MSU

LIBRARIES
4—!

 

 

 

BEIURNING MATERIALS:
P1ace in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped beIow.

 

 

"Mm—T

 

 

 

 

 

 

 

 

STRATIGRAPHIC PALYNOLOGY, VEGETATION DYNAMICS AND PALEOECOLOGY

OF THE FLORISSANT LAKE BEDS (OLIGOCENE), COLORADO

BY
Allan P. Hascall

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Geological Sciences

1988

ABSTRACT

STRATIGRAPHIC PALYNOLOGY. VEGETATION DYNAMICS AND PALEOECOLOGY
OF THE FLORISSANT LAKE BEDS (OLIGOCENE). COLORADO

BY
Allan P. Hascall

Palynological samples were analyzed from exposures of the Florissant
Lake Beds (Oligocene). Colorado. 78 palynomorph types are described. five
new to the flora. Most samples are dominated by pollen derived from plants
growing close to the lake shore. an observation in agreement with available
megafossil data.

291311192:th pollen appears to provide an indicator of relative water
depth. Its presence in the Lower Tuff suggests that this unit was deposited
after initial lake formation. Although the lakebed sequence represents basin
infilling by ash from proximal volcanic centers. there is no evidence for
significant disturbance of terrestrial communities. Spring turnover is
proposed to account for the alternating diatomite/sapropel couplets in the
laminated lake shales. Palynologic evidence suggests that the lake may have
dried at least once late in its history. Accumulation of ash finally ceased and

a period of erosion occurred prior to the deposition of the Upper Tuff.

ACKNOWLEDGMENTS

Professors Aureal T. Cross and Ralph E. Taggart have provided the
essential guidance to insure that this paper is scientifically accurate and
complete. Bach demonstrated a sincere personal concern for my academic
success and personal welfare. They exceeded the normal expectations of
thesis advisors and will always be my friends.

I am indebted to my father. Perry. my mother. Virginia. and my wife.
Nancy. for their unwavering encouragement. patience. and financial
assistance throughout this project. I appreciate their willingness to let my
work often inf ringe on their lives.

The National Park Service permitted me to excavate and take collections
from protected exposures at the Florissant Fossil Beds National Monument.
The assistance of the staff and provision of shelter on the property during
field study was much appreciated.

I would like to thank Professor Robert L. Anstey for participating on the

thesis guidance committee and preparing me for scientific research in the

library.

iii

TABLE OF CONTENTS

Page
LIST OF TABLES .............................................................................................. v
LIST OF FIGURES ............................................................................................ vi
LIST OF PLATES ............................................................................................. vii
INTRODUCTION ............................................................................................... l
GEOLOGY ............................................................................................................ 9
METHODS ........................................................................................................... 2 l
SYSTEMATICS .................................................................................................. 30
FLORAL COMPOSITION ................................................................................ 36
PALEOASSOCIATIONS .................................................................................. 47
STRATIGRAPHIC DISTRIBUTION OF PALYNOMORPHS ................... S4
PHYTOGEOGRAPHY/VEGETATION DYNAMICS .................................... 62
PERSPECTIVES ON REGIONAL HISTORY ................................................ 74
CONCLUSIONS ................................................................................................... 80
DESCRIPTIVE NOTES ..................................................................................... 84
REFERENCES CITED ......................................................................................... 134
PLATES ...... . ......................................................................................................... 142

iv

LIST OF TABLES

Page
Table 1. Known composition of the Florissant flora ...................... 37
Table 2. Fossil representation of the Florissant flora ................... 44
Table 3. Paleoassociations of plant taxa represented by
pollen and spores from Florissant ............................... 48

LIST OF FIGURES

Page
Figure 1. Map of Florissant area, Colorado ........................................ 2
Figure 2. Section 8/5/84-1 (taxa) ......................................................... 56
Figure 3. Section 8/23/84-1 (taxa) ...................................................... 57
Figure 4. Section 8/ 5/84-1 (paleoassociations) .............................. 60
Figure 5. Section 8/23/84-1 (paleoassociations) ........................... 61

vi

LIST OF PLATES

Page
PLATE 1. Illustrations I through 12 .................................................. 142
PLATE 2. Illustrations 1 through 16 .................................................. 144
PLATE 3. Illustrations 1 through 20 .................................................. 146
PLATE 4. Illustrations 1 through 33 .................................................. I48

vii

INTRODUCTION

A complex series of laminated lacustrine shales and associated deposits
near Florissant. Colorado. have provided insight into the mid-Tertiary
geologic and biotic history of a local intermountain area of the central Rocky
Mountains. The Florissant deposits are located approximately 65 km west of
Colorado Springs (Figure l). The Iakebeds are widely known for the
abundance and delicate preservation of diverse plant and insect fossils.
permineralized wood, and fossil vertebrates. Volcanic brecc1as and andesitic
and rhyolitic tuffs are situated above and below the lacustrine shales. A
portion of the area of distribution of the deposits was acquired by the
federal government in 1969 and designated the Florissant Fossil Beds
National Monument and placed under the auspices of the National Park
Service. The monument was established "to preserve and interpret for the
benefit and enjoyment of present and future generations the excellently
preserved insect and leaf fossils and related geologic sites and objects at the
Florissant lakebeds" (Root. 1983).

These deposits. of Oligocene age. are currently situated at an average
elevation of ”2530 m above sea level near 39' N.. 105' W. The region is
included in an extended eastward arm of the Thirtynine Mile volcanic field.
The Front Range rises from the High Plains 65 km to the east. dominated by
Pikes Peak (el "4300 m. 24 km east). The Florissant basin is arcuate in
shape. '16 km long NW-SE and "3-4 km wide. Topographic relief is
relatively moderate as compared to surrounding mountainous terrain: the

rolling hills within the formation vary in elevation by ” 150 m.

 

 

ANA " 53
. ,..o It "s ;' Paved highway

. of" m“ 3.16;; trail
~ Tozns

e I
..: 3 "
g. 8'. ‘, Ranch buildings
\ . U I bu.-."-
_ Latte George glg Rivers, streams

  
    

CRETE:
Generalised outline ot
the Florissant Laue Beds

IF”;

 

 

 

 

 

 

as m
satessee

 

 

  

 

 

UV.”

 

 

 

 

 

 

 

 

 

FIGURE 1. Map of Florissant area. Colorado
adapted from McLerey end Anderson. 1966. p. 606.
I - section 8/5/84 I 2 - section 8123/84 I

HISTORY OF RESEARCH

Excellent preservation and intricate detail of the fossil plants. insects and
other invertebrates and the geology of these lakebed deposits have attracted
paleontologists. sedimentologists. and fossil collectors for more than a
century. Extensive collections. both amateur and professional. have been
made over the years.

Lesquereux (1872) provided the first known published account of the
formation and its fossil flora which initiated scientific interest in the fossils
of the. deposits and contributed substantially to the identification of the plant
material. Between 1872 and 1890. numerous studies were conducted by the
United States Geological and Geographical (Hayden) Survey of the Territm'ies:
the most comprehensive contributions are Peale (1874). Cope (1875).
Scudder (1882). Lesquereux (1883) and Wadsworth (1883). f

The fossil insects of the Florissant deposits have been a particular focus
of study. Between 1876 and 1900. Scudder described the insect and plant
taxa in more than 25 publications. Cockerell produced more than 100 papers
on the fossil insects between 1906 and 1941. Brown (1986) has been
continuing work on insects.

F. H.1(nowlton of the United States Geological Survey published 10
studies devoted to the Florissant plant fossils between 1898 and 1926 and.
in 1916. described 121 types held by the United States National Museum
(Knowlton. 1916). By the mid-1930‘s. collections had become extensive but
were scattered in museums and other repositories throughout the United
States. Need for a systematic revision of this flora was apparent because
more than 250 plant species were recorded, many only tentatively. H. D.
MacGinitie began an investigation at this time with an interest in integrating.

revising. and interpreting the available materials. as well as making

4

additional collections. His publication (1953) is the most comprehensive
study of the fossil plants and regional paleoecology to date.

A number of studies have documented the geology of the Florissant
deposits. A petrographic study of several thin sections of the lake shales
was conducted by Wadsworth (1883). He concluded that the lacustrine
deposits resulted from wave redistribution of ash falls and mudf lows. Cross
(1894) named and defined the formation and published the first United
States Geological Survey geologic map of the region. The lakebed laminations
represent a period of pulsating volcanism in the region during the Oligocene.
according to McLeroy and Anderson (1966). who investigated the lithology
and texture of the volcanic sediments as well as their varved interlaminae of
biogenic origin. The study focused on the lake chemistry. seasonal variations
in deposition. regional climate and total elapsed time of deposition of the
lakebed sequence. Epis and Chapin (1974) published a study of the
stratigraphy of the Thirtynine Mile volcanic field which includes the
Florissant Lake Beds. Their radiometric analysis dates the Florissant
deposits at "34 Ma. A 15-minute United States Geological Survey geologic
map which included the entire area of distribution of the Florissant Lake
Beds was prepared by Wobus and Epis (1978). In a stratigraphic survey for
the National Park Service. Root (1983) refined the delineation and
description of units which comprise the lakebeds and differentiated them on
a geologic map of the National Monument.

Peale (1874) briefly described the regional physiography and
sedimentology and collaborated with Lesquereux to estimate the age of the
lakebed deposits as upper Tertiary. possibly Pliocene. Study of the fossil
fishes was initiated by Cope (1875. 1879) who suggested an age of late

Eocene or early Miocene (prior to recognition of the Oligocene). Cockerell

5

(1906) and Knowlton (1916) subsequently agreed with Lesquereux's initial
estimate of late Miocene based on comparisons with other fossil f loras of
North America and Europe. A single mammal. (Mathew sp.. a marsupial)
recovered from the lake shales. was described by Gazin (1935). who also
listed 9 previously described species of fish and 3 species of birds. lie
suggested an early Tertiary age since marsupials were not known in the
Tertiary of North America after John Day time (late Oligocene-early
Miocene). More recently. MacGinitie (1953) has supported an early-to-
middle Oligocene age which is consistent with Epis and Chapin's (1974)
radiometric age of "34 Ma.

Two studies of the palynology of the Florissant sediments have provided
preliminary surveys of the pollen flora. W. S. Ting examined six samples
from the varve couplet interlaminae (three of the diatomite and three of the
sapropel) for the McLeroy and Anderson (1966) study. These data were
used to confirm the seasonality of the varves as postulated by McLeroy and
Anderson. The only known illustration with identification of the pollen and
spores from the Florissant Formation was prepared by E. B. Leopold (1969 a).
Another pollen study. of importance to analysis of the Florissant flora. is the
recent discovery of pollen in a fossil flower of the extinct Eampsis longifglia

(Manchester and Crane. 1983). a taxon prominent in the megaflora.

ADVANTAGES AND LIMITATIONS OF PALYNOLOGICAL ANALYSIS

Current paleoecological interpretations of the Florissant sediments are
based primarily on analysis of plant megafossils. mostly leaf compressions.
Palynological analysis of the sediments has the potential to provide
additional information to supplement that available from previous

megafossil studies. The pollen representation of a sample must be

6

interpreted in light of several major criteria which are differentially
expressed in the various fossil plant taxa: volume and timing of pollen
production; pollen dispersal mechanisms and transportation to depositional
site; resistance of the wall to natural and laboratory deterioration; precision
of assignment to modern plant taxa: and inferred ecological/geographical
relationships. These criteria are directly and indirectly addressed
throughout this paper.

A study of the stratigraphic distribution of pollen and spores in the
Floristh deposits provides a valuable new dimension for the analysis of the
composition of the flora and the structure and dynamics of the source plant
communities. Any study of fossil plant materials must necessarily deal with
parts of plants which represent samples of the regional botanical inventory.
Relative to megafossils. particularly leaves. pollen and spores typically
cannot be identified with equivalent precision. Under normal conditions. it is
possible to document morphotypes within extant or form genera but in the
case of some palynomorphs. delineation of morphotypes within a complex of
genera or even a group of families is the best that can be expected. In the
case of angiospermous pollen. there is also a distinct bias favoring
representation of anemophilous forms at the expense of entomophilous
forms.

Despite the limitations noted above. palynological data do have several
advantages:

(1) Pollen grains and spores are dispersed in much greater numbers
than larger plant parts such as leaves.

(2) Palynomorphs tend to be dispersed over greater distances than
larger parts. thus improving the geographic sampling of the source

flora.

.7

(3) Sufficient numbers of palynomorphs are typically recovered over a
much wider stratigraphic range and from more diverse
sedimentary environments than plant megafossils.

Fossil pollen grains and spores are subject to representational biases as
are any other fossil entities. These biases include the relative production of
material in the source flora. the relative importance of taxa in specific
communities. the distribution of plant communities. the variations in
transport vectors. depositional biases (variable taphonomy). and post-
depositional factors. A complete discussion of the interaction of these factors
is beyond the scope of this study although specific considerations will be

introduced where they seem to be particularly significant.

OBJECTIVES OF RESEARCH
This palynologic study of the Florissant sediments utilized a stratigraphic

approach. It involved close sampling through the stratigraphic sections
studied. Three objectives or areas of research were planned:

(1) Inventory of floral composition represented by pollen and spores.

(2) Analysis of regional paleoecology (vegetation structure.

paleoclimate. paleogeography).
(3) Interpretation of phytogeography and vegetation dynamics

(disturbance effects on plant succession).

Floral Composition:
Given the preservational and transportational advantages of pollen and
. spores over leaves. it was expected that additional taxa would be recognized

in the current study. MacGinitie's (1953) systematic list had been extended

8

prior to this study by Ting (in McLeroy and Anderson. 1966) and Leopold
(1969 a) by pollen representing additional plants.

Regional Paleoecology:

The composite floral list of megafossils and palynomorphs represents a
number of distinct source vegetation communities that intergraded
geographically. A uniformitarian approach can be used to subdivide this list
into ecologically-distinct source plant assemblages. suggesting a regional
mosaic of plant communities that might contribute to both paleoclimatic and

paleogeographic reconstruction.

Phytogeography/ Vegetation Dynamics:

Comparison of geographically separated florules can increase
understanding of the distribution of plant communities within the region.
An advantage of stratigraphic palynology over megafossil studies is that
productive pollen/spore samples are more common than productive
megafossil layers in most stratigraphic sequences. making it easier to
deter mine changes in the source vegetation over time. Floral assemblage
changes throughout the stratigraphic column provide the opportunity to look
for patterns of succession and shifts in dominance in the communities

through the time interval represented by the Florissant deposits.

GEOLOGY

Detailed descriptions of the stratigraphic units of the Florissant deposits
have been included in several studies (MacGinitie, 1953; McLeroy and
Anderson. 1966; Root. 1983). Description of the internal characteristics of
each unit has been quite consistent. However, there are differences of
opinion on the stratigraphic position of adjacent beds. Recent studies (Epis
and Chapin. 1974; Root. 1983) have provided additional information for the

interpretation of stratigraphic relationships.

Earlier Deposits

Currently. the Florissant Lake Beds are known to overlie three earlier
rock units at different places. In most areas. the Florissant deposits lie,
unconformably on the irregular erosion surface of the Precambrian Pikes
Peak Batholith (Root. 1983). This extensive crystalline basement is
composed of pink-to—red, medium- to coarse-grained. biotite-hornblende
granite.

Two depositional sequences of inter mediate age have been identified as
sporadically occupying the position between the Pikes Peak granite and the
Florissant Lake Beds (Root. 1983). The Eocene Echo Park Alluvium has been
identified in two small patches on the southeastern section of the Monument.
The Wall Mountain Tuff (early Oligocene) forms scattered outcrops
throughout the valley margins and marks the onset of the mid-Tertiary
volcanism which dominated the region and subsequently provided the
sediments of the lake beds. Epis and Chapin (1974) dated the tqu at 34 and
36 Ma using the potassium-argon method. These rhyolitic erosional
remnants were once part of a continuous 9,100 km2 ash deposit of the

Thirtynine Mile volcanic field.

10

Florissant Lake Beds

MacGinitie (1953) and Root (1983) recognize four stratigraphic units in
the Florissant Lake Beds. They include: a basal arkosic breccia. a Lower
Tuff. the lacustrine shales. and an Upper Tuff. The term "lakebeds" is used
throughout this paper to represent the Lower Tuff and the Lake Shales
together.

Basal Arkosic Breccia

This unit was first described by MacGinitie (1953) and has been
recognized by Root (1983) in two isolated localities in the east-central
portion of the Monument. At the exposures of the lower contact. the basal
arkosic breccia lies unconformably on the Pikes Peak Granite. or. on the Wall
Mountain Tuff. where erosional remnants of that formation are still present.
The basal arkosic breccia is made up of homogeneously distributed angular
to subangular fragments of partially weathered Pikes Peak granite that
range in size from sand to gravel (up to 2.5 cm in greatest dimension). The
fragments have been incorporated into a fine-grained. greenish tuff matrix
forming a well-indurated rock. MacGinitie (1953) noted that hand
specimens are difficult to distinguish from the f iner-grained facies of the
Lower Oligocene Castle Rock Formation located near Colorado Springs. None
of the sections in this palynological study included this basal unit.

Lower Tuff

The Lower Tuff. a series of andesitic beds. is in abrupt contact with the
arkosic breccia below (MacGinitie. 1953). More than 11 m of this unit is
exposed in one of the sections selected for the present study (8/5/84-1). but

its lower contact was not observed because the slope terminates in the wide.

Il

nearly horizontal. valley floor. Considerable lateral variation in exposed
thickness of this unit exists from outcrop to outcrop with a maximum of
‘15 m (MacGinitie.1953).

The Lower Tuff is divided into two units. The lower half is a complex of
alternating fine and coarse beds with gravel lenses composed of andesite
cobbles. basaltic lava fragments and partially-weathered granitic material.
Cross-bedding is frequently apparent. Hand specimens are predominantly
grey in color and irregularly iron—stained throughout. Fossil plant fragments
are abundant. but none is of satisfactory size and quality for identification.
A f ining-upward trend was observed within some units (up to 0.5 m thick)
and throughout the lower sequence. Blocky surf ace weathering and north-
south trending vertical fractures are visible in outcrop. The upper half is
more uniform and massively bedded. Lithologic breaks for sampling were
thus selected arbitrarily at “0.3 m intervals. The consistently brown matrix
is fine-grained. well-indurated and fractures conchoidally. McLeroy and
Anderson (1966) described the composition as predominantly glass.
plagioclase. and clinopyroxene minerals. Fossil tree stumps of putatiVe
Sequoia spp. and several species of hardwoods are preserved in the upper
part of the Lower Tuff. many in their original upright positions. They vary
from ”0.5-4.5 m in diameter.

Size sorting and cross-bedding of the Lower Tuff sediments indicates
fluvial transport and deposition. However. MacGinitie (1953) attributes
these deposits to mudf lows which followed river channels and spread out
over the basin. Several sources suggest that deposition of this unit may have
ended coincident with the obstruction of the basin drainage system by a lava
flow or lahar. choking of the valley with ash f ill. or by faulting at its

southern end. producing an extensive ponded area known as Lake Florissant

12

(McLeroy and Anderson. 1966; Root, 1983: Epis and Chapin. 1974). This
postulation may be supported by the abrupt lithologic transition of the
Lower Tuff unit to the lacustrine shale sequence. above which indicates
sudden alteration of depositional conditions.

There is palynological evidence for standing water in the basin during
deposition of the upper part of the Lower Tuff, indicated by strong
representation of Eotamggetgn pollen in several of the samples studied. The
sediment-laden stream(s) may have been ponded at one or more points as it
meandered across the broad valley. Alternatively. it is possible that the
onset of Lower Tuff deposition was coeval with the obstruction feature to
basin drainage since both required an episode(s) of relatively energetic

volcanism. The lake would then have formed during Lower Tuff time.

Lacustrine Shales .

This highly variable sequence of laminated shales contains the plant and
insect compressions that are notable in the Florissant Lake Beds. In the two
sections of this study. total thickness of the lake shales is "4.5 m and “5.8 m.
Root (1983) identified the total thickness to vary between " 1.5 m and " 10.4
m. McLeroy and Anderson's (1966) petrographic study of the Florissant
deposits describes four fundamental types of laminae. The volcanigenic
sediments are divided into pumice laminae and graded tuff laminae: the
biogenic deposits are diatomite and sapropel couplets. In all cases.
transitions between laminations are abrupt with no transitional lithotypes.

The pumiceous laminae are the most conspicuous of the lake deposits.

. They appear yellow in outcrop and vary in thickness from several
millimeters to several centimeters. averaging ”1.5 cm. Petrographic analysis

(McLeroy and Anderson, 1966) indicates that composition is primarily

l3

vesicular. rhyodacitic. glass-froth pebbles averaging 3 mm in diameter. with
occasional well-rounded andesitic and trachytic pebbles. Inverse graded
bedding is evident when pebbles can be differentiated. The lower bedding
plane surfaces are relatively smooth as opposed to upper surfaces which are
irregular due to the concentration of the coarser pebbles on top. No leaf or
insect compressions have been observed in the pumice laminae.

In contrast with the pumice laminae. the graded tuff laminae are gray in
outcrop and contain an abundance of leaf and insect compressions. They
grade from a concentration of subrounded to subangular quartz and
ferromagnesian minerals at the base to an extremely fine-grained vitric ash
at the top (McLeroy and Anderson. 1966). The rhyodacitic ash comprises
90% of the volume of these layers. Vertical distribution is erratic and
roughly alternating with the pumice. Like the pumice. the graded tqu is
equally variable in thickness. averaging “1 cm. .

McLeroy and Anderson (1966) postulate distinct depositional
mechanisms to account for the two types of volcanic lamination and their
unique physical characteristics. These depositional mechanisms have
winter-spring flooding in common as the underlying cause. The inverse
graded bedding and differences in regularity of the upper and lower bedding
surfaces of the pumiceous sediments are attributed to delayed deposition of
the larger pebbles. Because of their vesicular nature. the larger pebbles
were temporarily buoyant upon introduction to the drainage basin.
Subsequently. they were deposited over the finer sediments after becoming
waterlogged. In contrast. the graded tuff laminae always fine upward
because the larger grains lacked vesicles and differentially settled before the

ash.

