Palinología

Review of Palaeobotany and Palynology
Elsevier Publishing Company, Amsterdam - Printed in The Netherlands
P A L Y N O L O G Y OF TERTIARY SEDIMENTS FROM TROPICAL AREAS
J. H. G E R M E R A A D , C. A. HOPPING ANI~ J. MULLER ~
Bataafse Internationale Petroleum Maatschappij, The Hag,e (The Netherlands)
(Received February 8, 1968)
SUMMARY
This article deals with the most important aspects of nearly twenty years of
intensive study of the pollen-and-spore content of Tertiary sediments in some parts
of tropical South America, Africa and Asia.
For a proper evaluation, the character of the data, including the selection
and preparation of the samples, the diversity of previous recording and the statis-
tically achieved uniformity in presentation of these basic data needs a full expo-
sition, given in the introduction. This is directly followed by an explanation of the
process of elimination of all stratigraphically unimportant species. The resulting
interpretation of climatic and topographical influences on the dispersal of pollen
and spores is illustrated with examples from the fossil record. The disturbing
effect of redeposition forms a problem, which in some cases can be solved. Now
that the main ways of dispersal of pollen and spores are understood, the charac-
teristics of the three major depositional environments can be distinguished by
purely statistical analysis, without necessarily having any botanical information
from probably related Recent plant species. Additionally botany and palae-
ontology may bring supporting evidence. This many-sided approach leads to the
discrimination between local and regional features of environmental or time-
stratigraphical significance which is needed for the evaluation of long-distance
correlation.
As a result the marker species can be classified into: (1) a restricted number
of pantropical marker species; (2) a larger number of marker species which oc-
curred in both the South American and west African regions, tropical today
(transatlantic distribution); and (3) a still greater quantity of species which are of
significance only within a single botanical province (intracontinental distribution).
Thus a broad stratigraphical framework on a pantropical scale is established,
which may be further subdivided regionally. These three systems of subzonation
are compared with independent zoopalaeontological time-stratigraphical cor-
relation and discussed in great detail, with special emphasis on the Carribean data.
1 Present address: Rijksherbarium, State University, Leiden (The Netherlands).
Rev. Palaeobotan. Palynol., 6 (1968) 189-348 189

The major palynological changes marking the boundaries of the pantropical
subzonation are thought to reflect the evolution of new groups of plants. They are
mostly marked by a gradual incoming of pollen types. Extinction of plants is
stratigraphically of less value, since they may have survived longer in one area
than in another.
Climatic boundaries are next in importance, but in general they are more
restricted to specific regions. Similarly the immigration of plants, although pro-
ducing sharp and useful boundaries, is only of regional value. Of least significance
for regional correlation are the locally restricted boundaries which are caused by
changes in habitat or dispersal. They may still be valuable for studies within one
basin.
An intriguing aspect of the palynological studies is formed by the possible
affinity of the fossil type with Recent botanical species. Such affinities are obviously
present in many fossil types. Whereas most are restricted to the level of family
relationship, some interesting cases of much closer affinity are recorded here. In
exceptional cases the morphogenetic development and migration of a restricted
group of related pollen types can be traced.
In the final section of this paper the species selected for this study have been
formally described and illustrated; they include several new ones. The study is
further documented by distribution charts and sections showing the stratigraphical
significance of the marker types, as discussed in detail in the stratigraphical section.
INTRODUCTION
This paper presents some of the results obtained in twenty years of palynolo-
gical investigation of Tertiary sediments from tropical areas by companies of the
Royal Dutch/Shell Group. The primary purpose of this investigation was to
arrive at a better interpretation of the stratigraphy of those Tertiary sedimentary
basins where other means of correlation failed. Of course it is not possible in
this paper to do more than present an outline of the large amount of work done
and to select for discussion a few topics which appear to be of more general
interest. Documentation will also be restricted to selected examples.
MATERIALS AND METHODS
The restriction of the subject of this paper to the Tertiary of tropical areas
is due to a combination of factors. Firstly, the tropics represent a natural geobotan-
ical unit. Secondly, palynological research by Shell was initially mainly concerned
with the Tertiary sediments of northern South America, Nigeria and Borneo.
Thirdly, the amount of detailed information on the Upper Cretaceous palynolo-
190 Rev. Palaeobotan. Palynol., 6 (1968) 189-348

gical succession in these areas is still very scanty and the base of the Tertiary
represents, therefore, at the present state of knowledge, a logical lower limit. Pleis-
tocene strata have been studied too, but Holocene sediments have been left out
of consideration because of lack of information.
Geographically the main emphasis will be put on the South American data,
which are by far the most extensive, covering Colombia, Venezuela, Trinidad and
the Guianas. From Africa only the Nigerian Tertiary succession is adequately
known. In the Far East, Borneo has supplied valuable information, but here the
stratigraphical coverage is incomplete, data on the Eocene being very limited.
The number of samples on which the study is based, is very large. Only
samples collected at the surface and cores and sidewall samples taken in wells have
been studied. As far as possible sections were chosen which contained, or were
situated in close proximity to, samples dated by other means.
Tropical Tertiary pollen floras are very rich in species and the average type
collection may easily contain 800-1,000 different species. For stratigraphical pur-
poses generally less than 200 are of importance per area. For a comprehensive
review, such as this, a further reduction is desirable and only 49 species are dis-
cussed.
These are, firstly, the species used to establish the major zonation, and some
which are of importance for elucidating local correlation problems. In addition a
few species of intrinsic botanical interest are discussed.
Fossil species have as far as possible been referred to published ones 1.
The introduction of new names has thus been kept to a minimum.
Preparation of samples was mainly by standard techniques, modified ac-
cording to local needs. For Neogene material treatment with HF, followed by
bromoform separation, is generally sufficient, although when much plant debris
is still present acetolysis may be helpful. For Palaeogene sediments slight oxidation
with Schulze's mixture may be advantageous.
The majority of the sediments examined consisted of clay and shale, silt or
siltstone with varying amounts of carbonaceous material. Sandy sediments or
coals were generally avoided.
The basic data are presented on distribution charts accompanied by all
pertinent lithostratigraphical and additional palaeontological information (Fig.l-
14). Again it must be emphasized that only examples are given. A complete
documentation would have been impossible because of lack of space. However,
the range chart (Fig.15) presents a compilation of all the data assembled for this
study.
Unfortunately, when palynological work was started in the different areas,
no uniform system of recording data was adopted. Sometimes percentages were
1 In this connection we should like to express our gratitude for the facilities extended to us by
Dr. T. van der H a m m e n (Amsterdam) and his staff, which enabled us to study the type collections
not only of already described species but also of types to be described in future publications.

calculated on the basis of a pollen sum; in other cases a rough estimate of the
abundance of the species was given.
In order to create uniformity in the presentation of these quantitative data
on our distribution charts, the occurrences are expressed as a value for the proba-
bility of re-observation, i.e., if a specified number of additional specimens from
a new sample of the same rock or stratum were to be investigated. This re-obser-
vation of the species may be in any number of specimens (grains), including one
only.
This chance of re-observation is computed according to the following
formula:
a M
P=I--(1-- N )
where a : the amount, that is the number of grains of the single species observed
in the old sample; N : the sum of the amounts of all species, that is the total
number of grains or pollen sum observed in the old sample; M : the sum of the
amounts of all species in the new sample.
Depending on the accuracy required by the palynological investigation and
the time and sample material available, the value of M may be taken larger or
smaller than 100. For technical reasons related to the type of computer at present
in use for palynological data processing, the value of M is here taken to be 88.
This conforms approximately to the current sample size of 70-I00 specimens of
the selected species. The advantage of such a relatively small sample size lies mainly
in the possibility of examining large numbers of samples within a short time, which
is essential for routine work, both stratigraphically and statistically, and is far
more informative than a few samples with large pollen sums.
In general the counting of the marker species takes place in two phases.
First, an analysis is made of the whole flora in which the dominant species like
Rhizophora are also counted. The second part of the microscopical investigation
is concerned only with the more rare but important markers, among which, e.g.
Asteraceae (Compositae). To save time dominant species are no longer counted,
but statistically it is not permissible to multiply the amount counted to get the
corresponding figure. Therefore, the probability of re-observation of, for instance,
Rhizophora pollen is calculated from the first pollen sum, but the probability of
re-observation of the rare markers like Echitricolporites spinosus from the second
pollen sum.
Example." I - - c o u n t o f t h e w h o l e f l o r a , i n c l u d i n g , i n t e r alia, Rhizophora -- 100 g r a i n s .
II : c o u n t o f t h e s p e c i a l s e l e c t i o n , e x c l u d i n g Rhizophora : 84 g r a i n s .
A m o u n t o f Rhizophora in ' T ' : 8 0 g r a i n s , p r o b a b i l i t y o f r e - o b s e r v a t i o n = ! - - (1 - -
8 0 / 1 0 0 ) ss : 1.00.
A m o u n t o f Echitricolporites sphwsus in " I " = l g r a i n , p r o b a b i l i t y o f r e - o b s e r v a t i o n =
1 - - (I - - 1/100) ss == 0 . 5 9 0 .
A m o u n t o f Echitricolporites spinosus in " l I " = 4 g r a i n s , p r o b a b i l i t y o f r e - o b s e r v a t i o n =
1 - - ( I - - 4 / 8 4 ) ss = 0 . 9 8 6 .

.~" GUAOUAS GROUP
.~ ;UADUASJ CACHO
SST BOGOTA~,~~ SAN FERNANDOSHALE MARGUA FORMATION ~
~D MEMBER 0
4~
'~ ~ 0
INFORMAL UNITS ---
n O
0 3" r-"
N ~ ~ ~ ,~,
{~, DEPTH OR SAMPLE NUMBER
I I I BUTTINIAANDREVI
~" I l J ~ PROTEACIDITESDEHAANI
STEPHANOCOLPITE$COSTATUS
= tit J J \ FOVEOTRILETESMARGARITAE
I J~ J LOHGAPERTITESVANEENDENBUKGi
%'1 RETIDIPORITESMAGDALENEHSI$
I I 0 ECHITRIPORITESTRIANGULIFORM]S
~o J J J ~ \ \ PROXAPERTITESOPERCULATUS
J J J %, 0 \ SPINIZONOCOLPITESGROUP
~" J J J CTENOLOPHONIDITESCOSTATU$
RETISTEPHANOCOLPITESWILLIAMSI
= I J J GEMMASTEPHANOCOLPITES GEMNATUS
I t J ANACOLOSIDITESCF LU|EOIDES
0 J J '% '% BOMBACACD
I TI EsPROXAC
PU
ERTSTIUESSANNAE
CTEHOLOPHONIDtTESLISAMAE ~--4
I I I %. FOVEOTRICOLPITESPERFORATUS
J I I ~ RETIBREVITRICOLPITESTRIANGULATUS
0 %. ',,, srelArelCOLPirEs CArAr(S~(ss
"~ J J I % \ RETITRtCOLPORITESIRREGULARIS ~- "~
I I I ,, O PSILATRICOLPORITESCRASSUS ~
I I I 0 ~ MONOPORITESANNULATUS
0 Q ~ 0 PSILATRICOLPORITESOPERCULATUS ~z r~
I I I HARGOCOLPORITESVANWIJHEI 0
~~ I I I 0 \ % \ R~ITRICOLPORITESGUIANENSIS ~
I~ I ] J ~ e ~ CICATRICOSISPORITES DOROGENS~S
~. 0 ~ 0 0 ~ PERISYNCOLPOR~TESPOKORNYI
~. J I ] \ % % VERRUCATOSPORITESUSMENSIS
STRATIGRAPHICAL DATA
RKER SPECIES SUBDIVISION
i-
~ ~ CARIBBEAN ATLANTIC PANTROPICAL
~ ~ ~ ~ x ~ ~ ~ ZONES ZONES ZONES
? JANDUFOURIA
SEANROGIFORNqS
/ MAGNASTRIATffES
/ HOWARD~
/ ! /
CICATRICOSISPORITES
DOROGENSIS
VERRUCATOSPORITES
USMENSIS
? RETtBREVITRICOLPITES
TRIANGULATUS
FOVEOTRICOLPITES
PERFORATUS
PROXAPERTffES
CTENOLOPHONIDITES RETIDIPORITES OPERCULATUS
LISAMAE NAGDALENENSIS
FOVEOTRILETES
MARGARITAE
)n of the symbols, see p.203.)

GEOLOGICAL DATA PALY
ROCK-STRATIGRAPHICAL FAUNAL DATA SELECTED POLLEN AND
UNITS
x
Z
142-164 / * *
229-245 / *
39G.414
476-496
561.578 * / /
857.670 ~
696-704 * O
709' /
773' / *
796.819 *
841.864
864 887
908' / /
954' * /
877' / /
1019'
1038'
1085'
1271.129 / * O * *
1292' / *
1337'
1382' / /
1453' O0 " 0
1454' *
1527'
1530' * *
1551' / / 0
1575' /
N
1602' / * /0
1625' //
o
1654'
1675' / *
1722' / * *
1732' / // *
1767' * * * 0
2ol9' / / /
2066' * * * /
2267' /
2312'
2335'
z
24~' *
z 2446'
2492'
Fig.2 (pp.195-196). Distribution chart well Chafurray-3, Colombia.

N O - S T R A T I G R A P H I C A L D A T A
SPORE M A R K E R SPECIES SUBDIVISION
-~ o~
CARIBBEAN ATLANTIC PANTROPICAL

= ~ ~ .~, ~- ~ o ~ ZONES ZONES ZONES
~o,
z~
/0 ./
GR~NSDALEA
I MAGNACLAVATA
/70 / /
OO O
.0
/O.
/ CRASSORET{TRILETES
VANRAADSHOOVENI
O O MULTIMARGINITES
VANDERHAMMENI
* o
/. /
//
//.
.O /
.O
0.
/0
00
0. /
O/ / /
t l l /
O
0 VERRUI"RICOI.PORITES
. RO[UNDIPORIS
MAGNA~TRIATITES
/ / flO~/ARDI
O0. /.
/O //
/0-
/ /
/ . /
/O
/O. /
I t /
tO
(ICATRiCOSISPORITES
DOROGENSIS
0.0
" VERRUCMOSPORITES
~ USMENSIS
/
G E L'O G IC A L DATA PA L Y N .ST R ATI G F
ROCK-STRATIGRAPHICAL FAUNAL DATA SELECTED POLLEN AND SPORE MARKER SPECIES
UNITS
s o +, ~ ~ +. ..~ o:
<
o ~ z - < ..-< =~
o g z g g i o~= _~5o
~o
~+
~o~
~c,< .~
- ~8 o~
RE 5452 /
545~ / El / /
i~ 544a .
~450 /
5447 /
54+6 /
5445
< 5443 /
= 61 49
53
58159 * * *
41 ll
RE 5433
5432 " /
61361 O0 0/'O
RE 5431
61 3~/
RE 5430 * /'
RE5429 ~ll / /
5428 / /
6J 30 / 0
29/" / O
RE 5427
6121 0 /
RE 5417 0~
5418 " " C *
/o 5416 0 *
~ 540q " / i l l
61 9 0
8 / 0 ,
7 O /lit
RE 5407
GI5 " O /OOQ /

=
~: RE5404 " / " O
< 5403 / *
s+o2 /leo
5401 O " / /
CATATUMBO 61 4 * / t
SANDS
RE 5397 /
5396 oo ooo.
NITO
JUAN 5380 / /
z 5374 / / /
5373 /
Fig.3 (pp.197-198). Distribution chart Rubio Road surface section, Venezuela.

~APHICAL DATA
SUBDIVISION
gz
ii! i ~
~ o ~-
CARIBBEAN
ZONES
ATLANTIC
ZONES
PANTROPICAL
ZONES
VERRUCATOSPORITES
USNENSIS
PSILATRICOLPORITES MONOPORITES
CRASSUS ANNULATUS
RETIBREVITRICOLPITE$
TRIANGULATUS
FOVEOTRICOLPITES
PERFORATU$
RETIDIPORITES
NAGDALENEN$1S
CTENOLOPHONIOffES
LISANAE
PROXAPERTITES
OPERCULATU$
FOVEOTRILETES
MARGARITAE
PROTEACIDITES
DEHAANI
~o
' \ 0 " 0 .0 --
\ 0
~ - . O" 0 \ . 0 ~ ~ 0 " "~
~ -0 \ 0 0 \ 0 \ \
. ~ - 0 \ \
o , o o o o o
0 0 0
oo
0 0 0 0
: : o : o : o !
~ 0 0 0 0 0 0
ol
0 0 0 0 0
oo
o o o o o o o o o o o , o , , o o o o o
0 0 0 0 0 0 0
o
0
I
>
SONVS JO dNO~9"3
I
o,c
o~
~c
\ - \ - \ 0 0 ~ \ 0 - 0 0 0 0 , ~
0 0 0 0 0 0 0 0 0 ~ 0 ~ . - 0 - 0 0 ~ . 0 0 ~ 0 0 - -
~ - . ~ . 0 ~ \ 0 0 . 0 0 0 . ~ -
\ , 0 . ~ \ O0 0 ~ 0 \ - ~ . \
0 ~ \ \ - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O \ \ \ 0 - ~ \ - \ 0
\ Q O ~ I I O G G I \ O I I I O t O I t l I t l O I , G O - ! 0 O , I !
. . \ . . O - ~ 0 0 - 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 , 0 0 0 0 0 0 0 0 0 0
~ 0
SaNVS 30 dNOEg"8
VOSIN
GROUP ~)
FORMATION
m
MEMBER -'C~ o
INFORMAL UNITS "1"

o
_
C
Z r--
--t
P ~" '~' ~' ~ ~ 'r' "~ ! DEPTH OR SAMPLE N U M B E R

o -~ ~ ~ ~ ~ ~ -~ I
J BUTTII41AANDREEVI
J PROIEACIDITESDEHAANI
SIEPflANOCOLPITESCOSTATUS
I FOVEOIRILETESMARGARITAE
J LONGAPERTITESVANEENDENBURGI
~ REIIDIPORITESNAGDALENENSIS
~ ECHITRIPORII"ESTRIANGULIFORMIS
Q Q * RROXAPERTIIES OPERCULATUS
0 0 (~ G 0 I ~ SPINIZONOCOLPITESGROUP
J CTENOLOPHONIDITESCOSTATUS
J GEMNASTEPHANOCOLPITESGEMMATUS
J ANACOLOSIDITESCF LUIEOIDES m
J PROXAPERTITESCURSUS r-
BOMRACACIDJTESANNAE ~)
J CTENOLOPHONIDITESLISAMAE
J FOVEOTRICOLPITESPERFORATUS
'% % " 0 I %" RETIBREVIIRICOLPIIES TRIANGULATUS
Q i STRIATRICOLPiTESCATATUMRUS
J RETIIRICOLPORITESIRREGULARI$ ~ "1~
J PSILATRICOLPORITES CRASSUS .~ ~.
J MONOPORITESANNULAIUS
PSILATRICOLPORITESOPERCULATUS 7~ t~
J MARGOCOLPORITESVANWIJHEI 0 "~
J. RETITRICOLPORITESGUIANEHSIS
J CtCATRICOSISPORIIESDOROGENSIS ~ Z
PERISYNCOLPORITESPOKORHYI C) 0
J VERRUCAIOSPORIfESUSMENSIS rn
J PERFOTRICOLPI[ESDIGITATUS "~
J PACHYDERMtTESD,EDERIX, ~
ZONOCOSTIIES RAMONAE
i, ECHIPERIPORII"ESESIELAE m
X
J ALNiPOLLENITESVERUS ~tO
J FLORSCHUETZIATRILOBATA ~ ~l>
JANDUFOURIASEANRO~IFORMIS ~ --I
J V~RRUTRICOLPORII'ESROTUNDIPORIS -- __
J MAGNASTRIATffES HOWARUI
i PSILADIPORITESMININUS ~'~
FLORSCHUEIZIASEMILOBA[A ~1~
i FLORSCHUETZIALEVIPOLI
J CRASSOREIITRILEI"ES VANRAADSHOOVENI ~1~
J MULTIMARGINlIES VANDERHAMMENI
GRIHSDALEAMAGNACLAVATA "I-
i FLORSCHUETZIAMERIDIONALIS
i ECHITRICOLPORli'ESSP~NOSUS --
i STRIASYNCOLPITESZWAARDI ('~
MULTIAREOLITES FORMOSUS
i FENESTRffESSPINOSUS
J ECHIIRICOLPORITESMCNEILLYI t'-"
I ~ n o
I ~.
I z
I
~z
O~
,J'l
Since Echitricolporites spinosus is part of the special marker selection, the probability
of re-observation of this species in I is of little interest and is not recorded on the distribution
charts presented. This example also shows the importance of further counting of the selected
species.
In order to avoid the impression of high accuracy and to facilitate visual

evaluation of the charts, the probability values on the distribution charts (Fig.l-14)
were grouped into classes and each class was indicated in the following way:
P 0.01-0.75 -- .
P 0.75-0.90 = /
P 0.90-0.95 = o
P 0.95-0.99 =
P 0.99-1.00 =
An interrupted vertical line on some distribution charts indicates a distri-
bution of a species which was not counted but which appeared common to abun-
dant in most samples of the interval examined.
INTERPRETATION OF' DATA
The practical purpose of our investigations can be briefly stated as the

establishment of a time-stratigraphical zonation, adapted to the scale of the
geological problem under investigation. To achieve this goal a discrimination
between stratigraphically reliable and unreliable pollen and spore types is necessary.
The problems may vary from detailed correlation between wells a few kilo-
metres apart in order to solve, for instance, tectonical complications, to broad
correlation between sediments of widely differing facies hundreds or thousands
of kilometres apart in different basins of deposition. It is of special importance to
evaluate, for every part of the zonation, the scale on which it is valid, together
with an estimate of the degree of accuracy.
The selection of the 49 species discussed and the zonation established in this
paper were preceded by a detailed study of the factors which influenced the regional
significance of these markers.
It is of course impossible to present a full account of this elimination process.
Instead an attempt will be made in this section of the present paper to outline the
principles involved with the aid of a few concrete examples. First the factors
influencing the composition of the pollen record in the rocks will be analysed.
Then the data from other disciplines, viz. botany, geology and zoopalaeontology,
which permit a better understanding of microfloral changes, will be discussed,
leading finally to the establishment of a zonation.
The pollen spectrum, obtained by calculating the percentages in which the
various species occur in a sample, is the end result of various factors which in-
fluence production and dispersal.
Rev. Palaeobotan. Palynol., 6 (1968) 189 348 203

Primary factors control the presence of the plants which produced the pollen
grains and spores, and secondary factors affect dispersal prior to fossilization.
Evohttionary change and migration
Primary factors are evolutionary change and migration or increases and

decreases in response to changes in climate or topography.
Evolutionary change may obviously be expected to yield the most reliable
criteria for the establishment of a time-stratigraphical framework of regional
significance. Examples are the gradual development of Poaceae (Gramineae)
and Asteraceae (Compositae), as reflected in the stratigraphical distribution of
the corresponding pollen types. Since these are large and varied groups of plants
with at present a virtually world-wide distribution, their first appearance and sub-
sequent development have proved to be of wide time-stratigraphical significance.
An example of evolutionary development on a subregional scale is the
evolution of Sonneratia pollen types in the Indo-Malesian area. As will be dis-
cussed in detail later, it was possible here to trace morphogenetic changes within
a single lineage.
However, such cases are exceptional, and pollen and spore species generally
appear fairly suddenly in the stratigraphical record without any indication of their
phylogenetic origin. Unless coinciding with a hiatus in the stratigraphy, these
relatively sudden appearances must have been caused by changes in the environ-
ment which enabled the parent plants involved to increase in numbers and extend
their geographical range. The main factor of interest here is climatological. The
problem of how to recognize the effect of changes in climate in the stratigraphical
distribution patterns of the many species studied can be approached in a direct
way by botanical identification of fossil pollen and spore types with living taxa,
which have well-defined climatological tolerances.
So far not much success has been achieved in this direction, since none of
the identified lowland species discussed in this paper are known to be indicative
of a specific climate except in so far as they are adapted to the tropical climate in
general. However, it is sometimes possible, by reference to climatologically sig-
nificant lithological changes, to detect the relative climatic requirements of certain
fossil markers. Such a case was present in the Oligo-Miocene (Magnastriatites
howardi Zone) of western Venezuela. Here a local accumulation of anhydrite in
shales, contrasting strongly with coal-bearing underlying strata, was associated
with a peculiar pollen flora not encountered elsewhere inthiscomposition. Here
the inference of a climatological change towards greater aridity was possible.
Another case in which a climatological cause may be detected is the rather
sudden disappearance of certain pollen species. The extinction of Nypa-type pollen
at the Eo-Oligocene boundary in Venezuela is approximately contemporaneous
with a widespread increase in the incidence of mottled shales, indicating a change

to a climate with more pronounced seasonal rainfall. It is conceivable that such
a change could have caused Nypa to disappear, since the plant at present is mainly
restricted to the ever wet tropics in the Indo-Malesian area.
The often rather sudden immigration of certain species may also have been
caused by a change in climate which enabled the plants involved to extend their
range. An example which could be explained this way is the rather sudden appear-
ance in the Eocene of western Venezuela of Retitricolporites guianensis and
Cicatricosisporites dorogensis. However, such phenomena may also be due to the
development of new pathways of migration. A good example of this is the south-
ward migration along mountain chains of montane elements such as Alnus. In
Borneo this could be attributed to a mid-Tertiary phase of mountain building. In
the Caribbean area the cooling of the climate at the beginning of the Pleistocene
must have facilitated the migration of Alnus along existing mountain chains from
the north.
An analogous case is the dispersal across oceans of plants which are adapted
to a coastal environment, such as mangroves.
If such floral changes are considered for the definition of a regional zonation,
the question of dispersal speed must be taken into account. Although little is
known quantitatively from Recent plants, it must be assumed that differences in
this respect exist and that they depend on factors such as adaptation and viability
of the seeds.
The conspicuously sudden appearance and subsequent rapid increase of
Asteraceae (Compositae) pollen types in the Lower Miocene of northern South
America, Nigeria and the Far East could reflect the adaptation to easy dispersal
of Asteraceae seeds.
In contrast one distinct case of delayed dispersal has come to light: Sym-
phonia pollen appears earlier in the stratigraphical record in Africa than in South
America and this obviously is a consequence of the difficulty of crossing the
Atlantic Ocean. It will be clear that such cases can only be detected when inde-
pendent dating is available, and they should of course not be utilized for the
definition of a regional time-stratigraphical zonation.
The majority of the floral changes observed on the distribution charts will,
however, have been influenced to a varying degree by local facies factors.
For instance, the varying abundance in which mangrove pollen occurs in
the sedimentary record, once it has made its first appearance in the area, largely
reflects the acreage covered by mangrove, which is determined by the local topog-
raphy.
It will now be clear that, if climatological and topographical influences are
of restricted lateral extent, the interplay of the environmental conditions on the
one hand, and the variable tolerances of different species on the other, combine
to produce complicated stratigraphical distribution patterns in which time-strati-
graphical correlation lines are not easily detected.

Dispersal of pollen grains and spores
In addition, the secondary factors controlling pollen and spore dispersal

may further obscure the patterns. The majority of the samples investigated are
deltaic-marine silts, clays and shales, which means that the individual grains were
generally transported, mixed, sorted according to size and possibly redeposited
once or several times.
Before giving a few examples of such complicated patterns, it is necessary
to discuss the influence of the secondary factors in greater detail. Relatively soon
after starting the investigation of Tertiary clastic sediments in the tropics, it was
realized that the conventional picture of a fairly homogeneous pollen rain, failing
impartially in various sedimentary environments, was inadequate for an under-
standing of the complicated stratigraphical distribution patterns encountered.
Studies of Recent tropical sediments have since shown that, in a humid environ-
ment at least, water transport may be of far greater quantitative importance than
wind transport (MULLER, 1959).
In the dispersal picture which has now evolved all species pass initially
through a phase of air transport. This phase may be of very short duration as, for
instance, in tropical peat swamp forests, where most of the pollen produced by
the trees was found to be deposited within a very restricted area. If the peat formed
here is preserved in the geological record as coal, no further transport takes place,
except when lumps of coal are later eroded and transported as pebbles.
At the other end of the scale, the pollen grains and spores produced by the
montane vegetation in the tropics are known to be transported in large quantities
over considerable distances by high-altitude atmospheric currents. This is especially
true of wind-pollinated trees such as Alnus, Pinus, and of certain ferns.
Sooner or later, however, mainly depending on wind strength, turbulence
and wash-out by raindrops, all pollen grains and spores land on the earth's surface.
If they are not soon carried away by water, they are mostly destroyed by corrosion
and escape fossilization.
A different fate, of course, awaits that portion of the pollen rain which is
caught in river systems or the sea. It makes a great difference whether further
transport takes place by river systems or whether pollen grains and spores settle
directly on the surface of the sea with further transport exclusively by sea currents.
In the first case, a small proportion of the pollen gathered from the entire
drainage area will be deposited in the alluvial and coastal plain sediments in the
delta area, but most of it will be concentrated at the river mouth and discharged
together with the sedimentary load into the sea. In this way concentrations of
pollen in fluviomarine sediments close to river mouths are formed, such as have
been found in front of the Orinoco delta (MULLER, 1959) and in the Gulf of Cali-
fornia (CRoss, 1966). Size sorting will hardly be noticeable during this phase and
tidal currents and turbulence will generally cause efficient mixing of the final

GROUP ~)
AMBROSIO FORMATION ~
m
MEMBER ET~ 0
~;O f"
ANOMALINA ZONE INFORMAL UNITS .~ 0

~ _
f3
C
Z ~
II
).
BUTIINIA ANOREEVI
PROTEACIDITESDEHAANI
STEPHANOCOLPITES(OSTATUS
FOVEOTRPLETESMARGARITAE
LONGAPERTffESVANEENDENBURGI
RETIUIPORITESMAGUALENENSIS
~ ~ t ECHITRIPORITESTRIANGULIFORMIS
" O " ~ \ % O " % % PROXAPERTITES OPERCULATUS
" 0 * ~ " ~ " * " 0 "~. ~ "~. O SPINIZONOCOLPITES GROUP
CTENOLOPHONIUITESCOSTATUS
RETIEi"EPHANOCOLPITESWILLIAMSI
GENMASTEPHAHOCOLPITES GENHATUS
ANACOLOSIDITESCF LU~'EOtDES
PROXAPERTITESCURSUS r-
BONBACAODITESANNAE ~)
CTENOLOPHONtDITESLISAMAE
FOVEOTREOLPITESPERFORA~'US (~
O O ~ Q O RETIBREVITRICOLPITESTRIANGULATU$
0 ' O " ~ ~ ~1 '% STRIATRICOLPITE$CATATUNBU$ (~
I~ * " O '~. RETITRICOLPORITESIRREGULARIS r-- "(2
r"
0 " 0 G ~ ~ PSILATRICOLPORITES CRASSUS m
z >
O '% MONOPORITESANNULATUS
~, %. PSILATRICOLPORITES OPERCULATUS 77
MARGOCOLPORITESVANWIJHEI ~ "~
%. %. 0 ~ %. G RETITRICOLPORITESGUIANEHSIS u'l
O ~ CICATRICOSISPORITES DOROGENSIS "o Z
'%. " O Q " PERfSYNCOLPORffESPOKORNYI ~ C~
V
VERRUCATOSPORITESUSNENSIS m
PEREOTRICOLPITESUIGITATUS
PACHYDERMITE$DIEDERIXI ~
ZONOCOSTITE$ RANONAE ~
ECHtPERIPORIEESESTELAE rn ~0
AENIPOLLENITESVERUS
FLORSCHUETZIATRILOBATA
JANDUFOURIASEANROGIFORMIS m
VERRUTRICOLPORITESROTUNDIPORIS n ~ n I i
HAGNASTRIATITESHOWARDI C~
PSILADIPORITESNINIMUS (4")
FLORSCHUETZIASEMILOBATA ~0
FLORSCHUETZIALEVIPOLI
CRASSORETITRILETESVANRAADSHOOVENI
NULTINARGINITE$ VANDERHANMENI
GRINSDALEANAGNACLAVATA
FLORSCHUETZIAMERIOIOHALIS
ECHITRICOLPO~ffESSPINUSUS 1
STRIASYHCOLPITESZWAARDI
NULllAREOLITES FORMOSUS
FENESTRITESSPINOSUS
ECHITRICOLPORtTESNCNEIELYI f"-
N~
N~
o; ~
Z
MICACEOUSSANDSTONE
B.GROUP OF SANDS
| O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 B O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
. . . . . . . . . 0 ~ . 0 . . . . 0 - , *
. - 0 ~ . . . . . . * 0 . 0 . , 0 0 0 ~ 0 . . 0 0 0 0 0
. . . . . . . . . . . . . * . . . .
) 0 ~ 0 0 0 0 0 . 0 0 0 0 0 0 0 0 0 ~ 0 0 0 0 0 0
. 0 , 0 . . ,
2o
bOO 0 0
D O Q 0 0 O t . - . - . O . . - O Q . O Q O Q Q O Q . .
D . . . . . . . . . Q "
_ _ .
O O O O Q O 0 0 0 0 0 0 [ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 . 0 0 ~
P .
. . . . , . 0 . . .
~o
0 0 0
0 0
, O O O 0
. O 0
e ~
~0
I
E
I.A ROSA LAGUNI~LAS
LOWER LAGUNJLLASSAND LAGUFIA BA(HAC)UER~
;a
N \ N @ \ ~
0 0 \ 0 \ \ ~ 0 ~ 0 0 \ 0 0 0 \ \ \ *
\ 0 0 0 0 - 0 0 0 0 0 ~
* 0 \
~ 0 0 0 0 ~ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 ~ 0 0 ~ 0 0 0 \ 0 ~
\
% 0 0 0 0 0 0 ~ \ 0 \
0 0 . \ " \ * ~ \
\ * 0 0 0 0 0 0
0 @ 0 0 0 0 ~ 0 \
\
0 0 0 0 \
O*
zz ~
o~
~ g
GROUP
LA
PUERTA FORMATION ~
m
MEMBER CZ~ 0
~ o
I N F O R M A L UNITS ~ t~)
1"3
C
Z r"
~. ~. ~ ~ ~ , DEPTH O R S A M P L E N U M B E R
RUTTINIAANDRERVI
PROTEACIDJTESDEHAANI
STEPHANOCOLPI~ESCOSTATUS
FOVEOTRILETESMARGARITAE
LONGAPERrI/ESVANEENDENgURG]
RETIDIPORIrESMAGDALENENSIS
ECHfTRIPORITESTRIANGUL!FORMr$
%. * PROXAPERTITESOPERCULATUS
SPINIZONOCOLPITESGROUP
CI'ENOLOPHONIDITE$COSTATUS
GEMNASTEPHANOCOLPffESGEMMATUS
ANACOLOSIDITESCFLUIEOIDES
PROXAPERTITESCURSU$ r-
BOMRACACJDITESANNAE ~)
CTENOLOPHONIOITE$LISAMAC
FOVEOTRICOLPITESPERFORATUS
RETIBREVITRICOLPITESTRIANGULATUS
G STRIATR~COLPITESCATATUMBUS (~
O '% RETITRICOLPORITESIRREGULARIS ~ "10
PSILATRICOLPORITESCRASSUS m 3>
NONOPORITES ANNULATUS
Q PSILATRICOLPORITES OPERCULATUS ~ r'-
MARGOCOLPOR~TESVANWIIHEI 0 "-~
\ \ RETITRICOLPORITESGUIANENSIS 7
~) CICATRICOSISPORITE$ DOROGENSIS "~
PERISYNCOLPORITESPOKORNYI ~ O
VERRUCATOSPORITESUSMENSIS m
PERFOTRICOLPITESDIGli'ATUS "~
PACHYDERMITESDIEDERIXI ~
ZONOCOSTITE$ RAHONAE ~ "-4
%. ECHIPERIPORITESESTELAE ~.~ ;10
ALNIPOLLENITESVERUS
FLORSCHUETZIATRILOBATA 3>
JANDUFOURIASEAMROGIFORMIS m .-4
VERRUTRICOLPORITESROTUNDIPORIS ~
MAGNAS)'RIAT ITES HOWARDI ~'~
PSILADIPORITESHINIHUS (~)
FLORSCHUETZIASEMILOBATA
~t CRASSORETITR~LETES VANRAADSHOOVENI 3>
%. MULTJHARGINIrE$VANDERHAMMENI
GRIMSDALEANAGNACLAVATA -I~
FLORSCHUETZIANERIDIONALIS
%, ~ ECHITRICOLPORITES SPINOSUS
STRIASYNCOLPITESZWAARDI (~)
MULTIAREOLIrES FORHOSUS
~. ~ FENESTRiTESSPINOSUS 3>
ECHITRiCOLPORITESMCNEILLYI I"
~ z
.= m> 3>
Z
wN 0
Z
s
N m'~
~N
XI
GROUP 0
C)
CATSHILL FORMATION ~, ~l
MEMBER ZC 0
INFORMAL UNITS ~ 0
~ _
Cl
r. >
C
Z r-"
~ ~ ~ ~ _ . . ~. ~ ~ ~. ~ ~ ~ .
o ~ ~ ~_ ~. . DEPTH O R SAMPLE NUMBER
BUTTINIA ANDREEVI
PROTEACIDllrES DEHAANI
STEPHANOCOLPITES COSIATUS
FOVEOTRILETES MARGARITAE
LONGAPERIITES VANEENDENBURGI
RETIDIPORtTES MAGDALENENSIS
ECHITRIPORITS TRIANGULIFORHIS
PROXAPERTITES OPERCULATUS
SPtNIZONOCOLPITES GROUP
CTENOLOPHOIqlOITES COSTAIUS
RETISTEPHANOCOLPITES WlL[IANSI
GEMMASTEPHANOCOLPITES GEMMATUS
ANACOLOSIDITES CF [UTEOIDES
m
PROXAPERTITES CURSUS
BOMBACACIDITES ANNAE
CTENOLOPHONIDITES LISAMAE
FOVEOTRICOtPLIES PERFORATUS
RETIBREVITRICOLPITES TRIANGULATUS "0
STRIAIRICOLPIIES CATAIUMBUS 0
%. %. " 0 * REIITRICOLPORIIES IRREGULARIS r--r'-
* PSILATRICOLPORITES CRASSUS m
MONOPORIIES ANNULATUS Z ~I>
). r--
0 %. '%. '%, %. PSILATRICOL PORITES OPERCULATUS Z
MARGOCOLPORITES VANWIJHEI ~ "~
RETITRICOLPORITES GUIANENSIS ' u~
CICATRPCOSISPORITES DOROGENSIS "~ z
(~ 0 " \ 0 %. ~E) %. 0 0 PERISYNCOLPORITES POKORNYI 0
I= VERRUCATOSPORfIES USHENSIS m
PERFOIRICOLPPTES DIGIIATUS -~
" "_ PACHYDERNffES DIEDERIXI ~I~ (a~
ZONOCOSTITESRAHONAE ~ "--I
ECHIPERIPORITES ESTELAE rn 3D
AI.NIPOLLENITES VERUS
FLORSCHUE/ZIA TRILOBATA
),
IANDUFOURIA SEAMROGI~ORMIS ~ .--t
0 \ \ ~ VERRUTRICOLPORITES ROIUNDIPORIS ~ m
%, %. %. O * %. %. NAGNASTRIAT IrES HOWARDI

