unequivocal megafossil of the Asteraceae, however,
types, the Helianthoid, the Senecioid, the Arctotoid, the Crepet & Stuessy (1978) presented evidence to suggest that
Anthemoid type and pollen types without designated this fossil may represent a plant organ from a variety of
patterns. These pollen types are variously distributed other families. Despite the precociousness of the extant
among the Cichorioideae and the Asteroideae. Pollen of Asteraceae the megafossil record is meager, and provides
the Barnadesioideae is similar to the echinolophate pollen of little insight into the time and origin of the family. The
the liguli¯orae-type, but lacks the spines. Comparison of the dispersed pollen record of the Asteraceae suggests that the
pollen types of extant taxa in the Asteraceae with dispersed family was widespread by the Upper Oligocene and
pollen may provide insight into the time and place of origin subsequently diversi®ed.
of the Asteraceae.
This paper reports the occurrence of two asteraceous
#
2000 Taylor & Francis. ISSN 0017-3134
Grana 39 (2000)

40
M. S. Zavada and S. E. de Villiers
dispersed pollen types, Tubuli¯oridites antipodica Cookson and transmission electron microscopy (TEM). For light microscope
slides, residues where dehydrated in an alcohol series and mounted
1
the Upper Paleocene Eocene of South Africa. This
947 ex Potonie 1960 and T. viteauensis Barreda 1993 from
±
with D.P.X. (BHD Chemicals, UK), the coverslips were sealed with
nail polish and were examined using a Zeiss microscope. For SEM
the Koingnaas material was mounted on SEM stubs with DAG-580
and coated with gold-palladium and viewed with an Amray-1000
scanning electron microscope in the University of the Witwaters-
rand Electron Microscopy Unit.
constitutes the earliest unequivocal occurrence of asterac-
eous pollen in the fossil record. We amend the descriptions
of these taxa to include pollen wall ultrastructure and
discuss the importance of these taxa to the evolution of the
Asteraceae.
To isolate individual pollen grains from the pollen residues for
SEM and TEM, the pollen residues were ®rst dehydrated in an
alcohol series (10%, 50%, 95%, 100%, acetone, xylene) followed by
acetone and ultimately brought into xylene. The residue was then
mixed with a polystyrene based mounting medium ProTexx and
smeared on a microscope slide and allowed to polymerize without a
cover slip. The slides were then scanned, and individual pollen
grains were cut out of the hardened polystyrene and placed on a
MATERIALS AND METHODS
Samples were collected from two localities, one offshore locality and
one onshore locality. Samples were collected offshore on the
research vessel Shearwater Bay, approximately 50 km due west of
Orangemund and Alexander Bay (Text-Figure 1). They were
initially used for a micropalaeontological study (McMillan &
Dale, unpublished data) and those which appeared to contain
suf®cient organic debris were submitted to the Bernard Price
Institute (BPI) for palynological analysis. The onshore samples were
taken from a paleochannel at Koingnaas (Text-Figure 1). The site
consisted of a quarry as described by de Villiers & Cadman (1997).
The two sample suites are referred to in this paper as Shearwater
Bay meaning those of Eocene age derived from the continental shelf,
and Koingnaas meaning the onshore paleochannel sediments of
Paleocene-Eocene age. The maceration of the samples follow Faegri
0.05 mm pore Millipore ®lter that was on several sheets of ®lter
paper. Xylene was then dripped over the polystyrene square
containing the pollen grain until all of the polystyrene was removed.
The pollen grain was then transferred to an SEM stub that had been
previously prepared by melting a thin coating of the high vacuum
wax Apiezon W-100 on the stub (Wachtel 1980), on which a
numbered grid was inscribed. The pollen grain was placed within a
grid square. This was repeated until the 16 grid squares each
contained a single pollen grain. The SEM stub was coated with
gold-palladium and the pollen was viewed on an Amray-1000. After
photographing the pollen viewed with the SEM, the individual
&
Iversen (1992), Gray (1965), and Doher (1980).
pollen grains were then transferred to
a small agar block,
One portion of the residue was used to prepare slides for light
microscopy and scanning electron microscopy (SEM) at BPI.
Another portion was sent to Providence College for further SEM
dehydrated in an alcohol series, embedded in Spurr's low viscosity
resin, sectioned on a MT-2b ultra-microtome and viewed, without
staining, on a Philips EM-300.