14

The diatomite and sapropel laminations occur as couplets averaging
”1.0 mm in thickness (McLeroy and Anderson. 1966). Each lamination varies
in thickness from "0.04-22 mm. The diatomite is white and composed of
95% opal; a fine silt of orthoclase. plagioclase and quartz contributes 5%. The
presence of diatom frustrules is evident at a magnification of 1000 x. In
sharp contrast. the sapropel is primarily composed of unidentifiable black
organic matter. Up to 50% of the remaining components is a mixture of
sponge spicules. pollen. diatoms, and silt particles of quartz. plagioclase and
biotite. Limonite and hematite from altered pyrite are common in outcrop
samples.

McLeroy and Anderson (1966) suggest the biogenic couplets represent
seasonal deposition on the lake bottom. They reason that the diatomite
resulted from accumulation after a diatom bloom during spring floods.
Diatom growth was favorable at this time because of the nutrient
replenishment provided by winter and spring rainfall and subsequent runoff
into the ponded basin. Illumination. abundance of dissolved silica. water
chemistry and water temperature were also especially suitable for diatom
growth during the spring. Once the diatoms depleted the nutrient content in
the late spring and the waters had warmed, summer and autumn plankton
flourished under the new conditions. The deposition of these organisms
resulted in a dark organic layer immediately over the diatomaceous
accumulation; therefore. these diatomite/sapropel pairs are varve couplets.

Pollen analysis has been used to substantiate the spring flooding
depositional model (W. S. Ting. in McLeroy and Anderson. 1966). Pollen is
absent in the diatomite but abundant in the sapropel. McLeroy and

Anderson (1966) suggest that this indicates very rapid deposition of the

15

diatoms in the early spring, followed by relatively slow deposition of the
sapropelic mud in the summer and fall months.

If significant flooding did occur in the spring. reworked sediments would
have been introduced into the lake. and turbidation would have disturbed
the fine sediment layering on the bottom. Neither of these is present in the
laminated shales examined in this study. In addition. absence of pollen in
the diatomite layers may be preservational. and many diatomite deposits are
barren of pollen through extensive exposures of section (Prof. R. E. Taggart.
pers. comm.)

An alternative explanation for the origin of the spring diatom maximum
is overturn of the lake waters during seasonal temperature cycles.
Temperate lakes are commonly thermally stratified in the summer and
winter. A ther mocline during the summer acts as a barrier to the mixing of
warm surface and cool bottom waters due to vertical density differences.
Atmospheric warming in the spring causes thermal destratification in lakes.
allowing surface and bottom waters to mix freely under the influence of
tributary stream or wind-produced currents accompanying winter-spring
storms. Such mixing would permit the temporary nutrient enrichment of
surface waters that would promote extensive phytoplankton growth. During
the warm summer months when thermal stratification would be restored.
water below the thermocline would become essentially anoxic and sapropelic
debris would thus be protected from degradation. Seasonal turnover would
be possible under MacGinitie's estimated average annual range of
temperature (”20°F absolute minimum: warmest month “80'F) for a warm-
temperate Florissant paleoclimate. Breakdown of thermal stratification with
subsequent turnover would occur twice a year--spring and fall.

Phytoplankton blooms are virtually universal in lakes during turnover,

16

particularly in the spring. and this is a likely source for the diatomite (Prof.
.R. E. Taggart. pers. com m.).

The majority of the lacustrine shale sequences are dominated by
alternation of the pumice laminae and the graded tuff laminae. with the
biogenic varve couplets as a minor constituent. Two unique bedding types
were observed in the measured sections of this study which may not be
defined under the four fundamental types of lamination: calcareous layers
and homogeneous mudstones.

Previous studies have indicated an absence of calcium carbonate in the
lacustrine sediments (e.g., McLeroy and Anderson. 1966). Several calcareous
layers were found in disjunct sections of this study. They vary from ~1.5-5
cm in thickness. In all cases. these deposits occurred within 2 m of the
Upper Tuff. Thin section analysis of one sample shows vertical fibrous
crystals of calcite, the mineralogy supported by x-ray analysis (Prof. D. F.
Sibley. pers. comm.) Occasional mud-silt laminae temporarily interrupt the
crystal growth. There exists marked lateral variability in thickness and
compositional layering in these units. even within a hand specimen. The
presence of carbonate deposits may be used to support MacGinitie's (I953)
suggestion of periods of aridity in the region which resulted in gypsum
layers. It is possible that they are caliche layers which formed 1—2 m below
the erosion surface at the Upper Tuff contact (Prof. A. T. Cross, pers. comm).

Thin laminae ~3 mm thick of interlocking laths of gypsum were found
intercalated between volcanic laminae and as f racture-f ill in one section.
This gypsum is most likely secondary. Gypsum layers as thick as 2.5 cm
have been described in previous studies (MacGinitie. 1953; Root. 1983). but

none was recognized in the sections of this study.

I7

Layers of indurated volcanic ash occupy positions between the pumice
and tuff laminae concentrated in the upper parts of the measured sections.
These layers are highly variable in thickness; most ranging from several
millimeters to several centimeters. with a single occurrence of more than a
meter. They are extremely homogeneous in texture and lack f issility.
Larger sediment particles and plant fragments are absent. All are light gray
in color. well-sorted and fracture conchoidally. Faint transitions in color of
the thick ash unit define the planes of deposition and very fine grading is
occasionally present. Cross-bedding. turbidity and other primary
sedimentary structures are not evident. Given these characteristics, these
massive mudstone layers are considered to be direct airf all ash deposits

produced by volcanic eruption.

Upper Tuff

At most lakebed exposures. a highly—indurated breccia lies
unconformably on the lake deposits. It is composed of angular-to-
subangular fragments of granite. basalt. andesite. and lake shales. The clasts
range in size-from sand to boulders and no sorting is evident. A light gray to
white rhyolitic matrix cements the highly heterogeneous assemblage into a
very hard mass up to ”7.6 m thick. This caprock is largely responsible for
preservation of the lake shales which are otherwise highly susceptible to
erosion upon exposure. Presence of this unit usually indicates the presence
of underlying lake shales. Due to its internal characteristics and extensive
lateral distribution over most of the basin. the suggestion of a lahar or series
. of lahars following the lake phase is feasible (Root. 1983).

18

Later Deposits

One or more of several deposits may overlie the Florissant Lake Beds.
MacGinitie (1953) discussed two such deposits under the category ”Later
Eruptive Rocks". One of these is the continued deposition of the Thirtynine
Mile volcanic series that. where present. increases in thickness toward the
southern end of the basin and southwestward toward the Guffey volcanic
center. MacGinitie (1953) refers to these mudf low deposits as basic
(alkaline) agglomerates. The second series. trachytic lava flows. also
originating from the southwest after a period of erosion. followed river
channels to the basin where they now overlie the Upper Tuff or basic
agglomerates in an irregular pattern. A small patch of one of these lava
flows was observed above the Upper Tuff at the southernmost collection
locality (8/23/84-1) of this study. _

Following an extended period of erosion, Quaternary and Recent
alluvium and colluvium formed a surf icial cover above the Florissant Lake
Beds outcrops and throughout the valley floodplain (Root. 1983). These
deposits consist of up to ”3.7 m of redistributed clay. mud. sand and gravel.
A layer of unconsolidated granitic grus. a partially-weathered gravelly
product of eroded Pikes Peak granite. frequently overlies the Upper Tuff

unit.

Structure

Field relations of the basin deposits and contiguous rocks suggest that
the slopes of the current Florissant valley region can be used to approximate
the original outline of Lake Florissant. The highest extent of the paleo-water
level and original accumulation of lakebeds is unknown because an

undetermined amount of erosion occurred prior to deposition of the Upper

l9

Tuff. MacGinitie (1953) and McLeroy and Anderson (1966) suggest that the
current distribution of the Florissant Lake Beds film is in part a
consequence of post-depositional faulting of the region and that the outline
of the outcrops probably does not accurately represent the ancient lake
margin. Tilting of the sections of this study appears to be minor: no
consistent strike directions were observed and dips ranged from 2" to 8°
(av ‘4'). There is not substantial evidence for significant change in
Florissant basin morphology since the early Oligocene.

A number of depositional and post-depositional factors preclude
lithologic correlation of disjunct outcrops. The nature of lake sedimentation
was variable. While the sections studied had a common upper stratigraphic
reference point. the beds in the 2.4 m immediately below each differed
substantially with respect to thickness of the component lithologies. .
Considerable lateral change in thickness of layers and even pinching out of
some layers was generally evident within the 0.5 m wide collection trenches
and even within hand specimens. Wind. water or mass movement
transported the volcaniclastic material to the basin from different sources
and directions. Deposition of the fluvial sediments in the form of deltas was
probably greatest near the lake margins at the intersection of tributary
streams. This variability and its stratigraphic implications have been
discussed in previous studies (MacGinitie. 1953; Root. 1983). It is possible
that the lake may not have been a continuous entity. but rather a series of
ponded areas which varied in position and time in response to sedimentation
and evaporation (MacGinitie. 1953).

Outcropping of the deposits indicates two depositional settings. Most of
the deposits appear to represent regions of the lake relatively close to the

paleo-shoreline. which may be extrapolated into the basin slopes. Lakebed

20

deposits which completely encircle granitic cored hills suggest the presence
of islands during the lake phase (Scudder, 1881). Following deposition and
prior to lithification. several factors (subsidence. soft-sediment deformation.
compaction) caused distortion of the original bedding and caused local
folding and faulting of the strata with displacements as much as 3-4.5 m
(Root. 1983).

The first post—depositional regional faulting probably occurred prior to
the close of the Oligocene. The basic agglomerates of the Thirtynine Mile
volcanic series which immediately overlie the Florissant Lake Beds. at the
southern end of the basin. possess the same structural trends as the lakebed
units (MacGinitie. 1953). They are considered to be early Oligocene in age.
These structures are absent in the trachytic lava flows of the late Oligocene.

Subsequent folding and faulting of the sediments due to regional
tectonism during the later Tertiary has produced two complex synclinal
structures in which the lake basin sediments are preserved (MacGinitie.
1953). Regional tilting probably reversed the direction of flow of the
drainage from southeast to northwest. the direction of flow of Grape Creek
today (Scudder. 1882).

METHODS OF STUDY
Field Collection

Collection of samples for palynological analysis was made during August.
1984. Two localities were selected; both are located on the Lake George.
Colorado. U. S. G. S. 7.5" series topographic quadrangle map and are plotted

on the regional map in Figure l:

8/5/84-1

CTR. Wl/2. NWl/4. SW1/4. SEC. 13. T. 13 S.. R. 71 W.. Teller County.
Colorado.

north-northeast side of “Scudder Hill":

section thickness - 16.8 m

87 samples for palynological analysis

8/23/84-1

NWl/4. SWl/4. SWl/4. SEC. 19. T. 13 S.. R. 70 W.. Teller County. Colorado.

"150-200 in south of Lower Twin Rock Road along northwest-southeast
trending outcrop; ‘

section thickness - 3.8 m

71 samples for palynological analysis

Locality 8/5/84-1 was selected because it was the site of the Scudder
insect collections and because it is a relatively thick sequence. The upper
third of this unit had been previously excavated in 1877 by S. H. Scudder for
the collection of insect and plant megafossils. Description of the stratigraphy
of the deposits in his (1882) paper was used for comparison with
observations in this study. The exposure is situated on the north side of an
isolated hill in the interior of the Florissant basin and is commonly referred

to by monument staff as the “Scudder site". Since laminated lakebed strata

21

22

completely surround this hill. which consists of a granite core that crops out
on the top of the hill. it may represent a former island.

Locality 8/23/84-1 was located “2.6 km south-southeast of locality
8/ 5/ 84-1 . Field relations suggest that it represents deposition near the
eastern paleo-shoreline of the lake. This section was selected primarily for
lithological and palynological comparison with the Scudder site to examine
the geographic variability in sediment deposition and palynological
assemblages.

These two sections have a common reference point in the form of an
unconformable contact between the Upper Tuff and the lacustrine shales.
Total thicknesses of the two sections are: 8/5/84-1. 16.8 m: 8/23/84-1.

3.8 m. In section 8/5/84-1. the thicknesses of the lakebed sequence and the
exposed Lower Tuff are ”5.8 m and "11 m. respectively. Near surface
distortion of the lakebeds in contact with the Lower Tuff was encountered.
apparently caused by gravitational displacement of spoils left behind during
Scudder's collections further upslope. The entire 3.8 m of section 8/23/84-1
are lakebeds; the Lower Tuff beds are absent because the base of the
lakebeds is concealed below by the alluvium/colluvium on the valley floor.
Accounting for the unobservable Lower-Tuff contact. extended lakebed
thickness in this section is estimated at “4.5 m by measured thickness of the
upper and lower bounded lakebed sequence at a nearby outcr0p ~100 m
north. The thickness discrepancy of 1.3 in between the two collection
localities is reasonable in light of the effects of significant lateral variation in
sedimentation and erosion.

The slopes of the eroded exposures average ”18'. with a maximum of 25°
near the top flattening to 10' near the base. Weathering of outcrops is rapid

and the shales are generally masked by a layer of loose material. making it

23

necessary to excavate the selected exposures in order to minimize the effects
of weathering on the palynological samples. Thickness of the loose material
and soil averages 3-6 cm on these hillsides. Water has penetrated the
fissures and bedding planes to more than a meter below the surface. A
trench was excavated with mattock and shovel to a depth of ”0.4-0.7 m.
Since the beds are approximately horizontal in aspect. a terraced approach to
excavation was employed because it was least destructive to the lakebed
deposits and was convenient for sample collection. The lower end of each
trench was commenced at the base of the exposed bedrock above the
alluvium/colluvium of the valley; the upper part was terminated at the
lakebed-Upper Tuff contact in each section. After sampling. each trench was
backfilled and restored to contour.

Following excavation. each trench was measured for thickness from base
to top with a 10 X hand level and engineer's rule used as a rod. Vertical
control for sampling was maintained by reference stakes placed at regular
intervals.

Samples were collected continuously from base to top of each section.
Sample thickness was governed in part by lithologic breaks in the sequence.
Single lithological units greater than 0.3 m in thickness were usually divided
into 0.3 in sampling intervals. Material for each sample was collected
throughout the thickness of the sample unit and across the width of the
trench. A small plastic bag was filled for each sampling unit. In addition.
hand specimens of unusual lithologic units were collected intermittently for
possible petrographic analysis in the laboratory. The thickness of sample
units ranged from 3 mm to 50 cm . averaging “'5 cm in the lakebeds and ”29

cm in the tuff beds. A total of 158 samples were acquired. Each sample was

24

assigned a number (number 1 at base of section) preceded by the section
number.

Since this is primarily a palynological study. megafossils were collected
only incidentally during sampling or reconnaissance work. Due to the nature
of excavation. the fossiliferous shales of the trenches were generally reduced

to small fragments and few megafossil specimens were recovered.

Laboratory Preparation

Given the limited scope of this project. 158 samples were more than
could be adequately prepared and studied in detail. 69 selected samples
were prepared for palynological analysis based on several criteria: (a) even
distribution throughout the entire thicknesses of the stratigraphic columns
and minimizing the gaps between samples; (b) covering all lithologies; (c)
continuous representation of layers which may provide information about
disturbance and succession.

Prior to processing. selected samples were recorded in the laboratory
catalog and assigned a unique number preceded by 'Pb'. This maceration
number is distinct from the sample number assigned in the field.

The objective of chemical maceration was to retrieve the maximum
number of palynomorphs preserved in the rock through dissolution of the
mineral and organic matter which entombs them. It is desirable to remove
as much mineral and extraneous plant material as possible. but a level of
tolerance of such matter must be accepted so that loss of smaller
palynomorphs and destruction of relatively fragile ones are kept to a
minimum during chemical maceration of the rock. Therefore. the treatment

employed was the minimum necessary to free the palynomorphs.

25

The following physical and mechanical treatments were used for
macerating the samples. Due to the variability in lithology. organic
composition and grain size, the course of treatment differed considerably
from sample to sample. Steps were eliminated whenever possible and time
of each treatment was highly variable. Samples were often split during
processing to apply different subsequent treatments for maximum results.
Procedures were modified from Doher (1980 ). Funkhouser and Evitt (1959).
and techniques recommended by Professor A. T. Cross.

Well-consolidated sediment samples were washed in distilled water
prior to processing to remove extraneous material. These and other samples
were crushed in a mortar with a pestle until the largest fragments were
approximately 5 mm in diameter.

7 to 25 gm aliquots of these crushed samples were generally sufficient to
produce the desired quantities of pollen residues. Some processed samples
were barren. Laminated shales were the most productive and contained the
best-preserved grains. Coarse sandstones and massive mudstones. on the
other hand. required relatively large samples to obtain satisfactory residues.
Carbonates and finer grained pumices and sandstones were intermediate in
productivity and preservation. Unproductive samples were usually coarse-
grained or highly weathered.

Dilute hydrochloric acid (10% HCl) was applied to all crushed samples for
up to 1 hr to dissolve any carbonate sediments and cements that might be
present. Some carbonate sediments required 8 hr contact time. using 5% HCl
for the first 5 hr. Samples were thoroughly washed and centrifuged
between each chemical treatment. Following HCl. hydrofluoric acid (52% HF)
was added to dissolve the silicate minerals. Crushed ice was used to control

the temperature and speed of the initial reaction. Solution digestion usually

26

was completed in 3-5 hr. cm 12-24 hr treatment was necessary in some
cases. If solution of carbonates was initially incomplete. a white. granular
precipitate (Can) often formed during treatment with HF. If this occurred.
the sample was immersed in 20% HCl for up to 2 hours in a steam bath to
dissolve the precipitate.

Some samples contained varying amounts of coallfied organic material.
It was generally desirable to remove or disperse this. If the rock was n0t

significantly weathered. the residue was immersed in a strong oxidizing
agent (Schulze's solution: ratio 5 parts nitric acid (HNO3) to 1 part potassium

chlorate (KC103)) for 1-2 minutes to disaggregate it. Since Schulze's solution

is highly corrosive to palynomorphs. especially the thin-walled types. its use
was limited. In most Florissant samples. it was found that this treatment
Could be avoided. Following treatment with Schulze's solution or if organic
material was weathered (oxidized). 5% potassium hydroxide (KOH) was
applied for a few minutes to dissolve the humic residues.

After release from the organic and inorganic matrix. the palynomorphs
were physically separated from undissolved extraneous matter. Ultrasonic
agitation was occasionally employed to first loosen f locculates. Specific
gravity separation in zinc chloride solution (ZnClz. sp. gr. 1.95) was quite
successful in the separation of the mineral fraction. Alternatively.
differential settling in a wetting agent (Calgon. Multiturge‘) was sometimes
used. If clays were abundant. sieving through 7.10. or 15 micron fiber
screens was effective in their removal. In all cases where the residue was
sieved. it was necessary to check the 'waste' fraction to verify that small
pollen grains or spores had n0t passed through the sieves.

Residues were stained with Safranin 0. Some of the residue from each

sample processed was suspended in a drop of HEC (Hydroethylcellulose) and

27

evenly distributed over a 22 mm coverslip and allowed to dry. A consistent
distribution density of grains on these strewn slides was attempted to
facilitate qualitative and quantitative analysis. A drop of Clearmount resin
was applied to the cover slip, when dry. A 75 by 25 mm microscope slide
was then placed over the inverted coverslip and allowed to harden for two

weeks.

Examination ‘

The slides were studied using a Leitz Ortholux research microscope
(M.S.U. 66 2667) with a metric calibrated stage. The positions of particular
grains were recorded as horizontal and vertical coordinates. For translation
of these coordinates to other microscopes. the coordinates of the upper right
hand corner of the coverslip of each slide were recorded.

The first step involved scanning the slides randomly. observing and
sketching the different morphological types as they were encountered.
Subsequently. photographs of the (most representative specimens of each
type were taken using a Leitz Orthomat automatic microscope camera in
conjunction with the above microscope. Kodak Panatomic-X film (ASA 32)
produced a fine-grained image with good contrast. Most photographs were
taken with the 95 X (oil) microscope objective for maximum resolution of the
structural and sculptural details of the pollen walls. Photographs at two or
three planes of focus of a single grain were obtained for a three-dimensional
perspective in many cases. since any particular plane of focus provides
limited information about the whole specimen. A 54 X objective was used in
. photographing larger specimens. Photographs were printed on Kodak
Polycontrast 11 Rapid R C paper. All photographs were enlarged so that all

specimens on the final plates appear at a magnification of 666 X. Specimen

28

dimensions. in micrometers, can be calculated by multiplying plate
dimensions in millimeters by 1.5.

Specimens were grouped into differentiated morphotypes and each
group was given a unique artificial code name until tentative taxonomic
designations were assigned.

The procedures involved in systematic determinations are discussed in

the Systematics section.

Counting

All counting was made with a 54 X oil immersion objective and
microscope stage graduated in millimeters. The 22 by 22 mm coverslip of
each slide was divided into 92 horizontal traverses so that the sum of
vertically stacked fields of view provided complete coverage of thegcoverslip.

For the purposes of this type of study. the necessary quantitative
information could be retrieved from a selected number of traverses fewer
than the total number possible for any coverslip. assuming the sample was
"productive". Each traverse was assigned a reference number so that it could
be selected arbitrarily without duplication or overlap with other traverses.
Six initial non-random traverses were selected for each slide to insure that
the information acquired represented the complete area of the coverslip.
Therefore. due to migration of grains of different sizes and densities during
slide preparation. areal biases within the coverslip area were accounted for.
Subsequent traverses were selected randomly with the aid of a random
numbers table until the total predetermined palynomorph count was
acquired for each slide.

Sample size was determined by techniques following those used by

Thompson (1969) and Gies (1972). The count should be sufficient to give a

29

good approximation of the relative abundance of the complete diversity of
types present in the sample. Raref action curves were prepared for several
samples and all exhibited pronounced flattening beyond counts of 300
grains. 400 grains was selected as a fixed sum to be counted for each sample
of this study. In order to overcome additional distributional biases involved
in slide preparation, at least two slides were counted per sample and no
more than 200 grains were counted per slide. No more than 100 grains were
counted on any single traverse. Several samples lacked the fossil pollen
recovery necessary to acquire the total count of 400 grains. but were
included in the quantitative data if 200-300 specimens could be tallied.

In order to avoid size bias. only those pollen grains and spores with
geometric centers within the field of view during the traverses were
counted. Grains of uncertain affinity were tallied in the lowest taxonomic
rank that they could be assigned to confidently. or included in morphological
classes. For counting purposes. the diversity was intentionally inflated.
especially early in the examination process. As intra-generic (or intra-typic)
variability became better known. previously defined groups were sometimes
later combined. The initial 8 samples counted were re-counted to allow for

practice and development of confidence in the procedures.