PSILAOIPORIIES HINIMUS (~
ELORSCHUETZIASEHILOBATA ~(3
FLORSCHUETZIA LEVIPOLI
CRASSORETITRILETES VANRAADSHOOVENI
HULTIMARGINITES VANDERHAMMENI
%. * ' %. %. %. " %. GRIHSDALEA MAGNACLAVAI'A -r
FI.ORSCHUETZIA MERIDIONA[IS
0 ~ ~ %. " " " ~. ~. 0 " %. %. ECHITRICOLPORITES SPINOSUS
O STRIASYNCOLPITES ZWAARDI (~
HULIIAREOL ITES FORHOSUS
FENESTRITES SPINOSUS
0 %. ECHITRICOLPORIIES NCNEILLYI
g >
c~ o
Z
8 02
~. wN
O"
. . . . . \ \
B . O 0 \ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 \ \ 0 - . ~ 0 \ . . . .
0 0 ~ 0 , . \ . - \ 0 \ - \ O Q
\ \
.
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
\ \ \ '~ 0 -
0 0 %.
0 0 0 -
0 0 0 \ 0 O \ 0 0 0 - 0 0 0 ~
\ 0 . 0 - .
z~
~m
8~
~0
~N
>- <,o
z ~- I
4( snSONIdS S3XIaIS3N33 %~ * J
SNSOH~O:I S~UIO3~IVlI1FIN
O IO~VVMZ S311dlO3N~SVI~JS J
__ SFISONIdSS311~OdlO)I~IIH]3 G ~ ~ O ~ ~ ~ ~ J ~' Q
SllVNOIGI~]H 1ZI3NH]S~OIJ j
"1- YIVAYI]VNgVH V]IYOSHI~9 ~*~ (~ ~ *
O.. IN]HHYH~]~NYA S3JINIg~VHIIlnH J
IN3AOOHSOVV~NVAS]/]II~JII3~OSSVH) J
IlOdlA31 1zI3nH)S~OIJ I
tV ViVgOlfH3S VlZI3nH]S~OI3
SnHIN1H S]II~OdlGVIlSd J
(~ ~ m~VMOH $3111VIHZSVNDV~ J ",, N
- - - - SISOdlGNnIO~S311~OdlO31~IN~3A J
sn~3A S3Z~N311Od~N1V %. J
oc ~
3V13IS3 S311~OdlH3dlH)]
IXI~3031QS311H~]OAH]Vd ~ ~ ~.
U'I ~ SNIVII91(] S31ldlO)l~lO~3d "%
O ('I~ SISN3HS~ S3/I~OdSOJV)~3^
IXN~O~OdS3UaOdlO)NXSla3d (~ (~ %, %. ~ ~ %,
Z ~ SISN390~Od S311~OdSISO31~XY3D
SISN]NVIff9 S]11~OdlO31~1113~ J
Z~ 13HIIMNV^ S311~OdlO3OgaVN J
.~ < SQ1Vlfi)g3dO S3U~OdlO}IH[YIISd Q "% ~ "% " %, %. " "
sn/VlnMNVS31RiOdONOH \ \ ~ I0 ~
< z snssva~ s~uao,no)~a~v~s~ ",, "., J
n ~ SI~Vl~93~I S3Jl~OdlO]l~IlJ]~ " I"
S~8HQLVXY] S3IIdlO)lgIVl~JS ~ " (~
S~IVlngNVI~/S311dlO)l~LtA3~EI13~ J
~3 snzYuOJH3dsaltdlO)lUZO3AOJ J
]VHVSll S31101NOHdOION31) J
]YNNY S]llOI]V]V~HO9
S~S~N) S]III~3~VXO~ ~ J
$3~1031nl ~) S31IOISO103VNV I
SQIVHN39 S31~dlO)ONVH~31SVHH3~ J
ISHYllllM S3LIdlO3ONVH~3LSU3~
SNIVISO) S31101NOHdOION3[3 J
d~O~9 S]IldlO]ONOZINIdS [
SffIVlff)~3dO S]XIJ~3dVXO~d J
SIHHO~II~gNVI~XS311HOdI~IH)3
SISN]N31VOgVH S311HOdI~I[3~ J
19~SN]QN33NVA S3111H3dVgNO] J
3VJl~Vg~VH S3131I~I03AOJ \ J
SflXVlSO) S311dlO3ONVHd31S
INVYH30 S3UOI)V310~d I
IA33~ONV YINILLflg i
.,...-,N
<
k,-
~ Z
- ~
L'~ U S/INn "IVM~O::INI

o
,,,
o ~13BI~13 ~I J - ,vtav~
Z 01 NOIIISNYal
~ N O I I V I , q~lO:l vt~vd ,,]aim
dnO~O
\ \
\ . \ \ \
O 0 0 0 0 0 ~
O \
~ - . . \ \ ~ - 0 . 0 O \ ~
0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
, \ ~
\ \
0 0 0 0 0 ~ 0 @ 0 0 0 0 0 \ 0 0 \ 0 \ \ @ 0
~ \ ~ ~ . \
0 0 0 0 0 0 ~ 0 0 0 0 @ 0 O 0 0 0 0 0 0 0 0
\ O"
\ . .
\ \ . \
%. o
'] S~31d SV1
"~lnO,, ~ SV~"3ld SV1
1Yfl3Y)V5
G E O L O G I C A L DATA
ROCK-STRATIGRAPHICAL FAUNAL DATA

UNITS
T
Z
100J20'
120 140 ~ E El El
140.160' I~ El I~
160-180~ /
220 240' ~
240-260~ ~
260 280'
280300' .3
300320' 0!
340.360' ~
360 380' E ! El
I
380.400' C IO
400-420' I~ ~ Q
420.440' El
445-448' * / / ,/" *
504 $24' ~ El El
524.544' *
640 660 r * El El
396 140; 0 i0
I
445.146( / El ,/
460147 *
525.153, /'1//
550-155~ * El
570-15~(
590161(1 ~
662.167; t~
I600.170( ~
700-172CEl
/
" *
/'
757.176~ El ~
830-185( ~ ~)
897-190~ /
990.201(
2046.204~ El
064-208a
122-212! El
2177'
INOCERAMUS SP. 438-245; ~/
Fig.9 (pp.217-218). Distribution chart well Egoli-l, Nigeria.

PALYNO-STRATIGRAPHICAL DATA
SELECTED POLLEN A N D SPORE MARKER SPECIES SUBDIVISION
.~i -~ ~ r
~--~ o
~z ~ ~ ~_:
~ ~
::, > o ~ .~ o ~~ ~ ~o:'~
o~ ;: ~z ~ ~ ~~: o' ~ z CARIBBEAN ATLANTIC PANTROPICAL
~ z ~ ~ N ZONES ZONES ZONES
~o~ ~o ~g ~ ~
RETIDIPORITES
MAGDALENEHSIS
PROXAPERTITES
OPERCULATUS
PROTEAC[DITES
DEHAANI
.~ OROUP
~ n
1140 SHALE AMEKI IOGWASHI FORMATION X
.ASA~A
I
MEMBER ~ ~ m
o~"
" -~ 0
0
~ ~ INFORMAL UNITS ~-
= ~, ~> _
o ~
<o o~c Z r--
-. ~ ~
z >
-~ z -~ ~ ~ ~ ~.
o o o ~ ~ ~ ~ o ~ D E P T H O R SAMPLE N U M B E R
~" BU]TINIA ANDREEVI

PROIEACIDIIESDEHAANI
SIEPHANOCOLPiTESCOSfATUS
FOVEOIRILEIESNARGARI[AE
~' Q * ~ "
~
~
LONGAPERIITESVANEENUENBURGI
REIIDJP()RITESMAGDALENENSIS
0 (~ " '~. ~. " ~ ~. " 0 N. 0 ECHIIRIPORIIES IRIANGUUFORMIS
~ " ~ ~ Q '%. (~ 0 PROXAPERTITESOPERCULAIUS
SPINIZONOCOLPIIESGROUP
'%, " C1ENOLOPHONIDIIES COSTAIUS
~. '~ * ~) " " ~ ' RETISTEPHANOCOLPITESWILLIAMSI
GEI'4MAS1EPHAHOCOIPiTESGEMMATUS
+ ANACOLOSIOIIESCF LUTEOIDES
%. %. (~ " ~ * PROXAPERTITES(URSUS i--
BOHBACACIDITESANNAE ~)
CTENOLOPHONIDIIESLISAMAE
FOVEOIRICOLPIIESPERFORATUS
~ ~ " 0 0 %. 0 ~ RETIBREVITRICOLPIIES 1RIANGULATUS
" " 0 ~. %. STRIATRICOLPITES CATAIUMBUS 6
RETI|RICOLPORIIESIRREGULARIS ~r-
O . - ~ PSILAIRICOLPORITES CRASSUS ~ ~I~
" " O O ~ ; NONOPORITESANNULATUS
PSILAIRJCOLPORITESOPERCULAIUS ~ i'~
MARGOCOLPORITESVANWIJHEI (~ "~
REIITRICOLPORIIESGUtANENSIS
CICATRICOSISPORITESDOROGENSIS ~ Z
PERISYNCOLPORITESPOKORNYI
~) O %. VERRUCATOSPORITES USMENSIS m
0
PERfOIRICOLPITESDIGIIATUS '-~
PACHYDERNITES DIEDERIXI )" ('[)
ZONOCOST ITES RAMONAE ~ "4
ECHIPERIPORI/ESESIELAE rn
X)
ALN'IPOLLENITESVERUS
'~LORSCHUETZIATRILOBATA ~
}ANDUFOURIASEAMROGIFORMIS ~ ,--I
VERRUTRICOLPORITESROIUNDIPORIS I,
MAGNASTRIATITESHOWAROI
PSILADIPORIIESMINJMUS
FLORSCHUEIZIASEMILOBA'fA ~10
CRASSORETIIRILEIESVANRAADSHOOVENI
HULIIHARG~NITES VANDERHAMMENI "(3
6RIHSDALEAHAGNACLAVATA -r
FLORSCHUETZIAMERIDIONALIS
~CHITRICOLPORffESSPINOSUS
STRIASYHCOLPIIESZWAARDI {~
HULTIAREOLITES FORMOSUS
FENESTRITESSPINOSUS
ECHIIRICOLPORIIESHCNEILLYI
~ >
m
z
~ O~
GROUP
C~
NSUKRA
I INO SHALE
I ANEKI FORMATION
ME.BER
.~ t~)
;
RULLENNEWTON'S ~ 0
EOCENE NOLLUSC LOCALII'I INFORMAL UNITS ~ ~)
~ _
" I I N
I'~ RANGE Of CNILOGUEMBELINANARIINJ ~..
~ ~ z r-
I =~ '~= >r-
!~ ~ o
- l!i >-, >

DEPTH OR SAMPLE NUMBER
BUTTINIA ANDREEVI
PROIEACIDITES DEHAANI
STEPHANOCOLPITE$COSTA/US
FOVEOERILETESNARGARITAE
(~ " LONGAPERIiTES VANEENDENBURGI
~. RETIDJPORITESMAGDALENENSIS
~) '%. %`. ~) 0 '%. %` ~. 0 ECHJTRIPORITESTRIAHGULIFORMIS
~. (~ (~ %.. ~ ~. ~. PROXAPRRnTESOPERCULATUS
'%. " '%. 0 \ " 0 0 %`. CTENOLOPHONIDITES COSTATUS
%~. ~ ~ (~ RETISTEPHANOCOLPITESWFLLIAMSI
GENMASTEPHANO'COLPITESGENMATUS
[ - J ANACOLOSIDITRSCE LUTEOIDE$
j ".% ~ %, ~. ~,~1%` . O %` PROXAPERTITESCURSUS ~n
BOMBACACIDITES AHNA~ 0
t CTENOLOPHONIDITE$LISAMAE
I FOVEOTRJCOLPIIESPERFORATUS
J %` %` Q " %` 0 (~) \ %. ~ ~ '%. ~) Q ~ RETIBREVIIRICOLPITESIRIANGULATUS
j %` * ~ 0 " (~ (~ 0 ~11 ~) ~) %, (~ 0 "%` SIRIATRICOLPITESCATATUNBUS
RETIIRICOLPORITESIRREGULARIS r-
J 0 '%. %` PSILATRICOLPORITESCRASSUS ~-~
7
I ~. %` O \ %` %. (~ %, ' ~. O \ MONOPORITESANNULATUS
PSILATRICOEPORITESOPRRCULATUS ~. r--
J MARGOCOLPORII~S VANWIIH1 (~
J RETITRICOLPORITESGUIANENSIS
j CICATRICOSISPORnESDOROGENSIS .-~ Z
PERISYNCOLPORITESPOKORNYI 0 0
J %` E) (~ ~ ~E) (~ (~ VERRUCATOSPORITES
USMENSIS rn
J %` PERFOTRICOEPITESDIGITAIUS "~
. . . . . . . . . . . . . . PACHYDERMIIESDJEDERIXI )~"
I ZONOCOSTITES RAMONAE ~ "~I
1 ECHIP~RIPORIIESESTELAE ~:) ~0
J ALNIPOLLENIIES VERUS
J FLORSCHUETZIAIRILOBATA ..~ ~1~
JANDUFOURIA SEAMROGIFORMIS t~ --I
[ - - VERRUERICOLPORITES
ROTUNDIPORIS
J MAGNASTRIATIIES HOWARDI ~ --
I PSILADIPORITESMIN~MUS
FLORSCHUETZIASEMILOBATA ~10 .~
J FLORSCHUEIZIA LEVtPOLI ~>
J CRASSORETITRILETESVANRAADSHOOVENI ,~
J MULTIMARGINITES VANDERHAMMENI -0
GRIMSDALEAMAGNACLAVA[A "1-
J FLORSEHUETZIAMERIDIONALIS
JI ECHITRICOLPORIIESSPINOSUS
I STRIASYNCOLPIIESZWAARDI
MULIIAREOLITES FORMOSUS
(")
c/)
J FENESTRITESSPINOSUS ~
J ECHIIRICOLPORIIESMCNEILLYi r~
.~.
I ~ ~ o
~ o~ -
I ~ z *5
I _o
~ ~ ~ ~ z ..~
pollen suspension entering the sea. Also, the largest number of species will be
present in the fluviomarine sediments.
In the second case the composition of the pollen load is quite differently
affected. First of all sea currents act as slow transporting agents over wide fronts
with relatively less turbulence further offshore; the result is gradual settling and
deposition of pollen grains and spores, accompanied by size sorting. Wind-trans-
ported pollen may be added and this may considerably modify the composition
of the pollen load in areas far offshore. Another factor about which little is known
is selective corrosion of pollen and spores during seaborne transport.
One final complication has to be evaluated and this is the probability that
not all pollen is travelling in a single move from anther to final resting place, but
that temporary entrapment in sediments is followed by subsequent erosion and
that renewed transport takes place. This recycling of pollen grains and spores may
take place within what is geologically speaking an insignificant period of time, as
is the case in the average delta, where meandering streams may erode sediments
deposited only a short while before. Such recycling, of course, only increases the
degree of mixing of the pollen load eventually supplied to the sea and does not
affect stratigraphical distribution patterns to any appreciable degree. However,
when the time lag between deposition and recycling becomes greater, serious
difficulties may arise, culminating when older pollen-bearing formations are eroded
in the source area of a river system.
When a clear difference in preservation between such older reworked pollen
grains and spores and autochthonous ones is present, they can be easily excluded
from the pollen sum, but unfortunately this is usually the exception rather than
the rule. Generally there is no clear-cut difference, and the presence of many
slightly different corrosion patterns in one single sample may be the only clue
indicating the presence of several generations of reworked pollen grains and spores.
In such cases a detailed knowledge of the basic floral succession in the area of
investigation may enable one to spot the scattered presence of anomalous pollen
associations and to recognize them as reworked assemblages. Here general geolog-
ical considerations, indicating the probable origin of the sediment particles have
also to be taken into account. Recently VAN GIJZEL (1961) has demonstrated that
fluorescence microscopy may be able to discriminate between autochthonous and
reworked pollen.
In the Palaeogene sediments of western Venezuela, for instance, reworked
pollen was scarce because the bulk of the sediments deposited during that time had
been derived from the Guiana shield, where no pollen-bearing sediments were
eroded. In younger sediments the occurrence of reworked pollen of a Palaeogene
age proved to be clearly related to distinct unconformities.
In contrast, in the Neogene sediments of northwestern Borneo no such
clear-cut unconformities exist and since the sediments had been recycled many
times during the course of the Tertiary, a rather diffuse distribution of reworked

pollen was the result. It was not unusual to recognize in a single sample up to
three different generations of reworked pollen and spores, which created consider-
able difficulties in the interpretation.
This situation has practical consequences for the establishment o f a zonation,
since the oldest occurrence of species obviously is more reliable than the youngest.
In fact the regional zonation presented here is exclusively based on such oldest
occurrences. However, when cuttings have to be investigated for routine purposes,
the youngest occurrences may be of greater practical value.
Examples
The practical application of the above principles will now be illustrated with
a few examples.
The first case refers to base Monoporites annulatus Zone in northern South
America. Within the Eocene of western Venezuela, which was the first area inves-
tigated, two floral boundaries were apparent in approximately the same strati-
graphical interval: aal well-defined increase in Echitriporites trianguliJormis and,
slightly higher, a less conspicuous base of Monoporites annulatus. Later the absence
ofEchitriporites trianguliformis in the Eocene of Colombia was noted. This absence
could not be attributed to a stratigraphical hiatus or to a different climate, since
the remainder of the floral assemblage proved to be essentially similar. However,
the Colombian Eocene sediments are more terrestrial than those in Venezuela, and
it became clear that Echitriporites trianguliformis was derived from a coastal plant
with mainly waterborne dispersal of its pollen in seaward direction. This reduced
its value for regional correlation and, since Monoporites annulatus appeared to
have a wider distribution, because it was derived from a more inland environment,
its increase in the Early Eocene both in marine and more terrestrial deposits is
preferred as the regionally valid time-stratigraphical correlative horizon. The
marked increase in Echitriporites trianguliformis retains, however, its practical
correlative value within the restricted area of the Lower Eocene in the Maracaibo
basin.
In the case of the top in the occurrence of Proteacidites dehaani (Upper
Cretaceous) it was soon apparent that this event could be recognized both in
marine deposits in the northern Maracaibo basin and in alluvial plain deposits in
Colombia and the southern Maracaibo basin. Regional time-stratigraphical value
could, therefore, be assumed for this floral boundary.
In both these examples botanical identification of the pollen species was
impossible or did not provide sufficient evidence to permit an independent check
on the environmental requirements of the parent plants. The method outlined
above was the only feasible one for the recognition of time-stratigraphical value.
In the examples from the Neogene, described below, on the other hand,
this sort of conclusion is indeed supported by direct botanical evidence.

GEOLOGICAL DATA PALYNO-STRATIG
ROCK-STRATiGRAPHICAL
UNITS FAUNAL DATA ~ SELECTED POLLEN AND SPORE MARKER SPECIES
Z ~ _~
z ~ ~ ~ ~ ~ ~ ~ ~ ~
~- =o ~o o.~oo =~, ~-~
~6~o' / . .
z8~6, / . /
~;'~' 0
2931' .~ .
==- ~oo~, I /
3140' " ! "3
~4~o' .
3475' .
3723' / O O /
CHiEOGUEHBELIHA MARTINI/ 3775' / o ,/

COBENSIS ZONE
38~o' /
m ~ sgso, . . ..
3980' / O O /
40~5' " " " " 0
m%, O / ~ . .
4170' " " go 0/O

TRUNCOROTALOIDES
ROHRIZONE 4250' * * . . .
~= 43z5' , . . .
o oo, . . . / . . /
4415' "/ " " O
45,,0, / /0 - . /
4650, / " 0
4725' " Oi / /
4805'
CASSIGERINELLOITA 4880'
AMEKII ZONE
SOBS, / / / /
Fig.12 (pp.223-224). Distribution chart well Benin West-l, Nigeria.

RAPHICAL DATA
SUBDIVISION
~_~ _
Oz
>~ ~ ~1 ~ z ~ ! CARIBBEAN ATLANTIC PANTROPICAL
o ~ ZONES ZONES ZONES
az
VERRUTRICOLPORITES
RO1UNDIPOR~S
NAGHASTRiATITES
HOWARDI
CKCATRICOSISPORITES
DOROGENSIS
VERRUCATOSPORITES
USMENSIS
MONOPORITES
ANNULATUS
z
@ O - k
\ \ @ Q O O O . ~ @ O O O O O O O O ' O - @ @ O O O O @ O @
0 0 0 @ 0 0 0 @ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Q O Q O \ .
0 0 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \ - . 0 \ 0 0 0 - 0 0 0
\
0 \ ' . ' 0 0 " 0 " 0 - " 0 0 0 0 0 \ 0 0 0 - 0 0

0 - - \ \ ~ - . \ - ' - - " ' ~ 0 0 \ 0 0 0 \ \
0 0 0 0 0 O 0 0 0 0 0 O O O O - . ~ O O O O - O O O O k O
O O O @ \ @ @ O O O O O O O O O O O 0 0 0 0 0 Q O O O 0 \ 0 0 0
\
GROUP
BENIN FORMATION
A~AN CLAY MEMBER _Z - - 0

;-
INFORMAl U N I T S ~ O
~ _
c~
C
Z r-
0 0
DEPTH O R SAMPLE NUMBER
BUITINIA ANDREEVI
PROIEACIDITESDEHAANI
STEPHANOCOLPITESCOS~'ATUS
FOVEOTRtLETESHAR~ARITAE
LONGAPERTITESVANEENDENBURGI
RETIDIPORITESMAGDALENENSIS
ECHITRIPORII"ESTRIANGULFFORMIS
PROXAPERTITESOPERCULATUS
0 0 ~ CTENOLOPHONIDITES COSTATUS
RETISTEPHANOCOLPITESWrLLIAMSI
GEHNASTEPHANOCOLPITES GEMMATUS
ANACOLOSIDITESCF [UTEOIDES
rn
PROXAPERTITESCURSUS r-
BONBACACIDITESANNAE
CTENOLOPHONIDITESLISAMAE
FOVEOTRICOLPITESPERFORATUS
RETIBREVITRICOLPITESTRIANGULATUS
\ ~ STRIATRICOLPITES CATATUMBUS (~
REI'ITRICOLPORITESIRREGULARIS ~r-
(~ ~ '~. PSILATRICOLPORITES CRASSUS m
z )"
" 0 %. ' 0 ~. NONOPORITES ANNULATUS
'~, O ~ - PSILATRICOLPORITES OPERCULATUS 7'~ r--
MAEGOCOLPORiTESVANWUHEI (~ "-~
RETITRICOLPORffESGU[ANENSIS
CICATRICOSISPORITESDOROGENSIS z
PERISYNEOLPORITESPOKORNYI C) 0
VERRUCATOSPORfTESUSNENSJS m
PERFOTRICOLP~TESDIGITATUS "~
%, G PACHYDERMITES DIEDERIXI ~. IU~
ZONOCOSTITES RAMONAE ~ ""4
ECNIPERIPORITESESTELAE m ~XJ
ALNIPOLLENITESVERUS
FLORSCHUETZIAIRILOBATA -.~ ~"
JANDUFOUR~ASEAMROGIEOBMIS m --4
0 VERRUTRICOLPORITES ROTUNDIPORIS LJ~ --
(~ '~. ~ NAGNASTRIATITES HOWARD1
PSILAOIPORITESMINIMUS ~'~
FLORSCHUETZIASEMILOBATA
"%. ~ \ CRASSORETITRILETES VANRAADSHOOVENt ~1>
MULTIMARGINlIES VANDERHAMNENI
GRIMSDALEAMAGNACLAVATA "1"
FEORSCHUETZIANERIDIONALIS
ECHITRICOLPORITESSPINOSUS
STRIASYNCOLPII"ESZWAARDI C")
MULTIAREOLITES FORNOSUS
FENESTRfTESSPtNOSUS
ECN~TRICOLPORtTESNCNEILLY~ r--
z~ -~
o~ ~
zz ~
Z
~o ~_~ N z
~$ O~
o~
o
<m N
GEOLOGICAL DATA PALYNO-STRATIGR
ROCK.STRATIGRAPHICAL FAUNAL DATA SELECTED P O L L E N A N D SPORE M A R K E R SPECIES
UNITS
Z ~ i
- ~.,~ ~,
~ ~o ~.~-
3352' / ~ /
4~o, / / /
4376'
~ 44~,~,
4~so' O O
4'~o5'
49BO,
~o'
656o' 0 - -- --
6634' 'D ,
66S2' 0 .~
~,,s,~o, 0 / /
~e74' /
712e,
733o, /
7410' / /
7480' O "D O
/744' 0 00
B~os' /
8650, /
8,S~o'
8725' O ~ "
eeoo'
9~00' , .
q32o' 0 O
~370'
~4oo,
~zzo'
'~88o' / 0
10 000' go
Fig.14 (pp.227-228). Distribution chart well Lubara Creek-2, Nigeria.

APHICAL D A T A
SUBDIVISION
>
o -
~ N '~ CARIBBEAN ATLANTIC PANTROPICAL

z ~. ~ ~ ~ ZONES ZONES ZONES
~m~88 i
/
0
Q ECHII"RICOLPORII"ES
O SPIHOSLIS

0 Ib
/
0
/
f
0
? CRASSORETITRILErES
VANRAADSHOOVENI
/' 0
/,
CRASSOEETIrRLLETES
VANRAAOSHOOVENI
I GEOLOGICAL PALYNOLC
SUBDIVISION
zN
O Z PA N T R O P I C A L A TLANTI
J ZONES ZONES
z_,--
~~U
u
O~
~z
L 3
~ r
ECHITRICOLPORITES.
UPPER SPINOSUS
18
CRA$SORETITRILETE$
VANRAADSHOOVENI
LOWER
VERRUTRICOLP(
ROTUNOIP0f
26
MAGNASERIA!fTE$
HOWARD1
w
U CICATRICOSISPC
0 DOROGENSI
UPPER VERRUCATOSPORITES
USMENSIS
45_
U
0 MIDDLE MONOPORITES
ANNULATUS
49
LOWER RETIBRE~TRI(O
TRIANGULAT
54
U PROXAPERTIIES RETIDIPORI1
OPERCULATUS MAGDALEHEI
DANIAN
65
~
-- PROTEACID|
Fig.15 (pp.230-232). Range chart, z~ MAASTRICHTIAN DEHAAN
Ln
70
~GICAL SUBDIVISION I SELECTE
C I CARIBBEAN I BORNEO
ZONES ZONES
ALNIPOLLENITES
VERUS
ECHITRICOLPORITES
MCNEILLYI
FLORSCHUETZIA
NERWIONALIS
PACHYDERM~ES
DIEDERIXI
GRIMSDALEA
--
._MA~~_ACL._AV_.AT__A
-- LEYIPOLI
PSILADIPORffES
MINIMUS
,R~TES
)ANDUFOURIA
SEANROGtfORM~S
I FLORSCHUETZIA
~RITES
S I TRILOBATA
GUIANENSIS
OPERCULATUS
- ~I~r'~c'G'L~t~E~ "~
.PITES
US
FOVEOTREOLPITES
PERFORATUS
ES I CTEHOLOPHONIDITES
ISIS LISAMAE
FOVEOTRILETES
MARGARtTAE
D POLLEN AND SPORE MARKER SPECIES
In the Oligo-Miocene the first increase in the Rhizophora pollen type (Zono-
costites ramonae) can only be reliably observed in coastal and offshore marine
sediments, while the approximately contemporaneous first increase in the Ceratop-
teris type (Magnastriatites howardi) can be detected in terrestrial environments as
well. However, the small fresh-water ferns of the genus Ceratopteris can only
flourish in open vegetation and will hardly occur under a closed forest canopy,
which restricts the dispersal of the spores to water transport mainly. Ceratopteris-
type spores thus occur more regularly in inland environments, where Rhizophora
pollen can be extremely rare, and therefore their oldest occurrence has been taken
as the more reliable time-stratigraphical horizon. In practice of course both floral
changes support each other and are used in conjunction.
Another example from the Neogene of western Venezuela concerns the
recognition within the same body of strata of three sets of correlation lines:
(1) A topographically-influenced succession of mangrove dominance alter-
nating with more inland vegetation.
(2) A subregional succession, probably of a climatological nature.
(3) A top occurrence of Cicatricosisporites dorogensis.
The correlation lines derived from (1) cross at an angle those from (2),
while none of these were observed in the adjoining fully marine succession. How-
ever, the top occurrence of Cicatricosisporites dorogensis proved to be recognizable
regardless of environmental influence, and thus was judged to be of major time-
stratigraphical significance.
In general the principle may be formulated that any palynological change
which is paralleled by a facies change is of questionable time-stratigraphical value.
if, on the other hand, the pattern of palynological changes crosses the environ-
mental correlation lines derived from lithology or benthonic faunal distribution,
then that pattern is likely to be of time-stratigraphical value at least within the
area of study.
Statistical analysis of the palynological data
In the absence of positive information from lithology, faunal or botanical

affinities or climatic indications, statistical analysis of the palynological data can
be developed into a powerful tool for interpretation.
A first method is to measure the percentage variation, which is high in a
terrestrial environment where pollen dispersal is restricted and low in a marine
environment in which the pollen load has become thoroughly mixed. This variation
is best measured by calculation of the standard deviation, for which a minimum
of twenty samples is necessary. This calculation is, however, applicable only to
common species.
Secondly, an attempt can be made to separate statistically associations of
species which have a common source area. For instance, if alluvial-plain deposits
Rev. Palaeobotan.Palynol., 6 (1968) 189-348 233

are being investigated, the percentages of the species inhabiting this environment
will vary greatly, depending on the local edaphic factors, while species associations
from outside this environment and transported into it in a better-mixed composi-
tion will always show a constant ratio. If successful, such a statistical analysis
will enable the palynologist to separate, for instance, mangrove, river bank, peat
swamp, alluvial swamps and marshes, upland and montane associations. The
proportion in which these associations occur will then be mainly a function of
the general topography of the source area as a whole, while any changes occurring
within associations may have climatic or evolutionary significance and will be
of greater time-stratigraphical significance.
Association tests can be adapted for this approach.
Thirdly, the variation in occurrence between rarer types is best measured
with a run length test, which measures the variation in distance and length of
series of single occurrences. To be reliable a minimum of 80 consecutive samples
are needed for this test, which can also be applied to more common species when
a minimum percentage truncation is chosen.
All these statistical tests can be so devised that the significance of the results
is expressed in probability values. One disturbing factor may be the presence of
reworked pollen as erratic assemblages.
For large numbers of samples, calculation by means of mechanical data
processing is, of course, the only practical method of applying these statistical
tests.
Summary of interpretation experience
Summarizing twenty years of experience, it may be stated that for most cases
tools are available which allow discrimination between local and regional events
of time-stratigraphical value on the one hand and effects caused by time-crossing
events such as transgressions and regressions on the other. This information then
allows the establishment of a zonation adapted to the particular stratigraphical
problem at hand. For short distances and short time intervals the effects of striking
topographical events may provide the correlative framework; for correlation over
larger distances climatological changes will produce the best means, while on an
intercontinental scale correlation on evolutionary change appears to be the only
reliable method.
In general, however, the larger the distance, the more difficult it becomes
to define and trace sharp floral boundaries. In particular boundaries defined on
evolutionary change are notoriously hard to place accurately in thick continuously-
deposited sediments. This often poses a practical problem, since the geologist
generally requires correlation lines, while the palynological events on which the
zonation is based are by nature mostly gradual. In fact any sharp floristic boundary
is more probably indicative of a hidden unconformity than of anything else.

It is worth ending this chapter by listing below a few more of the limitations
encountered in applying palynological correlation methods.
(1) Determinable pollen and spores absent or scarce. This may be due to:
(a) original scarcity in sediments such as sands, conglomerates, limestones; (b)
destruction by oxidation shortly after deposition, e.g., mottled clays; (c) destruc-
tion by oxidation after weathering of outcrops, common in the more arid parts
of the tropics; (d) destruction due to high temperature effects caused by deep
burial, volcanic activity, proximity of intrusions, heat generated by tectonic
friction.
(2) Statistical errors. These fall mainly under two headings: (a) the use of
inadequate counts; (b) the use of an unsuitable selection of species which comprises
the pollen sum, if percentage ratios are calculated.
As regards (2a) the main practical problem is to achieve a maximum of
results with a minimum count. In practice a count of 70-100 selected marker
species with a sampling distance of 75-100 ft. has proved to be entirely adequate
for the solution of most correlation problems. Here, the number of grains counted
per sample has been sacrificed to the number of samples investigated.
If more than one investigator is counting the same material, the influence
of determination and counting errors should be eliminated by carefully designed
counting schemes, which will average out individual errors.
Errors grouped under (2b) are bound to influence the results especially when
floral changes are indistinct, as in the case of thick sedimentary series deposited
within a short interval of time, or when local influences in the pollen spectra obscure
the occurrence of scarce regional marker species.
Then it becomes imperative to test various percentage sums in order to
arrive finally at the combination of species which, in their changing ratios, will
reflect most distinctly the events of time-stratigraphical value on which the cor-
relation will be based.
This is an important point and it should be clearly understood that the
percentage calculation system employed in the present study is adapted to the
recognition of the zones described here. Other studies, for instance of a localized
area in a restricted interval only, will start with a different pollen sum.
In every area one of the principal problems in the early phase of developing
a zonation in the extremely rich tropical Tertiary microfloras has been to decide
which marker species to include in the pollen sum. Generally the marker species
have been divided into two groups. Group A comprises all species which are of
importance for the major regional zones, while group B contains those types
whose fluctuations are of more local significance. Again one of the advantages of
mechanical data processing is that this allows the rapid calculation, plotting and
comparison of percentages based on different pollen sums.