A total of 24 specimens of Tubuli¯oridites vitauensis and 19
specimens of T. antipodica were photographed and studied with
light microscopy. Approximated 16 ± 26 specimens of each species
were viewed with the SEM and six specimens of each species were
examined with the TEM.
RESULTS
Samples recovered from the Shearwater Bay locality on
western continental shelf of South Africa were dated as
Eocene (54 ± 38 Mya).This is based on the occurrence of the
foraminifera, Globigerinatheka index, Turborotalia centralis
and Eponides lisbonensis (McMillan & Dale, unpublished
data). The palynological analysis, based on known age
ranges of Clavatricolporites, Oacolpopollenites and Porosi-
pollis supports this age determination. The Shearwater Bay
pollen assemblage is dominated by pollen of dicotyledonous
angiosperms (de Villiers & Cadman 1997, de Villiers 1998).
The most abundant palynomorphs are of Rhoipites, a genus
accommodating reticulate tricolporate grains of generalized
morphology. This pollen type accounted for about 50% of
the assemblage. Other common palynomorphs include
Nyssapollenites (7% of the assemblage), Retitriletes (5% of
the assemblage), and low percentages of Cyathidites,
Laevigatosporites, Proteacidites, Arecipites, Liliacidites,
Podocarpidites and Cupaniedites. Pollen exhibiting features
of the Asteraceae were present in all of the Shearwater Bay
samples and comprise 14 ± 27% of the total palynomorph
content depending on the sample.
Another suite of samples derived from Koingnaas locality,
which are sediments that in®ll a paleochannel adjacent to
the coast were subject to routine palynological analysis, and
Text-Figure 1. Locality Plan.
Grana 39 (2000)

Paleocene-Eocene Asteraceae
41
were found to contain similar asteraceous pollen in only 1/3 (Figs. 11, 12). Although there are occasional inter-elemental
of the samples, and comprise 2% of the total palynomorph spaces in this wall layer (Fig. 11), the lack of spaces may be
content. The samples from Koingnaas are considered older due to compression during fossilization. More numerous
than the Shearwater Bay samples and are dated as spaces may have been evident between these rod-like
Paleocene-Eocene based on the age range of various elements in living material. The columellae are thick, stout
palynomorphs in the assemblage (de Villiers, in prepara- structures that rest on a relatively thin homogeneous
tion).
footlayer (Fig. 11). The footlayer rest on a thin, slightly
more dense, endexine (Fig. 11).
Comments. ± Barreda (1993) compared the Argentinian
specimens of T. antipodica with those of T. vitauensis. The
latter species was found to be larger with smaller spines. The
South African species show the same variances. Barreda
Systematic Palynology
Tubuli¯oridites antipodica Cookson 1947 ex Potonie 1960
Description. ± The pollen of this species recovered from the
Paleocene-Eocene of South Africa is tricolporate, spherical
to subprolate (Figs. 1, 4) with a circular to lobate amb
(
1993) also compared T. vitauensis with Tubuli¯oridites sp.
of Partridge (1978), and concluded that the two forms might
be the same species. The South African specimens also
resemble those from Angola.
(Fig. 1), and oval outline in equatorial view (Fig. 4). The
pollen averages about 21 mm in polar diameter and about
2
3
0 mm in equatorial diameter. The colpi are long, extending
/4 or slightly more of the polar diameter (Figs. 1, 4). The
DISCUSSION
pores are circular to slightly lalongate (Fig. 3). The
sculpturing is echinate with supratectal spines that are
evenly distributed, about 1 ± 2.5 mm high and averaging
The pollen recovered from the Paleocene-Eocene of South
Africa shows af®nity with Asteraceae based on exine
morphology. Extant pollen of this family is generally
divided into three types, these are, a) psilate, b) echinate
and c) lophate, the lophate pollen type being further divided
into the echinolophate type and the psilolophate type
4
mm from the tip of each spine to the tip of the neighboring
spine. The spines taper to a ®ne point (Figs. 2, 5). The
tectum is microperforate (Figs. 2, 5) except at the tips of the
spines where it is imperforate. The pollen wall is comprised
of an ektexine and endexine. The ektexine is comprised of a
thick tectum that is made up of thick interconnected rods
that grade into a homogeneous layer just above the
columellae (Figs. 6, 7). The columellae are thick, stout
structures that rest on a relatively thin homogeneous
footlayer (Figs. 6, 7). The footlayer rest on a thick, slightly
more dense, endexine (Figs. 6, 7).