SYSTEMATICS
Nomenclature

The nomenclature of Tertiary palynology is complicated by different
authors' use of two contrasting systems. Specimens may be assigned names
of extant plants in some cases. indicating relationship to modern taxa. or
they may be assigned to artificial groups indicating extinct taxa or uncertain
biological affinities. Some studies use both approaches to nomenclature.
leading to some confusion. Difficulty arises in the degree of morphological
similarity between fossil and modern palynomorphs. The resemblance of
Late Tertiary fossils to extant forms is often very close or identical.
However. many Early Tertiary and Late Cretaceous pollen morphotypes are
distinct from today's. The majority of palynomorphs in rocks of the
Oligocene epoch are morphologically similar to modern taxa.

The nomenclaturaI system to be utilized should be selected in light of the
goals of research and comparative utility to other studies. Conventionally.
MacGinitie (1953) based his taxonomic designations of leaves and other plant
organs on comparisons with modern herbarium material. Since it is a major
objective of this study to add light to the paleoecological setting of the
Florissant region at the time of deposition, nomenclaturelwill be
phylogenetically-based when the morphological evidence is substantial.

Some palynologists employ a purely artificial system of nomenclature in
Tertiary studies. It is entirely morphologically-based. and the names are
independent of any botanical relationships. Tschudy's (1957) prooosed
symbolic coding system is tidy, logical. and practical for stratigraphic
correlation studies and f acies comparison. Schopf (1964) supports this
informal system for such limited applications and adds that if identifications

are consistent. it may be the preferred starting point to taxonomic treatment.

30

31

Taxonomic classification requires exhaustive study of the literature and
critical comparisons of specimens. both fossil and living. to establish
potential phylogenetic relationships. In addition to distinctive features of
morphology and relating different organs to the same plant. assignment of
specimens to particular. taxa must be consistent with known geographical
and stratigraphic ranges of those taxa (Schopf. 1964). Phylogenetic
nomenclature is not confused or prejudiced by the purely artificial system
and translation from one system to the other is possible.

The rules for the formal treatment of naming the groups are guided by
the International Code of Botanical Nomenclature (ICBN). This system is in
part arbitrary and artificial because the evidence for phylogenetic
relationships is variable and incomplete (Schopf. 1969). "Since the context of
phyletic relationship is difficult to establish. the identification of a new taxon
is more readily justified than identification with a previously known species"
(Schopf. 1969). However. "there are a good many more names now in the
literature than the differentiation of material would actually demand"
(Schopf. 1964). Confidence in assignment of fossil specimens to modern taxa
decreases with antiquity of the associated source sediments because of the
variable rates of morphological and physiological change of different plant
parts associated with evolution. Nevertheless. in order to integrate the biotic
element in a paleoenvironmental interpretation of the source region. it is
necessary to assign fossil specimens to extant plant genera and species when
comparison of the available morphologic evidence strongly suggests a
relationship.

Many palynomorphs from mid-Tertiary or older sediments may be
assigned to extant families or higher taxonomic ranks. but supporting

evidence is generally lacking to assign them to extant genera. Form genera

32

and organ genera were created to avoid potential misalliances with eXtant
genera. Both have been accepted under the ICBN. but the organ genus
concept was removed in 1975 (Pocknall and Mildenhall. 1984).
Paleoecological application of these names has been practiced with flawed
logic (Tschudy. 1969). For example. a specimen may be placed under the
half -natural form genus Quencgidites, which implies affinity to the extant
genus Quench: based solely on morphological resemblance. As long as this
designation is restricted in its use to biostratigraphic zonation. as in Pocknall
and Mildenhall (1984). it may be useful. However. parallel ecological
inference based on relationship to the genus Quergts would be erroneous if
the author created the fossil genus out of uncertainty. Confusion could be
eliminated by assigning a purely morphologically-based generic name and
suggesting potential botanical affinities in the systematic description.

The phylogenetically-based nomenclature used in this study requires
identification of specimens to the lowest possible taxonomic rank that can be
determined with certainty. Accuracy of taxonomic assignment decreases
with age of specimens. Many angiosperm palynomorphs from Quaternary
sediments can be associated with modern plant species on the basis of
morphological criteria. Megafossil studies show that many modern genera
and families existed in the Early Tertiary. but there is little evidence that
extant species were established in the Paleogene. The Tertiary Period
represents a time of significant evolutionary change for the angiosper ms.
Grasses and forbs (e.g.. Compositae) are generally identified at the family
level. Since many modern genera. primarily forest trees. are well
. represented as megafossils during the Oligocene Epoch, it is reasonable and
paleoecologically essential to assign the associated palynomorphs to modern

genera where possible.

33

Megafossil specimens collected from the Florissant Lake Beds provide
supporting evidence for the taxonomic composition of the palynoflora. Leaf
compressions are especially useful because the complex morphological
characters preserved provide for identification of many fossils to the genus
or species level. MacGinitie's (1953) monographic treatment of the Florissant
plant megafossils provided a systematic list of l 13 plant species in 87
genera based mostly on leaf identifications. Most of these (99 species. 76
genera) are angiosper ms. Only one of the genera. Eaggpsjs, is known to be

extinct.

References

In this study. many references were used in the identification of fossil
specimens. It was useful to become familiar with pollen of the modern plant
equivalents to those included in MacGinitie's (1953) systematic list of
megafossils. The modern pollen reference collection in the paleobOtanical
laboratory at Michigan State University was an indispensable aid. Two
pioneers in pollen analysis. Wodehouse (1959) and Erdtman (1952; 1957:
1965: 1969; 1971). contributed modern pollen handbooks that were
important for differentiation and identification of the Florissant types. Other
useful texts which focus on North American pollen/ spore types are
McAndrews. et. a1. (1973). Hyde and Adams (1958). Lewis. et. a1. (1983) and
Sawyer (1981). For comparison with modern pollen/spore floras of other
parts of the world. several descriptive texts with abundant photographs
were used: Taiwan (Huang. 1972); Argentina (Markgraf and D‘Antoni. 1978):
and Chile (Huesser. 1971).

A plate of selected Florissant palynomorphs prepared by E. B. Leopold

(1969 a) was very useful for initial comparison of some specimens. There

34

have been no previous systematic descriptions of the Florissant pollen/ spore
flora published. Several other studies of North American floras similar in
age to the Florissant flora provided systematic descriptions and plates. The
late Oligocene Brandon Lignite of Vermont (Traverse, 1955) contains many
genera in common with Florissant and was especially useful for its
illustration of a diversity of Pings-types and a discussion on the
differentiation of taxodiaceous genera. An early Oligocene flora of the Fraser
River Fm.. central British Columbia was the focus of a palynological study by
Piel (1971). In it. he described and illustrated many genera. most notably
Mm that are found in the Florissant flora. Hopkins (1967) published
a study of palynofloras from several formations of the Coos Bay Area of
southwest Oregon that range from upper Eocene to middle Oligocene in age.
The accompanying plates illustrate many genera present at Florissant. .
Systematic lists and plates of pollen/ spores from other post-Paleocene f loras
of the Rocky Mountain region and other North American Eocene-Miocene
floras were compared: middle Miocene Succor Creek flora. Oregon-Idaho
boundary (Taggart. 1971); Miocene-Pliocene Skonun Fm.. British Columbia
(Martin and Rouse, 1966): middle Eocene lone Gravels. Chalk Bluffs.
California (Leopold. 1969 b); middle Miocene Suntrana Fm., Alaska Range
(Leopold. 1969 c): middle Miocene Troublesome Fm. Granby. Colorado
(Leopold. 1969 (1).

Organization of Descriptive Notes

Systematic descriptions are in phylogenetic order from the sub-class
level upward. and f amilieswithin each class or subclass are arranged
alphabetically. following Taggart (1971). The lowest taxonomic rank for

identifications in this study is the genus. and designated. for example,

3S

Quezon: sp.(p.). Genera are organized alphabetically within each family. If
generic affinity is uncertain. the grain is classified as Type 1.13m; 2. etc..
within the family. Differentiation into morphologically-distinct sub-generic
categories is also discussed as Type 1.1293. 2. and so on. This intra-generic
variation may. but not necessarily. represent fossil species. Speculation on
botanical affinities to modern equivalents is discussed in a paragraph along
with the description. I was not able to identify some differentiated
morphotypes to a modern or fossil genus. If specimens could not be
identified to any pre-existing taxonomic rank. they are listed at the end of
the systematic section, the type preceeded by a "UM" (unidentified
morphotype).

Each morphologically differentiated type is described and illustrated
with a photograph in one of the plates that follow. Additional information
includes: abundance in the mega- and pollen/spore- floras; ecological and
geographic indications (based on modern analogs of identified morphotypes).

and temporal and spatial distribution of the type during the Tertiary.

FLORAL COMPOSITION

113 plant species were identified (exclusive of 27 We gems),
representing 87 genera in 48 f amilies. in MacGinitie‘s (1953) megafossil
study. The species included 4 lower plants. 10 gymnosperms and 99
angiosperms. Two previous pollen studies of the lake shales have added 13
genera to the megafossil list (Ting, in McLeroy and Anderson. 1966; Leopold.
1969 a). Among the 78 pollen/ spore types differentiated in this study. 4
additional genera (one is a pair of generic alternatives that were not
distinguished) and l undifferentiated family type are proposed. All
identified and tentatively identified plant genera and families from the four
studies are listed in Table 1. The taxa/types in Table 1 are organized in the
same fashion as the Systematic Descriptions section.

None of the 5 proposed taxa was abundant in the palynological samples.
but specimens observed had the critical morphological characters to justify
tentative assignment to unreported genera or families: Mongemja
(Zygnemataceae). 2951mm: (Podocarpaceae). Milka/Lari;
(Pinaceae). “minimum: (Hamamelidaceae). and undifferentiated
Nymphaeaceae. Pollen of three megafossil taxa classified by MacGinitie
(1953) as m: sedjs were identified in this study: 2013111921190. flex. and
Compositae.

The plant specimens of MacGinitie's (1953) study were drawn from one
large sample which was the sum of his collections from several localities plus
previous collections held in museums and other institutions throughout the
United States. 56 species accounted for more than 96% of the megafossil
specimens. The 18 genera which contributed 1.0% or more to the total
number of megafossil specimens examined by MacGinitie (1953) are listed in

Table 2 in contrast to the 12 pollen/spore types which contributed 1.0% or

36

37

Table 1. Known composition of the Florissant flora.

Numbers in the ”Megafossil Taxa" column represent the number of taxa
described by MacGinitie (1953). Parentheses in this column indicate incertae
m determinations. The "Previous Palynology Records" column represents
a composite from Leopold (1969 a), Ting in McLeroy and Anderson (1966).
and Manchester and Crane (1983). The "Current Study" column represents
observations in this study. Parentheses in the latter two columns represent
a degree of uncertainty in the systematic determination. Occurrences at the
family level are tabulated only when specimens are noted (typically pollen
or spore taxa) which were not assigned to a specific genus.

Bracketed % figures in "Megafossil Taxa" and "Current Study" columns
represent contribution of the noted types to composite mega- and
pollen/spore floras; respectively. Noted types of ”Current Study” column
which do not display a % figure were included in "Other" category of pollen

counts or were observed in additional traverses.

Abbreviations for groupings of undifferentiated genera or families;
TCT - Taxodiaceae. Cupressaceae, Taxaceae

“5 - Imha/Scamanium

PIS/E - Proteaceae/Sapindaceae/Elaeagnaceae

U/Z-Iilmus/Zelkoxa

Imn
FUNGI

ALGAE
Characeae

8119.111

Zygnemataceae

BRYOPHYTA
Grimmiaceae

Hypnaceae
Sphagnaceae
LYOOPHYTA

trilete spores
Selaginellaceae

W
W

SPHENOPHYTA
Equisetaceae

P'I'EROPHYTA
monolete spores
Polypodiaceae .

221.191

GYMNOSPERMS
Ephedrsceee

Cupressaceae
, W
Pinsceee
Allie:
Elm
Emu: .
Mistress/lam
Iain
Podocarpaceae
Taxaceae

Taxodiaceae

Swim
Insulin-type

110.4%]
(1)

1
117.5%)

[2.8%]

(TCT )

NNN '

A I
5

(TCT)
(TCT )

(X)

(0.47.)

(X)
(X)

[2.17.]
3
(X)

2I0.3% I
(TCT “26.0% I

3Il.4%l
210.7%)
5110.67.)

1
(TCT )[26 .07. I

(TCT “26.07. I

39

Table l (cont‘d.).
MONOCUI'S
Cyperaceae - X -
1 - -
Gramineae - X 310.2% )
Stine l - -
Liliaceae
. 1 _ _
Potamogetonaceae
M21251: (1) - -
Weston (l) - (X)
W (010.4%) - 1118.8%I
Sparganiaceae
' - T/S Z T/810.9%1
Typhaceae
115.8%] T/S 2 T/S[0.9%l
DICOTS
Aceraceee
Am «(105%) X 410.1%]
' ' 110.3% 1 - (X)
Anacardiaceae
Aatmnium 110.5%) - -
99111111.: 11 1.0% l - -
Rhys 314.5%) - -
5mm 110.1% 1 - -
Aquifoliaceae
(1 ) - 1
Araceae .
' (l ) - -
Araliaceae
anm l - -
Bean (1) - -
Aristolochiaceae
. . 1 _ _
Berberidaceae
' 310.5% 1 - . -
Betulaceao
Alain - X -
M11 - X -
Minus 110.9% 1 - -
Bursersceae
m 111.0% 1 - -
Csprifoliaceae
Sanhum l - -
Celastraceae
110.2% 1 - -
Chenopodiaceae/Amaranthaceae - - 311 4% 1
mum-type - X -
Compositae - - 1
Am (1 ) - -
Convolvulaceae

gamma: 1

40

Table l (cont'd.).
(l P/S/E)

Elseagnoceae -
m

Ericales
Eucommiaceae

Eu phorbiaceae
Cretan .
Eunbcnbta

F agaceae
Casings
Eczema
99.229.11.11

Hamamelidaceae

Juglandaceae

Meliacoae

Installs
Moraceae

Ficus

Mom

Myricaceae
Myrteceae .

Nymphaeaceae
Oleaceae
Omnibus
Onagraceae
him-type
Matte-type
Platanaceae
21mm:

(1)
1

110.2%)

1131.4%]
9(1)12.0%I

111.3%]

1
110.2%]

1
-10.7%I
210.1%)
110.2%]
1

10.8%)

Fanny‘s-oun—

110.7%]
110.4%)

(3)
1

110.5%)

110.6%]

1

N NN'

' NNN '

NNEN

I
11<0.l%1
414.3%]

110.1%]
314.1%)

110.2%)
113.9%]

41

Table l (cont'd.).

Proteaceae - - (1 P/S/E)
Loam 110.9%) - -
Rhemnaceae
Wm 110.1%) - -
Rhamnitas 110.4%] - -
Zizxnlm 110.2%) — -
Rosaceae ,
Amclanshier 110.2%) - -
Gammon: 112.7%) - -
Emmy; 3(1)10.6%l - -
M11113 .2 - -
Emu: 110.1%) - -
Rm 1 - -
210.2%) — -
Rutaceae
1 - -
Saliceceae
29211111! 101150751 ~ 1
53.11; 411.1%) X 211.4%]
Sepindaceae - - (1 P/S/E)
Athmns . 112.3%) - -
Mammal 110.2%) X -
Mama 1 - -
mm 111.0%) - -
512111111111 111.1%) - -
Ihnuinia l - —
Saxifragaceae
112.412.1133 1 - -
21111951212111); . 1 - -
. 121m 1 - -
Simarubaceae
. l _ _
Staphyleaceae
112.7%] - -
Storculiaceae
[mantis - X -
811311111 (1) - -
Styracacoae
. 1 _ _
Thymelseaceee
mm 1 - -
Tiliscese
' ' 1 - 1
Ulmecese - U/Z 3 U/ZI4.5%I
129m: 1 X 212.3%]
1.11m: 111.3%] - -
29km 119.6%] - -
Verbenaceae
We 1 - -

2cm 1 - -

42

Table 1 (cont'd.).
Vitaceae 1
Ellis 1 - -

All others [3.2%] [15.9%]
multiporate (UM l) - - [0.4%]

Additional M 821215

s—ss—ss—s—s—sNe—N

Normapolles group -

V N
I

unidentified pollen/spore morphotypes -

43

more to the total number of specimens counted in this study. Since discrete
samples at known stratigraphic positions were prepared for this
palynological study. the relative % contribution of each taxon to the sample
sums can be compared and is reflected by the % range data accompanying
the list of pollen types. Variablity in taxonomic composition and abundance
between pollen samples is discussed in the Vegetation Dynamics section.

The generic diversity represented by pollen is considerably less than
that of megafossils (Table 2). The disparity is in part a result of the excellent
depositional conditions in the lake f or f ossilization of leaves. that is.
relatively quiet waters with a regular influx of fine clays and silts to entomb
the leaves between accreting laminae. Other factors include: limited
taxonomic resolution of many pollen/spore types (e.g.. TCI‘). production
differences for different organs. restricted dispersal potential of some
pollen/spore types (e.g.. entomophilous types. such as Sanindus). and fragile
pollen/spore types (e.g.. 29mins). Therefore. each organ assemblage
represents a unique and limited sample of the entire regional flora.

Further comparison of the two lists in Table 2 shows that there is also a
disparity between the dominant taxa for each category of fossils
differentiated. lingual: leaves contributed almost one-third of the'total
megafossil specimens. more than three times the frequency of Zelkgya. the
second on the list. Eaggpsjs never contributed more than 0.5% to the pollen
sum. The pollen of Eagopsjs resembles that of Quergus spp. (Manchester and
Crane. 1983). Quems is abundant in both the mega- and pollen/spore-
floras (2.0% of leaves; 4.3% of the pollen). The disparity of the pollen/leaf
records between Eaggmsis and Quezon; may be attributed to various factors.
Plants of both genera have coriaceous leaves which preserve relatively well.

The overwhelming abundance of Eaggpsjs leaves suggests that the trees

44

Table 2. Fossil representation of the Florissant flora.
leaves vs. pollen and spores (genera)

’ = taxon common to both lists

LEAVES: COMPOSITE % (MacGinitie. 1953)
30.1 MIFagaceae)
9.6 *Zelkovawlmaceae)
6.3 *QhammflCupressaceae)
5.8 unhaflyphaceae)
5.0 MuflSalicaceae)
Runs (Anacardiaceae)
W (Staphyleaceae)
We (Sapindaceae)
*Ssguoja (Taxodiaceae)
Mmmosaceae)
*Quemus(Fagaceae)
1’ jnu§(Pinaceae)
‘fllmuswlmaceaw
*Salix (Salicaceae)
Swinging (Sapindaceae)
* gyjuglandaceae)
99.111115 (Anacardiaceae)
Bummurseraceae)
um=80.7

u-tf-Iu-u-sr-u-eg—NNNNNA
ocbw—mmoL—Nw'flm

(I)

all other leaves < 1.0 each
sum = 10.8 (29 genera)
other organs: sum = 5.3

undesignated types: sum = 3.2

POLLEN: % RANGE ALL SAMPLES. COMPOSITE % (current study)
0-75. 26.0 *Taxodiaceae/CupressaceaefTaxaceae(TCT)

0-79. 18.8 W (Potamogetonaceae)

1-68. 10.6 *Eim (Pinaceae)

0-22, 4.5 *fllmufllelkova (Ulmaceae)

0.5-14. 4.3 *mlencus(Fagaceae)

1-16. 4.1 *m (Ju glandaceae)

0-57. 3.9 luglans (Juglandaceae)

0-14. 2.3 (“£1113 (Ulmaceae)

0-15. 2.1 monolete spores

0-4. 1.4 *Saljx (Salicaceae)

0-13, 1.4 Able; (Pinaceae)

0-5, 1.4 Chenopodiaceae/Amaranthaceae (Che no-Am )
sum = 80.8

all others « 1.0 each
identified: sum - 3.0 (11 types)
unknown. sum = 16.2

45

occupied (and possibly crowded) the lake margin. They need not have been
abundant in the region as a whole. Therefore. while there may have been
local concentrations of the trees which shed a great many leaves directly into
the lake. on a regional level the pollen produced by these plants may have
been low. relative to the total pollen volume that reached the lake. Oaks. on
the other hand. may have been widespread throughout the region. supplying
leaves to the lake through stream influx; fewer leaves would reach the lake
than those of Eaggpsia. The anemophilous pollen of the regionally numerous
oaks. however. carried by both streams and wind. may well have exceeded
the contribution of Eagopsis pollen into the lake. It is also possible that
pollen production per tree of Eagopsis was less than that of Quezon or that
the pollen of the two genera was differentially degradable. Eaggnsia being
more susceptible to deterioration. Therefore. a preservational bias may
distort the actual regional distribution patterns of plants of the two genera.

Six genera are common to both lists in Table 2: Finns. Illmus and chlmxa.
Queacya Gama. Sam. These are all arboreal members of the
Bottomland/Slope paleoassociation. Their species are wind-pollinated except
for some species of Salix

prha. the third most important megafossil numerically (5.8% of leaves).
is also present in the pollen flora. but averages >1.0% only in section 8/ 5/ 84-
I. Since its pollen is relatively infrequent in section 8/ 23/8 4-1. the percent
contribution of Luna to the composite flora is less than 1.0%. My: is
also an important megafossil but is rarely observed as a pollen grain. The
pollen of both genera. especially Bacillus. are relatively fragile and low
concentrations are to be expected in fossil pollen assemblages. even when

the plants reside close to the depositional site.

46

Numerically important megafossil taxa that are not represented (or
perhaps not recognized) in the pollen record include Rims. Stanhvlea.
Ammflemrwéammcmmandhum. knits andSapindus
are entomophilous. Plants which are pollinated by insects are generally
under-represented in the pollen rain. in part because they produce pollen in
much smaller numbers than do wind-pollinated types (Faegri and lverson.
1974). and in part because of the mode of distribution.

The TCT-type pollen is the most abundant. The large fossil stumps.
putatively Sammie, but quite possibly Chamaecypanjs (see TCT discussion in
Descriptive Notes for explanation). attest to the fact that the trees were very
large. The pollen disparity between TCT and Eagonsis is likely a consequence
of significant differences in overall pollen production despite the fact that
individuals of both types may have been locally abundant.

EQLamogeth pollen varies greatly in abundance and ranks second
overall in the-composite pollen sum. High levels may be explained by large
numbers of source plants and the fact that Egtamggatgn pollen was actually
produced in the lake. Its presence as a megafossil is uncertain. This .
disparity in records likely reflects rapid deterioration of the succulent

leaves.