ZONATION
The zones recognized in this paper are biostratigraphical units in the sense
of the AMERICAN COMMISSION ON STRATIGRAPHIC NOMENCLATURE (1961). They
are, therefore, not a priori to be considered as time-stratigraphical units and in
fact some of the boundaries delimiting the zones probably are not contempora-
neous over large distances. However, within the areas in which the zones are
recognized, their succession is identical in all sections investigated.
According to their lateral extent, the zones are grouped as follows: (1)
pantropical zones; (2) transatlantic zones; (3) intracontinental zones.
The definition and description of the zones will be documented by a restricted
number of type sections, mainly taken from northern South America (Fig.l-14).
The range chart (Fig.15) illustrates the ranges of all species discussed, based on
evidence from a much larger number of sections. The independent dating of the
zones is discussed in the section "Independent dating".
Pantropical zones
The oldest pantropical zone recognized is the Proxapertites operculatus

Zone characterized by the regular co-occurrence of Proxapertites operculatus,
Proxapertites cursus, the Spinizonocolpites echinatus group, and Eehitriporites
trianguliformis. Its base, which as yet could only be studied in Nigeria and Borneo,
is provisionally defined by the qualitative base of the occurrences of Echitriporites
trianguliformis and the Spinizonocolpites echinatus group. Further study is necessary
to define and evaluate more accurately the base, which is probably of Senonian age.
The zone is well-developed in the Rubio-road section (Venezuela) (Fig.3),
well Egoli-1 (Nigeria) (Fig.9), and in the Lundu-Kayan section (Borneo) (MULLER,
1968). The transition to the overlying Monoporites annulatus Zone is marked by
the first regular occurrence of Monoporites annulatus. This is not a sharply defined
event, since the increase in Monoporites annulatus is rather gradual with alter-
nating periods of higher or lower abundance.
In the Rubio-road section (Venezuela) (Fig.3) the base of M. annulatus is
rather well defined, but this is a stratigraphically condensed section. In well Icotea-I
(Venezuela) (Fig.4), the increase is much more gradual. Since it is known that this
well penetrates a very thick Eocene section without any marked gaps, the record
shown here is probably a good reflection of the rather irregular increase in Mono-
porites annulatus in this part of the world.
In Nigeria the M. annulatus Zone in the Ovim Bende section (Fig.ll) also
shows a well-pronounced base, but in the Imo-river section (Fig.10) the base is
less distinct. In Borneo the base has not yet been accurately fixed stratigraphically.
M. annulatus is absent from all Upper Cretaceous and Paleocene sediments
examined so far and is known to occur throughout the Neogene. Its base can,
therefore, be expected within the Eocene there also.

The boundary with the overlying Verrucatosporites usmensis Zone is defined
by the first occurrence of Verrucatosporites usmensis. However, there is a fairly
wide interval in which transitional forms between Verrucatosporites usmensis and
the presumed ancestral forms with a low verrucation occur. Moreover, there is
evidence that the increase in number of Verrucatosporites usmensis takes place
earlier in Nigeria than in the Caribbean area. Consequently this boundary is not
a sharp one. In the Caribbean area the base in the occurrence of Cicatrieosisporites
dorogensis appears consistently to be below the base of V. usmensis and serves as
a useful additional criterion.
In the Paz del Rio section (Colombia) (Fig.l), the Verrucatosporites usmensis
Zone is thus characterized by regular occurrences of Cicatricosisporites dorogensis
and scattered occurrences of Verrucatosporites usmensis.
In the Ovim Bende and Imo-river sections (Nigeria) (Fig. 10, 11), the bound-
ary between the Monoporites annulatus Zone and the Verrucatosporites usmensis
Zone is sharply defined as these sections are incomplete, while in Benin West-I
(Fig.12) the change is slightly more gradual.
In Borneo the boundary has not been sufficiently studied as yet.
The boundary between the Verrucatosporites usmensis Zone and the over-
lying Magnastriatites howardi Zone is marked by the first appearance of Magna-
striatites howardi.
In the Paz del Rio section (Colombia) (Fig.l) this boundary is sharply
defined, in Chafurray-3 (Colombia) (Fig.2) less so.
In Benin West-1 (Nigeria) (Fig.12) a well-defined boundary ispresent, but
in the Imo-river section (Fig.10) recognition of the Magnastriatites howardi Zone
depends on a rare occurrence of M. howardi in one sample only. In Borneo the
stratigraphical position of this boundary has still to be clarified.
The boundary with the overlying Crassoretitriletes vanraadshooveni Zone is
marked by the base of the regular occurrence of Crassoretitriletes vanraads-
hooveni.
In Chafurray-3 (Colombia) (Fig.2), this boundary is weakly pronounced,
but in B-188 (Venezuela) (Fig.6), a sharply defined base is present.
In Okoloma-1 (Nigeria) (Fig.13), the base is rather well defined. In Borneo
the base of the occurrence of Crassoretitriletes vanraadshooveni more or less coin-
cides with the base of Florschuetzia levipoli and is slightly older than in the Carib-
bean area.
The youngest boundary is the base of the Echitricolporites spinosus Zone
and is marked by the base of the regular occurrence of Echitricolporites spinosus.
In B-188 (Venezuela) (Fig.6) and in CO-85 (Trinidad) (Fig.7) the boundary is
well defined.
In Lubara Creek-2 (Nigeria) (Fig.14) the increase in Echitricolporites is
gradual and fairly irregular, and the location of the boundary is, therefore, difficult
to determine.

Transatlantic zones
The pantropical Proxapertites operculatus Zone can be subdivided in the

Caribbean area and in Nigeria into three transatlantic zones. The oldest is the Pro-
teacidites dehaani Zone.
This zone is characterized mainly by the co-occurrence of Proteacidites
dehaani and Buttinia andreevi, together with high percentages of Foveotriletes
margaritae. The boundary with the overlying Retidiporites magdalenensis Zone is
taken at the top occurrence of Proteacidites dehaani. In the Rubio-road section
(Venezuela) (Fig.3) this boundary is well defined.
In Egoli-I (Nigeria) (Fig.9) the boundary is also rather well marked, mainly,
however, by the decrease in Buttinia andreevi, which was absent in the previous sec-
tion, although it is known to occur elsewhere in the Caribbean area within the zone.
The Retidiporites magdalenensis Zone is mainly negatively characterized by
the absence of Proteacidites dehaani and Buttinia andreevi. Foveotriletes maragritae
is still fairly frequent, although it may diminish in abundance in the higher part of
the zone. The zone is further characterized in both areas by the regular presence
of Retidiporites magdalenensis, Echitriporites trianguliformis, and Proxapertites
operculatus.
The boundary between the Retidiporites magdalenensis Zone and the over-
lying Retibrevitricolpites triangulatus Zone is marked by the first appearance of
Retibrevitricolpites triangulatus, Striatricolpites catatumbus, and Psilatricolporites
crassus.
In the Rubio-road section (Venezuela) (Fig.3) this boundary is clearly
defined.
In the Ovim Bende section (Nigeria) (Fig.l l) the boundary is also fairly
distinct.
In western Venezuela this boundary is further marked by the first appearance
of Retitricolpites irregularis and the disappearance of Retidiporites magdalenensis,
Proxapertites cursus, Bombacacidites annae, Ctenolophonidites lisamae, and Foveo-
tricolpites irregularis. It is assumed that this pronounced floral change coincides
with an unconformity.
The pantropical Magnastriatites howardi Zone can further be subdivided
both in northern South America and Nigeria into two transatlantic zones, the
Cicatricosisporites dorogensis Zone and the overlying Verrutricolporites rotundiporis
Zone. The boundary between these two zones is marked by the more or less simul-
taneous decrease in Cicatricosisporites dorogensis and increases in Verrutricolporites
rotundiporis and in Zonocostites ramonae.
In Chafurray-3 (Colombia) (Fig.2) this boundary is well marked, but in
Benin West-1 (Nigeria) (Fig.12) only one sample was available, which could be
assigned to the Verrutricolporites rotundiporis Zone, and the evidence here is not
as strong.
238 Rev. Palaeobotan.Palynol., 6 (1968) 189-348

Preliminary results indicate that the major increase in Zonocostites ramonae
in Borneo takes place roughly at an equivalent stratigraphical level. The other
two marker species do not occur here.
Intracontinental zones
Caribbean area
In the Caribbean area the transatlantic Retidiporites magdalenensisZone can
be subdivided into three units. The oldest unit is the Foveotriletes margaritae
Zone, which is characterized by the co-occurrence of frequent Stephanocolpites
costatus, Foveotriletes margaritae, Longapertites vaneendenburgi, and Gemma-
stephanocolpites gemmatus, and by the absence of Bombacacidites annae and
Ctenolophonidites lisamae.
The boundary with the overlying Ctenolophonidites lisamae Zone is taken
at the first occurrence of Ctenolophonidites lisamae and Bombacacidites annae, and
this zone is further characterized by the regular presence of Gemmastephanoeolpites
gemmatus, diminishing quantities of Foveotriletes margaritae, and the regular
presence of Proxapertites cursus.
The base of the youngest Foveotricolpites perforatus Zone is defined by the
first occurrence of Foveotricolpites perforatus, which species is restricted to the
zone. Bombacaeidites annae and Ctenolophonidites lisamae are frequent in this
zone, while Stephanocolpites costatus, Foveotriletes margaritae and Gemmaste-
phanocolpites gemmatus have virtually disappeared.
This subdivision is clearly visible in the Rubio-road section (Venezuela)
(Fig.3).
The Monoporites annulatus Zone can be subdivided in the Caribbean area
into four units. The boundary between the Psilatricolporites crassus Zone and
the overlying Psilatricolpites operculatus Zone is taken at the lowest occurrence
of Psilatricolpites operculatus, as visible in Icotea-1 (Venezuela), (Fig.4) and the
Prevenci6n section (Venezuela) (Fig.5).
The next higher boundary between the Psilatrieolpites opereulatus Zone and
the Retitricolporites guianensis Zone is placed at the base regular occurrence of
Retitricolporites guianensis. This boundary is present in the Prevenci6n section
(Venezuela) (Fig.5).
The Verrutricolporites rotundiporis Zone can be subdivided into the Jan-
dufouria seamrogiformis Zone and the overlying Psiladiporites minimus Zone. The
boundary between these two zones is marked by the base of the regular occurrence
of Psiladiporites minimus, as is visible in B-188 (Venezuela) (Fig.6). The Cras-
soretitriletes vanraadshooveni Zone can be subdivided in northern South America
into the Multimarginites vanderhammeni Zone and the overlying Grimsdalea magna-
clavata Zone. The boundary between these two zones is marked by the base of the
occurrence of Grimsdalea magnaclavata. This is clearly visible in B-188 (Venezuela)
(Fig.6).

The Echitricolporites spinosus Zone can be subdivided into three zones.
The boundary between the Pachydermites diederixi Zone and the overlying
Echitricolporites mcneillyi Zone is marked by the base of the regular occurrence of
Echitricolporites mcneilly.
This boundary is distinct in CO-85 (Trinidad) (Fig.7) and Paria-1 (Vene-
zuela) (Fig.8).
The boundary between the Echtricolporites mcneilly Zone and the Alnipol-
lenites verus Zone is placed at the base of the regular occurrence of A. verus. In
Paria-1 (Venezuela) (Fig.8) this boundary is well developed.
Borneo
In Borneo a local subdivision can be established, based exclusively on the
evolutionary development within the genus Florschuetzia (Fig.16).
The lowermost zone, FIorschuetzia trilobata Zone, is characterized by the
presence of F. trilobata and the absence of younger forms. The zone probably
covers the upper part of the Verrucatosporites usmensis Zone and the lower part
of the Magnastriatites howardi Zone, but the stratigraphical position of its base
has still to be determined precisely.
The first development of Florsehuetzia semilobata and F. levipoli marks the
base of the next higher Florschuetzia levipoli Zone, which coincides approximately
with the weakly defined base of the Crassoretitriletes vanraadshooveni Zone. In
this zone FIorsehuetzia trilobata decreases in numbers, while F. semilobata dis-
appears. The average size of F. levipoli increases also from 30 /z to 35 /~. The
overlying Florschuetzia meridionalis Zone is characterized by the regular presence
of F. levipoli and, in increasing quantities, of F. meridionalis, while the average
size of both species shows further increases. Florschuetzia trilobata is still present
in the lower part of this zone, but disappears soon.
The boundary between the Florschuetzia levipoli and Florschuetzia meridio-
nalis Zones coincides approximately with the base of the Echitricolporites spinosus
Zone, and the first regular occurrence of E. spinosus and Florschuetzia meridionalis
can be taken as a criterion for its recognition.
Both species have the disadvantage that in the very thick Miocene sediments
of northwestern Borneo, their increase in number is gradual, which makes it
impossible to recognize a sharp boundary.
INDEPENDENT DATING
In the virtual absence of any age-indicating plant macrofossil evidence with

which the palynological data could be correlated, associated animal fossils provide
the only check on the age of the proposed zones. For this purpose most reliance
has been placed on the age-indicating Foraminifera and, to a lesser extent, on

~zRRRTT
--
~ oN CASEOLARIS TYPE ALBA TYPE
~- o i 00 o/o o 1% 0 100 Jo o IOio
I I I I I I I I
.(.<
8~
.....
N ....." ' ...................." ~ " ;"~ ~ ~
1o- ~ "/~:' :;- ,
3
I ,,z,
u ..Y
~o. ~
~ i ~~ ' " I : ~ . rX ''' T R A N S I T I O N A L
~o- ~, ~, - ..... -. ::.i ..,;):;~
U < ~ " ,
0 ~ ' =%
I/I
. //
, < ;~'=
tim PERCENTAGES CALCULATED ON TOTAL FLORSCHUETZIA COUNTED
Fig.16. D e v e l o p m e n t of Florschuetzia pollen types in B o r n e o .
the Mollusca. Not all areas investigated have produced a sufficient number of
check points and it has also only rarely been possible to tie a floral change closely
to a faunal change. In general, however, the available evidence indicates that the
floral zones recognized are not markedly diachronous over the areas in which
they are considered valid. The boundaries between the zones, however, can only
rarely be sharply defined, and may not be exactly isochronous over larger dis-
tances.
Future detailed studies will no doubt provide further evidence to narrow
down the gaps in knowledge which still exist.
In the account given below the floral zones will be discussed chronologically,
a list being given of the age-indicating animal fossils, the faunal zones of which
these are characteristic and, as far as possible, the standard time-stratigraphical
units with which the zones are correlated. For the Palaeogene this is reasonably
well established, but for the Neogene considerable uncertainty still exists, and no
attempt will be made to relate the palynological zonation to the European stages.
Instead the planktonic zonation established by BOLLI (1966) will be taken as yard-
stick and his subdivision in terms of Oligocene, Lower, Middle and Upper Mio-
cene and Pliocene is adopted as a fair approximation. For Borneo reference is
made to the well-known letter classification of Van der Vlerk and Umbgrove (ref.
LIECHTI, 1960). Nomenclature of Foraminifera is according to LOEBLICH and
TAPPAN (1964).
It was not possible to present a full documentation for this part of the
present paper, except for those sections shown for the purpose of defining the
zonation and in which the occurrence of age-indicating fossils is recorded. How-
ever, in all other instances where a palynological zone is mentioned as associated
with certain animal fossils, this is based on actual co-occurrence in the same
sections.
(a) Proteacidites dehaani Zone
In western Venezuela and Colombia this zone is associated with the smaller
Foraminifera GIobotruncana gansseri, G. lapparenti, G. stuarti, Guembelitria ere-
tacea, Siphogeneroides bramletti, and the ammonite Sphenodiscus sp. In Nigeria
it is associated with the smaller Foraminifera Bolivina afra, Rugoglobigerina rugosa,
and striate Heterohelix spp., with the ammonites Didymoceras sp., Libyoceras
ismaeli and Sphenodiscus sp., and with the lamellibranchiat Inoeeramus sp.
This faunal list clearly indicates a Maastrichtian age. So far no evidence of
an older age has turned up, but since the lower limit of the Proxapertites opereu-
latus Zone has not been properly studied, the lower part of the Proteacidites
dehaani Zone may of course be older than Maastrichtian.
Unfortunately, both in Nigeria and in the Caribbean area, the top of the
Proteacidites dehaani Zone is present in a predominantly coastal facies without

sufficient marine fossils, and, therefore, cannot be closely correlated to the top
of the Maastrichtian. However, the presence in Nigeria of Danian fossils in the
lower part of the overlying zone suggests that this top closely coincides with the
top of the Maastrichtian.
(b) Retidiporites magdalenensis Zone
The presence, in the lower part of the Retidiporites magdalenensis Zone of

Nigeria, of the smaller Foraminifera Globigerina compressa and G. daubjergensis,
which are markers for Stolk's Globigerina compressa range Zone, indicates a
Danian age for part of this interval (SToLK, 1963).
A slightly younger age is indicated by the presence in the higher part of the
Retidiporites magdalenensis Zone in Nigeria of the smaller Foraminifera Globo-
rotalia pseudomenardii, G. velascoensis and G. acuta, markers for Stolk's Globo-
rotalia acuta/GIoborotalia velascoensis range zone, of Paleocene age.
This and the fact that the lower part of the overlying Retibrevitricolpites
triangulatus Zone still is of Paleocene age, indicates a Danian/Paleocene age range
for the Retidiporites magdalenensis Zone.
In Borneo a Paleocene age for part of the Proxapertites operculatus Zone
is indicated by the presence of the smaller Foraminifer Globorotalia velascoensis
and the larger Foraminifera Miscellanea miscella and Nummulites nuttalli.
(c) Retibrevitricolpites triangulatus Zone
In Colombia the lower part of this zone is associated with the larger Fora-
minifer Actinosiphon barbadensis, indicating a Paleocene age. In Nigeria the lower
part contains the smaller Foraminifera Globorotalia velascoensis, and G. acuta,
also indicative of Paleocene, while in the higher part the smaller Foraminifera
GIoborotaliaformosa and Globorotalia rex, which are markers for Stolk's Globoro-
talia.formosa range zone, indicate an Early Eocene age.
In view of the fact that in Venezuela the overlying Psilatricolporites crassus
Zone carries already a Middle Eocene fauna, while in Nigeria the overlying Mono-
porites annulatus Zone is associated with a Lower-Middle Eocene fauna, it is
suggested that the Retibrevitricolpites triangulatus Zone ranges from Late Paleocene
to Early Eocene in age.
(d) Monoporites annulatus Zone (not subdivided)
In Nigeria this zone is associated with the smaller Foraminifer Cassigerinel-

loita amekiensis and associated planktonic Foraminifera indicative of Stolk's
Cassigerinelloita amekiensis range zone of late Early-Middle Eocene age.
In the Caribbean area a more detailed tie-in of the subdivision has been
possible, and this will be discussed separately for each subzone below.
Rev. Palaeobotan. Palynol., 6 (1968) 18%348 243

(e) Psilatrieolporites crassus Zone
In Venezuela the lower part of this zone is associated with the larger Fora-
minifera Linderinafloridensis, Helicostegina gyralis and Lepidocyclina sp. A, while
the upper part was found to contain Helicolepidina spiralis form C. According to
VAN RAAOSHOOVEN (1951), these faunal associations indicate an early Middle
and a late Middle Eocene age respectively.
(f) Psilatricolpites operculatus and Retitricolpites guianensis Zones
In Venezuela these zones are associated with Helicolepidina spiralis form C

and are, therefore, of late Middle Eocene age.
(g) Boundary Monoporites annulatus Zone/Verrucatosporites usmensis Zone
Since the lowermost part of the overlying zone in Venezuela still is of late
Middle Eocene age, this boundary must be situated in the Middle Eocene.
(h) Verrucatosporites usmensis Zone
In Venezuela the presence in the lowermost part of Helicolepidina spiralis

form C is still indicative of a Middle Eocene age, but for the upper part the larger
Foraminifera Lepidocyclina pustulosa, Pseudophragmina mirandana, Nummulites
striatoreticulata and Helicostegina soldadensis and, in Colombia the smaller Fora-
minifer Bulimina jacksonensis, indicate a Late Eocene age.
In Nigeria the presence of the smaller Foraminifera Chiloguembelina martini
and Truncorotaloides rohri, indicative of Stolk's Chiloguembelina martini-C.
cubensis concurrent range zone, indicates a late Middle-Late Eocene age.
The well-known Late Eocene mollusc fauna, originally described by Bullen
Newton (ref. REYMENT, 1965) from the higher part of the Bende-Ameki Group,
also falls within this zone.
In the absence of any evidence of an Oligocene age, an age range for the
Verrucatosporites usmensis Zone from Middle to Late Eocene, possibly including
the earliest Oligocene, appears reasonable.
(i) Cicatricosisporites dorogensis Zone
In the Caribbean area the presence of the smaller Foraminifera Globigerina

ampliapertura, G. ciperoensis ciperoensis, Globorotalia opima opima and G. kugleri,
and in Nigeria Globigerina ciperoensis ciperoensis, G. ciperoensis angulisuturalis
and Globorotalia kugleri indicates an age range for the Cicatricosisporites doro-
gensis Zone spanning the interval Cassigerinella chipolensis/Hastigerina micra
244 Rev. Palaeobotan. PalynoL, 6 (1968) 189-348

Zone-Globorotalia kugleri Zone of BOLLI (1966). Accepting Bolli's age assignment
of this interval, it is concluded that the Cicatricosisporites dorogensis Zone largely
coincides with the Oligocene.
(j) Boundary Florschuetzia trilobata Zone/Florschuetzia levipoli Zone
This well-marked boundary in Borneo was found to be situated at a level

slightly above the boundary Tertiary e 1-4/Tertiary e 5 on a combination of
evidence from larger and smaller Foraminifera. In the present paper the latter
boundary is taken as coinciding with the Oligo-Miocene boundary, but it is
realized that this needs further confirmation.
(k ) Jandufouria seamrogiformis Zone
This zone is associated in the Caribbean area with the smaller Foraminifera
Catapsydrax stainforthii and Globigerinatella insueta, and approximately covers
the Catapsydrax dissimilis, C. stainforthi and part of the Globigerinatella insueta
Zones of Bolli (Lower Miocene).
(l) Psiladiporites minimus Zone
In this zone the smaller Foraminifera Globigerinatella insueta, Globorotafia

fohsi barisanensis, G. fohsi fohsi and Orbulina universa indicate that part of the
Globigerinatella insueta, the whole of the Globorotalia fohsi barisanensis and part
of the Globorotalia fohsifohsi Zones of Bolli are covered (Lower Miocene).
(m) Verrutricolporites rotundiporis Zone (unsubdivided)
In Nigeria the presence of Globorotalia fohsi fohsi and Orbulina universa

indicates that Bolli's Globorotalia fohsi fohsi, Globorotalia fohsi lobata and Globoro-
talia fohsi robusta Zones may possibly be covered, but since in the Psiladiporites
minimus Zone of the Caribbean area no evidence of the latter two zones is present,
it seems likely that in Nigeria only the Globorotaliafohsifohsi Zone is represented
in the upper part of the Verrutricolporites rotundiporis Zone. Due to unsuitable
facies conditions, the Globigerinatella insueta and Globorotalia fohsi barisanensis
Zones have not been found in Nigeria.
Stratigraphic position of the base of Orbulina universa

In view of the considerable importance which is generally attached to this
datum, its location in the different areas will be discussed separately.
In the Caribbean area, as well as in Nigeria and Borneo, it has been found
to lie in the upper part of the Magnastriatites howardi Zone. In the Caribbean area
Rev. Palaeobotan.PalynoL, 6 (1968) 189-348 245

it can be pinpointed more closely and was found to lie in the Psiladiporites minimus
Zone. In Nigeria it is present in the Verrutricolporites rotundiporis Zone, while in
Borneo it was found in the middle part of the Florschuetzia levipoli Zone. In
Bolli's scheme the base of Orbulina universa occurs in the upper part of the Globi-
gerinatella insueta Zone.
(n) Multimarginites vanderhammeni Zone
The presence of the smaller Foraminifera Globorotalia fohsi fohsi and G.

fohsi Iobata is indicative of Bolli's Globorotalia fohsi fohsi and Globorotalia fohsi
lobata Zones (Lower Miocene).
(o) Grimsdalea magnaclavata Zone
This zone was found to contain the smaller Foraminifera Globorotalia fohsi
robusta, G. mayeri and G. menardii and thus covers approximately the Globorotalia
fohsi robusta, Globorotalia mayeri and Globorotalia menardii Zones of Bolli. The
boundary between Lower and Middle Miocene would thus be present within the
zone.
(p) Crassoretitriletes vanraadshooveni Zone (not subdivided)
In Nigeria the presence of the smaller Foraminifera Globorotalia fohsi fohsi

and Orbulina universa indicates a possible age range covering BoUi's Globorotalia
fohsi fohsi, Globorotalia fohsi lobata and Globorotalia fohsi robusta Zones. This
corresponds well with the more detailed evidence discussed under the two
preceding Caribbean subdivisions of the Crassoretitriletes vanraadshooveni Zone.
(q) Florschuetzia levipoli Zone
The age range of this Bornean zone is determined by the presence, in the
lower part, of the smaller Foraminifer Globigerinatella insueta, indicative of Bolli's
Catapsydrax stainforthi and Globigerinatella insueta Zones, in the middle part, of
the larger Foraminifer Flosculinella bontangensis, indicative of Tertiary f 1, and
of the smaller Foraminifera Globorotalia fohsi barisanensis, G. foshi fohsi and
Orbulina universa, indicative of Bolli's Globorotalia fohsi foshi Zone, and, in the
upper part, of Globorotalia fohsi lobata, G. fohsi robusta and G. menardii, indicative
of Bolli's Globorotalia fohsi robusta and Globorotalia menardii Zones.
(r) Echitricolporites spinosus Zone (not subdivided)
In Borneo the lower part of this zone carries the smaller Foraminifer Globo-
rotalia menardii, indicating an age younger than Bolli's Globorotalia fohsi lobata

Zone, while in the upper part Pulleniatina obliquiloculata occurs, which is generally
taken to indicate a Pliocene age.
In the Caribbean area the latter Foraminifer is not found below the Echi-
tricolporites mcneillyi Zone, suggesting that the Mio-Pliocene boundary may be
close to the base of the last-mentioned zone. The youngest boundary in the Carib-
bean area, between the Echitricolporites mcneillyi and Alnipollenites verus Zones
is, in view of the climatological interpretations, taken as closely coinciding with
the Plio-Pleistocene boundary.
Summarizing the results presented in the foregoing account, it may be stated

that the top Maastrichtian, top Paleocene, top Lower Eocene, top Middle Eocene,
top Upper Eocene and top Oligocene are well established in correlation with the
palynological succession. The subdivisions in the Miocene up to the Mio-Pliocene
boundary are less accurately known, mainly due to uncertainties remaining in the
correlation of BOLLI'S (1966) planktonic zonation and of Van der Vlerk and Umb-
grove's letter classification (re['. LI~CHTI, 1960) in the Far East with the standard
European succession.
On the general range chart (Fig. 15) the proposed correlation of the standard
time-stratigraphical subdivision with the palynological subdivision is presented.
Zones have been drawn approximately to an absolute time scale, derived from
FUNNEL (1964).
Two points should be mentioned here in connection with this chart.
In the first place, the bases of the occurrences of Verrucatosporites usmensis
and Crassoretitriletes vanraadshooveni have proven to be slightly diachronous,
as discussed earlier. In the former case, the base occurrence in Nigeria has been
taken as the lower limit of the Verrucatosporites usmensis Zone, while the time-
equivalent horizon in the Caribbean area is approximately given by the base of
Cicatricosisporites dorogensis. In the latter case, the base occurrence in the Carib-
bean area has been taken as the lower limit of the Crassoretitriletes vanraadshoo-
veni Zone, since this is the best defined change in relation to the planktonic sub-
division.
Secondly, the detailed subdivision of the Eocene and Lower Miocene reflects
local stratigraphical needs more than rapid floral change. The relatively long,
unsubdivided Oligocene interval probably indicates lack of information due to a
hiatus in the stratigraphical record of both the Caribbean area and Nigeria. In
Borneo the Eocene interval has not yet been adequately investigated and no sub-
division is possible at present.
STRATIGRAPHICAL APPLICATION
In this chapter the practical application of the palyno-stratigraphical evidence
Rev. Palaeobotan. Palynol., 6 0968) 189-348 247

will be discussed and illustrated with three geological cross-sections from the
Caribbean area (Fig.17-19).
The sections chosen have been constructed as far as possible on the basis
of wells or measured surface sections in which a fairly detailed palynological
succession had been established. In a few cases a composite section had to be
constructed by piecing together various scattered shorter sections, so that thick-
nesses of formations are consequently less reliable.
Nomenclature of geological formations follows as far as possible generally
accepted terminology. For Venezuela reference is made to the respective volume
of the International Stratigraphic Lexicon (SCHWARCKANGLADE, 1956), for
Colombia and Trinidad to more scattered literature.
The environment of deposition has been broadly classified into three units:
(1) upper coastal plain (fresh water), and more inland environments characterized
by the absence of any brackish or marine faunal elements; (2) lower coastalplain
(brackish water), deposits which may carry typical faunal associations indicating
low and variable salinity; (3) marine.
Section I (Fig.17) covers the Palaeogene from east-central Colombia to
western Venezuela, running along the northern slopes of the Andes, circling around
Lake Maracaibo to turn southwards along the Perija foothills as far as the Tarra
area. A large part of the area covered by this section is underlain by marine
Cretaceous deposits.
The Colombian sections have not reached these, but in Venezuela the Col6n,
Mito Juan and their lateral equivalent the La Paz Shales have been sampled and
found to be restricted to the Proteacidites dehaani Zone. In the north a slightly
younger marine deposit is the Guasare, covering the Foveotriletes margaritae and
Ctenolophonidites lisamae Zones.
The regression shown to have taken place by the change towards a tran-
sitional or even terrestrial environment has evidently not been contemporaneous,
a phenomenon earlier described in a generalized way by KUYL et al. (1955). South
of Lake Maracaibo, the sediments assigned to the Proteacidites dehaani Zone
were already deposited in part in a transitional environment, as indicated by the
Umir in Colombia and the lower part of the Orocu6 in Southwestern Venezuela.
Further north in the Dibujo area (Riecito Mach6), the change from marine to
terrestrial deposits takes place at the base of the Foveotricolpites perforatus Zone.
The effect of tectonic movements during the Paleocene is also clearly visible.
The interval Foveotriletes margaritae Zone-Foveotricolpites perforatus Zone is
thickest in the Lebrija section and shows very regular thinning to the south and
northeast.
Along the Andean foothills (Fig. 17, columns 3-7) and in the northern Lake
and Concepci6n areas (columns 8-10) the Paleocene has been truncated, indicating
a minor orogenic phase separating Paleocene from Eocene. This truncation is least
noticeable in the Riecito Mach6 section, where the possibility of continuous

1 2 3
PAZ DEL RIO RIO I.EBRIJA RUBIO ROAD
F 1~0 KM I 115 KM I 9(
C. DOROGENS,SZONE ~ I
V. U S M E N S I 5 Z O N E ) v. USt
........ = (A) P. C R A S S U S Z O N E
R. T R I A N G U L A T U 5 ZONE R . = ~.[
}~i;[i~:::..\ \ /
\\"
.......... ::i~-~:.iiii.-:.:iiiiii?::i;.-"iisi.:.-"~;i~iis
VERIICAL SCALE
:iiiiiiiiiiiiiiiiii~!iiiiiiiiiiiiiiii
:: ?iiiiiiiiiiiiiiiiiiiiiiiiiiiiii
METRES FEET :~i~i!i i i:ili~i i l!i !~ii i!i i i i;ili!i~i i i
0 - 0
~ii i!i:i!i!ililili!i!i!i~i~:i!i!i~i:i i i!i!i~i~i~i~i
~000 !!!ii!55i!.~:)!ii52i12251i:
5
):i:i:!:!:i!:::~:i:i:i:i:!:!:
i:::::.:.:.:
..... ~:ii!!i!i!
500 - ":::::::::::::: .-.
i!:::i:i:i:!:!:!::~'o:!!i!iiiiiii
2oo0 ::F:!:i?iiiii:?!F!i:~,~:??!ii!i
m 3000 :::::::::::::::::::::
*00o - : ::2:::::
' 4000
1500 -
5OO0
4 5 6 7 8 9
R I O DE O R O RIO PERDIDO CATANEJO- CONCEI

(TACHIRA) PREVENCION AREA
CIDA. LA M O R A RIO M U L L A P A S NORTH CENTRAL
LAKE A R E A
] KM [ 20 KM I 45 KM l 40 KM I 180 KM ~ 40 KM I 55 KM
o ~_
OGENSIS ZONE
"s:::...~i~
::~:>.::::::::::::::::::::::::::::::::::::::::t o / !t::~:~t:~.:~:..-,~:!~::~.~.~!~.-:.*,::!::~:~.,::#:i
u ii::::iTiV
.... N" .xe~....~.,.X.'eN ". -~ "v +'.~' ---'-....... :,
.......... ~ ~:.~:~::. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~2~.~.'~?~}~i~q~i~i!i.>)i~~}i~ :"
~,~,~oN~ ~:.~ii~i~ ~t,~i.-'.-ii~.~i~i~ '.'%'.~-~t'.':.'.c::;~.-:~i~tii.~ii~r.~i~'~- ~iiilt
~s}}i~:~:i!~!:~.:!::::::~:~.~i~!~:~:~:i:::::::::::::i:i:i::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
~ ' NENsis"ZO'NEi)iiii?
::':':"'( L ..-.-'-'-" :"iiii iiiii :::::::::::::::::::::::: : : 2:' :--:. :.:. :-:2:::::::::-:::: ::::::::::::::::::::::::::::::::::::::::::::::i.iiA 1
................................. . . . . . . . . . . . . . . . . ===================================================================
":!:!:~:~:~:~:!!i~i!~!~F~#~8:'.=~i~..!.i:~:~:!:~.J:i:~:.::~.!:~:~:K . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ".:.:.:.:-:.:-:-:-:.:-:.:.:.:,:.:.
. . . . . :.:.:-:-:-::.::-:-
. . tl :
i:i~::.. ~t..*:~~~~!~ t,*'.'~,~!=============================
~~:i:i:i:i:i:i:i:i~ :::::::::::::::::::::
:-~iii~~i~.~'~-~'.;.!~. ~ i :::::::::I:~........................~ ~ : . : : : : ~ I .......... .L
:, ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::: (3)
LOCATION HAP ~.:::::::..:~:::::::::~::::.~j:'.:::::::::~:~~:~:::::..:::~:::...:...~#~:::

::':~:~ ::::::::::::::::::::::::::::::::::::::::::: <::::::'"""" I~ :+:'::::
:~F::.-::',:.~.~.~.F:$~.:'...~:~).~: :::::::::::::::
.....'..,::,:,:::~::N ...... ~!::ili::ili::i::i:,i::i::iiiiii!:::i~::ii!iiii'/:ii:i!iiiii ~ :illii:ii:.i::!:::

,:::.~:~:~!~
.,:~.'.e..::,.~#.-.~ ::::::::::::::::::::::::::::::::::::::::::::::: :.:.:.:.:.:
::::::::::::::::::::::::::::::::::::::::::::::::: ....
:::::::::::::::
~!~'i~!~!~!~ii~! :::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::
"Ii! :'ii:::':':':':::::::':'::':'::: i:i!i!i!!i!i!i!
, .% :::::::::::::::::::::
::t ii:: ii !iii!!!ii!iiii~
% ~ ::: ti#iiitiii ~
,:::: :.:.:.:.:.:.:.: ,........
VENEZUELA
, .4 "IL ':,::tiii!i)~ ............~i::i~i::~i~i~~!ii!iiiiiiill

i iiii!!i!iii!!iiiiiii!iiii!i!iiiiii
:i:i:i:i:::i:i: i
. .... ,.,.,, ::tli!ii.%: ::::::::::::::::::::::::::::::::::::::::::::::::::
' '%!ii
o, ,,oo ~oo ~ ....

I
'::,:~
"-t ..
,,,,o~.
: Gu*s*~E!ii::i!~::C::
(5) ~ PAZ ~AL~
10 11 12 13
)CION-154 RIECITO RIO T U C U C O ODA. L A V I C T O R I A

MACH~
I 60 KM I ISO KM I 140 KM I
"( ":" ~ v. USMe~S
7 P. CRASSI
R. TRIANGL
i / ~ .............i!iiiii~i~"
/
/ / :.....!7'
/ / , :~:~....
.4 /'~ ~... ~ ~:.....
= ' ~ ii771i? [ I UP,ERCOASTALPLA...,R~S.WATER

:::!:. ~ LOWERCOASTAL PLAIN, BRACKISHWATER
-- :: ~ MARINE
:.>:
~ ~
ii!i77
:-:-::~i::" ~ CONSPICUOUSSANDSTONE8ODY
' ~ ..Y
~.:~:~r:~ SECTIO
iS ZONE
JSI5 ZONE
JS ZONE
fLATUS ZONE
ENENSIS ZONE
N I
Fig.17 (pp.249-252). Stratigraphical section I.
1
CHAFURRAY-3
<
G. MAGNACLAVATA ZONE
M. VANDERHAMMENI ZONE
ROTUNDIPORUS ZONE
C. DOROGENSIS ZONE
V. USMENSIS ZONE
.....
VERTICAL SCALE
METRES FEET
0 - - 0
* 1000
500
2000
3000
1000
4000
1500
5OOO
Fig.18 (pp.253-256). Stratigraphical section II.