(
Skvarla & Turner 1966, Skvarla et al. 1977). The psilate
pollen of the Asteraceae has sur®cial characters that can
easily be confused with pollen from other families, which
make recognition of the dispersed fossil pollen with this
morphology dif®cult to assign to the Asteraceae. Echinate
and lophate pollen, however, are characteristic of the
Asteraceae and pollen of these morphologies can be easily
recognized in the dispersed pollen record. Pollen wall
ultrastructural data have been examined for a number of
taxa in the Asteraceae. Skvarla et al. (1977) have
distinguished ®ve basic groups of wall ultrastructural
patterns, the Helianthoid pattern, the Senecioid pattern,
the Arctotoid pattern, the Anthemoid pattern and taxa
without designated patterns. When the sur®cial characters of
the exine are coupled with pollen wall ultrastructural data,
pollen can often be associated with taxonomic groups below
the family level. Pollen recovered from the Paleocene-
Eocene of South Africa have characteristics indicative of the
Anthemoid pollen wall pattern (Skvarla & Turner 1966,
Comments. ± The specimens found in this study show
similarity to specimens recovered from the Oligocene of
Argentina (Guerstein 1990, Barreda 1993), from the Upper
Cretaceous and Tertiary of Australia (Martin 1973, Stover
&
Partridge 1973, Dettman & Jarzen 1988), and the Late
Tertiary of Africa (Partridge 1978). This taxon ranges from
the Upper Cretaceous to the Pliocene, becoming abundant
in a wide geographical area in the Miocene.
Tubuli¯oridites viteauensis Barreda 1993
Description. ± The pollen of this species recovered from the Skvarla et al. 1977). The Anthemoid pollen type is generally
Eocene of South Africa are isopolar, tricolporate, echinate, characterized by echinate, non-caveate exines, that have a
subprolate to spheroidal (Figs. 8 ± 10). Amb is subcircular to wall structure composed of thick basal columellae (Figs. 6,
slightly trilobate (Fig. 10) and is oval in equatorial view 7, 11) which support shorter levels of columellae or rods
(Figs. 8, 9). The colpi are long extending well into the polar alternating with internal tecta. This latter layer is described
region (Figs. 8 ± 10). The pores are circular and relatively in this paper as a tectal layer comprised of thick
large. The spines, unlike those found in T. antipodica, are interconnected rods (Figs. 6, 7, 11, 12) that are not
more widely spaced (Figs. 8 ± 10), averaging 5.5 mm from the necessarily vertically oriented. One of the differences
tip of each spine to the tip of the neighboring spines. The between the fossil pollen and extant taxa with this pollen
spines average less than 1 um high and are rounded at their type is that the layer of columellae supported by the thick
apexes (Figs. 8 ± 10). The pollen wall is comprised of an basal columellae are more vertically oriented and more
ektexine and endexine. The ektexine is comprised of a thick regularly shaped in the extant taxa, than the interconnected
tectum that is made up of interconnected rods that grade rods observed in this layer in the fossil pollen (Compare 6, 7,
into a thick homogeneous layer above the columellae (Figs. 11 with the Figs. 27 and 28 in Skvarla & Turner (1966), and
11, 12). The homogenous layer above the columellae appear Plate 21, Figures A ± D in Skvarla et al. (1977)). The layer of
to be comprised of very thick compressed rod-like elements interconnected rods is more comparable to a similar layer
Grana 39 (2000)

42
M. S. Zavada and S. E. de Villiers
Figs. 1 ± 7. Tubuli¯oridites antipodica. (1) SEM showing circular amb, echinate sculpturing and the extension of the colpi well into the
polar region, 62,500. (2) SEM of the same grain as in Fig. 1 showing the perforate tectum and the psilate, tapered, distal ends of the
spines, 65,000. (3) SEM of a grain showing the colporate condition and the slightly lalongate ora, 65,200. (4) SEM of a grain show-
ing the oval outline in equatorial view, perforate exine and long colpi, 62,000. (5) SEM of the same grain as in Fig. 4 showing similar
exine and spine morphology to that observed in the specimen in Figs. 1 ± 2, 65,000. (6) TEM of the pollen wall of the grain in