PALEOASSOCIATIONS

Following the approach of Taggart and Cross (1980). the individual taxa
contributing to the pollen sums were grouped into paleoassociations
according to related ecological/distributional characteristics of extant
equivalents. The 6 paleoassociations selected and their constituent taxa are
listed in Table 3. 30 pollen/spore types are included at the genus. family, or
larger group level. In addition to the taxa included in the pollen diagrams.
the identified types described in the Descriptive Notes section which were
not quantitatively significant are listed within brackets to indicate their
ecological position in the regional vegetation.

Hypothetically. the paleoassociations together comprise the complete
vegetation community mosaic across the topographically variable landscape
of the Florissant region. The approximate spatial distribution of these
communities with respect to increasing distance and elevation away from
the depositional basin may be tentatively arranged: Aquatic/Marsh.
Lakeside Forest. Bottomland/Slope Forest. Pine and Xeric/Disturbed Scrub.
and Montane Conifer Forest. Like modern landscapes. ecotonal zones would
have been present and a dynamic mosaic of community types would have
been variably developed on the regional topography under the influence of
microclimatic and edaphic factors.

Treatment of data as multiple-taxa paleoassociations tends to buffer the
effects of single taxon fluctuations. This lumping is done at the expense of
some loss in ecological precision. for patterns of variation of individual taxa
are obscurred and their ecological ranges are not accounted for fully.

' However. shifts in relative importance of the paleoassociations should
represent coordinated contributions of the constituent taxa. Such shifts in

paleoassociation importance are likely to reflect real changes in the plant

47

48

Table 3. Paleoassociations of plant taxa represented by pollen and spores
from Florissant.

MQNIANEIDNIEEREQRESI
A1212:

Elma
[Eodogarpusl
INN/tam 1

RIM;
Pings

ERIC/[215111331212 SCRUB
Chenopodiaceae-Amaranthaceae (Cheno-Am pollen)
(Compositae)

Gramineae

WERE]:
Assn
Cam
lfaslaneal
Ensclharilia
lumen:
u
[Proteaceae/ Sapindaceae/ Elaeagnaceae]
Qumus
111113 1

Taxodiaceae, Cu pressaceae, Taxaceae (TCT )

93111: .
Leanna:
1229.111113]
Salix
Elms/221m

/
[Algae]
111st

[Nymphaeaceae]

Tuba
911111

spore-producing plants

unidentified morphotypes and unidentifiable pollen and spores

(bracketed entities above. except for algae. are included under this category in the
pollen diagrams)

49

community structure. especially when plant diversity within the
paleoassociations is high.

At best. the paleoassociation groupings are an approximation of
coexisting plants and are not mutually exclusive or precise ecologically or
distributionally. Some of the genera or families have broad ecological
amplitudes among their species today. For example. the Pine
Paleoassociation is completely artificial because in extant f loras. pine species
may occur in pure or mixed stands or may be scattered widely in open
parklands. as P. nonderasa currently is at Florissant. Pings spp. may be
present in montane conifer communities (e.g.. E. maxilla. E. cantata).
bottomland/slope communities (e.g.. 2. mega. E. slmbns). xeric/disturbed
communities (e.g.. 2. 911.1113, 12. banksiana). and in moist lowlands (e.g.. 2.
911101111). Pine pollen is ubiquitous. even in forest soils where it is not locally
growing. because of its abundant production and great dispersal potential.
Similarly. Romulus spp. are ecologically diverse. Aspen (Regulus
W) is the most widely distributed tree in North America (Harlow.
et. al.. 1979). It is an aggressive pioneer which invades sites in pure stands
after disturbances. Cottonwood (E delicides) is most commonly an occupant
of moist alluvial banks. However. open grassy forests of Pings pgndamsa
with scattered pockets of 290.0111: Was in moist bottomlands are
common in the Rocky Mountains today. Quemus spp.. a common element of
the palynoflora. may also be found in diverse habitats (e.g.. Q. palaatajs and
Q. nigra: moist bottomlands and swamp margins; Q. amoryi and Q. malizanij:
shrubby sclerophyllous types of the dry southwest United States and
northern Mexico; Q. alba and Q. 111112123 of mesic deciduous forests in eastern

United States).

50

The closest existing ecological analog to the Lakeside Forest
Paleossociation (dominated by Eagansis. Seguoia and mammals) is
probably a combination of the redwood (53.911913 W) forest of the
California "fog belt" and the Port-Orford cedar (Chamaegynarls lamina)
forest of the coastal Pacific Northwest. The leader in overall pollen
production is the TCT type. Placement of TCT pollen into particular genera is
problematic (cf. TCT discussion in Descriptive Notes section). The most
common associates of modern redwoods are We. Elm Allies.
Inna. Lilhmatmli (of. L. densiflorus (Fagaceael). Alnus. Arbutus. and
1.1111112111113113 (Harlow, et. al.. 1979). The association inhabits moist benches
and alluvial flats. Modern North American 9112mm spp. have
distributions restricted to the east and west coasts (Harlow. et. al.. 1979). C.
1111011115. the east coast species. is characteristic of freshwater swamps and
bogs. C. lamscniana of the Pacific Northwest prefers an abundance of soil
and atmospheric moisture. but also frequently occurs on more dry. sandy
sites "45-65 km inland. C. nmtkatensls occupies a broader. more northerly
coastal distribution which extends into southern Alaska. It grows most often
on sites with rich soil and an abundance of moisture. Associates include
many of the same genera as Seangla: Finns. Lam. 121931. Am Isusa. 101113.
The available evidence suggests that the Florissant TCT pollen types indicate
presence of a rich. moist bottomland environment adjacent to the shallow
water body(-ies). possibly affected by periodic flooding.

The topographically variable landscape that Lake Florissant occupied
permitted microclimatic and edaphic zonation. Therefore. diverse ecological
settings probably coexisted within relatively close proximity to each other.
Elana. Bicaa. and Ablea which are common associates of extant western
North American Saqngla and Chamaegyparls, are also represented by pollen

51

in the Florissant sediments. This constitutes a mixture of drier upland and
slope and moist bottomland plants. Abias pollen is generally
underrepresented in deposits. even when the trees are abundant regionally.
and 2199a is generally indicative of the local basal area of the existing trees
(Leopold. 1964). When the relative effect of TCT pollen is removed from the
pollen counts. these two taxa are still poorly represented. Since neither
Ablas nor Riga leaves or pollen are abundant in the Florissant sediments. it
is difficult to associate them with the taxodiaceous-rich shoreline. but it is
possible that they occupied a very minor. scattered position in these forests.
Consistently low leaf and pollen levels substantiate MacGinitie's (1953)
suggestion that they were cold/dry-adapted members of the montane
complex. accompanied by species of Aaea. Phlladelphas and IQmaya.

I The leaf margin of W densjfloms resembles that of Eaggnsls
(Harlow. et. al..l979) and their flowers and fruiting structures are
comparable (Manchester and Crane.l983). It is postulated here that
Eagaasla occupied a co-position in the Florissant lakeside forest with Seangla
and Chamaecxparis in a relationship analogous to the position of modern
Lithggamua. This postulation is in agreement with the megafossil
abundances. Low levels of Eagonsls pollen have been attributed to low
production coupled with limited distribution of the trees.

Additional members of the Lakeside Forest Paleoassociation are
suggested by MacGinitie (1953). These include: Regulus. Sallx. Zelkgya.
minus. Calm, Members of the Ulmaceae are relatively abundant in both the
pollen and leaf records. MacGinitie (1953) suggests a reciprocal relationship
for the abundance of 2021111111 and a scarcity of Sflix fossil foliage. attributed

to competition for the same sites. The inverse of this relationship was found

52

in the pollen record. perhaps. in part. a consequence of the relative fragility
of Panama pollen.

Strong representation of Eatamggaton in the pollen record documents its
widespread occurrence in the ancient lake. Nymphaeaceous plants were also
present but probably sparse; only a few grains of pollen and no leaves have
been recovered. The existence of cattail marshes is supported by a
substantial number of leaf specimens (5.8%. fourth-ranking megafossil) and
occasional T112111 pollen. The algal record consists of Nitalla (megafossil) and
Maugaotla zygospores. The Aquatic/Marsh category of Figures 4 and 5 is
almost entirely represented by Pgtamggatgn pollen.

The Bottomland/Slope Forest Paleoassociation occupied a position on the
better-drained flats and hills peripheral to the basin of deposition. On the
average. it probably resembled the mixed mesophytic forests of eastern
North America andeast Asia. dominated by broad-leaved deciduous trees.
Mesic species of oak and hickory likely were the co-dominants. on the bases
of abundant pollen and leaf records. Trees of secondary importance were
species of Agar, (maple). Linuldamlzar. (sweetgum). luglans (walnut). Castanea
(chestnut) and Engelhandia. Sweetgum and walnut may have been locally
abundant in pure stands after disturbances. This association was transitional
between the lakeside and more upland. drier sites. Hickory may have been
ecotonal to the lakeside forest and drier-adapted oaks were probably
concentrated on the sparsely vegetated. well-drained slopes.

Exposed areas (e.g. hilltops and steep. south—facing slopes) characterized
by poorly developed soils were the likely position within the regional
. vegetation mosaic for members of the Xeric/Disturbed Scrub
Paleoassociation, primarily grasses, members of the families Chenopodiaceae.

Amaranthaceae and Compositae. and Enhance. They are poorly represented

53

in the fossil record because of their distance from the sites of accumulation
and their limited dispersal potential. MacGinitie (1953) suggested that these
herbs and xerophytic shrubs (e.g.. Amalangnlea (lemmas. Manna.
1431191113. X11115. M) were intermixed with an open forest of pines and
dwarf evergreen oaks. a group characterizing Chaparral vegetation.

The ”Other" category is completely artificial. In the pollen diagrams. it
includes unidentified morphotypes. unidentifiable pollen and spores and all
of the bracketed taxa listed within the paleoassociations. because they
played a very minor role quantitatively. It also includes the spore-
producing plants (monolete. trilete) because the diverse ecological
adaptations of these undifferentiated plants precluded classification into any
one of the paleoassociations. More than three-fourths of the "Other" category
represents pollen and spores which were corroded or deformed beyond

identifiability.

Stratigraphic Distribution of Palynomorphs

Pollen diagrams are a useful graphic display for the tabulated sums of
the selected samples. They provide a condensed. comparative presentation
of changes in vegetation elements and community structure over a particular
stratigraphic interval. Two major types of pollen diagrams (or pollen
profiles) are used by palynologists: absolute frequency pollen diagrams and
relative frequency pollen diagrams.

Absolute pollen counts are useful if the sedimentation rate is known or
constant (Moore and Webb. 1978). This method is most appropriate for
peats accumulated since the most recent glacial maximum. because the
subsequent deposition rates often may be calibrated by incremental
radiocarbon dating. Pollen data are expressed as pollen inf lux. an estimation
of the density per unit time (units: 1‘ grains of 'a species/cm3/yr). The
primary application of this method is to identify real changes in the pollen
contribution of specific taxa over time. independent of the fluctuations of
other taxa.

Relative percentage data are the most widely used in Tertiary sediments.
since most allochthonous deposits are less-well time—calibrated. They are
used to demonstrate shifts in relative numerical contribution of taxa or
groups of taxa among samples. The proportions are calculated according to a
fixed pollen sum independent of the relationship of the density of
palynomorphs in the sediment. Their limitation is that a change in one
component will therefore produce compensatory changes in all other
components (Moore and Webb. 1978). This creates the problem of
separating real changes in the source vegetation from apparent ones. This

problem may be largely overcome by grouping individual taxa into

54

55

ecologically similar assemblages (paleoassociations). thus damping the effects
of single taxon fluctuations.

Due to production. dispersal and preservation biases. pollen analysts
have long been aware of the discrepancy between pollen representation in a
deposit and source plant abundance and distribution. Correction factors such
as Faegri and Iverson's multiplier and Davis' R-value model have been
devised to alleviate problems of over- and under-representation (Birks and
Birks. 1980). These methods are based on measurement of modern pollen
percentages in surface samples in relationship to existing vegetation
percentages. The major limitation using these techniques is that correction
factors may fluctuate for a single species in different areas; therefore
validity decreases when sampling large basins. In addition. these methods
are calibrated for modern environmental settings. and it is unlikely that R-
values have remained constant over time. especially as far back as the
middle Tertiary. Therefore. correction factors have not been used in this
study.

Relative pollen diagrams are used in this study. The palynological
samples are arranged vertically to stratigraphic scale and the plant
compositional categories are horizontally listed across the top. The sum of
the relative compositional contributions at any horizontal level equals 100%
for a particular sample. The data are displayed in two primary ways.
according to contributions of:

(a) individual taxa
(b) associations of taxa

Figures 2 and 3 are relative frequency pollen profiles of individual taxa.

including all taxa which contributed greater than 1.0% relative frequency to

the pollen sum of at least one sample of the sections. Taxa/morphotypes of

FIGURE 2. Section 8/5/84 I (taxa)

56

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

” " I | |
6'
4.29.2233. II I II. . |
3 47 .. . .
I:I-:: e I I g :
”$532272 u:-s.. I . s s e
79’” .
30 e o
Ls.
91:70,)": :: . : T:
«”0 e:II:I I I a I :
6905.3 : :

”P’Wvo: :. s . t : :
”33°63‘23“ . z i -
“c4522: 2 2 I :

40' -
'4? .
°° I 1
“so. I I 1 I
$7~Iqs3selu '4! I e e
119% :
5%.. ::

 

 

I\
k0

e l-
—
—
3
a
_
_
—
_

 

 

 

~94 . . . ’ :
.312. ; - .
400%“)? :: : z: - : : I

100 96

0 50

ee-<2%

RELATIVE FREQUENCY

LAMINATED SHALES E Homoceueous TUFF I

FIGURE 3. Section 8/23/84 I (taxa)

S7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ill
0

ee.<2%

RELATIVE FREQUENCY

HOMOGENEOUS TUFF I

 

.10--
b—u—d

-’.I
~—-‘

—'
p--

 

 

 

 

 

 

 

 

 

LAMINATED SHALES

58

lesser abundance are included in the "OTHER" category. While two taxa
(Eagapsls and Manila) contributed very little numerically to the pollen
sums. they are retained in the diagrams for their ecological significance
when present or absent. These diagrams are useful for comparing the
composite pollen spectra of the two sections. and were the starting point for
interpretation of vegetation dynamics.

Two individual taxa. TCT and Pmampgetpa, are disproportionately
represented in relation to all the others. Since these are major components
of the spectra. any shift in their contribution may consequently cause a
major change in the expressed percentages of the other contributors to the
pollen sum. Cross and Taggart (1982) remark that such consequential
changes may occur without actual changes in the source plant abundance or
distribution. Common taxa which may produce this shifting effect in relative
frequency diagrams are Pings. Alana. Belula and Comma (Cross and Taggart.
1982; Birks and Birks. 1980). These taxa are either voluminous pollen
producers with long-distance dispersal capabilities (e.g.. Pippa). or typically
grow in dense concentrations close to the depositional basin (e.g.. amps

thickets). The second explanation is probably the case for the abundance of

TCT and W pollen in the samples. W is a ubiquitous

aquatic and leaf megafossil data suggest that Chamaeaypafla and Segupja
(the probable source of TCT pollen) probably crowded the lake margin as

bottomland dominants. In order to mitigate these single taxon biases. the
pollen sums were adjusted with the exclusion TCT and Eptampgatpn;
counting was continued for the other taxa until the fixed sum of 400 was
reattained. The relative abundance of each type did not change appreciably

in either of the two sections after this adjustment.

59

The relative frequency pollen diagrams in Figures 4 and 5 are based on
the ecological associations of taxa described in the Paleoassociations section.
TCT pollen is separated from other lakeside taxa so that patterns of variation
in the Lakeside Forest Paleoassociation may be separated from the
minimizing effects of TCT dominance. The Aquatic/ Marsh Paleoassociation is
overwhelmingly dominated by Eptampgatpn; the role of mm does not
exceed 4% (av. 0.85%).

FIGURE 4. Section 8/5/84 I (paleoassociations)

60

100 %

O

ee-<2%

RELATIVE FREQUENCY

LAMINATED SHALES E HOMOGENEOUSTUFF -

61

$°°P 8 O *NV'ee

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I — _ — 55:85 m>=5mm - as» maomzmeozo: mmmmmm mméxm 822.25
I I a! I I m .111 a
'4 .. .u .L W.
If .. l o. e Wm F
If L J I 3 4 Ir .4 .1.th w
t I I. I n J “I n
I‘ If Ir 3 l L o.
O .r .
as... a... a...» a...» a... «o... 04%,
, a. m. no 07
m. a... to .. a.
.. a... a.

acofioogmooae _ Emma 820% .m #50:.

Phytogeographyl Vegetation Dynamics

The pollen diagrams which contrast the paleoassociations (Figures 4 and
5) are used for identifying the major shifts in vegetation dominance for the
stratigraphic intervals studied. They may also be used to compare the floral
composition at geographically separated sites. Figures 2 and 3 are useful for
determining the significance of specific taxa/types within each
paleoassociation during shifts in dominance.

Because of the complex dynamic nature of plant communities and the
limited information preserved in the fossil and sedimentary records. it is
difficult to resolve the variables that influence changes in fossil
representation. For relative frequency pollen diagrams. shifts in dominance
of types may be attributed to changes associated with the source vegetation:
(a) numerical abundance. and (b) distribution. The primary environmental
factors which control these variables are climate. physiography. and
disturbances such as fire. flooding and volcanism. The effects of volcanism
and short-term climatic change are recognized in the f ossilif erous deposits at

Florissant.

Numerical Abundance of Source Vegetation and Geographic Variation in
Plant Distribution

The percent range of fossil representation in Table 2 shows considerable
variability in relative abundance of the pollen taxa/types for different
samples. In all cases. the average percent contribution for each type was
much less than the maximum of the range. For several taxa/types. the
’ vertical range of variation within a section was more than 50%. Such
dramatic fluctuations should reflect significant changes in the ability of the

pollen/ spores to reach the depositional site under the prevailing conditions

62

63

(Smiley and Rember. 1981). Post-depositional changes may further affect
preservation.

The most obvious explanations for fluctuation of a single taxon/type in
successive samples are changes in proximity of the source plants to the
depositional site. changes in the numerical size of the source plant population
(sum of pollen production). and preservation in different types of
sedimentary environments. However. the relative representation of a
taxon/type in the pollen record may vary with time when these factors are
constant for that taxon/type. An additional variable that must be considered
is colonization or change in population size of other plants between the
taxon/type noted and the depositional site. These other plants contribute
pollen to the record. thus decreasing the relative importance of the
taxon/type noted. and they may act as a physical barrier to pollen transport
from parent plants that were formerly more accessible to the depositional
site. When making postulations about the influence of factors such as
volcanism and climatic change on source vegetation. these potential biases
must be considered.

The Florissant pollen and leaf records indicate that abundant plants of
the Lakeside Forest Paleoassociation were capable of acting as buffers to
transport of pollen and other organs from plants of the slopes and uplands to

the depositional sites.

Volcanism and Pollen/Spore Representation

An important goal of this study was to examine the influence of the
regional volcanism on the vegetation in and around the depositional basin.
Evidence may be derived from the fossil content and petrology of the basin

deposits. The relative frequency pollen profiles do not demonstrate

64

catastrophic destruction of established plant communities here by volcanic
activity. Volcanic activity appears to have played a minor role in vegetation
dynamics near the depositional basin. indicated by subtle variations in single
taxa or coordinated groups of taxa. Other factors may have been more
important in controlling the composition and distribution of the regional
vegetation.

The extensive deposits of the Thirtynine Mile volcanic field record a
highly active interval of volcanism during the early- middle Oligocene. _
Several types of volcanigenic deposits were observed in the lakebeds. The
main stream at the northern end of the basin and tributary streams from the
slopes surrounding the basin carried most of the materials into the lake. The
homogeneity. coarse character. and cross-bedding of the lower tuff s suggest
that loaded streams supplied sediments to the basin in great quantities
during early lake history. Relatively quiet volcanic conditions for the latter
part of lake history are recorded in the laminated shales above the lower
tuff s. The deposits were mainly interbedded pumices and graded tuffs.
With respect to overall thickness. direct airf all of volcanic ash played a
variable and minor role in lake sedimentation and is the only primary
volcanic deposit. Lava flows and ignimbrites are not recorded in the
lakebeds.

Volcanism has the potential to destroy or disturb plant communities
proximal to the volcanic source. Subaerial deposits of ash and accompanying
gasses or heat could temporarily or permanently eliminate some plants.
Direct effects include burial. mechanical overloading and suffocation. An
indirect but important effect could be alteration of edaphic conditions.

Studies of modern analogs demonstrate highly variable effects of

volcanism on vegetation. In studies of the Paricutin eruptions. which began

65

in 1943 in central Mexico. Eggler (1948) and Dorf (1951) showed that plant
devastation by accumulated thicknesses of ash is not as severe as might be
expected. Mortality of small trees. shrubs and herbs was most extensive.
especially in >20 cm of ash (Eggler. 1948). However. many larger trees of the
dominant pine-oak forest survived >2 m depth of ash. Grasses (Stipa
included as a Florissant megafossil) and some dicotyledonous herbs colonized
bare ash 50 cm thick within two years. Malde (1964) notes that ash falls
may even be beneficial to plant growth by enhancing the nutrient content of
soils. The immediate effects are usually detrimental. however. because
unlike soil. ash is sharply granular. deficient of clays and easily eroded
(Malde. 1964). Although episodes of volcanically induced forest destruction
have been noted in studies of the Miocene Succor Creek flora (Cross and
Taggart. 1982). no such events are evident in the Florissant sections studied.
The best opportunity to investigate the effects of volcanism on Florissant
vegetation is a homogeneous volcanic ash layer in the lakebeds of section
8/23/84-1. In Figures 3 and 5. it is represented by the homogeneous tuff
zone of the lithologic section. This deposit is "l m thick. is uniform gray in
color. fractures conchoidally and contains no fragmental plant material or
sediment particles larger than the ash. It grades slightly finer upwards
throughout its thickness and contains a series of intermittent
microlaminations that are evident in thin-section. Cross-bedding and
turbidity figures are not apparent in hand specimens or thin-section.
Contact with the underlying finely-laminated fossiliferous shale is abrupt
with no evidence of scouring or erosion prior to deposition. The underlying
shale is uniform in thickness and the upper laminae are continuous; current
activity does not appear to have affected it. The intermittent grading may
be explained by subaqueous settling of ash particles dropped by fluctuating

66

wind currents transporting the materials from the vent (Guffey volcanic
center.”25 km southwest) to the lake. Three palynological samples
representing the total thickness of this bed were barren. This evidence
strongly suggests that the accumulation resulted from airf all deposition of
volcanic ash into the standing body of water during a geological instant in
time. although definitive evidence against fluvial deposition was not found.
No counterpart to this deposit has been observed in other exposures of the
lacustrine shale unit. It is postulated that it was once a widespread deposit.
but has been eroded out of other exposures observed.