Fig.19 (pp.257-259). Stratigtaphical section III.
2
VORAGINE-1 RIO
0 KM { 375 KM
~ ., :-:-:. :::::::::::::::::::::::::::::::::::: :::::::: :::::::::::: ::::::i:i:i:i:::i: :::::::::::::::::::::::::::::::::::::::::: " .i~~:::'::i:::i

!:::~"
i:::~:~:~:~::.:i:~:
::::::::::::::::::::::::::::::::::::: ....................................................`.................................
: :i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:].:. :::::: :::::::::: ::;:::::;:::: ::::: :::::::::::::::::::::: :::: ::::::::::::::::::::::::::::::: ::: ;:::::::; :;:;:::;::::::::;:: :::::::::::::::::::: ::::::::::::::::::::::::::::::: :;::::;:::: :;:::::: :::::::::::: :::: :::::::::2;:: :;2;: ::::::::::::::::::::::::::::::: ;:::::;:::::::::2 :::::::::2 :::::::::::::::::::::::::::::::
::::::::::::::::::::::::::::::::::::: ~:::~:~::::~:::::::::::::::::::::::::::::::~:~::~:~:]:~:~:~::::::::~:~:~:~:::~:~:::::::::::::::::::::::~:~:~:~::~:::~::::::::::::::::::::::::::::::::::::~:~:~:~:~:~:~:::~:~::::~:::::::::::::::::::::::::::::::::~;~;;;~;;
================================= :::::::::::::::::: ::;:::::::::::;:;; ;;:::::::::::::::::::::::::::::::: :::::::::::: :::: ;:;:::;:::::::;;:::: ;:;;:::;::: ;:;:::::;:::::: ::::::::: :::::;:::::::::;::: ::::;::; ;;;:;:;;:::;:: :;:::;:::;:: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::: ::::::::. ::::::::::::::::::::::::::
8:.
::::::::::::::::::::::::::::::::::::::
.....................~:i:i:i i li:;:::i::!ii::ii i
3 4 5 6
CUBUGON LA FRIA A R E A CATATUMBO-1 B-188 OI
135 KM 130 KM I 90 KM J lE K M - -
~ ,,..,,~.,,~. ,,.,~. ~.,,,,,,~ ~"" ~ i "c
}!~::::: ,.~ .~ .:: ========================

-- z ~ ==============================
':::::::.:.... ~ ~..,
........................... !:::!: ~ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: .:
. . . . . c~ i ~ (2) LOWERLAGUNILLA! ~ i i i i i
_: (3) LA ROSA
(41 ICOTEA
J~ : :SEAMROGIFORMIS ZONE::,
b ' . ~ : ' : + ~:':"'~:':~::':'~:'>:':~':':':':':':':':':':': Z,:c,
I. . . . , - - , ' ~ ~ ~ i i ~
r-- b-~~~.~!::.:~~.::'~i~~.!.~ :" ============================ :::::::::..
.i':':~:'~:'~~:'::'b:':'~':~::'
'Ii~:~:'::'~:~i:~::~i'
'~.:.'""
: ~ ........
iiiii/}~illi~i!~,~!.....................iiiif:i........ LOCATION MAP
NARACAIBO
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::
/ V EN EZ U EL A
:::'::::::::: :: :: :: :: :: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: z :!:!:i:':':'::: :':':':':': ZONE

~i~::!:ii!ii!i!iiiiii!i!i!iiiiliiiii!iiiiiii!iiii!!!i!i!iii!iiiiiiiiiiiiiiii!i ! GEN$1S ~ 4
V. USMENS
S
I ZONE /
mBOGOIA
o 200 300 KM
I | I I
7 8 q
)A. UCA AREA NORTH CENTRAL FALCON EAST FALCON
120 KM J 1 O0 KM I
DIEDERIXI Z O N E
GLOBOROIAEIA
MENARDII ZONE
GLOBOROIAUA
MAYERI ZONE
G. M A G N A C L A V A T A Z O N E
" GLOBOROIALIA
FOH$1 ROBUSTA ZONE
GLOBOROTAUA
EOHSI EOBATA ZONE
M VANDERHAMMENI ZONE
.................................
i 6EOBOROTALIA
L. FOHSI FOH$1 ZONE
POZON ................................
GLOBOROTALIA
i
,~ P. H I N I H U S Z O N E FOHSI BARJSANENSISZONE
..................................
GLOBIGERINAEELLA
0 -- INSUETA ZONE
........................
o EATAPSYDRAX
z ~.o J. SEAMROGtFORMIS ZONE SIAtNFORTHfl ZONE
"( ~ ......................
CATAPSYORAXDISSIMILIS ZONE
GEOBOROEALIA KUGLERIZONE
GLOBIGERINA
C. DOROGENSIS Z O N E
(IPEROENSIS CIPEEOENSISZONE
LEGEND
l - - ] UPPER C O A S T A L PLAIN, FRESH WATER
LOWER C O A S T A L PLAIN, BRACKISH WATER
MARINE
SECTION II
1 2 3
SINU BASIN BAYUNCA CATATUN
I 200 KM I 400 KM I
P. DIEDERIXIZONE
ZONE
::::::::::::::::::::::::::::::::::::::::::::::::: z
METRESVERTISCALEFEE1
CAL ~...~................~..~::::::::::::::::::::::::::::::::::::::::::::::::
'~
0 0 :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
1000
2000 :~:~:~:~:~:~:~:~:!:~:~:~:~:~:i
"'"'""'"'"~" :~:~::-, ~ 1
3000
1000
4000
1.500 5000
4 5 6
IBO-1 RIO BUENA VISTA EAST FALCON
BlO O N I A
--BO KM I 120 KM t 300 KM I 520 KM
/
/
/ LOCATION MAP
/ o
~ CARIBBEAN SEA
COLOMBIA
//
GLOBOROTALIA
N~NARDfl ZONE
......... .. _.-'..._~..._-~, ,,,,,, ,,,,,, ,,,,

' - . . . . . . . , . . .
GLOBOROTALIA
NAYERI ZONE
b GLOBOROTAUA
0 FOHSI ROBUSTA ZONE
..............................
.... FOHSI LOBATA ZONE

==
.................................
GLOBOROTAUA
~ m FOHSI FOHSI ZONE ~ ~
~OZON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
GLOBOROTALIA
0 FOHSI BARISANENSIS ZONE
............................
~r GLOBIGERINATELLA m ~ ~ ~ ~ ~ ~ --
o - - ~ INSUETA ZONE
o ~ ~ ...........................
i ~~ ..-----""
.,~ "~ """
~:~
::~;~:mi ~
z
CATAPSYDRAX
o . . . . . . . ~_,~!,,0~E,,,._~oN~_. . . . .
:~::~'~'~I~1 GLOBOROTAUA KUGLERI ZONE
GLOBIGERINA
, , , ~ . .~. ~ CIPEROENSIS CIPEROENSrS ZONE
1 2 3
SINU BASIN BAYUNCA CAIAIUM
J 2 0 0 KM t 400 KH I
~
:~
:~
:~
:~
:
P. DIEDERIXI Z O N E
G. M A G N A C L A V A T A ZONE
M. VANOERHAMHENIZONE ~:;::~.:.
..................
:.:.:.:.:.:.:.. ................................. Z ,~
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
VERTICAL SCALE ""::':-:':':'::':-:':':':'::-:-:':-:':-:':-..-:::
MEFRE FEEl
0 -- 0 ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: <
p MiN,i,iC~sz6NEili:i:i:i:i:i:i:i:i:i:i:i:
1000
2000
J. 5E
3000
1 ooo ...............:::~:~:~:~!~;~:.{{~:::.::i:::.
4000
1500
5000
sedimentation is not excluded, although the rather sharp and distinct floral change
at the base of the Retibrevitricolpites triangulatus Zone still suggests the presence
of a hiatus. It increases southwards since, in Concepci6n, the Foveotricolpites per-
foratus Zone is missing, while in the North Central Lake area the Ctenolophondites
ilsamae Zone is also missing, and it reaches maximum values along the northern
Andean foothills. In Rio Mullapas and Rio Perdido the Proteacidites dehaani
Zone is found underlying the Eocene, and the whole Paleocene appears to be
missing.
In a southwesterly direction the magnitude of the hiatus rapidly decreases
again and in Quebrada La Mora the Foveotricolpites petforatus Zone is found
again. Further south and in the Tarra area the top of the Paleocene interval co-
incides with the base of the Mirador Sandstone, and it is likely on geological
grounds that a minor unconformity is present at this level (ScHAuB, 1948, p.225),
coinciding with the base of the Retibrevitricolpites triangulatus Zone. In Colombia
sampling gaps around this level preclude any definite statements on this problem.
The Eocene sedimentation cycle starts with the Retibrevitricolpites triangula-
tus Zone. (As mentioned in the previous section of" this paper the lowermost part
of this zone may be of Paleocene age.) Sediments assigned to this zone are thickly
developed in the northeast and in Colombia, but they are absent along the northern
Andean foothills between Rio de Oro and Rio Mullapas. The next higher Psilatri-
colpor#es crassus Zone is absent between Rio Perdido and Rio de Oro and in the
Colombian sections. The overlying Psilatricolpites operculatus Zone is generally
rather thin and its absence in sections with reduced sedimentation such as Rio
Mullapas and Quebrada La Victoria may be due to sampling gaps. In contrast
to the rather restricted areal distribution of the older Eocene zones, it is clear that
the overlying Retitricolporites guianensis and especially the Verrucatosporites
usmensis Zones have a much wider distribution. The former is thickly developed
in the Rio Mullapas and Rio Lebrija sections, while the latter is found in more or
less uniform thickness m all the northern Andean foothills sections as well as in
T~ichira. The absence of one or both zones in Rio Mullapas, North Central Lake
and Concepci6n is due to truncation by the well-known post-Eocene unconformity.
From these observations it may be deduced that during the Eocene progres-
sively younger sediments transgressed over an eroded Paleocene surface, first
localized in the deeper parts of the sedimentary basin, but rapidly transgressing
over virtually the whole area of investigation in the upper part of the Retitrico#
porites guianensis Zone and especially during the time of deposition of the Ver-
rucatosporites usmensis Zone. This last transgressive phase coincides with the
widespread deposition of the marine Pauji in the east, which sharply contrasts
with the transitional environment prevalent in the older Eocene.
It must be noted that in the southwestern part this transgressive younger
Eocene rests unconformably on Lower Eocene sediments of Retibrevitricolpites
triangulatus-Psilatricolporites crassus Zone age, indicating a minor orogenic phase,
Rev, Palaeobotan. Palynol., 6 (1968) 189-348 263

probably resulting in a slight subsidence accompanied by a southwestern tilt of
the depositional basin. Thus in Quebrada La Mora the Verrucatosporites usmensis
Zone is resting on the Ctenolophonidites lisamae Zone, in Rio de Oro on the
Foveotricolpites perforatus Zone and in the Rubio road section already on the
Psilatricolporites crassus Zone.
In general Eocene sedimentation was thickest in the northeast, as shown by
the Catanejo-Prevenci6n section. Continuation of the Eocene sedimentary basin
towards Falc6n is likely, but no palynological data are available from this area
due to carbonization of plant material.
At this point it may be mentioned that, although the strong vertical exag-
geration of the section suggests the presence of marked angular unconformity at
many places, in reality this can hardly ever be observed in the field. Moreover, the
similar lithologies of the formations on both sides of the unconformities and the
absence of other age-indicating fossils made it virtually impossible to unravel the
geological history in detail before the advent of palynological studies (cf. SCHAUB,
1948; MENCHER et al., 1953).
The situation was further complicated by the presence of various sandstone
bodies, which had been correlated merely on lithological similarities supported by
weak photogeological evidence. The three main sandstone bodies are the La Sierra
in the northwest, the Misoa in the southeast and the Mirador in the southwest.
Palynological studies revealed that the bulk of the Mirador Sandstone had been
deposited in the interval Retibrevitricolpites triangulatus-Psilatricolporites crassus
Zone and was, therefore, roughly contemporaneous with the C-sand group of the
Eocene terminology used in the Lake Maracaibo area, while both the La Sierra
and Misoa Sandstones are mainly deposited in the interval upper Psilatricolporites
crassus-Retitricolporites guianensis Zone, thus being markedly younger and equi-
valent to the B-group of sands in the Lake area. In addition, as already pointed out
by SCHAUn (1948,p.224), the Mirador Sandstone cannot be traced in the field
along the northern Andean foothills beyond T~ichira, and earlier attempts to do so
were based on miscorrelation either with underlying Paleocene sandstone bodies
equivalent to the Barco Sandstone member or with basal Upper Eocene sandstones
transgressing over the eroded Paleocene surface. As can be seen on section I in
the area between Rio de Oro and Rio Perdido, (Fig.17), due to the considerable
gap shown to be present by palynology, these sandstones are close together and
can easily be confused in the field or by photogeology.
A further complication is present in the Tarra area southwest of Lake
Maracaibo, where the onlapping La Sierra Sandstone comes into contact with the
underlying Mirador Sandstone. It has been shown by palynological dating that in
some areas, Quebrada La Victoria for example, the uppermost part of what had
been called Mirador Sandstone is in fact the age-equivalent of the lowermost
La Sierra Sandstone and that locally a minor hiatus may thus exist in the upper
part of the Mirador.

These observations also have some consequences for the stratigraphical
terminology. First of all, the boundary between the Carbonera and the Orocu6
Formations is distinct only when these formations are separated by the Mirador
Sandstone, as is the case in T~ichira and the Tarra area. However, the disappearance
of the Mirador in a northeasterly direction along the Andean foothills and the
similar facies of the Orocu6 and Carbonera make it virtually impossible to separate
these formations in the field. On the section the boundary between the formations
has in such cases been taken to coincide with the stratigraphical hiatus revealed by
palynology and indicated with a dotted line.
Similarly the lithological separation of the Mirador and La Sierra Sandstones
in the area of overlap is difficult, and recognition of these different sandstone
bodies may only be possible when shale intercalations carry sufficient pollen for
dating.
Recently GONZALEZ(1967) has studied the transition Los Cuervos-Mirador
in the Barco area (Colombia). He reports the presence of Cicatricosisporites
dorogensis rather low in the Mirador Sandstone, which suggests that a relatively
large part of this formation can be assigned to our Verrucatosporites usmensis
Zone. However, the lack of data from the upper 100 m of the Mirador Sandstone,
the absence of Retitricolporites guianensis and Perisyncolporites pokornyi, several
other anomalies in the floral succession and the rather restricted number of samples
of unknown quality preclude a more definite interpretation of his results.
The final phase of Eocene sedimentation, where preserved, indicates general
regression of the sea. In the Catanejo-Prevenci6n area the marine-deltaic Eocene
sediments are succeeded by the continental La Victoria Formation. In the area
between Rio Lebrija and Rio Perdido the Esmeralda and Carbonera Formations
show no sign of transgression and in the Paz del Rio section the transitional
San Fernando Formation, which is roughly time-equivalent to the marine Pauji
transgression phase, is succeeded by the Oligocene continental Margua Formation.
The regression here has taken place during the time of deposition of the
Cicatricosisporites dorogensis Zone and is, therefore, younger than in the Catanejo-
Prevenci6n area where the La Victoria Formation is restricted to the Verrucato-
sporites usmensis Zone. In the Rio de Oro and Rio Perdido sections the Le6n Shale
at the transition between the Verrucatosporites usmensis and Cicatricosisporites
dorogensis Zones indicates a localized more pronounced subsidence, without
marine influences.
Section II (Fig.18) covers the older Neogene part of the stratigraphical
succession and runs from the western Llanos area in Colombia to southwest
Venezuela, crosses Lake Maracaibo and traverses Falc6n from west to east.
The top of the San Fernando Formation, which in most sections can be
found around base Cicatricosisporites dorogensis Zone, appears in the Rio Cubug6n
to lie close to the top of the zone, thus indicating that the Le6n Shales of the La
Fria area may be a lateral equivalent of the upper part of the San Fernando. In

Falc6n marine deposits characterize the Cicatricosisporites dorogensis Zone, while
in well B-188 no evidence of the occurrence of this zone could be found. Possibly
the barren Icotea Formation represents this interval.
In the south the period of deposition of the San Fernando and of the
Le6n Formations is followed by a regressive period, starting earliest in Voragine-l.
No trace of this regression can be detected in Falc6n. Then, approximately in the
middle of the overlying Verrutricolporites rotundiporis Zone (Jandujburia seam-
rogiformis Zone, upper part), a rather pronounced transgressive phase can be
discerned. This is the well-known La Rosa transgression from the Bolivar coastal
area of Lake Maracaibo, which can be followed from the marine Agua Clara in
Falc6n, via the La Rosa of the Maracaibo basin, towards the Uracil fossil horizon
of La Fria, which here occurs intercalated in the terrestrial Guayabo Formation.
In Colombia this transgression is visible as a marine intercalation in the lower part
of the Cubug6n Formation of the Rio Cubug6n and in the lower part of the
Chafurray in well Chafurray-3. Voragine-I was situated more inshore and no
marine interval occurs here. It has already been pointed out by Scr~atm (1948)
and MENCHER et al. (1953) that this La Rosa transgression is nowhere connected
with any disconformity. Its contemporaneity shows that during a short time a
continuous seaway was present from east Falc6n as far south as Chafurray-3 in
the Colombian Llanos.
In the upper part of the Verrutricolporites rotundiporis Zone (Psiladiporites
minimus Zone) a regression took place which can be recognized from the south as
far as north central Falc6n in the transitional-continental deposits of the Chafurray,
Caja, Cubug6n, Guayabo, Lagunillas and Cerro Pelado Formations, respectively.
In the south, only in the Rio Cubug6n, a second transgression is indicated
in the Multimarginites vanderhammeni Zone, which may be connected with the
marine Querales Formation of north central Falc6n. In younger zones regressive
deposits predominate. Only in Falc6n does marine influence persist. While in
north central Falc6n the marine environment has been interrupted twice during
the Neogene by regressive phases, in east Falc6n sedimentation was continuously
marine, which made it possible to arrive at an accurate tie-in between the palyno-
logical zonation and Bolli's subdivision on pelagic Foraminifera. Unfortunately
the east Falc6n stratigraphical sequence is based on a combination of scattered
short sections and stratigraphical thicknesses shown are, therefore, probably not
more than rough minimum estimates.
As already mentioned, in contrast to the preceding Palaeogene section, no
pronounced unconformities could be detected in the Neogene sediments investi-
gated. Still, rather strong variations in thickness are obvious.
In Colombia the section penetrated by Chafurray-3 is much more reduced
than in Voragine-1. In Rio Cubug6n it is again slightly thicker than Voragine-1, but
in the La Fria and Catatumbo areas maximum thicknesses are observed. Remark-

able are the extremely reduced thicknesses observed in B-188, especially since the
floral succession appears complete, without any marked gaps.
Thicknesses increase again rapidly, approaching the Falc6n basin, where, in
north central Falc6n especially the Verrutricolporites rotundiporis Zone is exces-
sively thick. This second section shows again the great value of palynology as a
means of correlating from the marine facies, as present in Falc6n, to the mainly
continental environment in southwestern Venezuela and the adjoining part of
Colombia.
The correlation of the various formations shown on both sections closely
corresponds, for the Maracaibo basin, with the correlation chart presented by
MILLER et al. (1958).
In section III (Fig.19) the stratigraphy of the younger Neogene deposits is
traced from northwestern Colombia to Trinidad. Three separate sedimentary
basins are represented. In northwestern Colombia an investigation has been made
of the area between Sinu and Bayunca, which is separated from the southern
Maracaibo lake and Falc6n basin by the Perija mountain range on the border
between Colombia and Venezuela. The latter basin is in turn separated from the
eastern Venezuela-Trinidad basin by the mountains connecting the M~rida Andes
with the coastal range of Venezuela. In these mountainous areas the youngest
Neogene deposits have never been present or were eroded during uplift. In view
of this geographical separation, the section is best discussed from west to east.
In northwestern Colombia the marine sediments in the Sinu basin and near
Bayunca are correlated by the Pachydermites diederixi Zone. The total interval
covered by these two sections gives a complete succession from the Multimarginites
vanderhammeni Zone to the Alnipollenites verus Zone.
These isolated marine deposits in northwestern Colombia are correlatable with
terrestrial sediments in the southern Lake Maracaibo area. In well Catatumbo-1
the Grimsdalea magnaclavata and Alnipollenites verus Zones can be recognized,
but not the boundary between the Pachydermites diederixi and Echitricolporites
mcneillyi Zones, which is due to an unfavourable mottled clay facies and to lack of
samples. In general sediments are thicker in this area than in northwestern Colombia,
and especially in the Rio Buena Vista section the younger zones reach considerable
thickness. In contrast, sediment thickness in the older interval Jandufouria seam-
rogiformis Zone-Multimarginites randerhammeni Zone remains fairly constant,
irrespective of facies, as shown by the correlation of Catatumbo-1 with the Rio
Onia and Rio Buena Vista sections. This increase in thickness of the youngest
sediments in the southern Maracaibo basin is a reflection of tectonic events con-
nected with the approximately contemporaneous rise of the M&ida Andes.
As already shown in section II, the younger Neogene sediments in the
southern Maracaibo basin are continuous with those in Falc6n, but this sedimenta-
ry area was probably separated by a broad land area from the east Venezuelan
basin. However, the floral characteristics marking the palynological zones turn UP

in the same sequence and in a virtually identical expression in both areas, and
correlation between the marine Falc6n sequence and well OG-2, where a regressive
sequence is present, presents no difficulties.
From OG-2 to Trinidad the floristic boundaries again cross facies lines, and
in the Catshill-Ortoire area a second tie-in with the zonation on age-indicating
pelagic Foraminifera confirms the time-stratigraphical value of the palynological
zonation. The younger part of the sedimentary sequence in Trinidad shows rapid
changes in thickness and facies over short distances and palynology has been of
great value here for unravelling the complex geological history. Of special interest
is finally the great thickness attained by the Grimsdalea magnaclavata Zone in the
Catshill-Ortoire area of Trinidad. This, together with the presence of large
quantities of reworked material and the well-known Lengua boulder beds, reflects
the final upward movement of the Andean orogeny, resulting in the upthrust of the
Trinidad central range.
Of course subdivision of the main floral zones presented here will make it
possible to refine considerably the stratigraphical picture presented. This, however,
lies beyond the scope of this paper, which was aimed primarily at demonstrating
the value of palynological subdivision for regional correlation.
Summarizing the results obtained so far, it may be stated that palynological
evidence has:
(1) shown the presence and magnitude of previously unsuspected strati-
graphical gaps;
(2) produced positive evidence for correlation of many Tertiary formations
lacking age-indicating animal fossils;
(3) shown that pollen boundaries can cross facies boundaries and are then
eminently suited for palaeogeographical reconstructions.
BOTANICAL RESULTS
Although the primary purpose of the palynological studies described here

has been stratigraphical, the desirability to relate as far as possible the fossil pollen
and spore species to Recent plants has, of course, produced some interesting
botanical results. In this section of the present paper the most important results
in this respect will be briefly discussed.
First of all, the criteria applied in botanical identification must be evaluated.
In general three categories were distinguished:
(1) Fossil species of which the botanical affinity is wholly obscure, e.g.,
Buttinia andreevi. Possibly many of these species represent pollen or spores pro-
duced by extinct plants.
(2) Fossil species which can be referred to more than one, not closely related,
groups of plants, while ecological evidence is insufficient to decide which group
may have been the parent plant, e.g., Striatricolpites catatumbus.

(3) Fossil species which, because of some unique morphological feature and
corroborative ecological evidence, can be confidently assigned to some Recent
plant taxon, e.g., Verrutricolporites rotundiporis, derived from Crenea sp. (Lythra-
ceae).
In the account given below only the species of category (3) will be discussed.
Within this group it may be possible to assign the fossil species to a single taxon,
as in the example given above, or the pollen or spore type under discussion may
characterize more than one related taxon, e.g., Multiareolites formosus, which
pollen type occurs in at least nine genera of Acanthaceae. In the former case it
appears justified to use the term Crenea pollen as equivalent to the form species
Verrutricolporites rotundiporis. In the latter case the fossil form species can also
be referred to as Justicia-type pollen, one of the genera showing the pollen type
being selected as indicator. Again the term Justicia type is equivalent to the form
species Multiareolites formosus. Only the names of form species and form genera
have, of course, taxonomic status.
FILICALES
PARKERIACEAE
Ceratopteris ( Magnastriatites howardiJ
The source areas of this spore species are the alluvial plain and coastal
swamps where the parent plant grows in the shape of a small aquatic fern in
shallow water, bordering lakes and river banks. This environment is characterized
by rapid local facies changes and this is clearly reflected in the pronounced quan-
titative fluctuations in fossil coastal plain sediments. In marine sediments the
spores are much rarer. The taxonomic identification is thus well supported by the
distribution pattern of the fossil spore.
The frequency of Ceratopteris spores in the Neogene of all three areas
investigated and their absence in older Palaeogene and Cretaceous deposits, shows
that the present-day pantropical distribution of this freshwater fern may date only
from the mid-Tertiary. This is, however, somewhat in contrast to the isolated
systematic position and specialized morphology of the genus, which is assumed
to be of great antiquity. Its origin may have been local, but as long as no ancestral
spore forms have been found this problem remains unsolved.
SCHIZAEACEAE
Mohria type ( Cicatricosisporites dorogensis)
This spore type is found in most of the species of Mohria and a few of

Anemia (Schizaeaceae). Without further study specific determination is impossible.
The spores of this group are, however, clearly distinct from those produced by
ferns of the genus Ceratopteris (Parkeriaceae).
The Tertiary distribution of the Mokria type can best be discussed in con-
junction with what is known of the Cretaceous distribution pattern of their nearest
relatives. This discussion refers to the genus Cieatricosisporites s.l., recognizing
the considerable difficulties which crop up when specific distinction of the dispersed
spores is attempted (cf. HU~HES and MOODY-STUART, 1966).
In the Caribbean area the Mohria type is absent from the uppermost Creta-
ceous, Paleocene and Lower Eocene but MULLER (1966) reports its presence in
Cretaceous-Paleocene strata in northeastern Brazil. In the Middle Eocene the
ferns producing this spore migrated into northern South America, as shown by
its remarkably sharply defined base occurrence in Venezuela. It then remains
fairly abundant up to the Oligo-Miocene transition. In the Miocene it virtually
disappears from the record.
In tropical west Africa it is common in the Albian-Senonian, absent in the
remainder of the Cretaceous, and returns in the Upper Eocene, somewhat later
than in the Caribbean, but suffering a sharp decrease in the Caribbean at the top
of the Oligocene, and disappearing completely at the same level in Nigeria.
In Borneo the spore type is abundant in the Lower Cretaceous, decreases in
the Upper Cretaceous and becomes very rare in the Tertiary.
As is well known, in Europe the type is frequent in the Cretaceous and in
the Palaeogene, is still locally abundant in the Oligocene and disappears from the
record in the Neogene (cf. KRUTZSCH, 1957).
DRUGG (1967) has recently reported its presence in Late Cretaceous Pale-
ocene strata from California.
The Recent genera Anemia and Mohria are at present confined to the tropics
and subtropics of America, the West Indies, southern and eastern Africa, Mada-
gascar and the Mascarenes, where they occur often in a semi-arid habitat (CHRIST,
1910).
From these data it is clear that the group as a whole has had a rather cheq-
uered history. In the Early Cretaceous it had a virtually world-wide distribution,
but in the Late Cretaceous a contraction of its area occurred. In the Middle
Eocene a second expansion of its geographical area took place in Nigeria and in
the Caribbean, but the decline at the end of the Oligocene appears to have been
world-wide again, leading to the present scattered relic areas of occupation in
America and Africa.
Lygodium microphyllum type (Crassoretitriletes vanraadshooveni)
Lygodium microphyllum is a climbing fern, common in the humid marsh

and swamp forests of west Africa and the Indo-Malesian area, but absent today

in South America. The geological record of the Lygodium microphyllum type
indicates rapid pantropical expansion in the Neogene. In the Caribbean area
the fern apparently became extinct in the Miocene. The reasons for this rapid
expansion and local extinction are entirely obscure. No ancestral spore type is
known, but may be present in the insufficiently studied Eocene sediments of the
Indo-Malesian area, where the type occurs slightly earlier and may probably
increase more gradually during the Neogene than in Africa or in the Caribbean
area. The high values of this spore type observed in the latter areas have not been
found, however, in the Neogene of Borneo, which is rather in contrast to its
present-day abundance in vegetationally disturbed areas and in coastal swamp
forests.
ANGIOSPERMAE
POACEAE (GRAMINEAE) ( Monoporites annu/atus)
The present-day abundance of grass in the tropics is largely confined to

more open vegetation such as is found in coastal savannah's, river valleys, montane
areas and in the Caribbean area above all in the Llanos plains. In general a drier
climate with marked rainy seasons favours the development of extensive grass
areas more than a humid one.
The absence of grass pollen in Cretaceous and Paleocene sediments of the
tropics, its first occurrence in low percentages in the Eocene and the increase in
abundance in the Neogene probably reflect the gradual development of the Poa-
ceae. According to GOTHAN and WEVLAND (1964), the family has been recorded
from the Cretaceous of the present-day temperate areas on macrofossil remains.
However, DRtJC~ (1967) confirms the absence of grass pollen for the Upper
Cretaceous-Paleocene of California. The development of grass vegetation in the
present-day tropics may, therefore, have been somewhat later than in the temperate
areas.
Although specific and generic determinations are hardly possible, it is likely
that the increase in quantity of Poaceae pollen occurred simultaneously with an
increase in taxonomic variety. The significance of grass pollen for the Tertiary
stratigraphy varies according to the area. In Borneo short-lived increases could
tentatively be related to phases of igneous activity in the hinterland, resulting in
ash showers which may have temporarily destroyed the forest cover of large areas.
Any evidence for widespread occurrence of savannah's is lacking in this humid
area. In Nigeria major fluctuations in the grass curve probably reflect the shifting
boundary between forest and savannah in the lowlands.
The greatest abundance of grass pollen has been found, however, in the
Caribbean area. Here open vegetation with a dominantly grass-covered soil must

have been widespread in the Neogene, just as it is today. Source areas may have
been the ancient Llanos, coastal savannah's, or alluvial plains in river valleys,
such as described by WYMSTRA (1967) from the Magdalena valley in Colombia.
That large amounts of grass pollen are produced in such environments is clear
from the diagrams presented by WYMSTRA and VAN DER HAMMEN (1966) and
WYMSTRA (1967). Transport of pollen could be by river or by air currents. The
fact that in Recent Orinoco delta sediments grass pollen proved to be scarce
(MULLER, 1959) is due to the dominance of swamp vegetation lacking in grass in
the delta area proper.
Since human interference can be excluded in the Mio-Pliocene, this is an
indication that open vegetation rich in grasses may be originally of entirely natural
origin in the tropics, although its wide expanse nowadays is largely caused by man,
a conclusion also drawn by WYMSTRA and VAN DER HAMMEN (1966).
A parallel development may be seen in the mid-continent area of North
America, where the wide expanse of grassy plains is generally also supposed to
date from the Neogene.
ARECACEAE(PALMAE)
Nypa (Spinizonocolpites baculatus, S. ech&atus)
Of this genus, at present restricted to the mangroves of the Indo-Malesian

area, a fairly extensive fossil record of both fruits and pollen exists (ref. TRALAU,
1964; MULLER, 1964). The origin of Nypa must be sought in the Upper Cretaceous,
since the ancestral pollen type Spinizonocolpites baculatus proved to be present
already in the Senonian of all three areas investigated. In Sarawak the earliest
records date probably from the post-Turonian (MULLER, 1968). Neither place of
origin nor pre-ancestral type are known as yet.
It would appear that Nypa early reached a pantropical distribution, which
it maintained during the Paleocene and Eocene. The widest extension of range
to the north took place during exceptionally warm phases in the Eocene, when it
grew along the Tethys shores, as testified by the presence of fossil pollen and fruits.
The occurrence of fossil Nypa fruits at latitude 52N in the London Clay
has recently been discussed by VAN STEENIS (1962a,b), who believes them to
represent allochthonous drift material, and by TRALAU (1964), who considers the
fruits to have been of local origin. This problem is connected with the evaluation
of the climate as derived from the well-known London Clay flora. Although the
palynological evidence presented here has no direct bearing on this controversy,
it may be pointed out that the pantropical occurrence of Nypa in the Eocene makes
it likely that it grew at least along the southern shores of the Tethys sea, which,
during the warmest phase of the Eocene would certainly have been hot enough
for Nypa to grow there. It then remains to be decided by further micro- and macro-

botanical investigation whether the European fossil fruits were derived by drift
from the southern Tethys shore or grew locally for a short time also on the northern
shores.
The disappearance of Nypa from the Tethys area has probably been caused
by decreasing temperature, but the disappearance at the end of the Eocene, both
from the Caribbean and Nigeria, as shown by the pollen record, is more difficult
to explain. Possibly an increase in aridity, coupled with the development of a more
pronounced seasonal climate, may have been operative.
BETULACEAE
AInus (Alnipollenites verus)
Identification of AInus pollen grains in tropical deposits is always a striking

event for a palynologist familiar with temperate floras.
The stratigraphical distribution of Alnus pollen in the tropics has clearly
been proved to be related to the presence of mountains of sufficient dimensions
to provide the temperate climate required and to the connections with more
northerly mountain ranges which provided migration routes.
In Borneo Alnus pollen was found to be abundant in Oligocene-Lower
Miocene sediments associated with a mid-Tertiary orogenic phase. Its gradual
decrease in quantity during the course of the later Miocene and Pfiocene could
be related to a gradual levelling of the mountains (MULLER, 1966). Immigration
must have been from the north in the Palaeogene.
In Nigeria Alnus has not been recorded, no doubt due to the absence of
suitable north-south mountain connections in the western part of Africa.
In the Caribbean area the fairly sudden appearance of Alnus pollen in
sediments of Pleistocene age is evidence of the fairly recent immigration of the
genus into the Southern Hemisphere, where it occurs at present at altitudes above
1,000 m in the Andes as far west as the coastal range of Venezuela and as far south
as Peru. Its migration from the north over the comparatively low mountain ranges
of the--by then closed--Panama isthmus was certainly facilitated by the first
colder phase of the Pleistocene.
PROTEACEAE
Guevina type (Proteacidites dehaani)
The presence of this pollen type in Senonian sediments of Nigeria and the
Caribbean would indicate a formerly more extended area of certain genera of
Proteaceae nowadays mainly restricted to the Southern Hemisphere (Guevina and
Lomatia in Chili, Leucospermum in South Africa, Lomatia, Cardwellia and Steno-
Rev. Palaeobotan. PalynoL, 6 (1968) 189-348 273

carpus in Australia). It is noteworthy that neither the Guevina type nor any other
proteaceous pollen type was found in the Upper Cretaceous and Paleocene of
northwestern Borneo (MULLER, 1968). On the other hand, closely related pro-
teaceous pollen was described by BRATZEVA(1965)from the Upper Cretaceous of
eastern Siberia and by ANDERSON (1960) from the Upper Cretaceous of New
Mexico.
These observations would suggest that certain Proteaceae had been able
to migrate along the Pacific coast of South and North America, as far as eastern
Asia. Their subsequent disappearance from this area must then have taken place
at the Cretaceous-Tertiary transition.
OLACACEAE
Anacolosa type (Anacolosidites cf. luteoides)
The form genus Anacolosidites may, according to ERDTMAN(1952), comprise

pollen derived from the genera Anacolosa, Cathedra and Ptychopetalum (Olaca-
ceae). The first-mentioned genus has a predominantly lndo-Malesian distribution
(19 spp.) with only one species (A. unc~fera) described from central Africa and
one from Madagascar. The other genera occur in Brazil (Cathedra, 5 spp.) and
in tropical America and west Afl'ica (Ptychopetalum, 6 spp.). Pending a complete
palynological study of this pollen-morphologically well-characterized group, the
fossil types are united here under the Anacolosa type, although some variation in
ornamentation of wall is known to exist.
The fossil record shows the first occurrence for Australia to be in the Paleo-
cene (HARRlS, 1965). In Borneo the first occurrence is also in the Paleocene (MuL-
LER, 1968), and the type is regularly present throughout the Tertiary, although a
reduction in abundance can be observed in the Mio-Pliocene.
In Nigeria the first occurrence is also in the Paleocene, but it is regularly
present only UP to the Miocene, after which it becomes very scarce, leading appar-
ently to the present-day rarity of Anacolosa and Ptychopetalum in Africa.
In the Caribbean area Anacolosa-type pollen makes its appearance at the
Paleocene-Eocene transition and is common only in the Lower and Middle Eocene.
In Europe the type has been recorded from the Paleocene and Eocene by
ERDTMAN (1954) and KRUSZSCH (1959).
From these observations it is clear that as early as Paleocene time the group
of olacaceous plants producing the Anacolosa-type pollen (Anacolosa, Cathedra
and Ptychopetalum) were widely distributed in Australasia, Africa and Europe,
while South America was reached somewhat later.
Noteworthy is finally the contraction of area and gradual reduction in
abundance during the Neogene.