Grana 39 (2000)

Paleocene-Eocene Asteraceae
43
Figs. 8 ± 12. Tubuli¯oridites viteauensis. (8) SEM showing the long colpi which extend well into the polar region and the widely spaced
blunt spines, 61,800. (9) SEM showing the echinate sculpturing, the long colpus and oval outline in equatorial view, 61,800. (10)
SEM showing the long colpi, and widely spaced spines, note the columellate structures where the wall has been fractured (arrow),
61,800. (11) TEM showing the outer portion of the tectum (T) comprised of interconnected rods which is underlain by a thick homo-
geneous layer that appears to be comprised of robust rod-like structures (RL) note the occasional spaces among the rod-like (RL)
structures that rest on the short, stout columellae (C). The columellae are underlain by a relatively thin footlayer (FL) and a thin more
dense endexine (E), 617,000. (12) TEM showing the wall structure of the tectum and associated spine. The outer portion of the
tectum is comprised of interconnected rods that grade into what appear to be large, thick robust rod-like (RL) structures just above
the columellae layer. Note that the spine is comprised of a similar structure to the outer tectum with occasional tectal perforations
(
arrows), 624,000.
found in the Anthemoid-like pollen of the Valerianaceae, determined that the Anthemoid pollen wall is the primitive
Apiaceae, and Dipsicaceae (Skvarla et al. 1977), families type. Although this pollen wall type is named for the tribe
that have at one time or another been considered related to Anthemideae of the Asteroideae, it is more commonly found
the Asteraceae (Bremer 1994). The wall structure of the two in the Mutisieae of the Cichorioideae, a tribe that Bremer
dispersed pollen types from South Africa unequivocally (1994) consideres to be basal within the Cichorioideae and
establishes their af®nity with the Asteracaeae, and in the Asteraceae in general.
particular their af®nity with asteraceous taxa that possess
Determining the time and place of origin of the
echinate, non-caveate, Anthemoid-type pollen (mostly the Asteraceae based on the fossil record has been problematic.
Anthemideae and the Mutisieae). Skvarla et al. (1977), Some investigators place the origin of the Asteraceae in the
based on their studies of extant asteraceous pollen, Oligocene (Raven & Axelrod 1974, Muller 1981). Turner
Figs. 4 & 5 showing the tectum, which is comprised of interconnected thick rods that grade into a homogeneous layer just above the
columellae, the columellae lye on a relatively thin footlayer, which is above the thick slightly more dense endexine, the arrow indicates
the gold-palladium coating for SEM, 66,800. (7) TEM of the same grain (®gured in 4 ± 6), showing the nature of the tectal layer (T),
the thick, stout columellae (C), the thin footlayer (FL) and more dense endexine (E), arrow indicates the gold-palladium coating for
SEM, X 16,000.
Grana 39 (2000)

44
M. S. Zavada and S. E. de Villiers
(
1977), and Kemp & Harris (1975, 1977), however, opted for
Chenque Formation, Golfo San Jorge Basin, Southeastern,
Argentina. ± Palynology 17: 169 ± 186.
an Upper Cretaceous origin for the family, accepting many
of the reports of pre-Oligocene asteraceous pollen at face
value. Turner (1977) considered the Upper Cretaceous
origin important to explaining the cosmopolitan distribution
of asteraceous taxa as a vicariance event (in part). Bremer
Becker, H. F. 1969. Fossil plants of the Tertiary Beaverhead Basins
in southwestern Montana. ± Palaeontographica 127: 1 ± 142.
Bremer, K. 1993. Intercontinental relationships of African and
South American Asteraceae
±
±
a cladistic biogeographic
analysis. In: Biological relationships between Africa and
(1993, 1994) in a review of these reports considers a lower
Tertiary origin and does not consider the current distribu-
South AmericA (ed. P. Goldblatt), pp. 104 ± 135. ± Yale Univ.
Press, New Haven CT.
tion of the Asteraceae related to continental movements. Bremer, K. 1994. Asteraceae. Cladistics and Classi®cation.
There are a number of bona ®de Oligocene reports of pollen Timber Press Inc., Portland, OR.