In the fossiliferous shale immediately below the ash deposit. the pollen
spectrum is dominated by the Lakeside Paleoassociation (Figure 5: TCT >50 %:
other Lakeside plants "14%). All of the other paleoassociations are
represented at low levels. Immediately above the deposit. TCT-type and
Aquatic/Marsh pollen are absent; other Lakeside members ("37%). Pine
("24%). and Bottomland/Slope ("18%) codominate. Pollen of the
Xeric/Disturbed paleoassociation decreases after that sedimentary event.

It is apparent from the pollen spectra below and above the ash deposit
that rapid subaerial deposition of more than a meter of ash did not
drastically alter the diversity of plants in the basin region. The only plants
that were significantly reduced in pollen representation were TCT-types and
Bojampgatpn, Populations of modern W spp. are sensitive to
reductions in light intensity. It is postulated here that the Eptampgetpn spp.
were catastrophically reduced in numbers by ash burial. change in lake
bottom conditions. and decreased photosynthesis in cloudy waters. In the
. second youngest sample studied from above the ”11.811111158111211 recovered
from the disturbance and achieved its greatest pollen representation for the

section ("22%). The recovery may be attributed to one or more of several

67

factors of lake substrate conditions suitable to growth of rooted. sub merged
or floating aquatics. e.g.. compaction of the ash. alteration of glass and
feldspars to clay minerals. increase in organic content (recovery of bacteria.
algae. and bottom fauna). Algal spores of Maugeatia. rare in the samples
studied overall. were abundant in the sample immediately above the ash
deposit. Sudden increased nutrient content of water accounts for modern
algal blooms. and the nutrient contribution and turbidity associated with the
ashf all may have triggered the Mpugeptia bloom.

The effects of volcanism on the composition of the palynoflora may be
compared for the Lower Tuff and Lake Shale units in section 8/5/84-1
(Figures 2 and 4). In Figure 4 the plant diversity within the two units
appears to be roughly the same. The major difference is the presence of
Montane Conifer Forest types (Abies. Riga) exclusively in the lake shales.
but at very low levels. This similarity in diversity suggests that. for the time
period represented. the regional vegetation composition was not significantly
different during intervals of relatively high and low volcanic influence. or
that sampling was not on a fine enough scale. Although not quantitatively
significant. the presence of Montane Conifers in the lake shales may
represent the colonization of these trees in the uplands after volcanic
influences were somewhat mitigated. ,

Aquatic/Marsh pollen is more abundant in the laminated shales than in
the Lower Tuff of section 8/5/84-1 (Figure 4). Aquatic/Marsh pollen is
overwhelmingly dominated by W The turbidity. unstable water
level. and rapidly changing lake bottom associated with rapid sedimentation
during Lower Tuff time may have limited colonization of Eptampgaton The
subsequent reduction in volcanic influence may have led to increased
stabilization of the lake bottom and clearer water. permitting Patampgetpn to

68

reestablish itself as a ubiquitous aquatic. Weathered products of the
volcanic ash were capable of supplying abundant dissolved nutrients to the
water to support rapid growth of extensive beds of Batampgeton. This may
account for the abrupt relative increase in Aquatic/Marsh pollen in the

lower part of the lacustrine shales of section 8/5/84-1 (Figure 4).

Basin Infilling. Climate and Lake Level

Variations in lake level may have caused profound changes in the
abundance and type of vegetation in and immediately bordering the lake
while not significantly influencing the vegetation of higher. drier sites. It is
postulated in this study that during the later lake history. lake levels
decreased and became increasingly sensitive to climatic variability.

Following impound ment by a lahar or lava flow. accumulation of
sediment and water in the basin may have caused water levels to rise to the
top of the containment structure. Additional input of sediment and water
could cause water to exceed the capacity of the dam. and progressively erode
the top and downstream side. Reduction of dam height or sudden breaching
is then possible. causing shallowing of the ponded area. The rate of
backcutting would be dependent on the origin of the containment structure;
a lahar would be much less resistant to erosion than a lava flow. The
reduction in lake level would be permanent. unless a later obstruction of
drainage reinitiated the process. Progressive shallowing of the lake would
increase sensitivity to fluctuations in the precipitation/evaporation ratio.

Sedimentological evidence in both study sections records a relatively
dry climate during the later part of the recorded lake history. Several
calcareous layers are present in varying thicknesses near the top of the

exposures. In the Geology section. these deposits were suggested to be

69

caliche formed below the erosion surface prior to deposition of the Upper
Tuff. Near-surf ace crystallization of calcium carbonate in the soil is possible
under the warm-temperate climatic regime suggested by MacGinitie (1953).
The palynological record of a shale sequence which includes the
calcareous deposits in section 8/5/84-—I may indicate changes in aquatic
plant representation during a dry period. Reduction in fossil representation
the aquatic weed Batampgatpn may be considered an indicator shallow water
and/or decreased lake area. In the upper six samples of Figure 4.
Aquatic/Marsh pollen declines from a dominating “80% to ”2% and then
recovers to more than 50% relative abundance. The lowest pollen
representation of the Aquatic/ Marsh Paleoassociation occurs in one of the
thick calcareous layers. The other samples are of clastic origin. Relatively
low stream input is indicated by associated thin volcaniclastic laminae as
compared to those lower in the lake shales. The same samples in Figure 2
show that fluctuation of Botampgatpn pollen is responsible for this shift.
There is no Rolampgatpn pollen in the calcareous layer. Algal growth during
a period of high evaporation may have been responsible for the deposition of
the calcium carbonate. The Aquatic/Marsh pollen reflects 2.3% 12pm. the
maximum the genus attains in the lake shales. The Iypna spike may
represent an increase in marsh conditions created by the receding shoreline.
Diagenetic processes during caliche formation may have resulted in the
selective destruction of W pollen in the calcareous sample.
However. a gradual decrease of Eptampgatpn pollen in the samples
immediately below the calcareous layer followed by a gradual recovery of
W in the samples above could be used as an argument for real.
systematic changes in the parent plant population in response to a

temporary dry period.

70

An abrupt increase in TCT-type pollen accompanies the elimination of
W This increase probably does not reflect an increase in the
number of associated plants, but a reciprocal decrease in Potampgatpn alone
could account for this change. However. analagous to an increase in marsh
conditions. reduction in water level may have opened flat. moist areas for

invasion of TCT-type plants.

Relationship of Collection Localities to Lake Physiography

Field relations of rock units suggest that the two depositional sites
studied were situated at different distances from the basin edge. The
Tertiary volcanigenic rocks of the region lie unconformably on the extensive
and irregular surface of the Precambrian Pikes Peak granite. The granite
crops out above the Tertiary rocks sporadically within the basin and
dominates the surrounding upland areas. Collection locality 8/5/84-1 is
positioned in the basin interior. near the western edge. in the form of an
isolated hill. The granite is exposed at the top of the hill and the Tertiary
lake deposits circumscribe the slopes. The lakebed/Upper Tuff contact
extends around much of the hill's circumference near the top. The lower
part of the lakebeds completely surrounds the hill and is concealed at the
base by the valley fill. The draping of the lake deposits on the granite core
at this locality suggests that the hill was an island within the lake during
Florissant time. An edge of the lake sediments outcrops a few hundred
yards to the northwest.

The deposits of section 8/23/84-1 are part of an extensive irregular ring
of exposures that outline the slopes near the east edge of the basin. The
Upper Tuff is overlain by Quaternary alluvium that covers the gently sloping

surface for "0.8 km to the east where it intersects the exposure of the

71

granite. The exposure of granite is continuous eastward to Pikes Peak and it
encircles the upper elevations of the entire Florissant basin. This
relationship suggests that the deposits of section 8/23/84-1 possibly
represent an interior area of the lake. up to “0.8 km from the shoreline.

The implications of water depth at the two depositional sites may be
supported by palynological content of the sediments. Figures 2 and 3 show
that W pollen was significantly more abundant in the sample
suites at locality 8/5/84-1 than 8/23/84-1. Water depth requirement varies
from shallow to deep among modern Potampgaton species. Comparison of
the two localities in light of the above physiographic criteria suggests that
the Florissant species of W was adapted to shallow water. During
lake shale deposition. the island shoreline of section 8/5/84-1 was likely
relatively nearshore and shallow due to the close proximity of the granite
outcrop at the top of the island. Abundant Rotampgetpn pollen in most
samples supports this interpretation of water depth. At locality 8/23/84-1.
where the site of deposition was further removed from the shoreline. water
depth may have been greater. Deeper water could also be attributed to close
proximity (<2 km) of this locality to the postulated site of the dam to the
south that formed the lake. The low abundance and common absence of
W pollen in many samples supports a greater relative water
depth. The single spike of Pptammatpn pollen near the top of the section
may represent an interval of relatively shallow conditions following normal
infilling. Alternatively. it is possibly directly related to the sudden >1 m
deposition of ash. Temporary absence of Eptampgaton pollen in the sample
. immediately above the ash might be attributed to burial of plants or
turbidity of the water as a direct result of the depositional event. Decreased
depth may have influenced the subsequent recovery of W. The

72

relatively shallow depth requirements suggested for Florissant W
spp. are conjectural but consistent with available inferences regarding the
paleobathymetry of the lake basin. If 201.1111ng species adapted to
greater water depths were present. it is unlikely that they contributed
significantly to the composite pollen data. It should also be emphasized that
”shallow vs. deep" water postulations are strictly relative since the absolute
bathymetric distribution of rooted. sub merged aquatics such as W
is controlled by factors such as water turbidity. thermal stratification. and

seasonal oxygen flux. all relatively unknown for Lake Florissant.

Correlation

Lithologic and stratigraphic correlation of Florissant Lake Beds exposures
have been unsuccessful in earlier studies due to post-depositional faulting.
strong lateral variation. in sediment thickness. and differential erosion.
Trends in pollen profiles also have provided no correlation of the two
sections studied here.

The base of the Upper Tuff is a consistent time marker. It caps the
lakebed exposures in both sections. as well as at most other lakebed
exposures throughout the valley. but correlation of the lakebeds
immediately below the Upper Tuff is not known. The sequence and
thickness of the layers below the Upper Tuff is different in. the two study
sections. Varying amounts of the upper lakebed layers were probably
removed by erosion prior to the deposition of the Upper Tuff.

The two collection localities are approximately equidistant from the
center of volcanic activity ~25 km southwest (Guffey volcano). It would be
reasonable to expect that an ashf all deposit more than 1 m in thickness

(upper surface "0.6 m below Upper Tuff contact) in section 8/23/84-1 would

73

be laterally continuous for more than the ”2.6 km which separates the two
localities. although its thickness could be somewhat different. There is no
evidence that this unique deposit is present in the section exposed at locality
8/5/84-1. This suggests that one or both of the sections has had a significant
amount of material eroded away prior to deposition of the Upper Tuff. Other
studies (e.g.. MacGinitie. 1953) have suggested that the Upper Tuff was
deposited on an erosion surface because fragments of lake shales are
embedded in the Upper Tuff matrix and the contact is highly irregular.

The two sections do not necessarily represent different regions of the
same water body. If the paleovalley was broad and relatively flat across its
width. the lateral meandering of the sediment-laden interdigitating streams
would have distributed materials in different areas at different times. If
this was the case. it would have been possible for a number of ponded areas.
some possibly large. to exist in such a setting (MacGinitie. 1953). At times
they would infill with sediment or their containments would be breached
and the water would be displaced to a lower location to form a new body of
water temporarily. This scenario would account for different depositional
sequences of the same types of sediments in disjunct areas. I

Calcareous layers were observed in the upper 2 m of the two study
sections. They may be a time marker which represents temporary relative
aridity in the region. Other mechanisms for deposition of the mineral calcite
are possible (e.g.. caliche formation. surface weathering. circulation of vadose
water. algal activity) and would not necessarily occur in different areas of

the lake contemporaneously.

Perspectives on Regional History

Independent studies of the Florissant Lake Beds have provided a large
body of information about the biotic and geologic history of the Florissant
region 34 million years ago. The history of the region may be summarized
from two perspectives. First. the sequence of events that led to the diverse.
fossil-rich lake deposits over a moment in geologic time may be assessed.
Second. inferences about Florissant paleoecology. paleoclimatology and
paledgeography may be compared with inferences from studies of other
fossiliferous deposits of the Rocky Mountain region that existed from the

Eocene through the Oligocene.

Sequence of Events During Accumulation of the Florissant Deposits

Extrapolation of depositional rates of modern sediments similar to the
four types of lamination in the Florissant lacustrine shales was used by
McLeroy and Anderson (1966) to estimate the duration of the lake phase.
Based on the maximum known thickness of the deposits and the relative
proportion of each type of lamination. the estimate at one locality was 2500-
5000 yr. It would be increased if erosion of the lakebeds did not precede
deposition of the Upper Tuff and if the Lower Tuff represents an early stage
of the lake during a period of relatively rapid deposition. Nevertheless. the
estimate would probably remain less than 10.000 yr and the deposits may
be considered a window into a moment of the early Oligocene.

The volcanic center that supplied the sediments to the Florissant basin
was at a critical distance for the quality and quantity of fossil preservation
present. MacGinitie (1953) identifies the source as the Guffey "volcano". "24
km southwest of Florissant. The Lower Tuff represents a relatively active

period of volcanism and was deposited more rapidly than the Lake Shales.

74

75

During Lake Shale deposition, output of volcanic material was frequent
enough to supply fine sediments via stream input. and. to a lesser degree.
direct airf all. for the extraordinary preservation of leaves and insects.
However. the magnitude of the eruptions was not enough to destroy the
previous deposits via lahars. ignimbrites. and lava f lows. for perhaps
thousands of years.

Continuous. pulsating regional volcanism provided clastic materials that
preserved and calibrated the Florissant basin history. Below the volcanic
deposits lies the eroded Pikes Peak granite valley and associated alluvium
and colluvium. Prior to lake formation. the onset of volcanism is recorded by
welded tuff s and lava flows of the Thirtynine mile volcanic field. As the
volcanism continued throughout and beyond the lake phase. erosional
products of these materials from the uplands were transported to the
Florissant basin by the streams of its watershed and deposited as water-laid
tuffs. .

Formation of the lake was probably caused by a volcanic barrier to
drainage. such as a lahar or lava flow that crossed the lower end of the
valley. The timing of this event is not certain. Most studies have placed it at
the termination of the Lower Tuff. Palynological evidence of standing water
suggests the possibility that drainage was obstructed at the onset or during
the early stages of Lower Tuff deposition.

The paleoecological. paleoclimatic. and paleogeographic inferences during
lake deposition as postulated by MacGinitie (1953) are compatible with the
results of subsequent studies. The composite plant megafossil assemblage
indicates a warm temperate climate. with less than 50-60 cm of precipitation
and an average annual temperature of more than 18' C. Rainfall was

concentrated in the late spring/early summer. Absolute minimum annual

76

temperature was greater than -6.6' C. The regional terrain was
mountainous: local relief around the Florissant basin was low to moderate.
The lake was at an elevation of "300-900 m. and the surrounding volcanic
uplands were at "1500 m.

In contrast. MacGinitie (1953) noted the presence of megafossils of both
fir (Ahies) and spruce (Bicea) and both taxa are present in the pollen flora.
Although not common. both taxa were probably growing at no great diatance
from the lake. based on the megafossil occurrences. According to Fowells
(1965). no North American species of fir are known where precipitation is
less than 51 cm and no spruces occur where precipitation is less than 65 cm.
It may well be that MacGinitie‘s precipitation estimate is low. particularly
given his relatively high estimate for paleotemperature.

A similar climate and mosaic of vegetation exists today in the northern
Sierra Madre of northeastern Mexico (MacGinitie. 1953). MacGinitie's list of
fossil species indicates a mixed conifer forest. woodland. and scrub setting.
His list of similar living species in no way implies a tropical or subtrOpical
climate. despite the presence of some hardy subtropical elements that
adapted to more temperate climates upon regional cooling after their Eocene
northward migration (MacGinitie. 1953).-

Entomological evidence. primarily from the Diptera. supports this
paleoclimatic interpretation (MacGinitie. 1953). Other fauna that have been
recovered from the shales include birds (Cope. 1881; Allen. 1878). fishes
(Cope. 1879). and a marsupial (Gazin. 1935).

Probable dessication of the lake was followed by an indeterminate
. period of erosion prior to deposition of the Upper Tuff. The Upper Tuff
represents a relatively violent episode of volcanism where the basin slopes

accumulated large amounts of volcanic debris to the point of instability. The

77

subsequent lahar(s) capped the uneroded remnants of the lakebeds in the

Florissant valley.

Comparison With Other Early Tertiary Rocky Mountain Floras

A number of other floras that are close to the Florissant flora spatially
and/or temporally have been the subject of other studies. Comparison of
several of these studies with information from Florissant can broaden the
understanding of the development of Rocky Mountain floras and their
physical environments during the mid-Tertiary.

Two floras of southwestern Montana have the most striking floral (leaf
fossil) similarity to Florissant known to the author. The Ruby paper shales
also accumulated in a lacustrine basin. ”1100 km north-northwest of
Florissant. Several species are found in only the two f loras that have in
common 32 species and 36 genera. Given this similarity. Becker (1961)
suggested that despite their wide geographical separation. the two f loras
represent a common botanical province and must have been floristically
connected along the central and eastern Rocky Mountains. Becker (1961)
estimated the age of the Ruby flora to be late Oligocene and the age has since
been refined to 30.8-29.2 ma (Wing. 1987). Approximately 65 km further to
the north. the Beaverhead Basins flora is slightly younger ("32 ma) than the
Florissant flora and more taxonomically diverse. 40 species are common to
both floras (Becker. 1969). All three f loras share a strong representation of
members of the Pinaceae. Fagaceae. Ulmaceae. Rosaceae. Eagppsis longifplia
under a dry/warm-temperate climatic regime.

The Red Rock Ranch flora of the San Mateo Mountains. New Mexico ("625
km south-southeast of Florissant). represents an ecotonal broad-leaved

sclerophyll and mixed conifer forest. almost contemporaneous ("35 ma) with

78

Florissant and at a slightly higher elevation of ”1280 m (Axelrod and Bailey.
1976). The megafossil record shows considerable relationship to Florissant.
including abundant Elma and Pings; Eagppsis longifplia is also present.

The closest flora to Florissant in distance and time is the Silver Cliff flora.
”88 km to the south. dated “35 ma (Axelrod and Raven. 1985). Diversity is
limited to 5 species that represent a semiarid woodland-scrub of southern
Cordilleran affinity. It has a drier aspect than Florissant. which Axelrod and
Raven (1985) attribute to a site of western exposure.

All of the floras described have been attributed several common
characteristics. They represent a mixture of coniferous and broad-leaved
deciduous forests growing under a seasonally dry subhumid climate at an
intermediate elevation (Wing. 1987). They cover an age range of ~30-35 ma
(early Oligocene). The more diverse floras (Le-Florissant. Ruby. Beaverhead)
probably represent a broader region of source plants that were ecologically
divided into distinct paleoassociations (see Paleoassociations section).

Early to middle Oligocene f loras of the northwest United States and
southwestern Canada were generally of lower elevation and received
orographic rainfall on the lee side of the uplifting Rockies. as well as higher
average temperatures and a narrower annual range of temperature.
Therefore. they tend to have more subtropical elements which grew under a
relatively humid climate (e.g.. Frazer River Fm.. British Columbia; Coos Bay
and Goshen floras. Oregon).

Early Oligocene floras as a group record a shift to a cooler and drier
climate than that of the late Eocene and the boundary between the two
periods marks the transition to modernization of angiosperm genera in the
Rocky Mountain region (Wing. 1987). Throughout most of the Eocene. broad-

leaved evergreens were the dominant component of the forests; broad-

79

leaved deciduous and conifer stands primarily occupied the uplands (Wing.
1987). Examples of the subhu mid subtropical to temperate woodlands of the
middle to late Eocene are the Yellowstone National Park f loras and the Green
River flora (Axelrod and Raven. 1985; Wing. 1987). Subsequent climatic
deteoriation may be attributed to orographic effects of uplift and the
accumulation of thick volcanic piles. as well as southward movement of the

North American Plate (Axelrod and Raven. 1985).

CONCLUSIONS

Palynological analysis of the Florissant Lake Beds has contributed to the
reconstruction of the regional vegetation and environmental conditions
during the early Oligocene. The focus of this study has been primarily to
verify or amplify interpretations of former studies which have concentrated
on megafossil assemblages (e.g.. MacGinitie. 1953) and mechanisms of
sedimentation (i.e.. McLeroy and Anderson. 1966). The important

contributions of this study may be summarized:

(1) The composite palynological record of this study contains many of the
I plant genera reported in megafossil studies (primarily MacGinitie. 1953)
and other palynological studies (primarily Leopold. 1969 a). 78

pollen/spore types are described. Five unreported taxa have been added

to the systematic list of plants: Maugectia (Zygnemataceae). Padmamus
(Podocarpaceae). Milan/Lari; (Pinaceae). Liguidamhan
(Hamamelidaceae). Nymphaeaceae (undifferentiated). .

(2) Composite abundances of taxa differ significantly between megafossils
and pollen/spores (Table 2). Production. transportation and preservation
biases account for the disparity in representation of different plant
organs in the fossil record.

(3) Stratigraphic comparison of the relative abundance of pollen/spores
demonstrates considerable variability in dominance of different plants
over the course of the lake history. This variability is attributed to
several factors: volcanic disturbances. basin infilling. short-term climatic
changes.

(4) The aquatic plant association. dominated by Epiampgetgn, demonstrates

the greatest variability in pollen representation of all paleoassociations.

80

81

Variation in W spp. at different stratigraphic levels indicates
response to the marked dynamism of lake conditions (e.g.. sudden
episodes of deposition. variable lake levels. water quality) regulated by
regional volcanism and climate.