FABACEAE (LEGUMINOSAE)--Caesalpinioideae
Caesalpinia type ( Margocolporites vanw(]hei)
The earliest record of this highly characteristic pollen type dates from the
Middle Eocene in the Caribbean, where it remains present up to Recent. In Nigeria
the first appearance is in the Upper Eocene and it is remarkably restricted in its
range there, disappearing again in the Lower Miocene.
In Borneo it is absent in Cretaceous-Paleocene sediments, but its base is
not accurately known. In the Neogene it is fairly common, however. RAMANUJAM
(1966) has recorded similar pollen types from the Miocene of south India. Since
a number of genera may be represented, it is difficult to be positive about the
phyto-geographical significance of these facts.
It is possible that Caesalpinia bonduc, which prefers a coastal habitat, has
produced a large proportion of the fossil pollen of this type, but in the Caribbean
area Caesalpinia coriaria, which is common in the dry thorn forests along the
northeastern coast of South America, may also have contributed. Further detailed
studies of this group of pollen types will no doubt result in interesting conclusions.
uyACl~AE--Ctenolophoniideae
Ctenolophon
The following types can be distinguished: Ctenolophon engleri type (Cteno-

Iophonidites costatus), Ctenolophon type A (Ctenolophonidites lisamae), Cteno-
lophon parv(folius type (Retistephanocolpites williamsi).
The oldest known type at present is the Ctenolophon engleri type, which
occurs already in the Senonian of Nigeria, is absent from the Paleocene, but
reappears and remains regularly present in younger sediments. The parent plant
Ctenolophon engleri is still living in west Africa.
A closely similar pollen type, Ctenolophon type A, has a sharply defined base
occurrence in the Paleocene of the Caribbean area, but disappears from the record
at the Paleocene-Eocene transition.
The third pollen type, the Ctenolophon parvifolius type, is first found in
Nigeria in the Paleocene, disappearing at the Eocene-Oligocene transition, but
present in the Neogene of Borneo, where the parent species C. parvifo#us still
lives today.
Ctenolophon engleri in Africa and C. parv(folius in the lndo-Malesian area
are the only known species of the genus.
From these data it is now possible to deduce tentatively the following history
for the genus Ctenolophon. Origin probably in Upper Cretaceous time, possibly

in Africa, followed by an early differentiation into an engleri and parvifolius pollen
type, the engleri type remaining present with a short interruption throughout the
Tertiary and Quaternary of Africa. Migration of a species with pollen closely
related to that of Ctenolophon engleri to South America in the Paleocene, shortly
followed by extinction. Migration eastwards of the parv(folius type reaching
Malesia probably in the Eocene, becoming extinct in Africa. This history in com-
bination with the peculiar morphology of the pollen grains supports the taxono-
mical opinion that the genus Ctenolophon is both ancient and sharply differen-
tiated.
The pollen types described by A. R. Rao and K. P. Vimal from Indian
lignites from Travancore of probably Miocene age and tentatively compared by
ERDTMAN (1956) to C. engleri are left out of account here, since descriptions and
illustrations are insufficient for proper comparison.
MALPIGHIACEAE
Brachypteris type (Perisyncolporites pokornyi)
The fossil record for this characteristic pollen type, which occurs in several
genera of Malpighiaceae, dates from the Middle Eocene of the Caribbean area.
In Nigeria it appears slightly later and remains much more scarce than in the
Caribbean area. This is in agreement with present-day distribution of the family,
which is best represented in South America, in which continent its origin may have
lain.
Fruits of Malpighiaceae recorded from the Tertiary of Europe have so far
not been accompanied by fossil pollen. The less characteristic pollen of the Mal-
pighiaceae growing in Borneo today, has not yet been found fossil.
EUPHORBIACEAE
Alchornea (Psilatricolporites operculatus)
The highly characteristic pollen grains of Alchornea are first found in the
lower part of the Middle Eocene of the Caribbean area. In Nigeria they appear
somewhat later, in the upper part of the Middle Eocene. In Borneo their first
origin cannot yet be precisely given. It would appear that the present-day pan-
tropical distribution of the genus was achieved within a comparatively short inter-
val of time in the Eocene.
Amanoa type (Retitricolporites irregularis)
This pollen type appears at the base of the Eocene in the Caribbean area
and in Nigeria. It is most likely that the genus Amanoa, which occurs both in

South America and in Africa, produced the pollen grains, and this would indicate
migration across the Atlantic Ocean in Early Eocene times.
BOMBACACEAE
Bombax ceiba type (Bombacacidites annae)
The first occurrence of the genus Bombax can be dated as Paleocene in the
Carribbean area. No ancestral types are known, however, and the rather sharp
lower limit of occurrence suggests immigration from elsewhere. In Nigeria similar,
but not identical, pollen types are known from the Paleocene-Eocene transition
onwards. In the Caribbean area the Bombax ceiba type becomes rare during the
Eocene and has given rise to a host of related types which cannot be discussed
here because of their limited stratigraphical interest.
Catostemma (Jandufouria seamrogiformis)
The very characteristic Catostemma pollen grains are not found below the
Upper Eocene in the Caribbean area. They show strong local dominance especially
in the Oligocene and Lower Miocene, which may be taken as an indicator of the
presence of rain forest. The origin of this pollen type is not known, but related
tricolporate pollen grains referable to the genus Aguiaria occur already in the
Lower Eocene.
CLUSIACEAE (GUTTIFERAE)
Symphonia globulifera type (Pachydermites diederixi)
Symphonia globulifera occurs at present in coastal swamps of tropical Africa

and South America and its pollen is highly characteristic. It is the only species of
the genus with a relatively wide distribution, the other sixteen species being restrict-
ed to Madagascar, where they occur in a variety of habitats.
In Nigeria there is clear evidence of first appearance in the Middle Eocene,
but in the Caribbean area the first appearance is much later and can be fairly
accurately dated as Early Miocene.
It would appear reasonable, considering the concentration of endemic and
mainly non-coastal species in Madagascar, to assume Africa and probably Mada-
gascar to have been the centre of origin of the genus. At an unspecified time
Symphonia globulifera managed to reach the west coast of Africa, arriving in
Nigeria at the end of the Eocene and becoming rapidly a dominant species in the
coastal vegetation. After a delay corresponding to not more than approx. 25 mil-
lion years the species succeeded in crossing the Atlantic Ocean and reached the

Caribbean area. The means by which this dispersal took place are as yet wholly
obscure. At this late stage in the Tertiary the topography must have been similar
to present-day conditions, and the chances of Symphonia seeds crossing the Atlantic
must have been extremely small. That this nevertheless occurred can hardly be
doubted, however, and the long time delay is a silent witness of the numerous
unsuccessful attempts which must have preceded the final success.
The fact that Symphonia globulifera grows in coastal swamps may have
slightly increased its chances of long-distance seaborne dispersal.
Similar problems are involved in other transatlantic distribution patterns,
but the case of Symphonia has been discussed at length, since here the data are
most reliable.
LYTHRACEAE
Cuphea (Striasyncolpites zwaardi)
The typical pollen grains of the genus Cuphea, which is restricted to America,
first appear in the palynological record of the Caribbean area in the Middle Mio-
cene, but no ancestral forms are known.
Crenea ( Verrutricolporites rotundiporis)
In the Caribbean area Crenea pollen is first found in the Upper Eocene and
is especially common in the Lower Miocene, associated with dominant Rhizophora
pollen in coastal sediments. In Nigeria it first appears at the base of the Miocene
and is fairly abundant during the Lower Miocene, also in a coastal environment,
only to disappear from the stratigraphical record in the Middle Miocene. In the
Caribbean area a decrease sets in at approximately the same time, but Crenea
pollen remains present in small numbers up to the present day.
The palynological data thus provide a fairly complete record of the history
of the genus Crenea. Its origin must be sought in the Eocene of South America
and it apparently succeeded in crossing the Atlantic Ocean, probably during the
Oligocene. It then flourished for a short time on the western African coast in the
same environment as in South America. The reasons for its disappearance from
Africa and its almost simultaneous decrease in abundance in the Caribbean area
are unknown. As in the case of Symphonia, the coastal habitat of Crenea no doubt
was a principal factor which significantly increased the chances of dispersal.
SONNERATIACEAE
Sonneratia
The following types can be distinguished: (l) Ancestral types: (Florschuetzia

trilobata, Florschuetzia semilobata); (2) Sonneratia caseolaris (Florschuetzia levi-
poli) ; (3) Sonneratia alba (FIorschuetzia meridionalis).
The relations between the FIorschuetzia pollen types, which so far only have
been found in Borneo, and recent Sonneratiaceae are shown on Fig.16. From
this diagram it can be seen that FIorschuetzia trilobata ranges from probably Upper
Eocene to Middle Miocene. In its morphology this pollen type shows affinity to
certain lythraceous pollen types, notably Lagerstroemia, and it appears not un-
reasonable to postulate derivation from lythraceous stock during the Eocene.
During the Lower Miocene Florschuetzia semilobata enters the stratigraphical
record, which type, although transitional specimens are scarce, is assumed to have
been derived from F. trilobata by a morphological change from a continuous
tectum to a discontinuous one, resulting in a verrucate sculpture, equally distributed
over the whole grain, This species has a rather more restricted distribution than
F. trilobata and becomes extinct slightly earlier. It cannot be closely matched with
the pollen of a living species of Sonneratiaceae or Lythraceae, although there is
a certain resemblance with the pollen of Sonneratia gri~thii.
Within a relatively short span of time after the appearance of Florschuetzia
semilobata, a third related species is noticed, F. levipoli. Between these two pollen
types transitions are rather common, with a gradual transition towards a concen-
tration of the verrucate sculpture on the equatorial belt, the formation of tectate-
psilate polar caps and the loss of the trilobate condition. Florschuetzia levipoli
further shows a statistically significant increase in size during the Mio-Pliocene,
and ultimately can be connected with the pollen of the Recent species Sonneratia
caseolaris.
Slightly later again, in Middle Miocene times, a further differentiation takes
place and FIorschuetzia meridionalis makes its appearance, characterized by further
modifications in the direction of a meridionally differentiated sculptural pattern,
intra-areolate structure on the poles and, compared with F. levipoli, a further
increase in size. Transitions with F. levipoli occur in the early part of its range.
The youngest fossil specimens of Florschuetzia meridionalis are identical
with pollen of the Recent species Sonneratia alba.
The changing percentages of the Florschuetzia pollen types are shown on
the graph, and are thought to reflect the degree of dominance in the coastal vege-
tation for each parent species.
It would appear from this graph that the ancestral species Florschuetzia
trilobata remains present in significant numbers after having given birth to the
younger offshoots F. semilobata and F. levipoli. Its disappearance is distinctly
gradual. It is not certain whether the parent plant of F. trilobata actually grew in
the mangrove environment, but its abundance in Oligo-Miocene coastal sediments
of the type in which later Sonneratia and Rhizophora pollen become dominant,
strongly suggests this having been the case.
The youngest offshoot of the Sonneratia complex proved to be Sonneratia

alba, but whether this species originated from S. caseolaris or from the much rarer
S. ovata, of which unfortunately no fossil record exists, must remain uncertain
for the present. Also the relation to the Indian species S. apetala is unknown for
lack of fossil evidence.
In this connection it must be stressed that, although it is possible to arrange
the pollen types described here in transitional sequence, starting from the simply
built FIorschuetzia trilobata and ending with the complex Sonneratia alba pollen,
this in itself does not constitute proof of phylogenetic relationship. The fact that
this morphological sequence can be shown to have time-stratigraphical significance,
the more complex pollen type being the youngest, of course argues for a phylogenetic
relationship. However, the rather abrupt change between FIorschuetzia trilobata
and F. semilobata is hard to visualize without a sudden change in genetic consti-
tution of part of the F. trilobata population. Or, it may be that this change did
not actually take place in northwestern Borneo but elsewhere, and that the newly
evolved F. semilobata immigrated soon afterwards.
The inferred gradual replacement of the unknown parent plant of the fossil
FIorsehuetzia trilobata pollen species by the modern Sonneratia group might reflect
ecological competition within the mangrove environment. The development of
Sonneratia alba, on the other hand, can be seen as the later origin of a species
which is, of all Sonneratia species, best adapted to the marine environment and
thus represents a culmination of the evolutionary process. This holds good not
only for the ecological adaptation of S. alba, but also for its pollen type, which
is the most complex and largest of all Sonneratia pollen types.
The large degree of variability in size and morphology observed in the pollen
of the living populations of Sonneratia alba, caseolaris and ovata in different parts
of their ranges, would indicate that the potentialities for further evolutionary
change in pollen characters are by no means exhausted. However, it will be difficult
to predict the course of future changes in the absence of any knowledge as to
what is precisely the selective value of pollen characters.
In particular the size increase in the course of the Neogene observed both
in Sonneratia caseolaris and alba pollen deserves special attention in this respect.
It has been stated by COVAS and SCHYACK (1945) that pollen size is related
to length of style, but in the Sonneratiaceae this relation does not exist, since in
the genus Duabanga the style is as long as in Sonneratia, while the pollen grains
are much smaller. Nor are the size differences between pollen of Sonneratia alba,
caseolaris and ovata correlated with any difference in length of style, which is
approximately equal for all three species. With the scarce data available it would
appear possible that the increase in pollen size is linked with the gradual physiolog-
ical adaptation to the saline mangrove environment, since S. alba, which can stand
the highest salinities, also has the largest pollen.
It is further of interest that pollen fertility in the hybrid S. alba ovata
is higher than in S. alba caseolaris (MULLER and Hou-L1u, 1966) indicating

that S. alba is probably more closely related to S. ovata than to S. caseolaris.
A further elucidation of these problems will have to be deferred until the
discovery, probably in other areas, of fossil Sonneratia ovata, S. griffithii and S.
apetala pollen.
RHIZOPHORACEAE
Rhizophora type ( Zonocostites ramonae)
Included in this type are the pollen grains of the genera Rhizophora, Bru-
guiera and Ceriops. Although differences in pollen type are known to exist between
the genera and constituent species, it has proved impracticable to try to separate
them on a routine basis. The following account therefore relates mainly to the
history of the tribe Rhizophoreae, to which the genera mentioned belong and
which are all mangroves.
As far as known the Rhizophora pollen type is unique and cannot be confused
with pollen from other taxa (cf. also MULLER, 1964). The fossil record in Borneo
shows that the Rhizophora type is absent in Cretaceous and Paleocene sediments.
It probably occurs first in Eocene sediments, but due to poor preservation this
cannot yet be stated with certainty. In the Oligocene it is definitely present, but
in small quantities only; it gradually increases in abundance to become in the
Miocene-Pliocene the dominant element in the coastal microflora.
In the Caribbean area it is definitely known for the first time in Upper
Eocene sediments, being absent in older strata. Again it is only in the Miocene
that Rhizophora reaches the high percentages so characteristic of the Neogene and
Recent sediments of coastal or marine origin.
In Nigeria, in contrast, it is definitely absent from pre-Miocene sediments
and starts occurring rather suddenly in high percentages in the lowermost Miocene.
If this fossil evidence is combined with the present-day distribution pattern
of the tribe Rhizophoreae as discussed recently by VAN SVEENIS (1962a), the fol-
lowing tentative history can be reconstructed:
(1) Origin of the tribe Rhizophoreae in the Eocene of southeastern Asia,
since today the largest number of genera and species occur there, as well as inland
relatives (Carallia, Anisophyllea).
(2) Extension of the range of a few species of Rhizophora eastwards across
the Pacific Ocean to tropical South America, reaching the Caribbean area via the
gap in the Panama isthmus, most probably during the Eocene, followed by develop-
ment of local species.
(3) Crossing of the Atlantic Ocean in the Early Miocene and settling of the
American species of Rhizophora on the west coast of Africa.
(4) Extension of the range of the Indo-Malesian species of Rhizophora
westwards to east Africa at an as yet undetermined time.

VAN STEENIS (1962a) has concluded that the genus Rhizophora was unable
to migrate along the southern tip of South Africa and along the southern shore
of the Tethys sea, because of adverse climatic conditions, thus explaining the fact
that the east and west African coasts have no Rhizophora species in common.
For South Africa this has certainly been the case, but for the Tethys it would
appear unlikely.
Firstly, the Eocene Period was characterized, as already mentioned in the
discussion on Nypa distribution, by an expansion of the tropical belt to higher
latitudes and, even at its present location the southern Tethys shore could during
this period have been favourable for the growth of Rhizophora. Secondly, it is
generally assumed that the formation of the Alps was accompanied by a certain
amount of crustal shortening during the considerable compression and over-
thrusting of the ancient Tethys sediments and therefore the Eocene southern shore
could easily have been situated 100-200 km further southwards than its present-day
location.
It is, therefore, conceivable that the failure of the Indo-Malesian genera and
species of Rhizophoreae to reach west Africa is due not so much to a cool climate
along the Palaeogene Tethys shores but to the fact that these plants had not fully
evolved yet. By the time they had developed the present species and had started
to extend their range, i.e., in the Oligo-Miocene Period, the Tethys connection to
the Atlantic Ocean had either already closed or was situated too far north in a
cooler climatic zone. This alternative hypothesis has of course to be tested by
studying the pollen content of the coastal sediments along the ancient Tethys
shores.
ACANTHACEAE
Justicia type (Multiareolites formosus)
Although some of the genera which produce this pollen type have a pan-
tropical distribution, the occurrences in the Neogene appear to be restricted to
the Caribbean area and Nigeria. It is possible that the extension of the range to
the Indo-Malesian area took place relatively late in the Mio-Pliocene.
Trichanthera type ( Multimarginites vanderhammeni)
The genera which produce this pollen type are restricted at present to the
American tropics and accordingly the fossil pollen type is known only from the
Caribbean area.
From the fact that acanthaceous pollen types are only known from the
Neogene it might be inferred that this family is of relatively late origin.
282 Rev. Palaeobotan. Palynol,, 6 (1968) 189-348

ASTERACEAE (COMPOSITAE)
The following three Asteraceae pollen types have been distinguished: (l)
Tubul([torae type (Echitricolporites spinosus); (2) Liguliflorae type (Fenestrites
spinosus); and (3) Ambrosia type (Echitricolporites mcneillyi).
Future detailed studies on well-preserved fossil material will no doubt lead
to a further subdivision of these type groups.
The Tubuliflorae type appears to be the oldest and the most widespread of
the three Asteraceae pollen types. It has not yet been found with certainty in
pre-Miocene sediments.
In all three areas investigated the very gradual increase in abundance appears
to be contemporaneous, but the high percentages reached in the Mio-Pliocene of
the Caribbean area have not been found so far in Nigeria or Borneo.
The Liguliflorae type is much rarer and has only been regularly observed in
the Caribbean area, where it appears slightly later than the Tubuliflorae type.
The Ambrosia type, only known from the Caribbean area, appears still later.
Asteraceae are widespread in the present-day tropics, but are more common
in open vegetation types, such as the savannah or higher montane vegetation than
in the closed lowland rain forest. This largely explains the considerable quantities
of Asteraceae pollen observed in the Caribbean area Mio-Pliocene in comparison
with the relative scarcity in an area such as northwestern Borneo.
Earlier views (KUvL et al., 1955) that, on the basis of the palynological
record, Asteraceae emerge as one of the youngest developments within the angio-
sperms are thus confirmed. The place of origin is still obscure, however, and may
very well have been situated outside the tropical belt, the time of origin presumably
being mid-Tertiary.
The relatively quick and simultaneous extension of the range of Asteraceae,
especially of the Tubuliflorae type, as shown by the palynological data, probably
reflects the efficient seed dispersal which characterizes the family.
SUMMARY OF BOTANICALRESULTS
The above examples demonstrate, when an explanation of the stratigraphical

distribution patterns is attempted, that a number of problems relating to origin,
dispersal and extinction of plant taxa are involved. First of all it must be empha-
sized, that the data presented give direct evidence only on the development of
pollen and spore characters. The first development of a combination of characters
similar to that found in the pollen type of a recent taxon does not necessarily mean
that all the other characters of flower, fruit and leaves, which define this taxon,
also originated at the same time. However, the gradual transformation of a pollen
type from an extinct form into one known from a Recent taxon may be taken to

represent reasonable evidence for time and place of origin. On the other hand a
sudden appearance of a new pollen type in the stratigraphical record of one area
will in most cases be due to the presence of a stratigraphical hiatus or to immigra-
tion.
So far only a general dating of first appearance has been possible for some
taxa (e.g., Nypa, Ctenolophon, Asteraceae) but ancestral types mostly have not
been found, signifying no doubt that the precise place of origin has not yet been
sampled. Only in the case of Sonneratia evidence for continuity between an ances-
tral extinct pollen type and younger, still living offshoots was found. It is expected
that future detailed studies of many more, continuous sections will bring to light
more examples of this kind.
Once the palynological record has registered the origin of a new taxon, it
may also provide evidence for its dispersal by recording the extension of range
in time and space. The most striking examples of this are represented by those
cases where transatlantic migration has to be postulated (Symphonia, Crenea,
Amanoa, Rhizophora, Malpighiaceae).
The significance of these data can of course only be evaluated if reasonable
estimates of dispersal probabilities and reliable palaeogeographic information are
available. As regards the first point, transatlantic dispersal of the taxa involved
under present-day conditions appears virtually impossible. Also geological evidence
is not in favour of closer geographic connections between tropical Africa and
South America in the Tertiary. It would, therefore, appear justified to regard
the cases cited as indeed reflecting rare instances of long-distance dispersal of
diaspores, illustrating SIMeSON'S (1952) opinion that, given enough time even
the most unlikely events will occur. Nevertheless, at the same time, it must be
stressed that a general exchange of floral elements between Africa and South
America is definitely not visible in the Tertiary palynological record. The un-
doubted floristic affinities between these two areas, referred to by plant geographers
from ENGLER (1905) onwards must date, therefore, from earlier periods.
The palynological record also shows many examples of contraction of
range, leading in extreme cases to extinction. It is generally impossible to pinpoint
the causes, although in a general way climatological changes, which can be deduced
from large-scale changes in the palynological record, must play an important part.
Other contributory factors, which may well be responsible, are competition due
to immigration or the evolution of new floral elements.
PLATE I
1. Foveotriletes margaritae (VAN DER HAMMEN) nov. comb., Colombia.

2. Foveotriletes margaritae (VAN DER HAMMEN) nov. comb., Venezuela.
3. Crassoretitriletes vanraadshooveni nov. gen., nov. sp., Nigeria, holotype.
Magnification 1,000.

PLATE[
i i i Iii
Re v. Palaeobotan. Palynol., 6 (1968) 189-348 285

SYSTEMATICALPART
Division I Sporites H. POTONIL 1893
Class A Triletes (REINSCH, 1881) POTONIE et KREMP, 1954
Genus Foveotriletes (VAN DER HAMMEN, 1954) ex R. POTONII~, 1956
Foveotriletes margaritae (VAN DER HAMMEN, 1954) nov. comb.

(Plate I, 1, 2)
Literature: Triletes margaritae VAN DER H AMMEN, 1954, p.102, pl.17.

Remarks: The original short diagnosis is here enlarged. The identity of the
specimens described by us has been confirmed by Van der Hammen.
Description." Single grain, radially symmetrical, anisopolar with rounded
distal pole and slightly pointed proximal pole, in polar view subangular; shape
spherical-suboblate. Laesura trilete, relatively short (7-13/~), straight, sometimes
inconspicuous. Foveolate; foveolae circular-oval, ~-2/~ wide, 1-2 ~t apart, exine
1-2 ~ thick.
Dimensions: 44-56 # (equatorial diameter).
Variability: In size and coarseness of sculpture.
Comments: This species differs from Filtrotriletes nigeriensis VAN HOEKEN-
KLINKENBERG, 1966 in its thinner wall and shorter, less pronounced laesura.
Distribution: Lower part of the Proxapertites operculatus Zone in Nigeria
and northern South America, decreasing in frequency in the upper part of the
Retidiporites magdalenensis Zone. Extinct in younger zones.
Taxonomic affinities: Unknown, although some similarity exists with pub-
lished illustrations of Lindsaya orbiculata, Ophioglossum falcatum, O. concinnuum,
and Botrychium subbiJbliatum, none of which was available for comparison.
Genus Crassoretitriletes nov. gen.
Derivatio nominis: Name derived from the coarsely reticulate ornamen-

tation.
Diagnosis: Spherical, trilete, entirely coarsely reticulate with undulating
muff; thick-walled, laesura indistinct.
Type species: Crassoretitriletes vanraadshooveni nov. sp.
The type material of the new species described below has been deposited at
the Palynological Section, Botany Department, Municipal University, Amsterdam
(The Netherlands).

Crassoretitriletes vanraadshooveni nov. sp.
(Plate I, 3; holotype)
Derivatio nominis: Named in honour of Mr. B. van Raadshooven in recog-

nition of his contribution to Venezuelan palynology.
Holotype: Slide No. TC-151, well Egbema-l, 3015 ft., Nigeria.
Description: Single grain, radially symmetrical, anisopolar, with rounded
distal pole and slightly pointed proximal pole, in polar view almost circular.
Laesura trilete, indistinct, often covered by sculpture. Reticulate over entire sur-
face; muri undulating, 3-4/z wide, 2/z high, lumina 2-4/z wide, 6-12 /z long.
Exine 14-2 # thick.
Dimensions: 58-101 /~ (equatorial diameter).
Variability: Moderately variable in size and in coarseness of sculpture. In
Borneo a distinct variety has been observed in which the wall is finely perforated,
but this is included in the species.
Distribution: Pantropical approximately from the base of the Crassoretitri-
/etes vanraadshooveni Zone upwards, but disappearing in northern South America
at the base of the Echitricolporites spinosus Zone. First appearance in Borneo
slightly earlier than in the Caribbean area, in Nigeria even at the base of the
Verrutricolporites rotundiporis Zone.
Taxonomic affinities." Close correspondence virtually amounting to identity
exists with Lygodium microphyllum (Plate II, 1) (~- Lygodium scandens). This is
based on Recent material collected in Nigeria and Borneo. However, NAYARet al.
(1964) have recently described the spore of Lygodium microphyllum as tuberculate
distally and smooth proximally. Evidently a different species is involved and
checking of the original herbarium sheets is necessary to resolve this discrepancy.
Genus Cicatricosisporites POTON~ et GELLZTJCH, 1933
Cicatricosisporites dorogensis POTONII~et GELLETICH,1933

(Plate I1, 2)
Literature: Cicatricosisporites dorogensis POTONI~et GEt LETICH, 1933, p.522,

pl. 1, fig. 1-5.
Striatriletes susannae VAN DZR HAMMEN, 1956d, p.l 15, fig.5.
distal pole and pointed proximal pole; in polar view semi-angular to almost cir-
cular. Laesura trilete, short. Contact area of proximal face unsculptured, remainder
of wall striate or striate-rugulate. Striae 1~-2 # high, ~--1/z wide; grooves between
striae lz-2~/~
t l wide. Exine 1-2/~ thick.
Dimensions: 55-70 ~u (equatorial diameter).
Variability: Moderately variable in size and ornamentation. Compared to

Cretaceous representatives of this species the variability is much less in our Ter-
tiary material and the spores are certainly identical with the type as originally
described by POTONI~ and GELLEVlCH(1933), and later on from the same locality
by KDVES (1960, 1961).
Distribution: In northern South America ranging from the base of the
Verrucatosporites usmensis Zone to the lower part of the Magnastriatites howardi
Zone (top Cicatricosisporites dorogensis Zone), higher occurring sporadically~
possibly reworked. In Nigeria ranging from slightly above the base of the Ver-
rucatosporites usmensis Zone to approximately the same level in the Magnastria-
tites howardi Zone (top Cicatricosisporites dorogensis Zone). In Borneo virtually
absent during the whole of the Tertiary.
Taxonomic affinities: Similar spores are found in the genera Anemia and
Mohria (Schizaeaceae).
Genus Magnastriatites nov. gen.
Derivatio nominis: Name derived from the coarsely striate ornamentation.

Diagnosis." Spherical, trilete, coarsely striate, except on the proximal contact
area which is surrounded by a circular ridge. Striae continuous, grooves about as
wide as ridges, size around 100 ~t.
Type species: Magnastriatites howardi nov. sp.
Comments." The genus differs from Cicatricosisporites in the relatively smaller
number of coarser striae, the circular ridge surrounding the smooth proximal
contact area and the much larger size.
Magnastriatites howardi nov. sp.

(Plate IIl, 1; holotype)
Derivatio nominis: Named in honour of Mr. E. A. Howard in recognition

of his contribution to Trinidadian palynology.
Holotype: Slide No. TC-152, well Dificil-1, 1300-1308 ft., Colombia.
distal pole and more pointed proximal pole, in polar view nearly circular; shape
suboblate-spherical. Laesura trilete, costate; costae 2 # wide. Contact area of
PLATE II
1. Lygodium microphyllum R. BROWN, Recent.

2. Cicatricosisporites dorogensis POTONII~ et GELL~TJCH, Nigeria.
3. Verrucatospor#es usmensis (VAn DER HAMMEN) nov. comb., Nigeria.
4. Stenoehlaena palustris BEDDOME, Recent.
288 Rev. Palaeobotan. Palynol., 6 (1968) 189 348

~ ~i ~ ~ ~ !~ iii~i~%~!!! !iii~iiii~i'!~i!i~ iiiii~i~I~ ~ !!~,!~i~i~i~,~iii~i~ii~i!~!~i~i~!~!~i
~ili~,~i~,~i~i~i~!iiiiii!~i~i~, ~ ~iiii~ii~ii~ ~iiii ~ ~ ii~ ~!~ ~ ~ ~i!~ iii ~ ~ ~ i~i~
i:~'~i~ii~i~i~ji~!!'~:~~,,i .... ! ~i , i~i ~i~i :~:!~

!i ~,~iiii~i~i~i~ii!ill
~I!~~ i~ii!,~!!~ ~. .
~!~!~i~!!!!i~!i~'!!~i~:i!i~ii:~ii~i!!!!~ii~!!~'~! . . . . !!ii~ iil!ii ~!~,,~i~i i~i:!~i!i~i~~i~i~~ili~ii~iili
! ~ ~ ~
f~
proximal face psilate, surrounded by a circular ridge, which makes contact with
the striate ridge pattern at the points of the laesura. Remainder of wall coarsely
striate; striae 1-2/~ high, 2-3/z wide; grooves 1~-3 ~t wide. Exine 1~-2~
~ it thick.
Dimensions." 77 132/z (equatorial diameter).
Variability: There is some variability in size and sculpture; thicker striae
may occur in smaller number or much wider grooves. Also the proximal side of
the grain may be locally perforated.
Distribution: Regularly present from the base of the Magnastriatites howardi
Zone upwards, in all three areas. Its first appearance is remarkably sudden and,
as far as can be judged simultaneous. No ancestral forms are known.
Taxonomic affinities." Virtually identical with the spores of the tropical-
subtropical fresh-water fern genus Ceratopteris (species seen: C. thalictroides,
Plate IV, 1, and C. cornuta).
Class B Monoletes IBRAHIM, 1933
Genus Verrucatosporites (PFLuG, 1952)ex R. POTONI~, 1956
Comments: The spores described by VAN DER HAMMEN (1956d) under the
genus Verrumonoletes clearly fall within the circumscription of Verrucatosporites
and this name has priority.
Verrucatosporites usmensis (VAN DER HAMMEN, 1956d) nov. comb.

(Plate II, 3)
Literature: Verrumonoletes usmensis VAN DER HAMMEN, 1956d, p.116, fig.7.

Remarks: The paper cited above consists of a Spanish (VAN DER HA~MEN,
1956C) and translated English text (VAN DER HAMMEN, 1956d). In the former the
species name is given as "usmensoides", in the latter as "usmensis". In agreement
with Van der H a m m e n the name "usmensis" is taken as the correct one, referring
also to Plate I, 7.
Description: Single grain, bilaterally symmetrical, anisopolar, with convex
distal outline and straight or slightly concave proximal outline; in polar view
ellipsoidal. Laesura monolete. Geminate; gemmae 12-2~~ ~ # high and 17-2z
~ /z wide
at base, rather widely and variably spaced. Exine thickness I ,u.
P L A T E III
1. Magnastriatites howardi nov. gen., nov. sp., Colombia, holotype.

2. Grimsdalea magnaclavata nov. gen., nov. sp, Trinidad, holotype.
3. Monoporites annulatus VAN DER HAMMEN,Venezuela.
290 Rev. Palaeobotan. Palynol., 6 (1968) /89 348

>
Dimensions: 39-61 /t, including gemmae.
Comments: The species most probably has evolved from a verrucate ances-
tral type, since transitions are common at the base of its range in Venezuela.
Distribution: First regular occurrence marks the base of the Verru-
catosporites usmensis Zone. In northern South America the species then remains
present throughout the Tertiary, but in Nigeria a sharp decline occurs at the base
of the Verrutricolporites rotundiporis Zone. In Borneo the species remains very
common up to the Recent.
Taxonomic affinities." Although a verrucate sculpture is very common among
fern spores of varied taxonomic affinities, a geminate sculpture in which the gemmae
are separated from each other is much rarer and has so far only been found in
Stenochlaena palustris, an Indo-Malesian climbing fern (Plate I1, 4). It is, however,
possible that other members of this genus show the same pollen type, since only
a restricted number of species was available for comparison. Spores which show
a transition between verrucate and geminate are produced by Phlebodium aureum
and Histiopteris incisa.
Division II Pollenites R. POTONII~, 1931
Class lnaperturatae IVERSEN et TROELS SMITH, 1950
Genus Grimsdalea nov. gen.
Derivatio nominis." The genus has been named in memory of the late Dr.
T. F. Grimsdale, who launched the idea of applying palynology in oil exploration
in 1937.
Type species." Grimsdalea magnaclavata nov. sp.
Diagnosis: Spherical, aperture indistinct or absent, wall thin, intectate,
sculpture of two types, finely scabrate and coarsely clavate.
Grimsdalea magnaclavata nov. sp.

(Plate II1, 2; holotype)
PLATE IV
1. Ceratopteris thalictroides BRONGNIART,Recent.

2. Spinizonocolpitesbaculatus MULLER,Venezuela.
3. Spinizonocolpitesechinatus MULLER,Venezuela.

PLATE IV

Derivatio nominis: Name derived from the clavate ornamentation.
Holotype: Slide No. TC-153, well Fe-100, 2365 ft., Trinidad.
Description: Single grain, possibly bilaterally symmetrical, isopolar, almost
spherical. Aperture absent or indistinct. Wall intectate, densely covered with a
finely baculate-scabrate sculpture as well as with scattered large clavae. Clavae
8-10 # apart, 7-10 # long, 22!-3/~ thick, sunken into locally thickened endexine with
a conical base. Endexine 1 #, but underneath clavae 2-2 /~ thick. Columellae
# thick and long.
Dimensions: 40-62 # (excluding of clavae).
Variability." Some variability in size and wall thickness occurs.
Distribution." Restricted to the Caribbean area where it occurs from the
base of the Grimsdalea magnaclavata Zone upwards to the lower part of the
Alnipollenites verus Zone, where it apparently becomes extinct.
Taxonomic affinities: Unknown, possibly derived from an extinct palm
species.
Class Monoporatae IVERSENet TROELSSMI]'H, 1950
Genus Monoporites (CooKSON, 1947) ex VAN DER HAMMEN, 1954
Monoporites annulatus VAN DER HAMMEN, 1954

(Plate III, 3)
Literature." Monoporites annulatus VAN DER HAMMEN, 1954, p.90, pl.6, fig.4.
Monoporites unipertusus VAN DER HAMMEN, 1956b, p.82, pl.5,
fig.10.
Description." Single grain, radially symmetrical, anisopolar, almost spherical.
Single aperture small, circular, penetrating entire wall, costate; costa 4~-/~ wide,
slightly protruding. Endexine < /~ thick; columellae indistinct <: /~ long and
thick; tectum < # thick, psilate-very finely perforate or scabrate.
Dimensions." 38-43/~.
Variability: Considerable variability in size of grain and pore exists, but
the wall is always rather thin. The surface of the wall is mostly smooth but a finely
scabrate surface occurs occasionally. Since it is mostly unknown whether this
condition is due to corrosion or not, this sculpture type is included.
Distribution: In the Caribbean area and in Nigeria occurring regularly from
the base of the Monoporites annulatus Zone upwards, although with rather pro-
nounced fluctuations in numbers and generally more abundant in the Neogene
than in the Palaeogene. Rare occurrences below the base of the Monoporites
annulatus Zone are known. In Borneo base not yet determined, but absent in
Paleocene and Upper Cretaceous. The significance of the fluctuations of grass
pollen has been discussed in the section "Botanical results".

Taxonomic affinities: The fossil species closely resembles non-cultivated
Poaceae (Gramineae) pollen. Since the type is produced by many different genera,
further identification is not possible at present.
Class Monocolpatae IVERSEN et TROELS SMITtt, 1950
Genus Spinizonocolpites MULLER, 1968
Spinizonocolpites echinatus group

(Plate IV, 2, 3)
Literature: Spinizonocolpites echinatus MULLER, 1968.