±
of the Asteraceae and by the Miocene asteraceous pollen has Bremer, K. & Gustafsson, M. H. G. 1997. East Gondwana ancestry
of the sun¯ower alliance of families. ± Proceedings of National
Academy of Sciences 94: 9188 ± 9190.
a cosmopolitan distribution. There are a number of pre-
Oligocene reports of asteraceous pollen which are usually
Collinson, M. E., Bolter, M. C.
& Holmes, P. L. 1993.
based on the similarities of the exine morphology to the
morphology of extant asteraceous pollen, or by the assign-
ment of dispersed pollen to a form genus that is suggestive
of a Asteraceae af®nity, e.g. Tubuliforidites lilliei (Couper)
Farabee & Canright (1986). Whether or not these pre-
Magnoliophyta (Angiospermae). ± In: The fossil record 2 (ed.
Benton, M. J.), pp. 804 ± 841. ± Chapman & Hall, London.
Cookson, I. C. 1947. Plant microfossils from the lignites of the
Kerguelen Archipelago. ± BANZ Antarctic Research Expedi-
tion 1929 ± 1931, Rep. Ser. A 2: 129 ± 142.
Eocene reports of Asteraceae pollen are valid, awaits further Crepet, W. L. & Stuessy, T. F. 1978. A reinvestigation of the fossil
study of their pollen wall structure, however, the fact that all Viguiera cronquistii (Compositae). ± Brittonia 30: 483 ± 491.
of the earliest reports of asteraceous pollen are restricted to Dettman, M. E. & Jarzen, D. M. 1988. Angiosperm pollen from the
uppermost Cretaceous strata of southeastern Australia and the
Gondwanaland, suggest a southern hemisphere origin of the
Antarctic Peninsula.
Australian Palaeontologists 5: 217 ± 237.
±
Memoirs of the Association of
family. Bremer (1994) also suggests a southern hemisphere
origin for the family, and speci®cally places the origin in
South America. Based on the fossil record, however, an
Australian (Bremer & Gustafsson 1997) or African origin
de Villiers, S. E. 1998. Palynological report on the Shearwater Bay
sample suite. ± Internal Report. De Beers Marine, Johannes-
burg.
(
this study) is more likely with subsequent dispersal to
de Villiers, S. E. & Cadman, A. 1997. The palynology of Tertiary
sediments from a palaeochannel in Namaqualand, South Africa.
± Palaeontologia Africana 34: 69 ± 99.
southern South America, given that the earliest occurrence
of Asteraceae pollen in South America is in the Oligocene
(
Barreda 1993). This is possible especially given the close Doher, I. 1980. Palynomorph preparation procedures currently used
in the Paleontology and Stratigraphy laboratories, U.S.
Geological Survey. Geological Survey Circular 830, U.S.
proximity of the tip of South America with Antarctica, and
Antarctica with Australia maintaining a dispersal corridor
well into the Eocene. It is also likely that a South African
origin could lead to dispersal to any of these more southern
continents that form a dispersal corridor from South
America to Australia. It is imprudent at this time to suggest
that this early South African occurrence of primitive
asteraceous pollen represents the origin of the Asteraceae,
especially given the number of uncon®rmed or inadequately
studied dispersed pollen grains exhibiting asteraceous
features from as early as the Upper Cretaceous. It is
signi®cant, however, that the origin of this large cosmopo-
litan family does appear to be in Gondwanaland (Australia:
Bremer & Gustafsson 1997 or South Africa: this study), or
possibly South America (Bremer 1994) and it appears that
the Asteraceae has a longer evolutionary history than
previously suspected.
±
Dept. of Interior, Washington D.C.
Farabee, M. J. & Canright, J. E. 1986. Stratigraphic palynology of
the lower part of the Lance Formation (Maestrichtian) of
Wyoming. ± Palaeontographica B 199: 1 ± 89.
Faegri, K. & Iversen, J. 1992. Textbook of pollen analysis. 4th ed.
rptd. ± J. Wiley & Sons, Chichester.
Guerstein, G. R. 1990. Palinologia estratigra®ca del Terciario de la
Cuenca del Colorado, Republica Argentina. Parte I: Especies
terrestres de la perforacion Nadir No 1. ± Revista Espanola de
Micropaleontologia 22: 33 ± 61.
