more brackish environment. However, this conclusion 
must remain tentative, until statistical analyses are un- 
dertaken of Rhizophora pollen presently being deposited 
in various types of marine and brackish environments, to 
show whether the relative percentages of the two types 
of pollen actually reflect the two vegetational groups. 
The percentage of fungal spores to total pollen is con- 
sidered by van der Hammen to represent a possible index 
of ‘‘landinwards”’ conditions. Since the identification of 
individual grains as belonging to the Rhizophora Mangle- 
type or Rhizophora spp.-type may not always be certain, 
a few grains in each sample may have been misidentified. 
Therefore, percentages given in Table I are approximate. 
Sample I from Simojovel appears from several lines of 
evidence to represent deposition at the actual site of a 
mangrove swamp. The sample is lignitic in nature, and 
mangroves today are known to be active peat formers 
(Chapman, 1939; Spackman e¢ al., 1964). The total con- 
tent of Rhizophora pollen is very high (90%); associated 
pollen is large and heavy, and would not be expected to 
be carried far from its source plant. The fungal spore 
content is very low. The extremely high percentage of 
Rhizophora pollen also suggests that the source area was 
not anarrow fringing mangrove but an extensive forest. 
Rhizophora Mangle-type is present (Plate XL, fig. 4), 
but in negligible percentages, suggesting that the man- 
grove forest here was composed largely of individuals 
belonging to the Rhizophora spp. division (Plate XL, 
figs. 2 and 8) and grew perhaps under brackish conditions. 
The occurrence in this sample of pollen very closely 
resembling that of Pelliciera rhizophorae (Plate XLII, 
figs. 4-7), a highly specialized member of the Theaceae, 
is very interesting. Since the range of this monotypic 
genus today is restricted to mangrove swamps along the 
Pacific coast from Costa Rica south to western Colombia 
[ 313 ] 