(5) As in the megafossil record. Lakeside Forest vegetation dominates the
pollen/spore record. TCT-type pollen is the most abundant overall in this
study. Limited taxonomic resolution precluded differentiation below the
level of the TCT family-group; therefore. ecology based on the pollen is
problematic. Of the megafossil taxa present. the pollen most closely
resembles that of gnamaagypapis and some of the silicified stumps may
represent this genus rather than Sequoia. Although abundant as leaf
megafossils. characteristic Saqupia pollen was not encountered.

(6) Paleoassociations representing plants on the mesic bottomlands and
slopes of the basin did not fluctuate very much in abundance or diversity.
Subaerial deposition from regional volcanism apparently did not
significantly alter the terrestrial vegetation near the lake margin for the
time interval studied. There is no evidence for catastrophic destruction
of terrestrial plants or for succession on a disturbed landscape.

(7) Throughout its history the size of the- lake was affected by the efficiency
of the containment dam(s). water input from the paleowatershed. and
evaporation rates. Late in its history. following extensive inf illing. the
lake fluctuated in area/volume. The most probable cause for these
conditions was relatively dry periods where water input from the
watershed declined below the annual evaporative/leakage water loss
from the basin. Given the limited size of the watershed and a regionally
warm climate. indicated by the f lora. this need not have been a "major"

climatic event. If similar dry cycles occurred earlier in the lake history.

82

their effect would probably have been obscured by the buffering effect of
a larger lake water volume.
(8) Pollen trends in Eptamogetpn support inferences of relatively shallow
water adjacent to island shorelines and deeper water in the open lake
basin.
(9) An alternative model for annual diatom blooms. driven by spring
turnover in a thermally stratified lake. is proposed to account for the
diatomite/sapropel couplets noted by McLeroy and Anderson (1966).
(10) Palynological correlation of the two sections sampled was not possible.
(11)Eagppsis lpngifplia which constitutes >30% of the leaf megaflora. was
relatively rare in the pollen record (av. << 1 %/sample). This disparity
may be a result of low pollen production coupled with a high
concentration of plants near the lake margin 3

(12) Infilling/drainage of the lake(s) had been completed. followed by an
indeterminate period of erosion. prior to the deposition of the Upper
Tuff unit.

(13) The overall patterns of vegetation dynamics observed in the study
sections are compatible with the relatively short lake history suggested
by McLeroy and Anderson (1966).

(14) Presence of W pollen in the Lower Tuff raises the possibility

that the lake may have for med at the onset of or during Lower Tuff

time.

Further study of the Florissant Lake Beds should continue to focus on
plant succession and correlation of the disjunct exposures. Given the highly
dynamic physical conditions of the region during the lake phase. a more

detailed palynological analysis utilizing closer sampling. possibly integrated

83

with stratigraphic studies of the megaflora and megafauna. should provide
much more information. Interpretation of the geologic and biotic history
preserved in these deposits will be fragmented until the temporal

relationships of the exposures are better understood.

DESCRIPTIVE NOTES
The average overall abundance and frequency of occurrence of pollen in
samples for the described taxa/ morphotypes is given according to the
following definitions:
highly abundant
- observed in more than 50% of the samples
- average at least 10% per sample
abundant
- observed in more than 50% of the samples
— average 2-9.9% per sample
common
- observed in more than 50% of the samples
- average less than 2% per sample
infrequent
- observed in 20-50% of the samples
- average less than 2% per sample
rare
- observed in 10-20% of the samples
- average less than 0.5% per sample
very rare
- observed in less than 10% of the samples

- 10 or fewer grains observed in the entire study

84

85

CRYPTOGAMS

ALGAE
Discussion:

MacGinitie (1953. p. 20) made single mention of the common presence of
impressions of the fresh-water alga NIIEJIa (Characeae) but did not include
any algae in his systematic list. Algal filaments were observed several times
in the palynological samples studied. Their affinity is unknown. Spores
from Nitalla were not differentiated.

In: 1
(Plate 4-33) Pb 14137-1 40.9/ 120.8
Description:

Single strands of circular to elliptical cells; average diameter of the cells
decreasing from one end of a filament to the other: max. 10 microns. min. 4
microns. Observed length of filaments variable. 15-45 microns.

Discussion:

Filaments very rare.

ZYGNEMATACEAE
Mountain Agardh.
(Plate 4-29) Pb 14262-4 47.2/ 128.7
Description:

Zygospores quadrate with two sides more or less straight or concave and
two sides with V-shaped indentation in middle which appears to be an
incipient fold. Thickness of wall quite variable. psilate. Usually oval
depression in center which resembles a pore (dimensions 7 by 4 microns).

Size of spores 18-22 microns. square; often clumped together.

86

Discussion:

The Florissant spores resemble modern W3 spp. (cf. Bold and
Wynne. 1978. fig. 3.173 c.d) and Netrium spp. (cf. Bold and Wynne. 1978. fig.
3.174 b.c).

Cosmopolitan in distribution; presence of Mpugaotla spores may be
indicative of oligotrophic lakes in cool to cold climates (Round. 1965) or
stagnant. shallow. mesotrophic f resh-water habitats with optimum growth
temperature 10-15 °C (van Geel and van der Hammen. 1978).

This green alga is generally very rare but abundant in one sample of
section 8/5/84-1.

FUNGI
Discussion:

MacGinitie (1953) did not report any fungi in his study of the Florissant
megafossils. Fungal spores were rarely encountered in this study. When
present. they were usually clumped together into groups of 4 or 5.

Two morphotypes are recognized; however. few specimens and limited
morphological characteristics precluded concise descriptions and the two
groupings are generalized. Further study is necessary to develop
understanding of the true diversity and systematic relationships of the
Florissant fossil fungi. Their affinity is uncertain.

Ixncl
(Plate 1-2) Pb 14137-1 46.0/111.7

87

Description:

Spores circular in outline. usually folded with crease at center: diameter
10-15 microns; when separated. points of former attachment are not easily
identified.

Discussion:
Spores very rare.
T192 2
(Plate 1-3) Pb 14137-1 359/1156
Description:

Grains of this type are generally larger than those of Iy_p_e_ l and are
elliptical in outline. Dimensions 20 by 24 microns for the specimen
illustrated. Surface is uniformly psilate with random notches at the margins
or perforations of the wall. Former points of attachment to adjacent spores
are torn in two places on opposite sides along the long axis of the grain.
Discussion:

Spores very rare.

MONOLET E SPORES
Discussion:

Monolete spores are present in more than three-quarters of the samples.
usually contributing from 2-4% of the pollen sum (max. 15%). They are more
abundant. but less diverse morphologically. than trilete spores.

The monolete spores are probably members of the family Polypodiaceae.
The only fern that MacGinitie (1953) included in his systematic list is
Willi (Polypodiaceae). Common as a group.

88

1m 1
(Plate 1-12) Pb 14137-1 30.1/1219
Description:
Size range 17 by 29 microns to 20 by 36 microns. always folded (usually
one major fold). Psilate.
Discussion:
This type accounts for more than 90% of the observed monolete spores.
Tine 2
(Plate 1-4) Pb 14038-2 298/1286
Description:
Size range 30 by 35 microns to 42 by 50 microns. Psilate.
Discussion:
Grains of this type comprise approximately 8% of the monolete spores.
Imci
(Plate 1-9) Pb 14138-1 34.5/1160
Description: I '
Reniform. thick-walled and psilate. Dimensions 28 by 37 microns.
Discussion:
Only one specimen encountered. Possibly modern contaminant.
Tine 4
(Plate l-ll) Pb 14020-2 31.9/122.0
Description:
Reniform. conspicuously verrucate-coarsely reticulate. size 28 by 38
microns.
Discussion:
One specimen observed.

Possible botanical affinity:

89

Possibly a modern contaminant. Drums-2:15 .7 (Polypodiaceae)
(remark: MacGinitie included madam (Lesquereux) MacGinitie in

his systematic list.

TRILETE SPORES
Discussion:

Trilete spores are present in about 1/3 of the study samples. never
exceeding 3% of the pollen sum. While numerically less abundant than the
monolete spores. the trilete types are morphologically more diverse.
Leopold (1969 a) illustrated three trilete spores in a plate of selected
Florissant palynomorphs. She identified two as Selaginella spp.
Infrequent as a group.

Tins 1
(Plate 2-15) Pb 14137-1 354/1190
Description:

Trilete suture open. suture radii extending approximately 1/2 way to
equator. Amb circular; maximum equatorial diameter 18-23 (av. 22)
microns; exine coarsely rugulate.

Discussion:
Accounts for approximately 15% of trilete spores observed.
Possible botanical affinities:
Wfirhart ? (cf. Leopold. 1969 c. fig. 3)
1.1121212951111111 L. 7
ED! 2
(Plate 2-12) Pb 14137-1 39.4/1237

90

Description:

Trilete suture closed. rays undulatory (highly contorted) and extending
almost to equator; lip surrounds triradiate suture. Amb sub-triangular;
maximum equatorial diameter 20-24 microns; exine psilate.

Discussion:

Most common trilete spore (approximately 40% of total trilete spore
count).

Possible botanical affinitiy:

Sphagnum Erhart .7 (cf. Leopold, 1969 c. fig. 2)
1.2912125111111111 L. .7
In: 3
(Plate 2-6) Pb 14134-1 29.7/129.l
Description: _

Trilete suture closed. rays extending nearly to equator. Amb sub-
triangular with broadly rounded equatorial "apices"; maximum equatorial
diameter 25-32 microns. Surface psilate.

Discussion:
Approximately 10% of total trilete spore count.
I22: 4
(Plate 2-8) Pb 14137-1 36.4/117.7
Description:

Trilete suture open. rays extending approximately 3/ 4 distance to
equator. Amb sub-triangular with broadly rounded apices; maximum
equatorial diameter 23 microns. Exine psilate and very thin.
Discussion:

Spores very rare.

91

In: 5
(Plate 2-7) Pb 14157-1 299/1279
Description:

Trilete suture closed. rays extending approximately 3/ 4 distance to
equator. Amb circular; maximum equatorial diameter 54 microns; exine
sparsely pitted.

Discussion:
Spores very rare.
Ms 6
(Plate 2-10) Pb 14137-1 40.1/121.l
Description:

Trilete suture open. rays extending approximately 3/4 distance to
equator. Amb sub-triangular with broadly rounded equatorial "apices”.
Trilete rays narrow. abruptly to pointed ends; maximum equatorial diameter
23 microns.

Discussion:
Only one specimen observed. Possibly a modern contaminant.
11m 2
(Plate 2-11) Pb 141-37-1 30.0/121.4
Description:

Trilete suture closed. rays extending approximately 2/ 3 distance to
equator. Amb sub-triangular with broadly rounded equatorial "apices";
maximum equatorial diameter 40-55 microns. Surface psilate except on
proximal surface where it is granulate.

_ Discussion:

Accounts for approx 15% of trilete spores.

92

1X2: 8
(Plate 2-14) Pb 14038-3 402/1118
Description:

Grains with cingulu m; width of cingulum approximately 5 microns.
Relatively small trilete suture. closed. rays extending approximately 1/2 way
to equator. Amb sub-triangular with broadly rounded equatorial ”apices";
maximum equatorial diameter 37 microns (not including cingulum). Surface
granulate.

Discussion:
Spores very rare.
Possible botanical affinity:
Selaginella Spring (cf. Leopold. 1969 a. figs. 2.3; Martin and Rouse.
1966. pl. I. figs. 7.8)
119.: 9.
(Plate 2-16) Pb 14038-2 31.5/119.2
Description:

Grains with cingulum. width of cingulum approximately 2-3 microns.
Trilete suture closed. rays undulatory with thickened margins and extending
to equator. Amb sub-triangular with broadly rounded equatorial "apices";
maximum equatorial diameter 25 microns (not including cingulum). Surface
. finely-granulate.

Discussion:

Spores very rare.

Possible botanical affinity:

Selaginella Spring (cf. Leopold. 1969 .1133. 2.3; Martin and Rouse.
1966. pl. 1. figs. 7.8)

93

GNETOPHYTA

Ephedra L.
Discussion:

MacGinitie (1953) tentativtly identified one species of Ephadpa (Ephedra
mimeniaa Wodehouse) from the impression of a branch. although he noted
that the preservation of the specimen precluded certain identification.

Leopold (1969 a) illustrated one specimen of Bahama typical of those
encountered in this study. in her plate of selected Florissant palynomorphs.
Similar to modern E. nexadensis.

Modern Epheglra spp. are xerophytic shrubs restricted to temperate and
tropical areas of southwest United States. Mexico. South America and the Old
World. I

Pollen infrequent in section 8/5/84-1 and common in section
8/23/84-1. Maximum 1.3% in one sample.

In: 1
a (Plate 1-5) Pb 14142-3 322/1139
b (Plate 1-6) Pb 14151-1 34.9/116.1
Description:

Grains polyplicate with meridional folds; prolate. Shape of poles varies
from knob-like to evenly and broadly rounded. Ridges (more than 5
discernable) straight and often obscure; furrows branched and extending
almost to poles. Branching sometimes circuitous giving a reticulate
appearance. Size dimensions 15 by 42 microns to 24 by 58 microns. Two
grains are illustrated to show the variation in size and shape.

Discussion:

This type includes almost all of the observed Bahama specimens.

94

In: 2
(Plate 1-8) Pb 14136-2 39.0/123.1
Description:
Grains relatively small (14 by 41 microns). Furrows not branched and
terminate before reaching poles.
Discussion:

Pollen very rare.

95

CON I FEROPHYT A

PIN ACEAE
Able: L.
Description:

Grains vesiculate-bisaccate. They closely resemble specimens of Einua
spp. in the distinction between the bladders and the body of the grain. as
well as the angle of bladder attachment. Fir pollen. however. is generally
significantly larger. with coarser exine on the body. a thicker cap. and loosely
appearing attachment of bladders to the body. Exine is reticulate or
granulate-reticulate. A prominent "shoulder-like" indentation is evident
where the bladder is attached to the cap. Cap 3-6 microns thick. often
thinning near proximal pole; marginal ridge rough.

Discussion:

The seed of one species of Able: (A. lgngipaslpis Knowlton) was observed
by MacGinitie (1953. p.20). He compared the exserted bracts of the cone
scales to those of extant A. mum of California.

Axelrod (1976) discussed the resemblance of A. lgngimstzia seeds from
the Creede flora of Colorado to A. chaneyi Mason from the Mascall and
Stinking Water floras of Oregon and suggested a relationship to modern A.
dalaxayi of southern China.

Leopold (1964) noted that even when fir trees are abundant regionally.
their pollen tends to be under-represented in surface samples due to rapid
settling from the air. This suggests that if fir pollen is abundant. fir trees
must have grown quite close to the site of accumulation.

Fir trees occupy upland (montane) sites. generally associated with spruce

in boreal and alpine forests in North America.

96

Three types are described and illustrated to demonstrate the range of
variability of pollen within the genus. A gradient of types between these
three examples characterizes the specimens observed.

Pollen rare in section 8/5/84-1 and abundant in section 8/23/84-1.
Maximum 12.8% in one sample.

1212s 1
(Plate 1-10) Pb 14151-3 369/1160
Description:

Body dimensions 98 by 86 microns in equatorial view. appearing
squared-off. Bladders rounded-rectangular. the long axis of each oriented
approximately parallel with polar axis.

In: 2.
(Plate 1-7) Pb 14151-1 45.2/119.6
Description:

Body more elliptical and elongate in shape than that of Type 1;
dimensions 115 by 67 microns in equatorial view. Bladders attached to
body at approximately 45 degree angle to the polar axis.

m: 3
(Plate 1-1) Pb 14151-1 35.5/111.8
Description:

Body elliptical. dimensions 113 by 58 microns in equatorial view.
Bladders attached to body at approximately 45 degree angle to the polar
axis. Bladder outline diffuse at attachment to body. the two separated by a
straight boundary. A membrane connects the two bladders at the distal

. pole.

 

97
Pippa L.
Description:

Spruce pollen is distinguished from other bisaccate pollen by the
characteristic smooth transition between bladder and cap; the bladders and
the body together constitute a grain of continuous (uninterrupted) outline
which is overall oval-shaped in equatorial view. Body appears granulate
with smooth to reticulate bladders. Two morphotypes are illustrated.
differing mainly in size.

Discussion:

Two species of 12mm. laughing MacGinitie, P. magpa MacGinitie)
were differentiated by MacGinitie (1953). However. they together
comprised less than 1% of the composite megaflora.

Leopold (1969 a) illustrated two specimens of Egaa from the Florissant
sediments.

Spruce trees typically occupy montane and boreal environments and
some coastal lowlands. Associated trees are fir and sometimes birch and
aspen in modern high latitudes.

Pollen rare in section 8/5/84-1 and infrequent in section 8/23/84-1.
Maximum 1.2% in one sample.

I19: 1
(Plate 2-1) Pb 14151—2 35.9/1244
Description:
Grain length 130 microns.
I19: 2
("Plate 2-9) Pb 14019-1 382/1201
Description:

Distal polar view of grain (length 90 microns).

98

Pippa L.
Description:

Florissant pine pollen is here considered as a complex with several
uniting characteristics. Grains are vesiculate-bisaccate. Body size ranges
from 44 by 22 microns to 70 by 50 microns: body always larger than
bladders. Exine of body granulate to f mely-reticulate. bladders coarsely
reticulate. Body with cap 1 to 2 microns thick. Position of bladder
attachment to body is a major differentiating characteristic. Bladders
generally appear well fused to body. the margins at edges of attachment
usually diffuse. Five specimens are illustrated to show the morphological
diversity of the group.

Discussion:

MacGinitie (1953. p. 21) differentiated three species (12. ljlppissaati I
Lesquereux. P_. hamhaphl Kirchner. E. Mani Cockerell) on the basis of
leaves and winged seeds. and indicated that more could likely be identified
with further analysis. Material from the three species composed 1.6% of the
megaflora.

Leopold (1969 a) illustrated one pollen grain of anus. Pine pollen is
common in all productive samples of this study contributing from 1-68‘t (av.
“11%) of the pollen sum.

Extant pine species have a very broad ecological amplitude and
cosmopolitan distribution. Their anemophilous pollen is produced in large
quantities which may travel great distances from the source. Therefore. low
to moderate pine pollen levels suggest a relatively insignificant role for pines
in paleoecological reconstructions.

Pollen abundant in section 8/5/84-1 and highly abundant in section

8/23/84-1. Maximum 67.8% in one sample.

99

Israel
(Plate 3-6) Pb 14137-1 398/1228

Description:

Body highly elongate perpendicular to polar axis. dimensions 47 by 23
microns. Bladders small with respect to body and circular in outline:
diameter “22 microns; attached on flattened distal side of body and
extending very little beyond body sides.

Discussion:
Pollen very rare.
Time 2
(Plate 3-2) Pb 14144-1 33.0/115.2
Description: _

Body dimensions 50 by 35 microns; cap rounded and remaining three
sides somewhat polygonal. Bladders large with respect to body. diameter
'30 microns. attached at approx 45 degree angle to polar axis. Bladders
almost meet at distal pole.

Discussion:
Pollen very rare.
Bins 3
(Plate 3-5) Pb 14151-3 37.3/1158
Description:

Body dimensions 62 by 34 microns; distinct shoulder at attachment of
bladders. Bladder diameter "37 microns. Bladders meet at distal pole and
extend well beyond th width of the body.

Discussion:

Pollen very rare.

100

Ixnfi 4
. (Plate 3-3) Pb 14151-1 472/1195
Description:
Body dimensions 57 by 30 microns. Bladders circular in outline.
diameter ~35 microns. Bladders attached at approximately 35-40 degrees

off polar axis. well separated at distal pole.

Discussion:
About one-half of the Pinus grains observed.
T112: 5
(Plate 3-4) Pb 14157-1 36.0/129.0
Description:

Body dimensions 63 by 45 microns, Bladders distinct and elliptical in
outline; dimensions 50 by 30 microns. Long dimension of bladders parallel

with polar axis; well separated at distal pole.

Discussion: .
About one-half of the Eipus grains observed.
Esaudptsuga Carr/Larix Mill.
(Plate 2-13) Pb 14151-4 33.4/118.2
Description:

Ellipsoidal grains. often folded and/or ruptured to form a f urrow-like
area. Psilate exine. 85->110 microns in long dimension.
Discussiom

Since only two pollen specimens have been observed and the
distinguishing exine characteristics are very limited. it is difficult to assign
these grains to either W23 or Lam. Descriptions and plates of

modern species of these two genera usually ally these larger grains to

We spp.

101

There is no report of either Esaugptsuga or Lam in previous studies of
the Florissant fossil leaves or pollen and spores.

Douglas-fir is a common element of coastal and interior forests of
northwestern North America. It is also present in scattered forests of
southwest China. Japan and Taiwan. These trees are most commonly found
on well-drained. loamy sites where an abundance of moisture is present.
Some forest associates in the Pacific Northwest include Inuja. Isuga Pippa
and Allies.

Larch and tamarack are widely distributed in north temperate and
boreal forests of North America. Europe and Asia. In the southern part of
their range. these intolerant trees occupy cool swamps and bogs. Further to
the north. they are commonly found on moist benches and better drained
uplands. Common associates include Apias. Eppulua. Belula and some Einus.

Two specimens were observed in a single sample (8/5/84-1. 61) of this

study.
PODOCARPACEAE
Epdppappus L'Herit. ex Pers.
(Plate 3-1) Pb 14016-2 382/1174
Description:

Grains vesiculate-bisaccate. As compared to bisaccate members of the
Pinaceae. the body of Badmarpua-types is relatively small with respect to
the size of the bladders. Body dimensions 48 by 39 microns: bladder
dimensions 60 by 46 microns. Bladders appear loosely attached to body.

. Cap highly convoluted; bladders coarsely reticulate.

Discussion:

102

Fossils of Rodmrnus have not been noted in any previous studies of the
Florissant flora.

These evergreen trees and shrubs today are widely distributed in
tropical to warm-temperate moist woodlands and mountains worldwide. but
are concentrated in the southern hemisphere (Piel. 1971: Bassett. et. al..
1978). Studies suggest presence of 29.0mm fossils as indicative of upland
habitats because representation is generally limited to rare shoot fragments
and pollen (Hopkins. 1967; Smiley and Rember. 1981).