Spinizonocolpites baculatus MULLER, 1968.
Remarks: The distinction made by Muller between an echinate and a
baculate form was not made in all areas and consequently for the present review
the two species have to be grouped together.
Description: Single grain, radially symmetrical, slightly anisopolar, since the
grains are separated by a continuous equatorial colpus into two slightly unequal
parts; suboblate-spherical. Colpus ectexinous, somewhat irregularly bordered,
with a narrow indistinct margin of thinning ectexine. Total wall thickness about
1/~; endexine # thick, columellae < 1 /~ long, < /~ thick, covered by a thin,
finely reticulate tectum; lumina < 1 /~, muri # wide and thick. Spines scattered
on tectum, 5-13 # long, 10-12/~ apart, either echinate, conical with a rather blunt
to pointed apex and with the lower part often slightly expanded (Sp. echinatus)
or baculate with rounded or slightly pointed tips, straight or slightly curved (Sp.
baculatus).
Dimensions: 30-62/~.
Variability: The two species grouped here together are in general rather
distinct, but transitional specimens do occur occasionally. In both there is some
variability in thickness of wall and coarseness of reticulation.
Distribution: Occurs throughout Upper Cretaceous and Palaeogene of all
three areas, but disappears from the record in the Caribbean area approximately
in the higher part of the Verrucatosporites usmensis Zone. In Nigeria the species
disappears slightly higher since it occurs in the whole of the Verrucatosporites
usmensis Zone. In Borneo present up to the Recent. In Upper Cretaceous-Paleo-
cene sediments of all three areas, Sp. baculatus is the dominant form, while later
Sp. echinatus is found exclusively. This distribution together with the occurrence
of transitional forms suggest a phylogenetic connection.
Taxonomic affinities: The echinate form is identical with pollen of Nypa
fruticans (Arecaceae or Palmae, Plate V, 1) and cannot be confused with any other
pollen type known at present. The baculate form is not matched yet, and probably

represents an extinct ancestral form. Nypafruticans is at present restricted to the
mangrove environment of the humid Indo-Malesian tropics, from the Ganges
delta to northern Australia.
Genus Proxapertites VAN DER HAMMEN, 1956c
Proxapertites operculatus (VAN DER HAMMEN, 1956c)

(Plate V, 2)
Literature: Monocolpites operculatus VAN DER HAMMEN, 1954, p.89, pl.5,

fig.2-3.
Proxapertites operculatus VAN DER HAMMEN, 1956c, p.105, pl.l,
fig.3.
Description." Single grain, radially symmetrical, slightly anisopola~', since
the grains ,-,re separated by a continuous equatorial colpus into two slightly un-
equal parts; oblate, in polar view rounded angular to oval; colpus ectexinous,
somewhat irregularly bordered by a narrow, indistinct margin of thinning ectexine.
Total wall thickness 1-2/z; endexine < 1 /z thick; columellae indistinct, -: I ,u
long, ~ # thick, covered by a <: 1 /z thick finely reticulate-perforate tectum;
lumina 1 # wide, smaller along margin of colpus, muri it wide, < 1 ,u high.
Dimensions: 35-61 kt.
Variability." In size and shape.
Comments: The grains are here described as monads, while Van der Ham-
men suggests that they are dyads with the apertural faces touching. For a full
discussion of this question see MULLER (1968). The species differs from the closely
related Proxapertites cursus by its finer sculptural pattern, its generally isodiamet-
rical lumina and the finer columellae which are more circular in cross-section.
Distribution: Regularly present in the Proxapertites operculatus Zone in all
three areas, but apparently appearing in Nigeria later than in the Caribbean area.
In the younger zones the distribution is also different. In the Caribbean area
decreasing in number in the Monoporites anmdatus Zone, but irregularly present
also in much younger zones. In Nigeria its greatest abundance appears to fall in
PLATE V
1. Nypa fruticans THUNBERG, Recent.

2. Proxapertites operculatus (VAN DER HAMMEN), Colombia.
3. Proxapertites cursus VAN HOEKEN-KLINKENBER6, Nigeria.
4. Longaoertites vaneendenburgi nov. sp., Venezuela, holotype.
5. Psiladiporites minimus VAN DER HAMMENet WYMSTRA,Trinidad.
6. Sorocea ilicifolia MIQUEL, Recent.
7. Retidiporites rnagdalenensis VAN DER HAMMENet GARCIA, Venezuela.

~J '~i,~iii~i~!~i~i~~ ~ ~~~ ~ ~i~~ ~ ~
the Monoporites annulatus Zone; it is thereafter irregularly present, except in the
youngest zones. In Borneo only known from the Proxapertites opereulatus Zone
and probably disappearing from the stratigraphical record earlier than in the other
two areas.
Taxonomic affinities: In our opinion not derived from Astroca~3,um as
suggested by VAN DER HAMMEN (1956C), but from an extinct group of palms
related to Nypa (ref. MULLER, 1968).
Proxapertites cursus VAN HOEKEN-KLINKENBERG, 1966

(Plate V, 3)
Literature: Proxapertites cursus VAN HOEKEN-KLINKENBERG, 1966, p.43,

pl.Ill, fig.2.
Description: Single grain, radially symmetrical, slightly anisopolar, since the
grains are separated by a continuous equatorial colpus into two, slightly unequal
parts; oblate, in polar view oval; colpus ectexinous, somewhat irregularly bordered
by narrow, indistinct margins of thinning ectexine. Total wall thickness 2-4 /~;
endexine 1 # thick; columellae rather distinct, straight, 1-2 ~ long, round to oval
in cross-section, 1 /~ thick, covered by a l kt thick reticulate tectum; muri flat
to rounded, 1-1 /~ wide, 1 /t high, lumina up to 2 ,u in diameter, variable in
size and shape, generally elongated oval or curved.
Variability: Rather variable in coarseness of wall structure, but attempts to
split the type into smaller units according to the coarseness of the reticulum have
been fruitless.
Comments: As with the preceding species, in disagreement with its author,
the grains are here described as monads. Differs from Proxapertites operculatus
mainly in coarser reticulum with oval muri and distinct columellae.
Distribution: Regularly occurring in the upper part of the Proxapertites
operculatus Zone of all three areas, but base and top are strikingly different. In
the Caribbean area starting at the base of the Ctenolophonidites lisamae Zone and
disappearing from the record in the uppermost part of the Retibrevitricolpites
triangulatus Zone. In Nigeria the first occurrences are in the lowermost part of
this zone and the species is ranging upwards into the uppermost part of the Ver-
rucatosporites usmensis Zone. In Borneo base and top occurrences are not exactly
known yet, but the species is absent in younger zones.
Taxonomic affinities: Clearly related to Proxapertites opereulatus and, there-
fore, affinities with Nypa also likely.
Genus Longapertites VAN HOEKEN-KLINKENBERG, 1964
Longapertites vaneendenburgi nov. sp.

(Plate V, 4; holotype)
298 Rev. Palaeobotan. Palynol., 6 (I 968) [ 89-348

Derivatio nominis: Named in honour of Mr. D. van Eendenburg in recogni-
tion of his contribution to Bornean and Trinidadian palynology.
Holotype: Slide TC-154, well VLE-415, 13998-14037 ft., Venezuela.
Description: Single grains, bilaterally symmetrical, anisopolar, the distal
side arched and provided with a long colpus, the proximal side flat or slightly
convex; in polar and in equatorial view oval. Colpus ectexinous, wide, slightly
intruding, extending over about 2/3 of the greatest circumference of the grain.
Wall thin, < 1 #; columellae only visible in central part near colpus, almost /z
long, ~ thick; tectum evenly and finely perforate; muri ~/z, lumina # in width.
Dimensions: 59-83 ,u.
Variability: Rather uniform in appearance.
Comments: The species differs from Longapertites marginatus VAN HOEKEN-
KLINKENBERG in the absence of a coarser reticulate pattern on the proximal side,
and from Longapertites proxapertioides VAN DER HAMMENet GARCIA in the finely
perforate wall, with lumen size <~ 1 /z.
Distribution: In the Caribbean area restricted to the Foveotrico/pites mar-
garitae and the Retidiporites magdalenensis Zone, but in Nigeria already present
in the Proteacidites dehaani Zone and disappearing in the lower part of the Retidi-
porites magdalenensis Zone.
Taxonomic affinities: The grain is probably derived from a palm species, but
no distinct affinities can be listed. The wall structure of Pritchardia kamapuaana
is rather similar, but the colpus is much shorter here. An arched colpus of the
Longapertites type on the other hand is found in the genus Eugeissona, but here
the wall structure is different.
Class Diporatae |VERSEN et TROELS SMITH, 1950
Genus Psiladiporites VARMA et RAWAT, 1963
Psiladiporites minimus VAN DER HAMMEN et WVMSTRA, 1964

(Plate V, 5)
Literature: Psiladiporites minimus VAN DER HAMMEN et WYMSTRA,1964, p.233,

pl.I, fig.10.
Description: Single grain, radially symmetrical, isopolar, oval in equatorial
and polar outline; diporate, pores penetrating both ectexine and endexine, 3-5 #
wide, simple. Wall -~-/zthick, without distinct layers, psilate or very finely perforate.
Dimensions: 13-21 #.
Variability: Mainly in size and width of pores.
Distribution: In the Caribbean area not found below the Verrutricolporites
rotundiporis Zone with a distinct concentration in the Psiladiporites minimus Zone.

In Nigeria not observed, in Borneo mainly concentrated in the Eehitricolporites
spinosus Zone.
Taxonomic affinities: The genera Artocarpus, Ficus, and Sorocea (Plate V,
6, Moraceae) have very similar pollen and the fossil species is likely to have been
derived from several genera of Moraceae, but there is little likelihood of further
identification, since the differences between the genera in this family are small.
Genus Retidiporites VARMA et RAWAT, 1963
Retidiporites magdalenensis VAN DER HAMMEN et GARCIA, 1966

(Plate V, 7)
Literature: Retidiporites magdalenensis VAN OER HAMMEN et GARCIA, 1966,

p. 109, fig.9.
Description: Single grain, radially symmetrical, isopolar, oval in equatorial
and polar outline; diporate, pores penetrating both ectexine and endexine, circular,
6-13/z wide, simple. Endexine 1- < # thick; columellae -~i /~ long, < ,u
thick; tectum - < # thick, evenly reticulate-perforate, occasionally slightly
finer near pores; lumina < -~-2/t wide, muri - ~-1 ~t wide.
Dimensions." 24-51 /z.
Variability: Larger grains generally show a coarser ornamentation.
Comments." Retidiporites bengalensis described by VARMA and RAWAT
(1963) shows some similarity, but differs in possessing an endexine twice as thick
as the ectexine.
Distribution: Both in the Caribbean area and Nigeria regularly occurring in
the Proxapertites operculatus Zone, but in the former area extending some distance
in the Monoporites annulatus Zone, in the latter disappearing at the top of the
Retidiporites magdalenensis Zone. Unknown from Borneo.
Taxonomic affinities: Obscure, although some superficial resemblance exists
with Banksia collina and Dryandra hmgifolia (Proteaceae).
PLATE VI
1. Multiareolitesformosus (VAN DER HAMMEN)nov, comb., Venezuela, upper focus.

2. Multiareolitesformosus (VAN DER HAMMEN)nov. comb., Venezuela, lower focus.
3. Dianthera americana LINNAEUS, upper focus, Recent.
4. Dianthera americana LINNAEUS, lower focus, Recent.
5. Multimarginites vanderhammeni nov. sp., Venezuela, holotype, upper focus.
6. Multimarginites vanderhammeni nov. sp., Venezuela, holotype, lower focus.
7. Trichantheragigantea HUMBOLDTet BONPLAND,upper focus, Recent.

~!~,~ ~~,: ~ ~ ~ ~' I;i,~t/~ ~ ~ ~s!~~111
Class Dicolporatae new class
Diagnosis: Pollen grains with two ectexinous colpi combined with two
endexinous pores.
Genus Multiareolites nov. gen.
Derivatio nominis: Name derived from circular areas of thickened ectexine.

Diagnosis: Prolate, in polar view oval, dicolporate, colpi ectexinous, meri-
dional, intersubangular, pores endexinous. Ectexine thickened on intercolpate
areas and in rows of circular areas, 2-4/~ in diameter, bordering the colpi, columel-
lae coarser in thickened areas.
Type species." Multiareolites formosus ex VAN DER HAMMEN, 1956b.
Comments: Van der Hammen has described the species Dicolporites Jor-
mosus nov. gen. nov. sp. with a pollen grain derived from a Recent plant as type.
Because this procedure is invalid according to the International Code of Botanical
Nomenclature and because he has not since described a fossil grain of this type,
the opportunity is taken to create a new genus for this pollen type, in order to do
justice to its peculiar morphology. As genotype the fossil representative of Van
der Hammen's original species is chosen and its name adopted. The genus name
Dicolporites has been elevated in taxonomic rank and used to name the class
Dicolporatae.
Multiareolitesformosus (VAN DER HAMMEN, 1956b) nov. comb.

(Plate VI, 1, 2)
Literature: Dicolporites formosus VAN DER HAMMEN, 1956b, p.85, pl.6,

fig.16.
Description: Single grain, radially symmetrical, isopolar, prolate, in polar
view oval with colpi intersubangular. Dicolporate, colpi ectexinous, narrow with
straight borders and pointed ends, pori indistinct, probably endexinous. Endexine
< /~ thick, ectexine thickened on intercolpate areas and in single or double rows
of circular "areoli", 2-4 # in diameter; columellae straight, sometimes slightly
branched, longer and much coarser in areas where endexine is thickened, 2 ,u long
on intercolpate areas and areoli, shorter on poles and along colpi, slightly irregular
in outline and up to 1/~ thick on intercolpate areas, fine and probably isodiametric
on rest of grain. Tectum # thick, smooth or finely verrucate due to protruding
columellae, probably finely perforate.
Variability: Smaller grains show a single row of areoli, while larger ones
generally bear double rows. There is in addition variation in the distribution of
areas with longer and coarser columellae and in the presence or absence of distinct

pores. The material did not permit a further separation and in the counting of
samples all these variations have been grouped under the variable species Mul-
tiareolites formosus.
Distribution: In the Caribbean area and Nigeria restricted to the Echitri-
colporites spinosus Zone. Not observed in Borneo.
Taxonomic affinities: This very characteristic pollen type has only been
observed in the following genera of the Acanthaceae: Adhatoda, Anisotus, Belo-
perone, Dianthera, Jacobinia, Justicia, Kolobochilus, Monechma, and Rungia.
Justicia americana and Dianthera americana (Plate VI, 3, 4) come very close to
the fossil grains, but it is of course not possible to be sure just which genera or
species are represented.
Genus Multimarginatus nov. gen.
Diagnosis: Spherical, di- or tricolporate; colpi long, multimarginate, in

dicolporate grains crossing perpendicular (90 rotated bilaterally symmetrical), in
tricolporate grains colpi always normally orientated; wall thick, tectum perforate or
foveolate-fossulate.
Type species: Multimarginites vanderhammeni nov. sp.
Multimarginites vanderhammeni nov. sp.

(Plate VI, 5, 6; holotype)
Derivatio nominis: Named in honour of Dr. Th. van der Hammen in recog-
nition of his contributions to South American palynology.
Holotype: Slide TC-155, well Hervidero-l, 2770 ft., Venezuela.
Description: Single grain, 90 rotated bilaterally symmetrical, isopolar,
spherical; dicolporate; colpi very long, ectexinous, bordered by 3-4 strips, 4-5/z
wide, perpendicularly orientated, straight borders and pointed or rounded ends;
pori endexinous, transversely elongated, 4/z wide, 13 # long, constricted in the
middle, costate; costae strongly protruding, 1-2/~ thick. Endexine < 1 /z thick;
columellae 1 # long, /z thick, tectum 2/z thick, foveolate, lumina I~-2 # wide,
muri 2-3/z thick.
Distribution: Only known from the Caribbean area where it occurs from
the base of the Retitriletes vanraadshooveni Zone upwards.
Taxonomic affinities: Close resemblance exists with Sanchezia klugii and
Trichanthera gigantea (Plate VI, 7; Plate VII, 1). The latter grain, however, has
a considerably larger size (105 #!). Other multimarginate species of Acanthaceae
available are non-rotated symmetrical (Asteracantha longiJblia, Asystasia
vogeliana, M imulopsis solmsii).

Class Triporatae IVERSENet TROELS SMITH, 1950
Genus Florschuetzia nov. gen.
Derivatio nominis: The genus has been named in memory of the late Prof.
Dr. F. Florschfitz, the founder of palynology in The Netherlands.
Diagnosis: Subprolate, sometimes trilobate and with meridional ridges.
Pores equatorial, circular, distinct. Wall tectate, columellae generally indistinct,
sometimes fused. Tectum continuous, smooth or broken up into separate verrucae
and differentiated in polar and/or meridional direction.
Type species: F/orschuetzia tri/obata nov. sp.
Comments: The genus differs from Verrutrico/porites VAN DER HAMMEN in
the absence of colpi, the indistinctly columellate structure and the absence of a
supratectate verrucate sculpture. The diagnosis has been taken wide enough to
include the fossil sonneratioid pollen types, including the presumed ancestral one.
Some confusion may, however, be possible with a "sporotype" described in 1954
by COOKSON and PIKE and named Santa/umidites, the description of which has
subsequently been emended and accepted as a valid form genus by POTONI~(1960).
In both Cookson and Pike's and Potoni6's diagnosis it is clearly stated that the
maximum thickness of the wall is around the pores, as is the case in Santa/urn
pollen. No confusion with Florschuetzia appears likely, therefore, and the genus
would not have to be mentioned here, but for the fact noted before (MULLER,
1964) that the photomicrographs accompanying Cookson and Pike's paper strongly
suggest that fossil pollen of Sonneratia caseo/aris has been included in S a n t a & m #
dites (see especially their fig. 71, pl. 2, which is strikingly similar to F/orschuetzia
levipoli). This suspicion is strengthened by Cookson and Pike's discussion of the
affinities of Santalumidites cainozoicus in which the authors mention puzzling
differences with recent Santalum pollen and a large degree of variability as well.
If this would indeed prove to be the case, there would have been no choice
left, other than correcting Cookson and Pike's description of the type and accepting
P L A T E VII
1. Trichanthera gigantea HUMBOLDTet BONPLAND,lower focus, Recent.

2. Florschuetzia trilobata nov. gen., nov. sp., Borneo, holotype.
3. Florschuetzia trilobata nov. gen., nov. sp., Borneo.
4. Florschuetzia trilobata nov. gen., nov. sp., Borneo.
5. Lagerstroemia speciosa (LINNAEUS)PERSOON,Recent.
6. Florschuetzia semilobata nov. gen., nov. sp., Borneo. holotype.
7. Florschuetzia semilobata nov. gen., nov, sp., Borneo.
8. Florschuetzia semilobata nov. gen., nov. sp., Borneo.
9. Florschuetzia levipoli nov. gen., nov. sp., Borneo.

ii~i!!i~!Ui~
'i~i~ ~!,i~i! ~i~,!i!~,i',i~,i,~i!~.,'~ ~ ,,,~,,

that Santalumidites had priority above Florschuetzia. However, since no type
specimen has been preserved, nor type locality been designated by Cookson and
Pike, it is impossible to verify this assumption. The photomicrographs alone, on
which no structural details can be observed are considered inadequate also for
selection of a lectogenotype, the more so since some of the diagnostic characters
of the genus are not visible on the pictures.
In view of this it was decided to follow recommendation PB 6 F (p.56) of
the International Code of Botanical Nomenclature, which advises not to apply
names originally attached to inadequately described or figured specimens to well
preserved material. Hence the new genus Florschuetzia is established.
Florschuetzia trilobata nov. sp.

(Plate VII, 2; holotype; Plate VII, 3, 4)
Derivatio nominis: Name derived from the trilobate shape.

Holotype: Slide TC-156, surface sample Dz-4493, northwestern Borneo.
Description: Single grain, radially symmetrical, isopolar; prolate; in polar
view strongly lobate. Triporate, pores circular, 2-4/z in diameter, intersubangular,
ectexinous + endexinous; rarely, pores are combined with ~: colpoid grooves,
which may be flanked by relatively short secondary meridional thickenings. Total
wall thickness 2-3 /z on poles and meridional ridges, 1 /z on porate fields;
endexine 4- /t thick; columellae generally indistinct, -1/~ long; tectum psilate,
not clearly differentiated from underlying columellae, 1-2 ,u thick on poles and
meridional ridges, probably < 1 /z on porate fields.
Variability: In degree of wall thickening on meridional ridges and in the
occasional presence of colpoid grooves with secondary meridional ridges.
Distribution: Only known from Borneo, earliest occurrence in Oligocene
sediments (probably Verrucatosporites usmensis Zone), very abundant in the
Magnastriatites howardi and the Crassoretitriletes vanraadshooveni Zone, dis-
appearing from the record early in the Echitriporites spinosus Zone.
Remarks: The species is distinguished from the psilate variations of Ver-
rutricolporites rotundiporis by its larger size, trilobate condition, thickened merid-
ional ridges and the nearly massive wall in which generally only a faint columellate
layer is visible. The species differs from Florschuetzia semilobata in the psilate
wall and the often pronounced meridional ridges.
Taxonomic affinities: In general shape and wall structure, the species shows
resemblance to pollen of Lagerstroemia flos-regina (Lythraceae) from which it
differs, however, in the absence of distinct colpi. Lagerstroemia indica shows a
comparable general shape and aperture structure which consists of circular pores
without distinct ectexinous colpi, but differs in the verrucate-areolate sculpture
of the tectum. Lagerstroemia speciosa (Plate VII, 5) is also similar in pore and

wall structure and sculpture but is only weakly trilobate and has inconspicuous
meridional ridges, while indistinct colpi are also present.
Since there exists distinct affinity to the recent and fossil sonneratioid types
it is considered more likely that Florschuetzia trilobata represents the extinct
ancestor of the recent genus Sonneratia than that it is derived from a fossil Lager-
stroemia species.
FIorschuetzia semilobata nov. sp.

(Plate VII, 6; holotype; Plate VII, 7, 8)
Derivatio nominis: Name derived from the semilobate shape.

Holotype: Slide TC-157, surface sample Dz-5007, northwestern Borneo.
view lobate. Triporate, pores circular, 1-2 # in diameter, interlobate, endexinous
(in less well preserved grains appearing as endexine + ectexine). Total wall thick-
ness 1-2 # on poles and meridional ridges, probably < 1 # on porate fields;
endexine -< /~ thick, columellae generally indistinct; tectum 1 # thick, areolate-
verrucate, coarser on meridional ridges, finer on porate fields, on poles often
showing a transition to a psilate polar cap.
Dimensions: 23-24/z.
Variability: In size, coarseness and areal differentiation of sculpture. Some
specimens show a rather evenly distributed sculpture while others show a differen-
tiation along the meridional ridges or in poleward direction.
Distribution: Virtually restricted to the uppermost part of the Magnastria-
tires howardi Zone and the Crassoretitriletes vanraadshooveni Zone in Borneo
( Florschuetzia levipoli Zone).
Taxonomic affinities: The apertures and sculptural type clearly point to
affinity with the pollen of Sonneratiaceae, but matching with a Recent species
has not been possible.
FIorschuetzia levipoli nov. sp.

(Plate VII, 9; Plate VIII, 1; Plate VIII, 2; holotype)
Literature: ?Santalumidites cainozoicus COOKSONet PIKE, 1954, p.209, pl.2,

fig.71.
Derivatio nominis: Name derived from the smooth polar caps.
Holotype: Slide TC-158, surface sample Mq-ll91, northwestern Borneo.
view circular to rounded triangular, sometimes bulging at equator. Triporate, pores
circular, endexinous, in well preserved grains covered by thin ectexinous membrane,
2-8/z in diameter (depending on size grain). Total wall thickness ~z 2 #, generally
thicker on poles than on equatorial area; endexine ~ 1/z thick; columellae mostly

indistinct, slightly enlarged (intra-areolate) on polar caps: tectum < 1 /~ thick,
psilate on poles, broken up into a verrucate-areolate pattern on equatorial belt.
Verrucae rounded-polygonal, of varying sizes, uniform height, regularly distrib-
uted.
Dimensions: 30-50 /~ (averages) showing a distinct increase in size from
older to younger strata.
Variability: In thickness of tectum on poles and coarseness of verrucation
on equatorial belt. The visibility of the columellate layer is mainly determined by
the state of preservation. Occasionally the presence of a meridional ridge is sug-
gested by folding of the grain but this feature is not structurally different from the
remainder of the equatorial belt.
Distribution: Regularly present in Borneo from the higher part of the Magna-
striatites howardi Zone upwards (FIorschuetzia levipoli and F. meridionalis Zones).
Remarks: The species is distinguished from Florschuetzia meridionalis by the
absence of a structurally differentiated meridional ridge, the indistinct or even
absent intra-areolate pattern on the polar caps and the generally smaller size.
Compared to Florschuetzia semilobata, the psilate polar caps and circular or only
weakly triangular shape in polar view are diagnostic. The relation between size
and age will be discussed separately.
Taxonomic affinities." Apart from the generally smaller mean size the species
agrees in all essential characteristics with the pollen of Sonneratia caseolaris (Plate
VIII, 3), a mangrove species at present growing in estuaries along the Indo-
Malesian coasts,
Florsehuetzia meridionalis nov. sp.

(Plate VIII, 4; holotype; Plate VIII, 5)
Derivatio nominis: Name derived from the meridional orientation of the

sculptural pattern.
Holotype: Slide TC-159, surface sample Mq-1191, northwestern Borneo.
view circular to rounded triangular. Triporate, pores circular, endexinous, in well
PLATE VIII
1. Florschuetzia levipoli nov. gen., nov. sp., Borneo.

2. Florsehuetzia levipoli nov. sen., nov. sp., Borneo, holotype.
3. Sonneratia caseolaris (LINNAEUS) ENGLER, Recent.
4. Florsehuetzia meridionalis nov. gen., nov. sp., Borneo, holotype.
5. Florsehuetzia meridionalis nov. gen., nov. sp., Borneo.
6. Sonneratia alba J. E. SMITH,Recent.
7. Sonneratia alba J. E. SMITH, Recent.

. ~ =iii=!=!=!!(iii=i
==
preserved grains covered by thin ectexinous membrane, 5-12/, in diameter. Total
wall thickness ~: 3/~, generally thinner on porate fields; endexine < l #; columel-
lae distinct on meridional ridge where thin, uniform and rather dense, on polar
caps probably fused to form a typical intra-areolate pattern, not visible on porate
fields. Tectum 1-1 /, thick, psilate on polar caps, broken up into an areolate-
fossulate pattern on meridional ridges and into a verrucate pattern on porate
fields. The meridional ridges are fairly sharply outlined against the porate fields,
but merge gradually with the tectate-psilate polar caps.
Dimensions: 35-60 # (averages) showing a distinct increase in size from older
to younger strata.
Variability: In thickness of tectum on poles and in coarseness of sculpture
and structure.
Distribution: Occurring in Borneo only, from the middle part of the Cras-
soretitriletes vanraadshooveni Zone upwards (FIorschuetzia meridionalis Zone).
Remarks: The species is distinguished from Florschuetzia levipoli by the
presence of a structurally differentiated meridional ridge and the distinct intra-
areolate structure of the polar caps and generally larger size.
Taxonomic affinities: Apart from the generally smaller size of the Fossil
assemblages, the species agrees in all essentials with the pollen ofSonneratia alba
(Plate VIII, 6, 7), a mangrove species at present growing in estuaries along the
Indo-Malesian, Indian and eastern African coastline.
Genus Echitriporites VAN HOEKEN-KLINKENBERG, 1964
Echitriporites trianguliformis VAN HOZKEN-KLINKENBERG,1964

(Plate IX, 1, 2)
PLATE IX
1. Echitriporites trianguliformis VAN HOEKEN-KL1NKENBERG,Venezuela.

2. Echitriporites trianguliformis VAN HOEKEN-KLINKENBERG,Colombia.
3. Proteacidites dehaani nov. sp., Venezuela.
4. Proteacidites dehaani nov. sp., Colombia, holotype.
5. Guevina avellana MOLINA, upper focus, Recent.
6. Guevina avellana MOLINA, lower focus, Recent.
7, Lomatia ilicifolia R. BROWN, upper focus, Recent.
8. Lomatia ilicifolia R. BROWN, lower focus, Recent.
9. Anacolosidites cf. luteoides COOKSONet PIKE, New Guinea.
10. Anacolosidites cf. luteoides COOKSONet PIKE, Borneo.
11. Anacolosa frutescens BLUME, upper focus, Recent.
12. Anacolosa frutescens BLUME, lower focus, Recent.
13. Alnipollenites verus POTON1~, Borneo.
14. Alnus glutinosa (LINNAEUS)GAERTNER, upper focus, Recent.
15. Alnus glutinosa (LINNAEUS)GAERTNER, lower focus, Recent.

>
~:IIm!i~ ~ ~ i~ ~ ~:::~mm
rll~ 11mmmm
i~~ ~i~i~i~~ ~ i~~ ~i!i~i!i!ii!~
~::~i~!i,,,,~,~,~i~i~!~,~:~
~ , i~ ~,~,~,ii~i~,~ii~,
~?~ii~i~,~,j~i~iiii~ii:~:,~S~i~!~!~J~
~,~: ~ ~
Literature: Echitriporites trianguliformis VAN HO~KFN-KLINKEN~ERG,1964,
p.218, pl.47, fig.7.
Echitriporites trianguliformis BELSK, BOLTENHAGENet POTON~k,
1965, p.77, pl.13, fig. 22, 23.
Description: Single grain, radially symmetrical, isopolar, oblate; outline in
polar view semi-angular. Triporate, pores ectexinous and endexinous, circular,
2 # in diameter, costate, often slightly protruding. Endexine 4 1 /z , intectate-
echinate, echinae 1-2 ,u long, ~: 1 /z wide at base, conical or with rounded tops,
2/z apart.
Variability: Considerable variation exists in size and shape of the grain and
in size and density of the spines. The more triangular grains with fewer and smaller
spines are more common in the Upper Cretaceous of northern South America,
whereas the more rounded grains with slightly more and larger spines are more
common in the Eocene. However, since many transitional specimens occur, no
specific distinction was possible.
Distribution: Encountered in the Proxapertites operculatus Zone in all three
areas, but rare in Borneo and soon disappearing. In the Caribbean area and
Nigeria the species remains an important component of the Eocene microfloras
which has great practical value for local correlations. In both areas it disappears
from the record around the top of the Verrucatosporites usmensis Zone.
Taxonomic affinities: Despite intensive search, no positive identification has
been possible yet. This, and its gradual disappearance at the top of the Eocene
suggest that the taxon producing the pollen grains is probably extinct. A superficial
resemblance is shown by some Proteaceae (Embothrium, Garnieria, Persoonia,
Telopea). Only Telopea oreades and Embothrium mucronatum were available for
comparison but these species showed finely columellate-tectate pollen grains.
Genus Proteacidites (COOKSON, 1950) ex COUPER, 1953
Proteacidites dehaani nov. sp.

(Plate IX, 3; Plate IX, 4; holotype)
Derivatio nominis: Named in honour of Mr. R. de Haan in recognition of

his contribution to Venezuelan palynology.
Holotype: Slide TC-160, surface sample E 1362, Colombia.
polar view triangular. Triporate, pores ectexinous and endexinous, circular, 2~,/~
in diameter, costate. Total wall thickness ~ 1 ~t, endexine /z thick, columellae
< # thick and long; duplicolumellate, reticulate-foveolate tectum; muri -:~ /z
high, /z wide, fiat topped; lumina 1-1/z wide, of variable shape, coarser on
interporate fields, finer on poles and around pori.

Variability: Ornamentation varying from reticulate-foveolate to perforate.
Constant are size and shape, pore structure and the coarser sculpture on the
interporate areas and the species is well characterized by these features. A com-
parable but not identical type is known from the post-Eocene of western Venezuela.
Comments: The species appears different from reticulate members of the
genus described by COOKSON (1950), COUPER (1953) and HARRIS (1965) mainly
because of size, straight sides and duplibaculate muri. It is rather similar to Pro-
teacidites thalmanni ANDERSON, 1960, from which it differs, however, in its larger
size, absence of "notched" pores and presence of duplicolumellate muri.
Distribution: In the Caribbean area and in Nigeria restricted to the lower
part of the Proxapertites operculatus Zone and principal marker for the Prote-
acidites dehaani Zone.
Taxonomic affinities: Close resemblance exists with pollen of Guevina avellana
(Proteaceae, Chili, Plate IX, 5, 6). Other Proteaceae are more or less different:
Lomatia ilicifolia (Chili, Australia, Plate IX, 7, 8) has smaller lumina, Cardwellia
sublimis (Queensland) differs in its more rounded, non-costate pores. Stenocarpus
sinuatus (Australia) is less distinctly multicolumellate and has a much thicker wall,
Leucospermum hypophyllum (southern Africa) has smaller lumina and Symphionema
montanum (Australia) has a much smaller size.
Class Stephanoporatae IVERSENet TROELSSMITH, 1950
Genus Anacolosidites (CooKSON et PIKE, 1954) ex R. POTONI~, 1960
Anacolosidites cf. luteoides COOKSON et P1KE, 1954

(Plate IX, 9, 10)
Literature: cf. Anacolosidites luteoides COOKSON et P~KE, 1954, p.207, pl.1,

fig.47-50.
Description: Single grain, radially symmetrical, isopolar, oblate; subtrian-
gular outline in polar view; pores arranged in two groups of three, opposite each
other, each at a distance of :L 3 # from the equator, probably ectexinous and
endexinous, 4 /~ in diameter, equatorially elongated with slightly thickened
borders; endexine # thick at angles, up to 2/t thick at straight sides of equator;
columellae simple, < 1 /~ thick at angles, 1 ,u thick at straight sides of equator;
tectum uniformly < # thick, psilate.
Variability: In size and wall thickness.
Comments: It would seem that Cookson and Pike's species A. luteoides comes
closest to the specimens described here and this name is adopted provisionally,
pending further study.

Distribution: In Borneo and Nigeria already recorded from the Paleocene
part of the Proxapertites operculatus Zone, but in the Caribbean area appearing
later at the base of the Retibrevitricolpites triangulatus Zone (base Eocene). In
the Caribbean area the species disappears from the record already at the top of the
Retitricolporites guianensis Zone, in Nigeria it is found higher, up to the youngest
part of the Magnastriatites howardi Zone, while in Borneo it remains common
throughout the Tertiary.
Taxonomic affinities: Close correspondence exists with pollen grains of
Anacolosa (Plate IX, 11, 12), Cathedra and Ptychopetalum (Olacaceae).
Genus Alnipollenites R. POTONI~, 1931
A&ipollenites verus (R. POTONI~, 1931) ex R. POTONI~, 1934

(Plate IX, 13)
Literature." AInipollenites verus R. POTONII~, 1934, p.58, pl.2, fig. 17.

Stephanoporitesfornicatus VAN DER HAMMEN, 1956b, p.94, pl.10,
fig.3Y
Description: Single grain, radially symmetrical, isopolar, oblate; outline
in polar view subangular. 3-6 ectexinous and endexinous pores, 3-4 # wide,
slightly vestibulate; pores connected by arci; endexine <:~ 1 # thick, ectexine 1 /z
thick, columellae indistinct, tectum smooth or finely granulate.
Dimensions: 28-31 /~.
Variability: The pore number is the main variable, five being the average.
Distribution." In the Caribbean area occurring from the base of the Alnipol-
lenites verus Zone upwards. In Borneo present from the upper part of the Ver-
rucatosporites usmensis Zone to the top of the Crassoretitriletes vanraadshooveni
Zone (MULLER, 1966).
Taxonomic affinities: Close correspondence exists with Alnus pollen (Plate
IX, 14, 15).
Genus Pachydermites nov. gen.
Derivatio nominis: Name derived from the thick exine.

Diagnosis: Stephanoporate grains with 4-6 rather irregularly shaped aper-
PLATE X
1. Echiperiporites estelae nov. sp., Borneo, holotype.

2. Pachydermites diederixi nov. gen., nov. sp., Nigeria, holotype.
3. Symphonia globtdifera LINNAEUS (ill.), Recent.
314 Re v. Palaeobotan. Palynol., 6 (1968) 189-348

PLATE X

tures, which are larger in the endexine than in the ectexine, a very thick endexine,
which is finely perforated around apertures and a thin tectate-psilate ectexine.
Type species: Pachydermites diederixi nov. sp.
Pachydermites diederixi nov. sp.