Gray, J. 1965. Extraction techniques. ± In: Handbook of palaeon-
tological techniques, P. III (ed. B. Kummel, & D. Raup),
Freeman & Co., San Francisco CA.
Kedves, M. 1971. Presence de types sporomorphes importants dans
les sediments prequaternaires Egyptiens. ± Acta Botanica
Academiae Scientiarum Hungaricae 17: 371 ± 378.
Kemp, E. M. & Harris, W. K. 1975. The vegetation of Tertiary
islands on the Ninetyeast Ridge. ± Nature 258: 303 ± 307.
Kemp, E. M. & Harris, W. K. 1977. The palynology of early
ACKNOWLEDGEMENT
Tertiary sediments, Ninetyeast Ridge, Indian Ocean.
±
The authors thank Ian McMillian and Dienne Dale of de Beers
Marine (Pty) Ltd for permission to quote their micropalaeontolo-
gical results. Thanks are due to de Beers for providing sample
material and for permission to publish these results. Dr Ann
Cadman and Professor Louis Scott were involved in the initial
stages of this project
Palaeontological Association London Special Papers, Palaeon-
tology 19: 1 ± 69.
Lang, J. & Meon-Vilain, H. 1976. Contribution a l'analyse
pollinique des bassins intramontagneux cenozoiques de Bayan,
de Yakawlang et du Ghorband en Afghanistan central.
Geobios 9: 425 ± 480.
±
Martin, H. A. 1973. The palynology of some Tertiary and
Pleistocene deposits, Lachlan River Valley, New South Wales.
REFERENCES
± Australian Journal Botany, Supplementary Series 6: 1 ± 57.
Barreda, V. D. 1993. Late Oligocene? ± Miocene pollen of the Muller, J. 1981. Fossil pollen records of extant angiosperms. ±
families Compositae, Malvaceae and Polygonaceae from the
Botanical Review 47: 1 ± 142.
Grana 39 (2000)

Paleocene-Eocene Asteraceae
45
Partridge, A. D. 1978. Palynology of the Late Tertiary sequence at
Site 356, Leg 40, Deep Sea Drilling Project. ± Initial Reports of
the Deep Sea Drilling Project 40: 953 ± 961.
morphology in the Compositae and in morphologically related
families. ± In: The Biology and Chemistry of the Compositae,
(ed. V. H. Heywood, J. B. Harborne & B. L. Turner)
pp. 141 ± 248. ± Academic Press, New York.
Potonie, R. 1960. Synopsis der Gattungen der Sporae dispersae, III
Teil: Nachtrage Sporites, Fortsetzung Pollenites. Mit General- Stover, L. E. & Partridge, A. D. 1973. Tertiary and Late Cretaceous
register zu Teil I ± III. ± Beihefte zum Geologischen Jahrbuch
9: 1 ± 189.
Raven, P. H. & Axelrod, D. I. 1974. Angiosperm biogeography and
spores and pollen from Gippsland Basin, southeastern Austra-
lia. ± Proceedings of the Royal Society of Victoria 85: 237 ± 286.
Thanikaimoni, G. 1977. Principal works on the pollen morphology
of the Compositae. ± In: The Biology and Chemistry of the
Compositae (ed. V. H. Heywood, J. B. Harborne & B. L.
Turner), pp. 249 ± 259. ± Academic Press, New York.
Turner, B. L. 1977. Fossil history and geography. ± In: The Biology
and Chemistry of the Compositae, (ed. V. H. Heywood, J. B.
Harborne & B. L. Turner), pp. 21 ± 39. ± Academic Press, New
York.
3
past continental movements.
Botanical Garden 61: 539 ± 673.
±
Annals of the Missouri
Saporta, G. de. 1862. Etudes sur la vegetation du Sud-Est de la
France a l'epoque tertiaire. ± V. Masson et Fils., Paris.
Skvarla, J. J. & Turner, B. L. 1966. Systematic implications
from electron microscopic studies of Compositae pollen ± A
review.
20 ± 256.
Skvarla, J. J., Turner, B. L., Patel, V. C. & Tomb, A. S. 1977. Pollen
± Annals of the Missouri Botanical Garden 53:
2
Wachtel, A. W. 1980. Thermoplastic wax for mounting SEM
specimens. ± Scanning 3: 302.
Grana 39 (2000)