Dilcher (1969) identified Roam: cuticle from a leafy shoot in Eocene
deposits of western Tennessee. He noted that the pollen record of
depparpus has formerly not been considered to extend into Tertiary
sediments of North America. However. megafossil reports of 2131922132113 in
the Mississippi embayment area and wood from Eocene sediments in
Washington indicate that Wwas present in North America during
the early Tertiary. He concludes: "The Eocene Epdpcappua described in this
paper and the numerous reports of 2051mm pollen from Mesozoic
sediments in widely scattered areas of the northern hemisphere are
remnants of a much more northern extension of depcarpus rather than a
recent northernmost expansion of a southern ele ment" (Dilcher. 1969. pp.
300-301).

Subsequent reports of Purim pollen have documented its presence
in North America well into the Tertiary. In a palynological study of early
Oligocene Alexandria-Narcosli sediments from central British Columbia. Piel
(1971) identified three distinct types of Warm pellen. HOpkins (1967)
identified Epdpcamus pollen from Eocene-Oligocene sediments of the Coos
Bay area. Oregon. Epdmappua pollen has been recovered from sediments as

young as the early Miocene Clarkia Lake flora of northern Idaho (Smiley and

103

Rember. 1981) and the middle Miocene Succor Creek flora of Idaho-Oregon
(Taggart. 1971). It is possible that deppanpus spp. of western North
American Tertiary floras were derived from temperate Asian forms which
inhabited sheltered mesic lowlands rather than exposed upland sites as
many southern hemisphere types occupy today (Taggart. 1971).

Only three specimens were observed in the samples studied. limited to

different samples of section 8/5/84-1.

TAXODIACEAE. CUPRESSACEAE. TAXACEAE
Description:

Pollen of the families Taxodiaceae. Cupressaceae. and Taxaceae (TCT;
collectively referred in discussion as "taxodiaceous") are difficult to
differentiate with the light microscope. Based on observations of his
reference collection. Traverse (1955. p. 40) comments that the generic
differentiation of taxodiaceous pollen is not as difficult as some writers
indicate. and he assigns Brandon Lignite specimens to a new species of the
genus W (Taxodiaceae). His primary criterion for distinguishing
genera is the presence. size and shape of a germinal papilla.

This author's observation of Chamampapis lawspniana. Iaxpdium

11151191111111. and Sepupla semperxipens in the Michigan State University
paleobotanical pollen reference collection suggests that the papilla is a

relatively frequent characteristic (approximately 80%) of Sagupia
sempamipens and is absent in Chamaecypazis lawspniana and Iaxpdium
11151191111111. Also variable within each of these genera is the presence of
pores. V-shaped rupture. and elongation of the grains.

On Florissant specimens. the papilla is absent and the other

characteristics are highly variable. If the large fossil stumps are properly

104

identified as members of the genus Sequpia. it is possible that the pollen of
Florissant grains represents non-papillate Sequpia-type. However. the
Florissant pollen specimens are remarkably similar to Chem
lampniana of the Michigan State University reference collection. This is
corroborated by the recovery of nearly three times more leaf fossils of
mammals than Sequoia (MacGinitie. 1953). Therefore. until further
taxonomic study of the specimens is completed. assignment of Florissant
material to any particular taxodiaceous or cupressaceous genus must be
considered untenable: they will be referred to collectively as "TCT".

Leopold (1969 a) illustrated a typical Florissant specimen as
"Taxodiaceae. Iaxedium-type".

Discussion:

The leaf megafossil record includes 5311111213 31111115 (Taxodiaceae. 2.2%).
91131113292319.4113 linguasfplia (Cupressaceae. 6.3%) and Ippreya gepmetrpnum
(Taxaceae. < 0.1%): Median spp. are absent (MacGinitie. 1953). Therefore.
if pollen production volume per plant of each of the represented species is
similar and they were growing under the same ecological conditions.
Chamecxpm's pollen might be expected to be more abundant than that of
Secupia.

Most studies of the Florissant deposits include mention of the abundant
silicified Sequpiastumps (Scudder. 1882. p. 284; Cockerell. 1906. p. 175:
Knowlton. 1916. p. 243; MacGinitie. 1953. p. 21; McLeroy and Anderson.
1966. p. 607). While there appears to be a consensus on the generic identity
of these stumps. the evidence for this association seems to be based on a
unif or mitarian analogy (i.e.. size of coast redwood) rather than shared
anatomical characteristics. The California coast redwood. S. We

has similar foliage and cones to Florissant material (MacGinitie. 1953. p. 21).

105

These trees commonly reach 8-12 ft. in diameter and 340 ft. in height
(Wodehouse. 1959. p. 269). WWW of southwest Oregon

and northwest California is known to reach 16 f t. in diameter with a height
of 240 ft. (Harlow. et. al.. 1979. p. 202). Chamaesypanisfpsmpsensis Matsum.
of Taiwan can reach a diameter of 23 ft. (Harlow. et. al.. 1979. p. 199). The
large diameter of the fossil stumps alone is insufficient evidence for
identification. The abundance of Wis foliage and the comparable
size of the mature trees to the fossil stumps must qualify it for further
consideration. In the absence of comparative anatomical studies (i.e.. thin
sections) of a large number of Florissant wood specimens with modern
taxodiaceous genera. their designation to the genus Sequpia must be
tentative.

Sequpia semperxirens is confined to temperate fog-belt areas of west
coast North America and inhabits moist. but well-drained slopes and valleys.
Extinct Sequpia spp. were widespread in forests of the Northern Hemisphere
during the Tertiary Period. Chanmecxpapis spp. typically occupy freshwater
swamps and bogs or humid forests of coastal eastern and western North
America (Harlow. et. al.. 1979).

In: I
a (Plate 2-3) Pb 14142-1 43.9/115.5
b (Plate 2-4) Pb 14155-1 28.2/121.9
c (Plate 2-2) Pb 14155-1 31.8/118.9
d (Plate 2-5) Pb 14020-2 33.9/1148
Description:
Grains circular in outline. usually (90% of the time) with a ruptured

exine. The exine has a finely granulate surface. occasionally with

106

conspicuous. widely and irregularly spaced verrucae. Diameter ranges from
24 to 32 microns. averaging 30 microns.

Four specimens are illustrated to show the range of morphological
variability within the type. Type 1, (a) is overwhelmingly the most abundant
morphotype encountered.

Discussion:

This is the most abundant pollen type in the Florissant sediments. It is

found in almost all samples. reaching a maximum of "75% of the pollen sum

and averaging ”26%.

107

MONOCOTYLEDONAE

GRAMINEAE
Description:

Given the limited pollen morphological variability between the large
number of grass family genera. as well as the relatively broad ecological
occurrence of the group. this study will not attempt to identify below the
family. level.

Three types are differentiated and illustrated. All are monoporate with
a pronounced. thickened annulus surrounding the pore.

Discussion:

MacGinitie (1953) identified one species of grass represented by fossil

fruits. Sllpa 119nm (Knowlton) MacGinitie.
Grass pollen is common in section 8/5/84-1 and rare in section 8/23/84-1.
Maximum 1.5% in one sample.
Time 1
(Plate 3-10) Pb 14142-2 430/1163
Description:

These grains are psilate and circular in outline. accounting for the
majority of observed grass pollen. Diameter ranges from 19 microns to 25
microns (ave. 23 microns). Annulus is very well-developed; thickened.
Exine is psilate to faintly scabrate. All grains display one or two folds.

Tim 2
(Plate 3-7) Pb 14148-2 29.0/114.3
Description:
Only a few of these grains were observed. It differs from Iype l grains

in having a microreticulate exine.

108

POTAMOGETONACEAE
Rotamoxeton L.
(Plate 3-14) Pb 14023-2 436/1175
Description:

Aporate pollen grains with distinctive coarsely reticulate-rugulate exine.
Grains spherical. ranging from 30-37 microns in diameter (av. ”34 microns).
commonly with 3 or 4 folds. They often (approx. 40% of grains) have a
rounded to angular dark spot (occasionally 2 or 3) which sometimes appears
pore-like. up to 6 microns in diameter; a ghost of the exine generally appears
through the spot.

Discussion:

The genus Bojampgetpn is classified by MacGinitie (1953) inthe category
inpectae seals comprising 0.4% of the leaf megaflora. He also includes in this
category Naiadppsis and W

Pondweed is a floating or sub merged aquatic which may occupy
positions from shallow ponds to deep areas of lakes. sometimes forming
extensive beds. Preferred temperature of the water body is variable among
species. and some species are tolerant of calcareous or brackish conditions.
Because of these broad ecological requirements. evidence of Betampgetpn
pollen will not provide much insight into the depth. temperature or
chemistry of Lake Florissant. Its many species are distributed throughout
the northern United States and Canada.

Lewis. et. al. (1983) note that most eqampgeipn spp. are anemophilous
and produce small quantities of pollen.

Botampgelpn pollen occurs in most of the palynological samples.
contributing up to 79% of the pollen sum (av. 18.8%) and is co-dominant in

most pollen spectra with TCT-types.

109

The large quantities of pollen noted in this study are therefore
suggestive of extensive development of Pptampgetpn "beds" and possibly of

relatively shallow water and/or high water clarity.

TYPHACEAE (SPARGAN IACEAE)
Txpha L. (Spanganium L.)
Discussion:

129.114 W11 Cockerell comprises 5.8% of the leaf fossils. ranking
fourth in abundance among all plant species (MacGinitie, 1953). Since no
megafossil specimens of Spanganium have been reported. the Florissant
pollen is likely Iypha. For inventory and paleoecological discussions in this
study. the grains will be referred to Iypha even though. in the absence of
other evidence. pollen of the two genera is difficult to distinguish.

Leopold (1969 a) illustrated one specimen of prha/Spazganium in a
plate of Florissant palynomorphs which resembles a cross between the two
grains. illustrated here. .

Cattail can be found along riverbanks and in swamps and marshes in
temperate to tropical climates throughout the world. except south of the
equator in Africa. Species of bur reed are widespread temperate to
subtropical aquatics. occupying swampy margins of ponds. lakes and
streams. Some species may be found in deeper waters. Most extant western
North American species are tolerant of acidic. shallow. soft-bottomed.
temporary ponds.

Israel
(Plate 3-12) Pb 14143-1 354/1183

110

Description:

Grains monoporate. spherical and often folded. Pore margin indistinct.
Thin exine with reticulate thickenings. Diameter 25-30 microns.
Discussion:

These specimens occurred in about 1/2 of the palynological samples.
averaging about 1% overall and about 2% when present (maximum 4%).

Pollen common in section 8/5/84-1 and infrequent in section 8/23/84-1.
Maximum 4.0% in one sample.

In: 2
(Plate 3-13) Pb 14249-3 36.5/117.5

Description:

Tectate. circular grains. less commonly folded than Type 1. Diameter 30—
35 microns. Single pore usually evident but margin merges with sculpturing
pattern. Reticulation coarse. Closely resembles T. W (cf. Huang.
1972. pl. 177)
Discussion:

Pollen rare.

lIl

DICOTYLEDONAE

ACERACEAE
Apes L./Dlp_lemnia Oliv.
Description:

Grains tricolpate with furrows extending almost to the poles which are
indented. Exine is conspicuously striato-baculate or striato-granulate. the
striations generally oriented longitudinally. All grains are prolate. Three
types are differentiated.

Discussion:

MacGinitie (1953) recognized 5 species of Aceraceae in the leaf fossil
record: A. mlppadepse MacGinitie. A. flpsissami Kirchner. A. W
(Chaney) MacGinitie. A. ppegpnianum Knowlton. Dipteppnia insignis
(Lesquereux) Brown. Only one of these species. A. [19121531111 (0.5%). is
represented in his list of species occurrences contributing greater than or
equal to 0.1% of the total.

The family is equally low in abundance in the pollen record. Ages spp.
were infrequent in the palynological samples of both sections. Leopold
(1969 a) illustrated a specimen which resembles Type 3.

Maple pollen is generally highly under-represented in pollen records.
probably because most species are entomophilous and produce small
amounts of pollen per flower (Leopold. 1964; Meeuse. 1961).

Diplerpnia leaf fossils also occur in the Republic flora of Washington and
the Bridge Creek flora of Oregon. Its present distribution is confined to
' central and southern China. Erdtman (1952) describes and illustrates
modern pollen of the genus. The description cannot be differentiated from

Age: but the striations of his pen drawing appear to be less conspicuous than

112

striations on Apes. He also notes that pollen of the Aceraceae is similar to
that of the Sapindaceae.

Maple trees are a common element of the mixed mesophytic deciduous
forest of North America.

In: 1
(Plate 4-7) Pb 14154-4 287/1094
Description:
These grains measure 27-33 microns in the polar dimension and 19-23
microns equatorially. The striations are relatively coarse.
T1119. 2
(Plate 4-6) Pb 14249-2 456/1280
Description:
These grains measure 18-23 microns in the polar dimension and 16-19
microns equatorially. The striations are relatively fine.
11m 3
(Plate 4-8) Pb 14143-1 31.0/128.1
Description:

Only one of these grains was observed. In polar view. its equatorial
dimension is 45 microns. The colpi are very broad and open. The striato-
baculae are very narrow and randomly oriented.

Tim 4
(Plate 4-9) Pb 14144-1 456/1232
Description:

Tricolp(or)ate (?). spherical. diameter 35 microns. Characteristically
baculate (baculae up to 2 microns long). arranged in short striae.
Discussion:

One specimen observed; possibly a modern contaminant.

113

AQUIFOLIACEAE
11211 L.
(Plate 4-12) Pb 14143-1 434/1164
Description:

Grains tricolporate with wide furrows narrowing at the equator and
transversely elliptical ora. The exine is conspicuously clavate. the individual
clavae more than 2 microns in length. Polar diameter of these prolate grains
is 30 microns and equatorial diameter 20 microns.

Discussion:

MacGinitie (1953) has classified llex knigmiaefplia Lesquereux in the
category incentae sedis and it contributed less than 0.1% to the leaf fossils.

This is the first known occurrence of llex in the Florissant pollen record.
One specimen was observed; therefore, it maybe a contaminant. '

Modern holly consists of nearly 400 species of evergreen or deciduous
trees and shrubs distributed throuthout the world. except Australia. It is
entomophilous and typically occupies warm-temperate to subtropical bogs

and swamps (Gibson. 1901; Hopkins. 1967).

CHENOPODIACEAE/AMARANTHACEAE
Description:

Grains periporate (pores 15-35). pores strongly aspidate and evenly
dispersed over the entire grain. Specimens are spherical and rarely folded.
Grains are thickly tectate (up to 2 microns thick) with f inely-granulate
surface. Three types are differentiated and may represent individual

species.

114

Discussion:

Since pollen of the families Chenopodiaceae and Amaranthaceae are
difficult to morphologically differentiate with the light microscope. the
families are generally discussed together in pollen studies under the
designation "Cheno-Am". a convention that will be followed here.

Neither family was represented in the megaflora reported by MacGinitie
(1953).

Cheno-Am pollen occurs regularly in the samples studied. but is rarely
abundant: most is Type 1. It usually contributes to between 1% and 5% of
the pollen sum of a sample (av. 2%).

Lambs Quarters. Sappppatus spp.. and other members of these two
herbaceous families are generally characteristic of xeric sites and often

indicate the re-establishment stage of secondary succession.

Israel
(Plate 4-32) Pb 14151-1 46.7/122.8

I12: 2
(Plate 4-30) Pb 14138-1 29.6/119.4
Description:

Type 1 is differentiated from Type 2 based exclusively on size criteria.
Leopold (1969 a) illustrates two grains resembling these types and
designates them Chenopodiaceae. Swamps-type. The pores of both are
relatively large (dia. 2-4 microns). Histogram representation of size
variation of observed specimens shows a bimodal distribution clustering at
25 microns and 30 microns. Type l_ includes all specimens whose diameter
falls between 17 microns and 27 microns (av. 24 microns). Type 2 includes

grains within the 28 microns to 33 microns diameter range.

115

Discussion:
. Pollen common in both sections. Most observed Cheno-Am grains were
111291.
122: 3

(Plate 4-31) Pb 14154-4 33.7/122.5

Description:
The grains are small (dia. 17-22 microns) and have distinctly small (dia.

l- l .5 microns) pores whose annuli may be inconspicuous.
Discussion:

Pollen very rare.

COMPOSITAE
(Plate 4-16) Pb 14262-4 432/1261
Description:

Grains tricolpate. colpi deeply-incised in polar view. amb sub-triangular.
Diameter from side to opposing colpus 18 microns. Exine thick. very coarsely
echinate (2-3 microns long. spaced 1-2 microns.

Discussion:

Very rare. Possibly a contaminant.

FAGACEAE
Castanea Mill.
(Plate 4-23) Pb 14143-1 43.5/1 17.0
Description:
Grains tricolporate and prolate. 14-16 by 11-14 microns. Pores

transversely elliptical. Exine psilate.

116

Discussion:

MacGinitie recovered a small number (0.2% of total examined) of
Castanea deliphpphylla Cockerell leaves from the lake shales.

Modern chestnut trees may be found in mesic woodlands throughout the
northern hemisphere.

Pollen very rare in section 8/5/84-1 and infrequent in section
8/23/84-1).

1342911815 1002119111 (Lesq.) Hollick
(Plate 4-18) Pb 14249-3 362/1233
Description:

Grains tricolporate and spherical. furrows extending almost to poles and
generally appear as closed partings in the exine. Diameter 20-25 microns
(av. 22 microns); exine psilate to f inely-verrucate or granulate.

Discussion:

This extinct monotypic genus was first classified as Elanepa (Ulmaceae)
by Lesquereux (1883). Hollick (1909) reclassified and renamed it Eagppsis
lengifplia (Lesq.) Hollick and attributed it to the Fagaceae. MacGinitie (1953)
further classified it as a member of the Betulaceae based on "cones".

Leopold (in MacGinitie. 1969) and Manchester and Crane (1983) have
since confirmed its membership to the Fagaceae in detailed studies of the
leaves. flowers. fruits and pollen utilizing SEM and TEM. Manchester and
Crane (1983) isolated and described pollen removed from an anther of a
staminate inflorescence. This has provided evidence for the pollen
morphology and has therefore been particularly useful in this study.

Eagppsis leaves have also been found in the Republic flora of Washington
(early-middle Eocene). the Red Rock Ranch flora of New Mexico (early

Oligocene). the Beaverhead Basins (early-middle Oligocene) and Ruby Paper

 

117

Shales (middle-late Oligocene) floras of Montana. and in unspecified Miocene
sediments (Becker. 1961. p. 18; Becker, 1969. p. 22; Manchester and Crane.
1983. p. 1162: Axelrod and Bailey. 1976. p. 237).

Eaggpsis mm is the most abundant leaf fossil in the Florissant
megaflora. accounting for 30.1% of the specimens (MacGinitie. 1953).
Paradoxically. its contribution to the pollen flora is negligible (ab sent in
section 8/5/84-1; very rare in section 8/23/84-1). According to Manchester
and Crane (1983). the pollen of Eaggpsis longileia. although distinctive. is
quite similar to that of 993m spp. Quezon; pollen is quite abundant in the
microflora. averaging 4.3% of the pollen sum of the samples. Suggested
explanations for the leaf / pollen representational disparities of these two
genera are offered in the Floral Composition section.

Quencus L.
1'19: 1
(Plate 4-22) Pb 14249-3 28.4/119.6

I12: 2
(Plate 4-20) Pb 14138-1 40.6/1135
I12: .1
(Plate 4-26) Pb 14157-1 325/1232
Ixm 1
(Plate 4-19) Pb 14151—1 41.8/126.0
Description:
Grains tricol(por)ate. prolate with deeply dissecting furrows commonly
extending almost to poles and spread apart. Broad range of sizes varying
from 16 by 20 microns to 28 by 38 microns. Exine f aintly- to coarsely-

granulate or irregularly spaced warty verrucae.

 

118

There is no clearcut morphological division to separate the ob served
specimens into separate species. although it is expected that at least several
of the 10 oak species listed by MacGinitie (1953) are represented.
Therefore. (1112:9115 pollen is considered as a complex and four grains are
illustrated to show the gradient of sizes and sculpturing elements which
were encountered in the samples studied.

Discussion:

MacGinitie (1953) recognized 10 species of oak in the leaf flora.
contributing 2.0% of the total: Q. deQsoides MacGinitie. Q. knowltomana
Cockerell. Q. W15 Cockerell. Q. 1119110221111; Axelrod. Q. orhata sp. nov..
Q. mum Cockerell. Q. W MacGinitie. Q. 5991111 (Lesq.) MacGinitie. Q.
59115102111 Knowlton. 111mm; 5:511; Q. balanmorum Cockerell.

Leopold (1969 a) illustrated one specimen. of 9351931; pollen from the
Florissant shales.

The genus has relatively broad ecological amplitude and wide modern
distribution. Oak is commonly a dominant tree of the mesic deciduous
forests of the temperate northern hemisphere. Some dwarf evergreen
species are adapted to xeric habitats of northern Mexico and southwestern
United States. MacGinitie (1953) suggests that the Florissant oaks were like
the latter ecotype. occupying the high-ground of the region.

Pollen abundant in both sections.

HAMAMELIDACEAE
LinuidambarL.
(Plate 4-27) Pb 14154-1 37.1/120.8

 

119

Description:

Grains 8-15+ periporate; pores covered by characteristic persistent
membrane with verrucae; pores large (dia. 3-5 microns). irregularly
outlined. and unevenly but randomly scattered. Grains f mely-reticulate; dia.
27-38 microns (av. 29 microns). The distinctive appearance of these grains is
not easily confused with any of the other Florissant palynomorphs.
Discussion:

This is the first record of any organ of Liquidamhar in the Florissant
shales.

Pollen very rare in section 8/5/84-1 and infrequent in section
8/23/84-1.

Uranium fossils are also present in many other mid-Tertiary floras.

Sweetgum is often a component of oak-hickory forests of warm-
temperate regions of the northern hemisphere (Hopkins. 1967). It also
occurs in almost monotypic stands following clear-cuts in both Illinois and

montane Mexico (Prof. R. B. Taggart. pers. comm.)

JUGLANDACEAE
(Lama N utt.

Discussion:

Qamleaf fossils contributed 1.0% of MacGinitie‘s (1953) systematic list.

Pollen of the genus has also been reported by Leopold (1969 a. figs. 26.
27).

Hickory is commonly a dominant tree of the mesic deciduous forests of
the temperate northern hemisphere on bottomland or upland sites;
commonly co-dominant with oak.

Pollen abundant in both sections.

 

120

I12“
(Plate 3-16) Pb 14016-1 40.4/1173

Description:

Triporate pollen grains with relatively large; circular pores (dia. 3-5
microns). Pores heteropolar; equally spaced near the equator. Amb circular
to sub-triangular (pores near ”apices”); grain dia. 31-45 microns (av.
approximately 40 microns). Exine faintly scabrate to granulate.