(Plate X, 2; holotype)
Derivatio nominis: Named in honour of Mr. D. O. J. Diederix in recognition

of his contributions to Colombian and Nigerian palynology.
Holotype: Slide TC-161, well Ughelli-3, 5200 ft., Nigeria.
Description: Single grain, radially symmetrical, isopolar, oblate-suboblate.
4-6 porate, ectexinous apertures 6-10 /~ long, 4-9 /~ wide, almost circular and
irregular in outline, endexinous apertures of identical shape, but slightly larger
and somewhat vestibulate. Endexine 3-6 # thick, irregularly perforated around
apertures. Interior surface irregularly verrucate; columellae /~ thick and long;
tectum # thick, psilate.
Variability: In size and coarseness of wall.
Distribution: In Nigeria known from the base of the Monoporites annulatus
Zone upwards, but in Venezuela occurring only from the base of the Grimsdalea
magnaclavata Zone up wards.
Taxonomic affinities." Identical with the pollen of Symphonia globulifera
(Guttiferae, Plate X, 3). As far as known this pollen type does not occur in other
genera and the identification has a very high degree of probability. Unfortunately
the pollen of the Madagascar species of this genus have not yet been described.
Class Periporatae IVERSENet TROELSSMITH, 1950
Genus Buttinia BOLTENHAGEN, 1967
Buttinia andreevi BOLTENHAGEN, 1967

(Plate XI, 2, 3)
Literature: "type S 231", JARDINI~and MAGLOIRE, 1965, p.208, pl.V, fig. 8.
Buttinia andreevi BOLTENHAGEN, 1967, p.342, pl.lI, fig.5-7.
PLATE XI
1. The~pesiapopulnea COgREA,Recent.
2. Buttinia andreevi BOLTENHAGEN,Colombia, upper focus.
3. Buttiniaandreevi BOLTENHAGEN,Colombia, lower focus.
316 Rev. Palaeobotan.PalynoL, 6 (1968) 189-348

P L A T E XI

Description: Single grain, centro-symmetrical, isopolar, spherical. Apertures
20 or more, ectexinous and endexinous, rounded to angular, 7-10 # wide, 3-5/~
apart. Endexine 2 / , thick; columellae <-. I # long, < # thick, only visible in
well preserved grains; tectum reduced to rounded ridges, /~ thick, 1-2 # wide,
between the pores, psilate.
Dimensions." 40-55/z.
Variability: There is only slight variability in size and wall thickness.
Comments: This species has already been described and figured without
naming from the Senonian of western Africa by JARDIN~ and MAGLOIRE(1965)
and was recently named by BOLTENHAGEN(1967). None of these authors has
observed the presence of columellae and tectum, which in our opinion proves
angiospermous affinities and permits the assignment to the class Periporatae.
Distribution." Both in Nigeria and the Caribbean area restricted to the
Proteacidites dehaani Zone. In Senegal and Gabon also restricted to the Senonian.
Not recorded from Borneo.
Taxonomic affinities: Unknown.
Genus Echiperiporites VAN DER HAMMENet WYMSTRA,1964
Echiperiporites estelae nov. sp.

(Plate X, I; holotype)
Derivatio nominis: Named in honour of Mrs. Estela Bradley de Giacomo in

recognition of her contribution to Venezuelan palynology.
Holotype: Slide TC-162, surface sample MQ-1189, northwestern Borneo.
Description: Single grain, centro-symmetrical, isopolar, spherical; periporate,
pores 20-24, ectexinous and endexinous, slightly annulate, annuli 1-2 # wide;
pores 4-6 # wide, 10-12/~ apart. Endexine 1 ~ thick; columellae ~z /~ thick,
~ 1 /~ long; tectum smooth, -:< 1/~ thick, thickened at roots of spines; spines 6-9 #
long, 2-4/~ thick at base, conical, blunt, 6-10/~ apart.
Dimensions." 55-87/~, excluding spines.
Variability: The species as defined here shows considerable variation in
size, coarseness of wall structure and spines, density of spines and in number of
pores.
Distribution: The oldest occurrences are known from the Caribbean area
where the species starts in the middle of the Eocene (Verrucatosporites usmensis
Zone). In Nigeria the species is very rare and known virtually only from the
Magnastriatites howardi Zone upwards. In Borneo it is occurring regularly in the
Echitricolporites spinosus and Crassoretitriletes vanraadshooveni Zones and is
probably absent in earlier periods.
Taxonomic affinities: Due to the rather large variability of this species it is
difficult to indicate the taxonomic affinity with certainty. The majority of the

specimens agree well with pollen of Thespesia populnea (Malvaceae, Plate XI, 1),
but larger grains could be more related to Hibiscus tiliaceus (Malvaceae). Ipomoea
pollen grains have generally more pores and are more densely spinose, but other-
wise belong to this group of species.
Class Fenestratae IVERSEN et TROELS SMITH, 1950
Genus Fenestrites (VAN DER HAMMEN, 1956b)
Lectogenotype: Fenestrites spinosus VAN DER HAMMEN, 1956b.
Remarks: Since Van der Hammen has invalidly indicated a recent pollen
grain as type specimen, his genus is hereby legalized by selecting as lectogenotype
the fossil species as described below.
Fenestrites spinosus ex VAN DER HAMMEN, 1956b

(Plate XII, 1)
Literature: Fenestrites spinosus VAN DER HAMMEN, 1956b, p.97, pl.12, fig.38.
Description: Single grain, radially symmetrical, isopolar, spherical; outline
In polar view almost hexangular. Colpi and pori indistinct, probably tricolporate.
Exine differentiated into a pattern of intectate lacunae (fenestrae), 8-11 /~ wide
and tectate-columellate cristae. Columellae 1 # long, /z thick; cristae 1-1 #
wide, 1/t high, bearing single rows of spines, 2 # long, 1 # wide at base.
Dimensions: 30-38 # (including cristae, excluding spines).
Variability: In size and coarseness of ornamentation. Sometimes undulating
cristae have been observed.
Distribution: Restricted to the middle and upper part of the Echitricolporites
spinosus Zone. More frequently observed in the Caribbean area than elsewhere.
Taxonomic affinities: This species is typical for the liguliflorae pollen type
of the Asteraceae (Compositae), which is produced by a large number of genera.
Of the many species which produce this pollen type, the following only will be men-
tioned as coming close to the fossil grains: Elephantopus angustifolia, Rolandia
fruticosa (Plate XII, 3), Vernonia canescens, Vernonia remotiflora (Plate XII, 2).
Class Tricolpatae IVERSEN et TROELS SMITH, 1950
Genus Striatricolpites (VAN DER HAMMEN, 1956b) ex GONZALEZ, 1967
Striatricolpites catatumbus GONZALEZ, 1967

(Plate XII, 4)
Rev. Palaeobotan.Palynol., 6 0968) 189-348 319

Literature: Striatricolpites catatumbus GONZALEZ, 1967, p.30, pl. VIII, fig.7.
Description: Single grain, radially symmetrical, isopolar, prolate; outline
in polar view trilobate to spherical. Tricolpate, colpi ectexinous, long, intruding
with straight simple borders and pointed ends. Pores absent or indistinct. Endexine
1 # thick, columellae simple, # long and /~ thick, slightly longer on poles,
linearly arranged underneath striae of tectum; tectum striate; striae 1 # thick,
1-1/~ wide, 1 # apart, subparallel or slightly anastomosing; between striae tectum
very thin, sometimes finely perforate.
Variability: Some variability exists in size and in coarseness of wall structure
and presence or absence of indistinct pores. The specimens described here agree
in all respects with Gonzalez' type material. Tricolpate striate types described
by MULLER (1968) from the uppermost Cretaceous of Borneo are considered to
be specifically distinct.
Distribution: Both in the Caribbean area and in Nigeria occurring from the
base of the Retibrevitricolpites triangulatus Zone upwards. In Borneo the lower
limit is not exactly known, but a related species occurs already in the Paleocene.
Taxonomic affinities: Closest resemblance is with pollen of the genus Crudia
(Fabaceae, Plate XII, 5, 6), but Anthonotha and Isoberlinia (Fabaceae) have
similar pollen. Macrolobium bifilium (Fabaceae) has a much coarser striate pattern
but is otherwise similar. It is not yet possible to be more precise about the affinities
of this rather variable pollen species.
Genus Perfotricolpites GONZALEZ, 1967
Perfotricolpites digitatus GONZALEZ, 1967

(Plate XII, 10)
Literature: Perfotricopites digitatus GONZALEZ, 1967, p.34, pl.VI, fig. 1.

Description: Single grain, radially symmetrical, isopolar, prolate. Tricolpate:
PLATE XII
1. Fenestrites spinosus VAN DER HAMMEN,Venezuela.
2. Vernonia remotiflora L. C. RICHARD,Recent.
3. Rolandia fruticosa (LINNAEUS)KUNTZE,Recent.
4. Striatricolpites catatumbus GONZALEZ,Colombia.
5. Crudia amazonica SPRUCE, upper focus, Recent.
6. Crudia amazonica SPRUCE, lower focus, Recent.
7. Foveotricolpites perforatus VAN DERHAMMENet GARCIA,Venezuela, upper focus.
8. Foveotricolpites perforatus VANDERHAMMENet GARCIA,Venezuela, lower focus.
9. Foveotricolpites perforatus VANDERHAMMENet GARCIA,Venezuela.
10. Perfotricolpites digitatus GONZALEZ,Venezuela.

PLATE XII
r I " ' ~ 2 ' 3
4 !~-T~~i~ 5 ii ~' 6
i ~i~,iiiii!ii~i)!!~!!~i~i~~i~~i~ ~!!ii!iilJii!iif
~ l O

colpi ectexinous, long, strongly intruding, with straight borders and pointed ends.
Pores very indistinct, if present endexinous. Endexine /~ thick; columellae
digitate, 2-2/~ long, 1 p thick at base, I~ thick at top of branches; tectum thin,
finely reticulate-perforate, lumina ~ ~ in width.
Variability: Occasionally in Venezuela pericolpate specimens are found,
which because of the very typical wall structure should be included in the species.
Whether some specimens observed in Nigeria with large round endexinous pores
should be included is more doubtful.
Distribution: Both in the Caribbean area and Nigeria occurring for the
first time approximately at the base of the Verrucatosporites usmensis Zone, but,
while continuously present in the former area up to Recent, in Nigeria disappearing
from the record in the upper part of the Magnastriatites howardi Zone. In Borneo
present at least from the base of the Echitricolporites spinosus Zone upwards.
Taxonomic affinities: The pollen grains of Merremia glabra (Plate XIII, 2)
and M. umbellata (Convolvulaceae) are rather similar. Identification of the fossil
dispersed pollen with Merremia appears fairly reliable. Branched columellae are
also found in Scaevola (Goodeniaceae, Plate XIII, 1) the pollen of which differs
in the presence of costate endexinous pores and much thicker wall on poles.
Valerianella stenocarpa (Valerianaceae) has also slightly similar pollen.
Genus Foveotricolpites VAN DER HAMMEN et QARCIA, 1966
Foveotricolpites perforatus VAN DER HAMMEN et GARCIA, 1966

(Plate XII, 7, 8, 9)
Literature: Foveotricolpites perforatus VAN DER HAMMEN et GARCIA, 1966,

p.l12, fig.18.
Description: Single grain, radially symmetrical, isopolar, subprolate. Tricol-
pate, colpi ectexinous, medium long, with straight borders and pointed ends.
Endexine < 1 # thick; columellae -1 # thick and long, tightly packed in an
irregular pattern; tectum 1 # thick, foveolate; lumina rounded oval, 1-2/~ wide,
coarser, up to 3# wide on poles, approx. 3/~ apart.
PLATE XIII
1. Scaevola plumieri VAI4L,Recent.
2. Merremia glabra HALUER(ill.), Recent.
3. Stephanoeolpites eostatus VAN DER HAMMEN, Venezuela.
4. Tabernaemontana attenuata URBAN,upper focus, Recent.
5. Tabernaemontana attenuata URBAN,lower focus, Recent.
6. Gemmastephanocolpites gemmatus VAN DER HAMMENet GARCIA,polar view, Venezuela.
7. Gemmastephanocolpites gemmatus VAN DER HAMMENet GARCIA,equatorial view, Venezuela.

......... ~'~iiiiiii~!~T~i~~ ~ ,~i~i!~i~i~ ~,~
,~,~iii,~ ~
x
Dimensions: 44--64/z.
Variability: A well characterized species with some variability in size and
coarseness of wall structure.
Distribution: Restricted to the Foveotricolpites perforatus Zone of the Carib-
bean area.
Taxonomic affinities: Unknown.
Class Stephanocolpatae IVERSENet TROELS SMITH, 1950
Genus Stephanocolpites (VAN DER HAMMEN, 1954) ex R. POTONII~, 1960
Stephanocolpites costatus VAN DER HAMMEN, 1954

(Plate XIII, 3)
Literature: Stephanocolpites costatus VAN DER HAMMEN, 1954, p.92, pl.7,

fig.9.
Remarks: The original short diagnosis is here enlarged. The identity of the
specimens described by us has been confrmed by Van der Hammen.
in polar view circular. Apertures 5-6; ectexinous, 10-12 # long colpi and one fused
circular endexinous aperture which is distinctly costate. Endexine < p thick,
columella < # long, < /~ thick; tectate-foveolate, tectum < /t thick, lumina
oval to circular up to 1 # in diameter, coarser on poles, muri rounded, 1-1/~ wide.
Variability: Only slightly variable in size and sculpture and distinctness
of colpi.
Distribution: Restricted to the lower part of the Proxapertites operculatus
Zone in the Caribbean area (top in Ctenolophonidites lisamae Zone).
Taxonomic affinities: Unknown. A similar aperture combination has, how-
ever, been found in Tabernaemontana attenuata (Apocynaceae, Plate XIII, 4, 5),
the wall structure of which is much finer reticulate, while the equatorial endexinous
aperture is much wider and less costate.
Genus Gemmastephanocolpites VAN DER HAMMENet GARCIA, 1966
Gemmastephanocolpites gemmatus VAN DER HAMMENet GARCIA, 1966

(Plate XIII, 6, 7)
Literature: Gemmastephanocolpites gemmatus VAN DER HAMMENet GARCIA,

1966, p.110, fig.12.

in polar view circular. Apertures 5-6; ectexinous, 10-12 /t long colpi and one
fused circular endexinous aperture which is distinctly costate. Endexine < 1 /~
thick, columellae < 1 /~ long, /~ thick; tectum smooth, covered with verrucae or
gemmae 1-2/~ in diameter, 1 /~ thick, generally situated on top of columellae, but
rather irregularly distributed.
Distribution: Restricted to the Ctenolophonidites lisamae and the Foveo-
tricolpites perforatus Zones in the Caribbean area.
Taxonomic affinities: No comparable Recent plant species is known, but
the distinct resemblance with Stephanocolpites costatus and its occurrence at a
stratigraphically higher level suggest a phylogenetic relationship between the two.
Genus Retistephanocolpites LEIDELMEYER,1966
Retistephanocolpites williamsi nov. sp.

(Plate XIV, 1, 2; holotype)
Derivatio nominis: Named in honour of Mr. R. W. Williams in recognition

of his contribution to Nigerian palynology.
Holotype: Slide TC-163, well Ubulu-1, 3000 ft., Nigeria.
Description: Single grain, radially symmetrical, isopolar, oblate. 6-7 colpate,
colpi ectexinous, 14 # long. Wall 221--4/z thick; endexine 1-2/z thick, columellae
1 # long and thick, irregularly arranged and forming a spongy structure; tectum
< 1 /z thick, reticulate-foveolate; lumina oval to circular 1-2 # in diameter,
muri 1 # wide.
Dimensions: 4 1 4 7 #.
Variability: There exists considerable variation in the coarseness of the
sculpture and some in the thickness of the endexine.
D&tribution: Absent in the Caribbean area, in Nigeria occurring from the
base of the Proxapertites operculatus Zone upwards to the middle of the Ver-
rucatosporites usmens& Zone. In Borneo known from the Neogene, base not
exactly known.
Taxonomic affinities: The combination of 5-6 short apertures, spongy
columellate structure and reticulate-foveolate tectum is so far only known from
Ctenolophon parvifolius (Ctenolophonaceae, Plate XIV, 3, 4) which differs only in
the 2-3/z wide muri. Identification with this Indo-Malesian species appears highly
probable.
Genus Cteno/ophonidites VAN HOEKEN-KLINKENBERG,1966

Ctenolophonidites costatus (VAN HOEKEN-KLINKENBERG, 1964) ex VAN HOEKEN-
KLINKENBERG, 1966
(Plate XIV, 5, 6)
Literature: KUYL et al. (1955, p.2, fig.10, 12, 13).

Stephanocolpites costatus VAN HOEKEN-KLINKENBERG, 1964,
p.221, pl.4, fig.10.
Ctenolophonidites costatus VAN HOEKEN-KLINKENBERG, 1966,
p.42.
Description: Single grain, radially symmetrical, isopolar, spherical-oblate;
6-colpate; colpi ectexinous, 24 # long, slightly costate equatorially. Ectexine locally
thickened into a pattern of radial costae, meeting near the poles and forming
ring-like ridges, occasionally with extra ridges inside these rings; ridges 1-2 #
high, 5 p wide. Columellae and tectum not differentiated, wall psilate, 3-4 p thick
at equator and apparently densely and finely perforated in all directions, creating
a spongy effect which diffuses the light considerably.
Variability: In size and coarseness of wall structure.
Distribution: Only known from Nigeria where the species already occurs in
the Proteacidites dehaani Zone, is temporarily absent in the higher part of the
Proxapertites operculatus Zone, to be present again from the base of the
Monoporites annulatus Zone up to Recent.
Taxonomic affinities: Agrees in all essential details with the pollen of Cteno-
lophon engleri (Ctenolophonaceae, Plate XV, 1,2) except its smaller size. This pollen
type is not known from any other plant as yet (cf. also SAAD,1962).
Ctenolophonidites lisamae (VAN DER HAMMEN et GARCIA, 1966) nov. comb.

(Plate XIV, 7, 8)
Literature: Scabrastephanocolpites lisamae VAN DER HAMMEN et GARCIA,

1966, p.ll0, fig.14.
PLATE XIV
1. Retistephanocolpites williamsi nov. sp., Nigeria, holotype, upper focus.
2. Retistephanocolpites williamsi nov. sp., Nigeria, holotype, lower focus.
3. Ctenolophon parvifolius OLIVER,upper focus, Recent.
4. Ctenolophon parvifolius OUVER,lower focus, Recent.
5. Ctenolophonidites costatus VAN HOEKEN-KLINKENBER~,Nigeria, upper focus.
6. Ctenolophonidites costatus VAN HOEKEN-KUNKENaERG,Nigeria, lower focus.
7. Ctenolophonidites lisamae VAN DER HAMMENet GARC~A,Venezuela, polar view.
8. Ctenolophonidites lisamae VAN DER HAMMENet GARCIA,Venezuela, equatorial view.

X
~ i ~i~!ii~
~ _.
~ ~:~i~ ~~:i~~!~
.... ~ ,i, ~i~

~ ~ i ij Ji' ~~ ~:~i ~,~,:~:~j~:, ~i~,~:~,:~:~ ~,~ii~i~ii~iii~i~ii~i,~:~,~i~ ,~:~,~i~,. . . . . . . . . . . . . . ~ ~ ~ ~ ,_~,~:~ ~J,,~ ~i:~'ii,~!:!i:ii:i~,i~ii:,ii!j:,~!~
~
1!
Description: Single grain, radially symmetrical, isopolar, spherical-sub-
oblate; outline in polar view stellate-circular. 5-colpate, colpi 14 .u long, ectexinous,
slightly costate equatorially. Ectexine locally thickened into a pattern of radial
costae, joining in pairs around the terminations of the colpi, not touching near
poles; ridges 12~--2bt high, 2-3 ,u wide. Columellae and tectum not differentiated;
wall finely scabrate, 3-4 # thick at equator, apparently densely perforated in all
directions, creating a spongy effect which diffuses the light considerably.
Variability: Some variability is present in the way in which the ridges unite
with each other.
Comments: In view of the close relationship which exists between this species
and Ctenolophonidites costatus, we have transferred it to the latter genus. The
identity of our specimens has been confirmed by Van der Hammen.
Distribution: Virtually restricted to the Ctenolophonidites lisamae and
Foveotricolpites perforatus Zones in the Caribbean area.
Taxonomic affinities: Has in all probability been derived from an extinct
species of Ctenolophon (Ctenolophonaceae).
Class Tricolporatae |VERSEN et TROELS SMITH, 1950
Genus Psilatricolporites (VAN DER HAMMEN, 1956) ex VAN DER HAMMENet WYM-
STRA, 1964
Psilatrieolporites operculatus VAN DER HAMMENet WYMSTRA, 1964

(Plate XV, 3)
Literature: Psilatricolporites operculatus VAN DER HAMMEN et WYMSTRA,

1964, p.236, pl.1, fig. 13.
polar view circular to slightly semi-angular. Tricolporate; colpi long, ectexinous,
PLATE XV
1. Ctenolophon engleri MILDBREAD, upper focus, Recent.
2. Ctenolophon engleri MILDBREAD, lower focus, Recent.
3. Psilatricolporites operculatus VAN DER HAMMENet WYMSTRA, Surinam.
4. Alchornea cordifolia MUELLER-ARG., Recent.
5. Alckornea obovata PAX et HOFFMANN, Recent.
6. Zonocostites ramonae nov. gen., nov. sp., Borneo, holotype.
7. Zonocostites ramonae nov. gen., nov. sp,, Borneo, holotype.
8. Rhizopkora rnucronata LAMARCK,upper focus, Recent.
9. Rhizophora mucronata LAMARCK,lower focus, Recent.
328 Rev. Palaeobotan.Palynol., 6 0968) 189-348

,~ ~,~..... ~i~'~i~iiiiii!i~ ', ~,
~ ~ii!iiiiil !i
i~i~i~i i~,~ii~i~!i~i~ii~,~ ~......... ~ ~'~!ii~'i~i!ill

~''!I'~'~ ~' ~!~ ~,~ ~,~ ' '~'~''
. . . . ~ ~i~iii~ii~,~'~' i~,~!,~i~!~ii~!!~ii,'
d d m L ~ ,, ~. ~., ~ . ~. ~ . .I F . . . ~
~ i ~ ~~i~i~ii~!~i!~!i~,~,,,,,,,,,,i,,i, ,~ !~iii~i~iii~iii~,'~iii,~'i~i~ili~,i'~'~!i~,,i~
! ~~ CO ,, ~i~!"~'~! q~l (~ ~,~......

, ....,~,i ~ "~'~i'~!! ' u, %'. . . . ~iiiiiiiii
. 4~ ~'~~'~i,~i~ ~ !,~!!!~~,~,~,~
ii~ii~,~i,~!~ii
...........i' ~'
marginate and provided with distinct opercula of identical structure as found in
margins; opercula 1 # wide at equator; pores distinct only in equatorial view,
endexinous, equatorially elongated with indistinct borders. Endexine ~ /~ thick;
columellae < # thick and long, slightly larger at margins of colpi; tectum ~( #
thick, finely perforate, but ~ /~ thick and less finely perforate at margins of colpi.
Variability: Margins of colpi vary in extent and thickness.
Distribution: In the Caribbean area present from the base of the Psilatricol-
porites operculatus Zone upwards. In Nigeria the first occurrence is slightly higher,
at the base of the Verrucatosporites usmensis Zone. In Borneo this species is
regularly present in the Neogene, but its base has not yet been defined.
Taxonomic affinities: This species shows a striking resemblance to the pollen
of the genus Alchornea (Euphorbiaceae). The pollen of Alchornea cordifolia (Plate
XV, 4) is very similar, that ofA. obovata (Plate XV, 5) and A.javanensis has smaller
colpi and opercula, but larger margins.
Psilatricolporites crassus VAN DER HAMMEN et WYMSTRA, 1964

(Plate XVI, 1, 2)
Literature: Psilatricolporites crassus VAN DER HAMMEN et WYMSTRA, 1964,

p.237, pl.1, fig. 1-4.
Description: Single grain, radially symmetrical, isopolar, spherical to sub-
prolate; outline in polar view circular. Tricolporate; colpi medium long, ectexinous,
straight with pointed ends; pores endexinous, equatorially elongated, oval or slit-
shaped, 15-19 # long, distinctly costate, costae 5 # wide and up to 4 # thick,
parallel to the pores. Endexine 1/z thick, columellae # thick and 1 /z long,
tectum psilate to finely perforate-foveolate, 1-1/z thick.
PLATE XVI
1. Psilatricolporites crassus VAN DER HAMMENet WYMSTRA,Nigeria.
2. Psilatricolporites crassus VAN DER HAMMENet WYMSTRA,Venezuela.
3. Hura crepitans LINNAEUS, upper focus, Recent.
4. Hura crepitans LINNAEUS, lower focus, Recent.
5. Verrutricolporites rotundiporis VAN DER HAMMENet WYMSTRA, Trinidad.
6. Verrutricolporites rotundiporis VAN DER HAMMENet WYMSTRA,Surinam.
7. Verrutricolporites rotundiporis VAN D~R HAMMENet WYMSTRA, Nigeria.
8. Crenea maritima AUBLET, Recent.
9. Crenea maritima AUBLET, upper focus, Recent.
10. Crenea maritima AUaLE'r, lower focus, Recent.
11. Echitricolporites spinosus VAN DER HAMMEN, upper focus, Venezuela.
12. Echitricolporites spinosus VAN DER HAMMEN, lower focus, Venezuela.
13. Riencourtia glomerata CASS, upper focus, Recent.
14. Riencourtia glomerata CASS, lower focus, Recent.

PLATE XVI
Variability: This species is very variable in size, thickness of wall, especially
the endexine, dimensions of columellae and width of the perforations. It is, more-
over, susceptible to corrosion of the tectum, which may lead to pseudo-ornamen-
tations such as areolate sculpture with widely spaced irregular grooves, geminate
sculpture with deep grooves, verrucate sculpture, densely intersected by grooves
or scabrate sculpture with the columellae protruding through membrane-thin
remnants of the tectum.
Distribution: In the Caribbean area and in Nigeria occurring from the base
of the Retibrevitricolporites triangulatus Zone upwards, in the Caribbean area up
to the Recent, but in Nigeria not known at present (except introduced). Absent
from Borneo.
Taxonomic affinities: Closest resemblance is with pollen of the tropical
American genus Hura (Euphorbiaceae) in which the tectum is not perforate-
foveolate as in the fossil species, but psilate. Hura polyandra is very similar in
its apertures, but in the pollen of Hura crepitans (Plate XVi, 3, 4) the ectexinous
colpi are much longer than is found in Psilatricolporites crassus. The pollen types
found in Sapium (Euphorbiaceae) are more different. It is possible that the fossil
species may represent an extinct species of Hura or that the pollen type of this
genus may have changed its appearance slightly. The dominance of Hura crepitans
at present in the Caribbean coastal environment forms a strong additional argu-
ment for assigning the fossil species to Hura. However, the absence of Hura in
the present day natural vegetation of Nigeria (Hura senegalensis is, according to
HUTCHINSON and DALZIEL (1954) probably imported) casts doubt on the affinities
of the Nigerian grains assigned to the species. Recently LANGENHEIMet al. (1967)
have described a fossil pollen closely resembling that of Pelliciera rhizophorae
(Theaceae) from the Oligo-Miocene of Chiapas (Mexico). The resemblance to
some variations of PsilatricoIporites crassus is striking indeed and the possibility
that this monotypic mangrove genus had formerly a much wider range is worth
further investigation.
Genus Zonocostites nov. gen.
Derivatio nominis: Name derived from the costate endexinous apertures.

Diagnosis: Spherical to subprolate, tricolporate; colpi relatively short,
endexinous apertures equatorially elongated or almost fused, distinctly costate;
wall tectate, finely perforate, thin, often almost psilate on equator, or thicker at
poles, due to longer columellae.
Type species: Zonocostites ramonae nov. sp.
Remarks: This genus is intended to accommodate fossil dispersed pollen of
the Rhizophora-Bruguiera type.

Zonocostites ramonae nov. sp.
(Plate XV, 6; holotype; Plate XV, 7)
Literature: Rhizophora-type MULLER 1964, p.37, pl.I, fig.3.

Derivatio nominis: Named in honour of Miss Ramona Henriquez in recog-
nition of her contribution to Venezuelan palynology.
Holotype: Slide TC-164, well CO-77, 3600 ft., Trinidad.
Description: Single grain, radially symmetrical, isopolar, spherical. Tricol-
porate, colpi ectexinous, medium long, straight with pointed ends, slightly costate,
endexinous apertures equatorially elongated to almost fused, distinctly costate, in
polar view slightly vestibulate. Endexine ~ /z thick; columellae < # thick
and high; tectum < # thick, densely perforate, coarser on poles and finer to
almost psilate on equatorial belt, perforations < /~ wide.
Dimensions: 16-19 bt.
Variability: This is a very variable species, especially in size and coarseness
of structure and perforations on the poles. Attempts to separate subtypes corres-
ponding to Recent species of Rhizophoraceae have been unsuccessful.
Distribution: The earliest positive occurrences of Zonocostites ramonae are
known from the Verrucatosporites usmensis Zone in the Caribbean area, with
doubtful ones from the Monoporites annulatus Zone. In Nigeria the first appear-
ances are later and occur around the base of the MagnastriatiteshowardiZone. In
Borneo Eocene sediments unfortunately do not carry well enough preserved
microfloras to permit positive identification, but in the Verrucatosporites usmensis
Zone the species does occur. These pre-Miocene occurrences are all in low frequen-
cies. Approximately at the Oligo-Miocene transition (lower part of the Magna-
striatites howardi Zone) in all three areas a distinct quantitative increase can be
observed and the species remains dominant in the coastal and marine sediments
of the whole tropics up to Recent.
Taxonomic affinities: The rather wide limits adopted for this species include
the pollen types produced by the various species of the genera Rhizophora (Plate
XV, 8, 9), Bruguiera, Ceriops, and Carallia (Rhizophoraceae). Specific differences
exist, but in general the preservation of the fossil grains does not allow their
recognition. Confusion with pollen of other taxa appears unlikely, however.
Genus Verrutricolporites VAN DER HAMMENet WYMSTRA,1964
Verrutricolporites rotundiporis VAN DER HAMMENet WYMSTRA,1964

(Plate XVI, 5, 6, 7)
Literature: Verrutricolporites rotundiporis VAN DER HAMMEN et WYMSTRA,

1964, p.237, pl.I, fig.14.

Description: Single grain, radially symmetrical, isopolar, spherical to sub-
prolate; outline in polar view lobate. Tricolporate, apertures interlobate; colpi
ectexinous, indistinct, medium long, intruding; pores endexinous, circular, 2-3/z
in diameter, sharply outlined. Total wall thickness approx. 1 /~; endexine ~ ~/~
thick, columellae thin, but generally clearly visible, straight, /z long; tectum
# thick, psilate or covered by low, irregularly shaped verrucae, 1-2 # in
diameter and up to /~ high.
Dimensions: 15-24 ~u.
Variability: Mainly in the development of the verrucate sculpture. All
transitions between almost psilate and strongly verrucate grains can be observed
and it has proved impossible to arrive at a separation between a psilate and a
verrucate type, in conformity with Van der Hammen and Wymstra's experience.
The ectexinous furrows are only rarely clearly outlined and, when the grains are
folded together, mostly hidden between the lobes.
Comments: Our description agrees well with the one given by VAN OER
HA~aMEN and WVMSa'RA (1964), but we would stress in addition the trilobate
condition, which often causes the compressed grains to assume a characteristic
folded appearance with the pores often half hidden by the folds. There is evidence
for a gradual increase in degree of verrucation from older to younger strata, but
the phenomenon is not clear enough to be used for stratigraphical interpretation
without detailed statistical analysis. A comparable degree of variation may exist
in the genus Crenea which has probably produced the fossil species. There exists
some superficial similarity with FIorschuetzia trilobata, the points of difference
being the smaller size, distinct columellae, tectate-verrucate sculpture with equal
wall thickness over the entire grain in Verrutricolporites rotundiporis.
Distribution: In the Caribbean area the first occurrences are around the base
of the Magnastriatites howardi Zone; it is very abundant in the coastal and marine
sediments of the Verrutricolporites rotundiporis Zone and afterwards decreases in
numbers. In Nigeria the first occurrence is at the base of the Verrutricolporites
rotundiporis Zone. In the Caribbean area present up to the present day, in Nigeria
scattered rare occurrences from the base Echitricolporites spinosus Zone. Absent
from Borneo.
Taxonomic affinities: The pollen of Crenea maritima (Plate XVI, 8, 9, 10)
agrees in all essential details with Verrutricolporites rotundiporis, but is more
variable. Investigation of Crenea pollen from various localities revealed a vari-
ability in degree of verrucation which is paralleled in the fossil material. However,
in view of the stratigraphical trend towards increased verrucation, one might expect
the recent relatives to show the highest degree of verrucation and this is definitely
not the case. It may be mentioned here that, according to information kindly
supplied by Prof. F. P. Jonker, Utrecht, several species of Crenea described from
the tropical South American coastline are considered by him to be synonyms of
Crenea maritima, with the possible exception of Crenea patentinervis STANDLEV,

1947, of which unfortunately no material was available for comparison. The
dominance of Verrutricolporites rotundiporis in coastal and marine sediments is
in agreement with the present day habitat of Crenea, growing in swampy places
along river courses and in the mangrove vegetation. Some superficial resemblance
exists with the pollen of Sonneratia apetala, but this is considerably larger, has no
ectexinous colpate apertures and shows a graded development in the verrucate
sculpture which is coarser on poles.
Genus Echitricolporites VAN DER HAMMEN, 1956b.
Lectogenotype: Echitricolporites spinosus VAN DER HAMMEN, 1956b
Remarks: The genus is here validated by selecting a lectogenotype, since

Van der Hammen has invalidly indicated the pollen of a Recent species as genotype.
Echitricolporites spinosus VAN DER HAMMEN, 1956b

(Plate XVI, 11, 12)
Literature: Echitricolporites spinosus VAN DER HAMMEN, 1956b, p.92, pl.10,

fig.30.
Remarks: Van der Hammen's name is here validated by selecting a fossil
grain as type.
Description: Single grain, radially symmetrical, isopolar, spherical. Tricol-
porate; colpi ectexinous, straight with pointed ends, fairly long; pores indistinct.
Endexine < # thick; columellae # long (-1/z long underneath spines), < -#
thick; tectate-echinate, tectum < # thick, spines 3-6 # long, 2-4 /z thick at
base, sharply pointed, rather densely spaced.
Dimensions: 23 # (excl. spines).
Variability: A rather large amount of variability is found in size and in
degree of spinosity. However, the spines are always separated and not connected
by a reticulate pattern of muri.
Distribution: In all three areas investigated virtually restricted to the Echi-
tricolporites spinosus Zone. In the Caribbean area scarce occurrences are known,
however, from the Crassoretitriletes vanraadshooveni and upper part of the Magna-
striatites howardi Zones. Most numerous in the Caribbean Mio-Pliocene sediments,
rare in Nigeria, fairly common in Borneo.
Taxonomic affinities: This is the tubuliflorae type of the Asteraceae (Compo-
sitae). Very similar pollen types are found among the genera Espeletia, Mikania,
Pectis, Riencourtia (Plate XVI, 13, 14), Wedelia, and Wulffia and closer identi-
fication will prove to be very difficult. Also many extratropical genera of Asteraceae
show this pollen type. For a further discussion of its possible origin see the section
on "Botanical results".

Echitricolporites meneillyi nov. sp.
(Plate XVII, 1; holotype)
Derivatio nominis: Named in honour of Mr. D. G. R. Mcneill in recognition

of his contribution to Trinidadian palynology.
Holotype: Slide TC-165, well CO-73, 333 ft., Trinidad.
in polar view trilobate. Tricolporate; colpi ectexinous, fairly long, straight with
pointed ends; pores indistinct. Endexine * /2 thick; columellae ~--1 /2 high, /2
long (slightly longer underneath spines), . /2 thick; tectate echinate, tectum
# thick, spines < 1 /2 long, < /2 thick at base, 2-3/2 apart.
Dimensions: 23/2.
Variability: Slight variability in size and coarseness of sculpture.
Comments: The species differs from Echitricolporites eehinatus in its much
smaller and more widely spaced spines.
Distribution: Only found in the Caribbean area, where it occurs from the
base of the Echitricolporites mcneillyi Zone upwards.
Taxonomic affinities: This is a small-spined subtype of the tubuliflorae pollen
type of the Asteraceae. It is found in the genera Ambrosia (Plate XV[I, 2, 3),
Crassocephalum, Ira, and Xanthium. Of these Xanthium forms, with its coarser
sculpture, a transition to the larger spirted tubuliflorae type. Ambrosia and lva
come very close to the fossil grains.
Genus Retitricolporites (VAN DER HAMMEN, 1956) ex VAN DER HAMMEN et WYM-
STRA, 1964
Retitricolporites guianensis VAN DER HAMMEN et WYMSTRA, 1964

(Plate XVII, 8, 9)
P L A T E XVII
1. Echitricotporites mcneillyi nov. sp., Trinidad, holotype.

2. Ambrosia cumanensis KUNTH, upper focus, Recent.
3. Ambrosia eumanensis KUNTH, lower focus, Recent.
4. Retitricolporites irregularis VAN DER HAMMENet WVMSTRA,Venezuela, upper focus.
5. Retitricolporites irregularis VAN DER HAMMENet WYMSTRA,Venezuela, lower focus.
6. Amanoa oblongifolia MUELLER-ARG.,upper focus, Recent.
7. Amanoa oblongifolia MUELLER-ARG.,lower focus, Recent.
8. Retitrieolporites guianensis VAN DEn 1-1AMMENet WVMSTRA,Surinam.
9. Retitricolporites guianensis VAN DER HAMMENet WYMSTRA,Surinam.
10. Bombaeaeidites annae (VANDER HAM~E~),Venezuela.
11. Bombax eeiba LINNAEUS,upper focus, Recent.
12. Bombax eeiba LINNAEUS,lower focus, Recent.

t'-'
>
~1 l~+i+!+?
+++++++ + m, +.
Literature: Retitricolporites guianensis VAN DER HAMMENet WYMSTRA,1964,
p.235, pl.IH, fig.l-2.
Description: Single grain, radially symmetrical, isopolar, prolate. Tricol-
porate; colpi ectexinous, long and intruding with straight, marginate borders and
pointed ends; pores endexinous, circular, 1-2 # wide, rather indistinct. Endexine
/z thick; columellae /z thick and long, regularly distributed and rather closely
spaced; tectum # thick, reticulate; muri /~ thick and high, multicolumellhte,
but not always distinctly visible, lumina of angular shape, slightly elongated,
3-4/z wide, diminishing in size towards colpi.
Dimensions: 28-47/z.
Variability: Size and shape of this species are rather variable, the smaller
specimens showing a finer ornamentation. The species can, however, be easily
recognized by its loose-meshed reticulate sculpture with angular shape of the
lumina and thin, low muri. In well preserved specimens the regularly distributed
columellae are characteristic.
Comments: Inspection of type material has shown the identity of the speci-
mens described here with Retitricolporites guianensis. Our description, however,
differs slightly from the one given by VAN DER HAMMEN and WYMSTRA(1964).
Distribution: Only known from the Caribbean area where it occurs from
the base of the Retitricolporites guianensis Zone upwards.
Taxonomic affinities: A certain resemblance exists with pollen of Firrniania
colorata and Hildegardia barteri (Sterculiaceae), but a striking difference is the
finely reticulate appearance of the fairly broad muff in these types. To a lesser
degree this is the difference also with pollen of Glossosternon bruguieri (Sterculia-
ceae). The pollen of Pterocyrnbiurn beccarii has finely reticulate muri and in
addition a thicker wall. Pollen of Sterculia rnexicana has a thicker wall and is finer
reticulate on the poles. Trichospermurn pollen (Tiliaceae) has distinctly equatorially
elongated pores, although the wall structure is strikingly similar.
In view of this situation it is not yet possible to come to a firm decision
which taxon has produced Retitricolporites guianensis pollen. However, Ster-
culiaceae and Tiliaceae are the only families in which this pollen type has been
found so far.
Retitricolporites irregularis VAN DER HAMMENet WYMSTRA,1964

(Plate XVII, 4, 5)
Literature: Retitricolporites irregularis VAN DER HAMMENet WYMSTRA,1964,

p.235, pl.III, fig. 9, 10.
in polar view circular. Tricolporate; colpi ectexinous, long, strongly intruding
with straight costate borders and pointed ends; costae 2/z thick and wide; pores
endexinous, oval, 2-3 # wide, equatorially elongated, slightly costate. Endexine
338 Rev. Palaeobotan.Palynol., 6 (1968) 189--348

< 1 # thick, columellae <~ /~ thick and long, arranged in a reticulate pattern and
supporting in single rows a curvimurate reticulum; lumina of irregular shape, up
to 5 / , wide, finer along colpi.
Variability: There is considerable variation in the coarseness of the ornamen-
tation. Occasionally four colpi are present.
Distribution: In the Caribbean area present from the base of the Retibrevi-
tricolporites triangulatus Zone upwards. In Nigeria the first scattered occurrences
are in the middle part of this zone.
Taxonomic affinities: The closest resemblance is with Amanoa oblongifolia
(Euphorbiaceae, Plate XVII, 6, 7), less so with Pseudolachnostylis glauca (Euphor-
biaceae) which has non-costate apertures and a finer reticulation. Since the former
species is, according to LINDEMAN(1953), abundant in scrub wood along creeks
on peaty mud in Surinam, it is a likely candidate for producing the fairly large
amounts of Retitricolporites irregularis found throughout its range in the Carib-
bean area.
Genus Retibrevitricolpites VAN HOEKEN-KLINKENBERG,1966
Retibrevitricolpites triangulatus VAN HOEKEN-KLINKENBERG,1966

(Plate XVIII, 1, 2)
Literature: Retibrevitricolpites triangulatus VAN HOEKEN-KLINKENBERG,

1966, p.40, pl.II, fig.2.
polar view subtriangular. Tricolporate; colpi 4 /* long, ectexinous, marginate,
margins 1 # wide, with thickened tectum; pores endexinous, equatorially elon-
gated and vestibulate, 8/~ long; vestibulum 5 # wide, 1/~ high. Endexine { # thick;
columellae 3 # thick and long, slightly longer around pores; tectum # thick,
near colpi 1-13 # thick, finely reticulate; lumina 13 # wide, rather angular in
shape; perforate on poles; muri /~ thick and wide, on poles perforations 3 # wide
and 3 # apart.
Variability: Rather pronounced variability in coarseness of sculpture, less
so in size and thickness of wall.
Comments: The description presented here differs from the original type
description presented by the author, but agrees with GONZALEZ'description (1967).
Inspection of type material has convinced us that our specimens can and should
be incorporated in Brevitricolpites triangulatus VAN HOEKEN-KLINKENBERG.Since
pores are present, this genus is here placed in the Class Tricolporatae.
Distribution: Both in the Caribbean area and in Nigeria the first occurrence
is at the base of the Retibrevitricolpites triangulatus Zone and disappearance at

the top of the Verrucatosporites usmensis Zone. Absent from Borneo.
Taxonomic affinities: Unknown, apparently extinct.
Genus Bombacacidites COUPER, 1960
Bombacacidites annae (VAN DER HAMMEN, 1954)

(Plate XVII, 10)
Literature: Tricolporites annae VAN DER HAMMEN, 1954, p.96, pl. 9.

Retitricolporites (Bombacites) annae VAN DER HAMMEN et
GARCIA, 1966, p.112.
Bombacacidites annae LEIDELMEYER, 1966, p.55.
Description: Single grain, radially symmetrical, isopolar, oblate; in polar
view rounded-intersubangular. Tricolporate; colpi 14/~ long, ectexinous, slightly
protruding, costate; costae 2 # wide; pores indistinct, endexinous, oval, equato-
rially elongated. Total wall thickness 1~-2/~; endexine # thick; columellae present
under muff of reticulum only, /z thick and long; tectum reduced to a reticulum;
muri multicolumellate, z-17/z
~ 1 thick, -1 # high; lumina 1#-2/~ wide on poles
and near colpi, much smaller on intercolpate areas, where the sculpture is almost
perforate; transitions in sculptural distribution are gradual.
Variability: Size and coarseness of ornamentation are variable, but the
multicolumellate structure underlying the muff is a fairly constant characteristic,
although in badly preserved grains this may have been reduced to a finely granulate
pattern.
Comments: Our specimens agree with the type material, but in order to
differentiate the species from other closely related bombacaceous grains, the recom-
mendation is made to take the multicolumellate structure underlying the muri,
PLATE XVIII
1. Retibrevitricolpites triangulat~s VAN HOEKEr~-KUNKENBERG,Venezuela, upper focus.

2. Retibrevitricolpites triangulatus VAN HOEKEN-KuNKENBERG,Venezuela, lower focus.
3. Margocolporites vanwijhei nov. sp., Venezuela, holotype.
4. Caesalpinia coriaria WILDENOW,upper focus, Recent.
5. Caesalpinia coriaria WILDENOW,lower focus, Recent.
6. Jandufouria seamrogiformis nov. gen., nov. sp., Guyana, holotype.
7. Catostemma altsonii SANDWITH,upper focus, Recent.
8. Catostemma altsonii SANDW1TH,lower focus, Recent,
9. Striasyncolpites zwaardi nov. gen, nov. sp., Trinidad, holotype.
10. Cuphea aequipetala CAVANILLIES,Recent.
11. Perisyncolporites pokornyi nov. gen., nov. sp., Venezuela, holotype.
12. Brachypteris ovata SMALL,Recent.
Magnification x 1,000.

0
m .... V , ~ ~ ~ ~ ~ ~ ~ .... ~ ~ ~,~,~
~ ~ili~i~ii~i ~ ~i
i
which is not mentioned by VAN DER HAMMENand GARCIA(1966), as an additional
typical feature.
Distribution: In the Caribbean area restricted to the Ctenolophonidites
lisamae and the Foveotricolpitesperforatus Zones. Absent from Borneo and Nigeria.
Taxonomic affinities: Closest resemblance is with certain species of the genus
Bombax (Bombacaceae). Especially close are Bombax ceiba (Plate XVII, II, 12),
B. rhodognaphalon, and B. pubescens, which all show the multicolumellate structure
in varying degrees. Bombax mexicana differs in the pronounced triangular shape.
No other genus of Bombacaceae shows this pollen type and identification with
a, possibly extinct species of Bombax appears well founded.
Genus Margocolporites RAMANUJAM, 1966
Margocolporites vanwijhei nov. sp.

(Plate XVIII, 3; holotype)
Derivatio nominis: Named in honour of Mr. D. H. van Wijhe in recognition

Holotype: Slide TC-166, Venezuela.
polar view circular. Tricolporate; colpi ectexinous, 10 # long, costate and sur-
rounded by wide, finely baculate margins with straight edges and pointed ends
which are or are not connected at the poles; costae 2 # wide; width of margins in
equatorial plane 20/~; pores endexinous, round, 4-5 # in diameter, often slightly
protruding. Endexine # thick; columellae 1 # long, /~ thick; tectate-reticulate
outside colpate margins, muri multicolumellate, 1 # thick and high; lumina 2-2 #
wide at equator, 1-11 # wide on poles. Colpate margins intectate, finely baculate
because of densely spaced, evenly distributed columellae.
Variability: Size and coarseness of sculpture are rather variable, syncolporate
and colporate grains are included, because in Recent relatives these variations
occur within one and the same flower.
Distribution: In the Caribbean area occurring approximately from the lower
part of the Monoporites annulatus Zone upwards, in Nigeria restricted to the
Verrucatosporites usmensis and Magnastriatites howardi Zones, and in Borneo
occurring at least from the Magnastriatites howardi Zone upwards, base not yet
determined, but absent from the Paleocene and Upper Cretaceous.
Taxonomic affinities: Among the Fabaceae (Leguminosae)the following
genera show strikingly similar pollen: Adipera, Brasilettia, Caesalpinia, Hae-
matoxylon, Mezoneuron, and Poincianella. Without further detailed studies it is
impossible to come to a closer identification of the fossil material. Especially
similar, however, is the pollen of Caesalpinia bonduc and C. coriaria (Plate XVIII,
4, 5).
342 Rev. Palaeobotan.Palynol., 6 (1968) 189-348

Class Stephanocolporatae [VERSENet TROELSSMITH, 1950
Genus Jandufouria nov. gen.
Derivatio nominis: The genus has been named in honour of Prof. J. Dufour,
who successfully stimulated the early palynological investigations in Venezuela.
Diagnosis: Oblate, 4-6 colporate, colpi ectexinous, slightly costate, reaching
approximately halfway the poles, endexinous pores indistinct, equatorially elon-
gated, wall rather densely columellate, tectate-densely perforate, perforations
evenly distributed, lumina < /~ in width.
Type species: Jandufouria seamrogiformis nov. sp.
Comments: This genus accommodates fossil pollen of the Catostemma type.
It can be distinguished from Retistephanocolporites by its evenly and densely
perforated tectum.
Jandufouria seamrogiformis nov. sp.

(Plate XVIII. 6; holotype)
Derivatio nominis: Name derived from the resemblance in shape to the leaf
of the Irish shamrock (Trifolium repens).
Holotype: Slide TC-167, well Aurora-l, 651 ft., British Guyana.
Description: Single grain, radially symmetrical, isopolar, oblate. 4-6 col-
porate; colpi ectexinous, 36/z long, reaching half-way the poles, slightly costate
with straight borders and pointed ends, costae 1~-2 # wide; endexinous apertures
equatorially elongated, faintly costate. Endexine 1 # thick; columellae -~-1 # long,
# thick; tectum -1 /~ thick, densely and evenly perforated.
Variability: There is some slight variability in size and in wall thickness, but
as a rule the species is highly characteristic because of its shape.
Distribution: Only known from the Caribbean area, where it occurs from
the base of the Verrucatosporites usmensis Zone upwards.
Taxonomic affinities: Close resemblance exists with the pollen of the genus
Catostemma (Bombacaceae), especially with C. altsonii (Plate XVIII, 7, 8), C.
sclerophyllum, and C. fragrans. C. commune has distinctly digitate columellae, a
feature which has not been observed in the fossil material so far. The pollen of
Aguiaria is similar, but tricolporate. The widespread occurrence of Catostemma
in the rain-forests of tropical South America supports the identification.
Class Pericolporatae IVERSENet TROELSSMITH, 1950
Genus Perisyncolporites nov. gen.

Derivatio nominis: Name derived from perisyncolporate apertures.
Diagnosis: Spherical, often slightly angular grains. Ectexinous apertures
syncolpate, arranged basically in a hexahedral pattern, with variations to higher
orders (hepta- to dodecahedr,ql) or lower orders (penta- to tetrahedral) or any
incomplete arrangement of the foregoing or absent. Endexinous apertures peri-
porate, in syncolpate grains generally located eccentrically on colpi. Wall thick,
tectate.
Type species: Perisyncolporites pokornyi nov. sp.
Comments: This genus is intended to include the very characteristic fossil
malpighiaceous pollen grains listed below.
Perisyneolporites pokornyi nov. sp.

(Plate XVIII, 11 ; holotype)
Derivatio nominis: Named in honour of Mr. G. Pokorny in recognition of

his contribution to Colombian palynology.
Holotype: Slide TC-168, well OG-1, 5828-5850 ft., Venezuela.
Description: Single grain, radially symmetrical, often centro-symmetrical,
isopolar, spherical; outline often angular in conformity with aperture pattern.
Ectexinous apertures syncolpate, or nearly so, arranged in a complicated pattern,
in general according to the ribs of a hexahedron, with variations to higher orders
(heptadodecahedral patterns) or lower orders (penta- and tetrahedral patterns) or
any incomplete arrangements of the foregoing, deviating in a zonate pattern or
totally absent; colpi generally straight or rounded syncolpate, then often very
wide and with relatively small intercolpate areas, sometimes very shallow, margins
generally slightly costate, costae 1 gLthick, 3 p wide. Pores endexinous, generally
circular to oval, 2-4 t~ wide, if colpi present located eccentrically, frequently less
pores than colpi. Endexine 2-3 p thick; columellae < 1 It long, < F~ thick;
tectum p thick, slightly thickened along colpi, psilate.
Variability: As discussed above, variation is mainly in aperture number and
arrangement. The wall structure is constant, but corrosion often results in a pitted
or scabrate surface.
Comments: It would be possible, on the base of aperture configurations to
split this pollen type into a large number of species, but, since these smaller groups
have no restricted ranges, they are united here under one large polymorphic
species.
Distribution: In the Caribbean area occurring from the base of the Ver-
rucatosporites usmensis Zone upwards, in Nigeria known from the middle part
of the Verrucatosporites usmensis Zone upwards. Unknown from Borneo.
Taxonomic affinities: This very peculiar pollen type is so far known only from
Malpighiaceae. The pollen grains of the following genera correspond best with

the fossil species: Brachypteris (Plate XVIII, 12), Bunchosia, Hiraea, Mascagnia,
Stigmatophyllum, and Tetrapterys. The pollen grains of the Malesian genera of the
Malpighiaceae do not belong to this type.
Class Syncolpatae IVERSENet TROELSSMITH,1950
Genus Striasyncolpites nov. gen.
Derivatio nominis: Named after striate ornamentation and syncolpate

apertures.
Diagnosis: Oblate, tricolporate; colpi connected at poles, pores protuding;
wall thin, finely striate, generally coarser at equator.
Type specimen: Striasyncolpites zwaardi nov. sp.
Comments: Identical with the pollen type in Cuphea (Lythraceae).
Striasyncolpites zwaardi nov. sp.

(Plate XVIII, 9; holotype)
Derivatio nominis: Named in honour of Mr. A. L. P. Zwaard in recognition

Holotype: Slide TC-169, surface sample HO-24, Trinidad.
Description: Single grain, radially symmetrical, isopolar, oblate; outline
in polar view rounded triangular. Tricolpate, colpi ectexinous, connected at poles;
pores endexinous, small, protruding over a distance of 2 #. Endexine < /z thick;
columellae < # thick and long; tectum < /z thick, striate in meridional direc-
tion; striae /z wide, /~ apart, coarser at equator.
Variability: Mainly in coarseness of striate sculpture.
Distribution: Only known from the Caribbean area, and restricted to the
Echitrieolporites spinosus Zone.
Taxonomic affinities: This pollen type is only known from the genus Cuphea
(Lythraceae). Especially close are C. aequipetala (Plate XVIII, 10), C. pinetorum,
C. platycentra, and C. wrightii.
ACKNOWLEDGEMENTS
In the compilation of this paper the authors were able to draw on the accumu-
lated experience of a large number of colleagues, whose assistance and criticism
is gratefully acknowledged here. Thanks are also due to the Management of the
Bataafse Internationale Petroleum Maatschappij and the Shell-BP Petroleum
Development Company of Nigeria Ltd. for permission to publish this paper,

to Creole Petroleum Corporation and Mene Grande Oil Company (Venezuela)
for allowing the publication of well data received in exchange, and to Prof. Dr. C.
G. G. J. van Steenis (Leiden) for critically reading the manuscript.
REFERENCES
AMERICANCOMMISSIONON STRATIGRAPHICNOMENCLATURE,1961. Code of stratigraphic nomen-

clature. Bull. Am. Assoc. Petrol. Geologists, 45: 645-665.
ANDERSON, R. Y., 1960. Cretaceous-Tertiary palynology, eastern side of the San Juan Basin
New Mexico. New Mexico, Bur. Mines Mineral Resources, Mere., 6:58 pp.
BELSKY, C. V., BOLTENHAGEN,E. und POTONI~, R., 1965. Sporae digpersae der Oberen Kreide
yon Gabun, A.quatoriales Afrika. Paliiontol. Z., 39: 72-83.
BOLLI, H. M., 1966. Zonation of Cretaceous to Pliocene marine sediments based on planktonic
Foraminifera. Bol. Inform. Asoc. Venezolana Geol. Mineria Petrol., 9: 3-32.
BOLTENHAGEN, E., 1967. Spores et pollen du Cr6tacd sup6rieur du Gabon. Pollen Spores, 9(2):
335-355.
BRATZEVA, G. M., 1965. Pollen and spores in Maastrichtian deposits of the Far East. Tr. Geol.
Inst. Akad. Nauk S.S.S.R., 1 2 9 : 5 4 2 (in Russian).
CHRIST, H., 1910. Die Geographic der Fame. Gustav Fischer, Jena, 357 pp.
COOKSON,I. C., 1950. Fossil pollen grains of proteaceous type from Tertiary deposits in Australia.
Australian J. Sei., Set. B, 3: 166-177.
COOKSON,I. C. and PIKE, K. M., 1954. Some dicotyledonous pollen types from Cainozoic deposits
in the Australian region. Australian J. Botany, 2: 197-219.
COUPER, R. A., 1953. Upper Mesozoic and Cainozoic spores and pollen grains from New Zea-
land. New Zealand, Dept. Sci. Ind. Res., Geol. Surv. Palaeontol. Bull., 22:77 pp.
COVAS, G. y SCHNACK,B., 1945. El valor taxonomico de la relacion "longitud del pistilo: Volumen
del grano de polen". Darwiniana, 7: 88-90.
CROSS, A. T., 1966. Source and distribution of palynomorphs in bottom sediments, southern
part of Gulf of California. Marine Geol., 4(6): 467-524.
DRUGG, W. S., 1967. Palynology of the Upper Moreno Formation (Late Cretaceous-Paleocene),
Escarpado Canyon, California. Palaeontographica, B, 120: 1-71.
ENGLER, A., 1905.15bet floristische Verwandtschaft zwischen den tropischen Afrika und Amerika,
sowie iiber die Annahme eines versunkenen brasilianisch-~ithiopischen Continents. Sitz.
Ber. Kdnigl. Preuss. Akad. Wiss. Berlin, 1905: 180-231.
ERDTMAN, G., 1952. Pollen Morphology and Plant Taxonomy (An Introduction to Palynology, 1).
Angiosperms. Almqvist and Wiksell, Stockholm; Chronica Botanica Co., Waltham,
Mass., 239 pp.
ERDTMAN,G., 1954. On the occurrence of olacaceous pollen grains in Eocene deposits in Germany.
Svensk Botan. Tidskr., 48: 804-885.
ERDTMAN, G., 1956. Pollen grains of cf. Ctenolophon from Tertiary deposits in India. Grana
Palynologiea, 1: 5-7.
FucHs, H. P., 1967. Pollen morphology and its relation to taxonomy and phytogeography in the
family Bombacaceae. Rev. Palaeobotan. Palynol., 3: 119-132.
FUNNEL, B. M., 1964. The Tertiary Period. In: W. B. HARLAND,A. GILBERTSMITH and B. WIL-
COCK (Editors), The Phanerozoic Time Scale. A Symposium Dedicated to Professor A.
Holmes.--Suppl. Qaart. J. Geol. Soc. London, 120: 179-191.
GONZALEZ GUZMAN, E., 1967. A Palynological Stud), on the Upper Los Cuervos and Mirador
Formations (Lower and Middle Eocene," Tibt~ Area, Colombia). Brill, Leiden, 129 pp.
GOTHAN, W. und WEYLAND, H., 1964. Lehrbuch der Paliiobotanik. Akademie Verlag, Berlin,
594 pp.
HARRIS, W. K., 1965. Basal Tertiary microfloras from the Princetown area, Victoria, Australia.
Palaeontographica, B, 115: 75-106.
HOPPING, C. A., 1967. Palynology and the oil industry. Rev. Palaeobotan. Palynol., 2: 23-48.

HUGHES, N. F. and MOODY-STUART, J., 1966. Descriptions of schizaeaceous spores taken from
Early Cretaceous macrofossils. Palaeontology, 9: 274-289.
HUTCHINSON, H. and DALZIEL, J. M., 1954. Flora of West Tropical Africa. Crown Agents Over-
seas Governments and Admini~,trations, London, 2nd ed. revised by R. W. J. KEAY,
l(I): 1--294; 1(2): 297-828; 2(1963): 1 544; supplement by A. H. G. ALSTON" Ferns and
Fern-Allies of West Tropical Africa, 1-89.
JARDINI~, S. et MAGLOIRE, L., 1965. Palynologie et stratigraphie du Cr6tac6 des bassins du S6n6gal
et de C6te d'lvoire. MOrn. Bur. Rech. GOol. MiniOres, 32: 187-245.
KEDVES, M., 1960. F.tudes palynologiques dans le bassin de Dorog, 1. Pollen Spores, 2:89-118.
KEDVES, M., 1961. l~tudes palynologiques dans le bassin de Dorog, 2. Pollen Spores, 3: 101-153.
KRUTZSCH, W., 1957. Sporen- und Pollengruppen aus der Oberkreide und dem Terti/ir Mittel-
europas und ihre stratigraphische Verteilung. Z. Angew. Geol., 3: 509-548.
KRUTZSCH, W., 1959. Mikropal/iontologische (sporenpal/i.ontologische) Untersuchungen in der
Braunkohle des Geiseltales. Geologie (Berlin), Beih., 3 (21-22): 425 pp.
KUYL, O. S., MULLER,J. and WATERBOLK,H. T., 1955. The application of palynology to oil
geology with reference to western Venezuela. Geol. Mijnbouw, 17: 49-76.
LANGENHEIM, J. H., HACKNER, B. L. and BARTLETT, A., 1967. Mangrove pollen at the depositional
site of Oligo Miocene amber from Chiapas, Mexico. Harv. Univ. Bot. Mus. Leaflet, 21(10):
289-324.
LEIDELMEYER, P., 1966. The Paleocene and Lower Eocene pollen flora of Guyana. Leidse Geol.
Mededel., 38: 49-70.
LIECHTI, P., 1960. The geology of Sarawak, Brunei and the western part of North Borneo. Bull.
Geol. Surv. Dept., Kuching, 3: 1-360.
LINDEMAN, J. C., 1953. The Vegetation of the Coastal Region of Suriname. Thesis, State Univ.,
Utrecht, 135 pp.
LOEBUCH, A. R. and TAPPAN, H., 1964. Treatise on Invertebrate Paleontology, part C, Protista--
Geol. Soc. Am. Spec. Papers, 1(2): 1 900.
MANTEN, A. A., 1966a. Half a century of modern palynology. Earth-Sci. Rev., 2(4): 277-316.
MANTEN, A. A., 1966b. Some current trends in palynology. Earth-Sci. Rev., 2(4): 317-343.
MAYR, E. (Editor), 1952. The problem of land connections across the South Atlantic with special
reference to the Mesozoic. Bull. Am. Museum Nat. Hist., 99, 79-258.
MENCHER, F., FlCHTER, H. J., RENZ, H. H. and WALLIS, W. E., 1953. Geology of Venezuela and
its oil fields. Bull. Am. Assoc. Petrol. Geologists, 37: 690-777.
MILLER, J. B., EDWARDS, K. L., WOLCOTT, P. P., ANISGARD, H. W., MARTIN, R. and ANDEREGG,
H., 1958. Habitat of oil in the Maracaibo Basin, Venezuela. In: L. G. WEEKS (Editor),
Habitat of Oil. Am. Assoc. Petrol. Geologists, Tulsa, Okla., pp.601-640.
M/2LLER, H., 1966. Palynological investigations of Cretaceous sediments in northeastern Brazil.
In: J. E. VAN HINTE (Editor), Proceedings of the Second West African Micropaleontological
Colloquium, lbadan, 1965. Brill, Leiden, pp. 123-136.
MULLER, J., 1959. Palynology of Recent Orinoco delta and shelf sediments. Micropaleontology,
5: 1-32.
MULLER, J., 1964. A palynological contribution to the history of the mangrove vegetation in
Borneo. In: L. M. CRANWELL (Editor), Ancient Pacific Floras. Univ. Hawaii Press,
Honolulu, pp. 3 3 4 2 .
MULLER, J., 1966. Montane pollen from the Tertiary of northwestern Borneo. Blumea, 14:
231-235.
MULLER, J. and Hou-L1u, S. Y., 1966. Hybrids and chromosomes in the genus Sonneratia
(Sonneratiaceae). Bhlmea, 14: 337-343.
MULLER, J., 1968. Palynology of the Pedawan and Plateau Sandstone Formations (Cretaceous-
Eocene) in Sarawak, Malaysia. Micropaleontology, 14:1-37.
NAYAR, B. K., LATA, P. and TIWARI, L. P., 1964. Spore morphology of the ferns of west tropical
Africa. Pollen Spores, 6: 545-582.
POTONI~, R., 1934. Zur Morphologie der fossilen Pollen und Sporen. Arb. Inst. Paliiobotan.
Petrog. Brennstein., 4: 5-24.
POTONI~, R., 1956-1960. Synopsis der Gattungen der Sporae dispersae, 1-3. Geol. Jahrb. Beih.,
23(1956): 1 103; 31(1958): 1-114; 39(1960): 1 189.

POTONI~, R. und GELLETICH, J., 1933. Lrber Pteridophyten-Sporen einer eoc/inen Braunkohle
aus Dorog in Ungarn. Sitz. Ber. Ges. Naturforsch. Freunde Berlin, 1932: 517-528.
RAMANUJAM, C. G. K., 1966. Palynology of the Miocene lignite from south Arcot district,
Madras, India. Pollen Spores, 8: 149-203.
REYMENT, R. A., 1965. Aspects of the Geology of Nigeria. Univ. Press, Ibadan, 144 pp.
SAAD, S. I., 1962. Pollen morphology of Ctenolophon. Botan. Natiser Lunds Botan. Fdren.,
115(1): 49-57.
SCHAUB, H. P., 1948. Outline of sedimentation in Maracaibo Basin, Venezuela. Ball. Am. Assoc.
Petrol. Geologists, 32: 215-227.
SCHWARCK ANGLADE,A. (Editor), 1956. Lexico Estratigrafico de Venezuela. Bol. Geol., Publ.
Especial., Mitt. Minas e Hidrocarbon, l : 740 pp.
SHORT, K. C. and STXUBLE, A. J., 1967. Outline of geology of Niger delta. Bull. Am. Assoc.
Petrol. Geologists, 51: 761-779.
SIMPSON, G. G., 1952. Probabilities of dispersal in geologic time. In: E. MAYR (Editor), The
Problem of Land Connections across the South Atlantic with Special Reference to the
Mesozoic--Bull. Am. Museum Nat. Hist., 99: 163-176.
STOLK, J., 1963. Contribution h l'6tude des correlations microfauniques du Tertiaire inf6rieur de
la Nigeria m6ridionale. Dans: Colloque International de Micropalkontologie, Dakar--Bur.
Rech. G~ol. Minikres Mdm., 32:247 275.
TRALAU,H., 1964. The Genus Nypa VAN WURMB. Kgl. Svenska Vetenskap. Akad. Hand/., 10(1):
1-29.
VAN GIJZEL, P., 1961. Autofluorescence and age of some fossil pollen and spores. Konh~kl. Ned.
Akad. Wetenschap., Proc., Ser. B, 64: 56-63.
VAN DER HAMMEN, T., 1954. El desarrollo de la flora Colombiana en los periodos geologicos, 1.
Maastrichtiano hasta Terciario mRs inferior. Bol. Geol. (Bogota), 2(1): 49 106.
VAN DER HAMMEN, T., 1956a. Nomenclatura palinologica sistematica. Bol. Geol. (Bogota), 4:
23-62.
VAN DER HAMMEN, Y., 1956b. A palynological systematic nomenclature. Bol. Geol. (Bogota),
4(2]3): 63-101.
VAN OER HAMMEN, T., 1956c. Descripcion de algunos generos y especies de polen y esporas
fosiles. Bol. Geol. (Bogota), 4: 103-109.
VAN DER HAMMEN, T., 1956d. Description of some genera and species of fossil pollen and spores.
Bol. Geol. (Bogota), 4(2/3): 111- l 17.
VAN DER HAMMEN, T. and GARCIA DE MUTIS, C., 1966. The Paleocene pollen flora of Colombia.
Leidse Geol. Mededel., 35:105-116.
VAN DER HAMMEN, T. and WVMSTRA, T. A., 1964. A palynological study of the Tertiary and
Upper Cretaceous of British Guiana. Leidse Geol. Mededel., 30:183 241.
VAN HOEKEN-KLINKENBERG, P. M. J., 1964. A palynological investigation of some Upper Creta-
ceous sediments in Nigeria. Pollen Spores, 6: 209-231.
VAN HOEKEN-KLINKENBERG,P. M. J., 1966. Maastrichtian, Paleocene and Eocene pollen and
spores from Nigeria. Leidse Geol. Mededel., 38: 37~8.
VAN RAADSHOOVEN, B., 1951. On some Paleocene and Eocene larger Foraminifera of western
Venezuela. World Petrol. Congr., Proc., 3rd, The Hague, 1951, Sect. I, pp.476~89.
VAN STEENIS, C. G. G. J., 1962a. The distribution of the mangrove genera and their palaeogeo-
graphical implication. Koninkl. Ned. Akad. Wetenschap., Proc., Ser. C, 65: 164-169.
VAN STEENIS, C. G. G. J., 1962b. The land-bridge theory in botany. Blumea, 11(2): 235-372.
VARMA,C. P. and RAWAT,M. S., 1963. A note on some diporate grains recovered from Tertiary
horizons of India and their potential marker value. Grana Palynologiea, 4:130-139.
W1LSON, J. T., 1965. Evidence from ocean islands suggesting movements in the earth. In: P. M. S.
BLACKETT, E. BULLARDand S. K. RUNCORN (Editors), A Symposium on Continental Drift--
Phil. Trans. Roy. Soc. London, Ser. B, 1088: 145-167.
WYMSTRA, T. A., 1967. A pollen diagram from the Upper Holocene of the Lower Magdalena
valley. Leidse Geol. Mededel., 39: 261-267.
WYMSTRA,T. A. and VAN DER HAMMEN,T., 1967. Palynological data on the history of tropical
savannas in northern South America. Leidse Geol. Mededel., 38:71-90.

Palinología

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Palinología

Uploaded by

Copyright:

Available Formats

Review of Palaeobotany and Palynology

Elsevier Publishing Company, Amsterdam - Printed in The Netherlands

P A L Y N O L O G Y OF TERTIARY SEDIMENTS FROM TROPICAL AREAS

J. H. G E R M E R A A D , C. A. HOPPING ANI~ J. MULLER ~

Bataafse Internationale Petroleum Maatschappij, The Hag,e (The Netherlands)

(Received February 8, 1968)

1 Present address: Rijksherbarium, State University, Leiden (The Netherlands).

Rev. Palaeobotan. Palynol., 6 (1968) 189-348 189

MATERIALS AND METHODS

190 Rev. Palaeobotan. Palynol., 6 (1968) 189-348

Rev. Palaeobotan. Palynol., 6 (1968) 189-348 191

192 Rev. Palaeobotan. Palynol., 6 (1968) 189-348

)n of the symbols, see p.203.)

Fig.2 (pp.195-196). Distribution chart well Chafurray-3, Colombia.

SPORE M A R K E R SPECIES SUBDIVISION

CARIBBEAN ATLANTIC PANTROPICAL

5418 " " C *

GI5 " O /OOQ /

Fig.3 (pp.197-198). Distribution chart Rubio Road surface section, Venezuela.

INFORMAL UNITS "1"

P ~" '~' ~' ~ ~ 'r' "~ ! DEPTH OR SAMPLE N U M B E R

In order to avoid the impression of high accuracy and to facilitate visual

INTERPRETATION OF' DATA

The practical purpose of our investigations can be briefly stated as the

Rev. Palaeobotan. Palynol., 6 (1968) 189 348 203

Evohttionary change and migration

Primary factors are evolutionary change and migration or increases and

204 Rev. Palaeobotan. Palynol., 6 (1968) 189-348

Rev. Palaeobotan. Palynol., 6 (1968) 189-348 205

In addition, the secondary factors controlling pollen and spore dispersal

206 Rev. Palaeobotan. Palynol., 6 (1968) 189-248

ANOMALINA ZONE INFORMAL UNITS .~ 0

LOWER LAGUNJLLASSAND LAGUFIA BA(HAC)UER~

%, %. %. O * %. %. NAGNASTRIAT IrES HOWARDI

L'~ U S/INn "IVM~O::INI

~ N O I I V I , q~lO:l vt~vd ,,]aim

'] S~31d SV1

"~lnO,, ~ SV~"3ld SV1

ROCK-STRATIGRAPHICAL FAUNAL DATA

INOCERAMUS SP. 438-245; ~/

Fig.9 (pp.217-218). Distribution chart well Egoli-l, Nigeria.

~" BU]TINIA ANDREEVI

- l!i >-, >

Rev. Palaeobotan. Palynol., 6 (1968) 189-348 221

222 Rev. Palaeobotan. Palynol., 6 (1968) 189-348

~- =o ~o o.~oo =~, ~-~

CHiEOGUEHBELIHA MARTINI/ 3775' / o ,/

40~5' " " " " 0

4170' " " go 0/O

4415' "/ " " O

Fig.12 (pp.223-224). Distribution chart well Benin West-l, Nigeria.

0 \ ' . ' 0 0 " 0 " 0 - " 0 0 0 0 0 \ 0 0 0 - 0 0

A~AN CLAY MEMBER _Z - - 0

DEPTH O R SAMPLE NUMBER

Fig.14 (pp.227-228). Distribution chart well Lubara Creek-2, Nigeria.

~ N '~ CARIBBEAN ATLANTIC PANTROPICAL

Statistical analysis of the palynological data

In the absence of positive information from lithology, faunal or botanical

Rev. Palaeobotan.Palynol., 6 (1968) 189-348 233

Summary of interpretation experience

234 Rev. Palaeobotan. Palynol., 6 (1968) 189-348

Rev. Palaeobotan. Palynol., 6 (1968) 189-348 235

The oldest pantropical zone recognized is the Proxapertites operculatus

236 Rev. Palaeobotan. Palynol., 6 (1968) 189-348