When grain diameters are plotted on a histogram. a unimodal curve
results with a peak in the 40 to 43 microns range. While the size range of
the grains is quite broad. the histogram data does not suggest that the genus
should be divided into distinct types based on size criteria.

This type accounts for most of the tabulated 9mm grains.

' Dane 2
(Plate 3- 18) Pb 14249-3 278/1285
Description: .

Grains triporate with very large pores relative to the size of the grain;
pore diameter 4 microns: strongly heteropolar. Grains thin-walled and often
folded into a distorted shape; however. unfolded grains are circular in
outline. Grain diameter 22-28 microns. Exine finely-granulate to psilate.
Discussion:

The few samples containing these specimens are primarily from section
8/23/84-1.

129:5
(Plate 3-17) Pb 14262-4 293/1243

121

Description:

Grains triporate. heteropolar. with large (dia. 3-4 microns). annulate
pores. Average grain diameter 28 microns. Amb circular: intectate; exine
granulate.

Discussion:

Specimens of this type have been observed in only two samples of
section 8/23/84-1.

Enzelhandia Lesch. (Mommies Wodehouse)
(Plate 3-20) Pb 14157-1 372/1244
Description:

Grains triporate. amb straight- to slightly convex- triangular with pores
at the rounded apices. Diameter from pore to opposite margin 20-28
microns. Exine surface finely-granulate to scabrate.

Some of these grains possess two white streaks resembling the
description of Momipites minim Wodehouse by Frederiksen (1980).
Occasionally. a trans-polar crease is evident as in 21am described by
Leopold and MacGinitie ( 1972 ).

Discussion:

Dilmy (1955) discusses the confusion of the spelling of the genus in the
literature (Engelhardia vs. Enflglhanmia). He notes the correct original
spelling to be Engelhandia.

W was not included in MacGinitie‘s (1953) systematic list of
megafossils.

Leopold (1969 a) illustrated a Florissant specimen with uncertain
. identity to W.

Extant species of the genus exist in tropical to subtropical climates of

Mexico. Central America. and southeast Asia (Piel. 1971 ). The pollen appears

 

122
in early Tertiary sediments when its geographic range was much more
extensive than that of today (northern S. America to arctic) (Graham.l972 ).

Pollen infrequent in section 8/5/84-1 and common in section 8/23/84-1.

luglans L.
(Plate 4- 17) Pb 14138-1 279/1228
Description:

Grains periporate (12-25 pores. av. 16) and spherical; heteropolar with
many of the pores concentrated near the equator: pores aspidate (annular
thickenings 5-7 microns across): pore dia. approximately 2 microns. Exine
surface f inely-granulate to psilate. Grain diameter 27-40 microns (av. 35
microns).

Discussion:

Juglans leaves do not occur in the megaflora (MacGinitie (1953).
However. walnut pollen appears in most samples of this study and usually
contributes about 4% to the pollen sum.

Leopold (1969 a. fig. 31) illustrates one grain of juglans from the
Florissant sediments.

Walnut trees are a common deciduous element of the mixed mesophytic
forest and are usually characteristic of rich soils.

Pollen abundant in section 8/5/84-1 and common in section 8/23/84-1.

In one sample. luglans accounts for almost 60% of the pollen sum.

NY MPHAEACBAE
(Plate 4-2) Pb 14249-2 45.6/128.0

 

Description:

Grains tri(tetra-?)porate. spherical. diameter 30-34 microns. Distinctly
echinate (spines 13 microns long). spines distributed evenly and spaced
approximately 3-4 microns apart.

Discussion:
Pollen very rare.
Potential affinities:
Lotus .7. Nelumbo .7. 11111121133 .7. 11111121131 .7
All are aquatic plants of cosmopolitan distribution in tropical to
temperate climates. All are entomophilous and therefore generally poorly
represented in palynological samples.

alternative family: Malvaceae?

PROTEACEAE/SAPINDACEAE/ELAEAGNACBAE
(Plate 4-1) Pb 14154-4 28.6/1150
Description: ‘

Grains triporate: oblate: amb triangular with irregular (undulating)
margins; indented at poles with creases extending from equatorial pores to
poles. Pores ragged and V-shaped from polar view and located at "apices" of
amb. Diameter from pore to opposite side ranges from 30—45 microns. Exine
surface reticulate-rugulate.

Discussion:

Pollen very rare.

Remark: MacGinitie (1953) lists two species of Sapindaceae. Athayana
(2.3% of leaves). Sapindns (1.1% of leaves).

Potential affinities:

 

124

-Proteaceae?: MacGinitie (1953) identified Lomatia Enema (Lesq.)
comb. nov.. from leaf fossils.

Most modern genera are xerophytic and restricted to the southern
hemisphere. indicative of a long annual dry season (Hopkins, 1967).

-Sapindaceae?: Leopold (1969 a). fig. 18. W1! L. (from
Florissant); Huang (1972). pl. 138 figs.l4-15.Cflt_d19_59_er_mllm.

-Elaeagnaceae?: Leopold (1969 d) figs. 34-35. mm Nutt.;
Frederiksen (1980). pl. 14. fig.l4. mm 11911111; McAndrews. et. a1.
(1973). fig. 13 B, WW.

SALICACEAE
12129111113 L.
(Plate 3-9) Pb 14143-1 39.0/128.7
Description:

Grains inaperturate; spherical but usually folded around edges. Diameter
20-24 microns. Exine scabrate-verrucate.

Discussion:

MacGinitie (1953) identified three species of poplar based on leaf
compressions: 201111111: crassalLesq.)Cockere11 (5.0% of leaves): 2912111119.:
hem (Lesq.) MacGinitie luggage aegis; and Poppins magnolia Kirchner.

The disparity between the pollen and leaf records may be a consequence of
the high susceptibility of pollen to destruction prior to f ossilization.

Aspen. poplar or cottonwood trees have a cosmopolitan distribution in
temperate to tropical climates. They may be found on riparian sites (lakes
or streams) or as opportunists on disturbed sites.

Pollen very rare.

 

125
53111 L.
Description:

Grains tricolpate. colpi extending nearly the full length of the grain.
Discussion:

MacGinitie (1953) includes four species of $3111 in his systematic list.
comprising 1.1% of the leaf fossils in sum: Sam libheu’ Lesquereux. 5.
12311131911 Cockerell. S. cglgmdica MacGinitie. S. 1311191191519; MacGinitie.

Leaf fossils are common in the middle and upper Tertiary. but rare in
the lower Tertiary (MacGinitie. 1953).

Leopold (1969 a) illustrated one specimen of $3111? pollen.

Willow often holds a dominant position along streams and lakes of

 

temperate regions worldwide.
1 Pollen common in both sections. More than 95% of the grains are
119: 1.
1299. 1
(Plate 4-11) Pb 14038-2 327/1145
Description:

Shape prolate; dimensions 17 by 22 microns to 22 by 32 microns. Exine
distinctive coarsely-reticulate between the furrows but diminishing within
them.

In: 2
(Plate 4-10) Pb 14157-1 334/1185
Description:
Same characteristics as Iype l but more spheroidal and larger (diameter

26-38 microns)

126

TILIACEAE
Illia L.
(Plate 4-14) Pb 14262-1 39.2/1 16.1
Description:

Grains tetraporate (tetracolpate?): pores U-shaped and equatorial. Amb
circular to rounded quadrangular. Diameter approximately 20 microns.
Exine reticulate.

Discussion:

MacGinitie ( 1953) identified Illia nopulifglia Lesquereux leaf
impressions which are extremely rare in the Florissant shales.

Alternative affinities:

lnmmmmpollennes 51mm Frederiksen (triporate, but discussed
as rarely tetraporate). Frederiksen (1980). pl. 14. fig. 17 (hot. aff. Tiliaceae)

Emmontia Torn? (Wm Cov.?) (Sterculiaceae) : Leopold
(1969 b. “33)

Basswood occurs in temperate deciduous forests and along streams
throughout the northern hemisphere.

Pollen very rare.

ULMACEAE
99.111: L.
Discussion:
MacGinitie (1953) recognized one species of hackberry in the leaf flora.
Celtis mm Lesquereux.
Leopold (1969 a. #30) illustrated a up: 1 specimen without systematic
designation.

Pollen common in section 8/5/84-1 and abundant in section 8/23/84-1.

 

127

Modern hackberry trees are often found on disturbed sites.
line 1
(Plate 4-21) Pb 14149-1 327/1192
Description:

Grains triporate: two of the pores heteropolar. the other pore equatorial.
Pores incipiently aspidate; pore diameter 2 microns. Amb sub-circular to
rounded triangular. Exine f inely-granulate to psilate. Grain diameter
approx. 22-28 microns (av. 25 microns). One or two folds typically cross the

grains.

Discussion:
More common than Type 2.
In: 2
(Plate 3-15) Pb 14151-3 37.8/119.2
Description:

Grains triporate (rarely tetraporate). slightly heteropolar. Pores large
(4-5 microns) and strongly aspidate with narrow but highly protruding
annulus. Amb circular. Exine psilate to finely-granulate. Grain diameter
28-33 microns. Most grains have small folds which do not affect shape.

111mm L./Zelkeya Spach.
Discussion:

Elm pollen is well represented in both the fossil pollen and leaf f loras.
The genera mines and mm are discussed together because their pollen
are difficult to differentiate morphologically.

MacGinitie (1953) reported one species of each genus. minus tenuinemis
Lesquereux (1.3% of leaf fossils) and Zelkm dumeja (Lesq.) Brown (9.6%
leaf fossils). Together. they contributed 10.9% of the Florissant leaf
megafossils. ranking second in abundance to Eageesls lengifglia.

 

128

mes/mm spp. are common elements of other western North
American mid-Tertiary f loras.

Leaf fossils of Zelkm spp. are also present in the Green River. Bridge
Creek, and Latah f loras.

Elm trees are widespread in temperate regions of the northern
hemisphere. occupying an often dominant position in lowland deciduous

forests: often early successional.
I Pollen abundant in both sections.
Type 1
(Plate 4-5) Pb 14252-2 373/1125
Description:

Grains 5-stephanoporate (common range 4-6); heteropolar; pores large
(3-5 microns) and sometimes slightly annulate; amb semicircular-subangular
(polygonal): exine surface coarsely rugulate. giving a distinctive faint
convoluted appearance; diameter 25-40 microns (av. 31-33 microns).
Histogram representation of the range in size shows no bimodal distribution
that might suggest separation into more than one type.

Discussion:
This type accounts for more than 95% of limes/lemma pollen grains in
the samples studied.
I193. 2
(Plate 4-4) Pb 14023-2 306/1133
Description:

Grains triporate and smaller (diameter av. 26 microns) than Type 1.
Characteristic Limits/lemma rugulate appearance.

Discussion:

Very rare.

 

129

I12: 3
(Plate 4-3) Pb 14019-3 37.0/1198
Description:

Grains tetraporate; pores situated along the equator; pore diameter 3
microns: slightly aspidate. Amb circular-quadrangular. Exine tectate and
very thick; the grains appear very rigid and dense. never folded. Surface
coarsely foveolate. Grain diameter 25-30 microns.

The surface of the specimens has a coarse topography which closely
resembles that of other lllmes/Zelkgxa types. but the muri are definitely
more circular than elongate (Le. not rugulate). The wall is thicker than the
other mmus/Zelkm types described and heteropolarity is not apparent.
Discussion:

Pollen limited to one sample.

130

UNIDENTIFIED MORPHOTYPBS (UM)

LLM 1(MULT1PORATE in pollen diagrams)
(Plate 4-13) Pb 14138—1 375/1132
Description: '

Grains 5-7 stephanoporate; heteropolar; aspidate to varying degrees
(ranging from thickened annulus to unraised darkened rim around pore);
pore large (dia. 4 microns). Grains spherical: diameter 28—24 microns: amb
often elliptical. Occasionally. the equatorial outline is angular when the
pores are thickly aspidate. Exine surface finely-granulate.

Discussion:
- One specimen observed in section 8/5/84-1 and common in'section.
8/23/84-1. '

Possible affinities:

gem? (cf. Traverse (1955). p. 45; Leopold and MacGinitie (1972).
pp. 181-182: Frederiksen (1981 ). p. 43. pl.81

Ulmaceae?

111211111?

Mesa? Leopold (1969 c). 1’42

see also Leopold (1969 a), p. 363. pl. 16-6 (undetermined specimen
from Florissant)

11M 2
(Plate 3-19) Pb 14135-1 40.0/117.0
Description:
Grains triporate, pores strongly aspidate with operculuml?) Amb

triangular; grain diameter approximately 30 microns.

131

Discussion:
Very rare.
Possible affinities:
Benin ?
11M 3
(Plate 4-15) Pb 14143-1 39.2/1128
Description:
Grains tricolpate; very coarsely reticulate between furrows. diminishing
within. Shape prolate. size 15 by 12 microns.
Discussion:

One specimen observed.

Possible affinities:

poss. Sell: .7 Engines L. ? (Oleaceae); CaprifoliaCeae?
. 11M 1
(Plate 4-28) Pb 14031-3 36.9/114.6

Description: 1 "

Diporate. oblate. with pores equatorial. Thickening of exine near pores
but not aspidate. Exine psilate-granulate; dimensions 19 by 22 microns.
Discussion:

One specimen observed.

Possible affinities:

991311111 (Corylaceae) [see Traverse (1955). pp. 46-471
1.1M 5
(Plate 3-11) Pb 14142-1 296/1175
- Description:
Grains aporate. spherical, and usually characterized by many jagged

fOlds. Exine psilate; diameter 22-32 microns:

132

Discussion:
Rare.
Possible affinities: .7
11M 6.
(Plate 4-25) Pb 14148-3 470/1239
Description:

Grain tricolporate, prolate and inflated at equator. Size dimensions 38
microns by 24 microns. Exine tectate: surface baculate (densely packed
baculae). ..

Discussion:

One specimen observed.

Possible affinities:

. ms?

UMZ
(Plate 4-24) Pb 14138-1 333/1195
Description: 2
Grain triporate(?); spherical; densely echinate: spines short (length less
than 1 microns). Diameter 15 microns.
Discussion:
One specimen observed.
Possible affinities:
bryophyte spore?
Compositae .7 (remark: no Compositae in MacGinitie (1953) systematic

list)

 

133

1.1M 3
(Plate 3—8) Pb 14137—1 315/1203
Description:

Spherical entities of unknown biological origin. Diameter 14-18 microns.
Linear surface features radiate from central ”pore". Pore outline unclear
(dia. approximately 3-4 microns).

Discussion:
Infrequent and limited to a few samples.
Possible affinities:

Wing segment of moth or other insect?

 

References Cited

Allen, J. A. 1878. Description of a fossil passerine bird from the insect
bearing shales of Colorado. United States Geological and Geographical
(Hayden) Survey of the Territories. Bulletin, 4: 443-445.

Axelrod. D. I. 1976. Evolution of the Santa Lucia fir (Abies bracteata)
ecosystem. Annals of the Missouri Botanical Garden. 63:24-41.

Axelrod. D. I. and H. P. Bailey. 1976. Tertiary vegetation. climate and
altitude of the Rio Grande Depression. New Mexico. Paleobiology. 2;
235-254.

Axelrod, D. l. and P. H. Raven. 1985. Origins of the Cordilleran flora. Journal
of Biogeography. 12: 21 -47.

Bassett. J.. C. W. Crompton and J. A. Parmelee. 1978. An Atlas of Airborne
Pollen Grains and Common Fungus Spores of Canada. Canada
Department of Agriculture. Research Branch. Monograph 18. 321 p.

Becker. H. F. 1961. Oligocene plants from the Upper Ruby River Basin.
southwestern Montana. Geological Society of America. Memoir 82.
127 p.

_. 1969. Fossil plants of the Tertiary Beaverhead Basins in southwestern
Montana. Palaeontographica. Abt. B. 127: l-142.

Birks. H. J. B. and H. H. Birks. 1980. Quaternary Palaeoecology. University
Park Press. Baltimore. 289 p.

Bold. H. C. and M. J. Wynne. 1978. Introduction to the Algae, 2nd ed.
Prentice-Hall. New York. 720 p.

Brown. F. M. 1986. The tsetse flies of Florissant. Colorado. Pikes Peak
Research Station. Bulletin 1. 12 p.

Cockerell. T. D. A. 1906. Fossil saw-flies from Florissant. American Museum
of Natural History, Bulletin. 22:499-501.

Cope. E. D. 1875. On the fishes of the Tertiary shales of the South Park.

United States Geological and Geographical (Hayden) Survey of the
Territories. Bulletin. 1:3-5.

134

135
_. 1879. The Amyzen Tertiary beds. American Naturalist. 13:332.

___. 1881. On a wading bird from the Amyzon shales. United States

Geological and Geographical (Hayden) Survey of the Territories. Bulletin.
6:83-85.

Cross. A. T. and R. E. Taggart. 1982. Causes of short-term sequential changes
in fossil plant assemblages: Some considerations based on a Miocene
flora of the northwest United States. Annals of the Missouri Botanical
Garden. 69:676-734.

Cross. W. 1894. Description of the Pikes Peak sheet. Colorado. United States
Geological Survey. Geological Atlas, Folio 7. 5 p.

Dilcher. D. L. 1969. Romnefrom the Eocene of North America. Science,
164:299-301.

Dilmy. A. 1955. Nama iang sjah untuk keluarga Bischoffia Bl. dan
_ Engelhardia Lesch. Rimba Indonesia. 4(1-2): 29-37.

Doher. L. I. 1980. Palynomorph preparation procedures currently used in
the paleontology and stratigraphy laboratories, U. S. Geological Survey
Circular 830. 29 p.

Dorf. E. 1951. Lithologic and floral f acies in the Paricutin ash deposits.
Mexico. New York Academy of Sciences Transactions. Series 2. 13:317-
320.

Eggler. W. A. 1948. Plant communities in the vicinity of the volcano El
Paricutin. Mexico, after two and a half years of eruption. Ecology.
29:415-436.

Epis. R. C. and C. E. Chapin. 1974. Stratigraphic nomenclature of the
Thirtynine Mile Volcanic Field. central Colorado. United States
Geological Survey, Bulletin 1395-C, 23 p.

Erdtman. G. 1952. Pollen Morphology and Plant Taxonomy 1, Angiosperms
(An Introduction to Palynology. 1). Almqvist 8r Wiksell. Stockholm.
539 p.

_. 1957. Pollen and Spore Morphology/Plant Taxonomy 11.
Gymnospermae. Pteridophyta. Bryophyta. (Illustrations). Almqvist &
Wiksell. Stockholm. 151 p.

 

136

_ 1965. Pollen and Spore Morphology/Plant Taxonomy IlI.
Gymnospermae. Bryophyta. (Text). Almqvist 8: Wiksell. Stockholm.
191 p.‘

_. 1969. Handbook of Palynology. Hafner Publishing Company. New
York. 486 p.

Faegri. K. and J. lverson. 1974. Textbook of Pollen Analysis. 2nd Revised
Edition. Hafner Publishing Co., New York. 237 p.

Fowells. H. A. 1965. Silvlcs of forest trees of the United States. United
States Department of Agriculture, Handbook 271. 762 p.

Frederiksen, N. O. 1980. Sporomorphs from the Jackson Group (Upper
Eocene) and adjacent strata of Mississippi and western Alabama. United
States Geological Survey. Professional Paper 1084. 75 p.

Funkhouser. J. W. and W. R. Evitt. 1959. Preparation techniques for acid-
insoluble microfossils. Micropaleontology. 52369-375.

Gazin. C. L. 1935. A marsupial from the Florissant beds (Tertiary) of
Colorado. Journal of Paleontology. 9:57-62.

Geel. B. van and T. van der Hammen. 1978. Zygnemataceae in Quaternary
Colombian sediments. Review of Paleobotany and Palynology. 25:377-
392.

Gibson. W. H. 1901. Blossom Hosts and Insect Guests. Newson and
Company. New York. 197 p.

Gies. T. F. 1972. Palynology of sediments bordering some Upper Cretaceous
strand lines in northwestern Colorado. Ph. D. Dissertation (Geology).
Michigan State University. East Lansing. Michigan. 355 p.

Graham. A. 1972. Outline of the origin and historical recognition of f loristic
affinities between Asia and eastern North America. p. 1-18, In A.
Graham (ed.). Floristics and Paleofloristics of Asia and Eastern North
America. Elsevier, New York. 278 p.

' Harlow. W. M.. E. S. Harrar and F. M. White. 1979. Textbook of Dendrology.
6th ed. McGraw-Hill, New York. 510 p.

137

Heusser. C. J. 1971. Pollen and Spores of Chile. University of Arizona Press.
Tucson. Arizona. 167 p.

Hollick. A. 1909. A new genus of fossil Fagaceae from Colorado. Torreya.
9: 1-3.

Hopkins Jr.. W. S. 1967. Palynology and its paleoecological application in the
Coos Bay area. Oregon. Ore Bin. 29:161-183.

Huang. T. C. 1972. Pollen Flora of Taiwan. National Taiwan University.
Botany Department Press. 297 p.

Hyde-H. A. and K. F. Adams. 1958. An Atlas of Airborne Pollen Grains.
MacMillan and Co.. London. 1 12 p.

Knowlton. F. H. 1916. A review of the fossil plants in the United States
National Museum from the Florissant lake beds at Florissant. Colorado.
with descriptions of new species and list of type specimens. United
States National Museum Proceedings. 51:241-297.

Leopold. E. B. 1964. Reconstruction of Quaternary environments using
palynology. p. 43-52. In J. J. Hester and J. Schoenwetter (eds). The
Reconstruction of Past Environments (Proceedings of the Fort Burgwin
conference on paleoecology: l 962. no. 3). Fort Burgwin Research Center.
New Mexico.

_. 1969 a. Selected pollen and spores of the (upper Oligocene) Florissant
Formation (plate of 40 grains). pl. 16-6. p. 363. In J. S. Penny. Late
Cretaceous and Early Tertiary Palynology, ch. 16. In R. H. Tschudy and
R. A. Scott (eds). Aspects of Palynology. Wiley-Interscience, New York.
510 p.

_. 1969 b. Selected pollen and spores from the (middle Eocene) lone
gravels. Chalk Bluffs. Calif. (plate of 43 grains). p1. 16-5. p. 358. In J. 3.
Penny. Late Cretaceous and Early Tertiary Palynology. ch. 16. In R. H.
Tschudy and R. A. Scott (eds.). Aspects of Palynology. Wiley-
lnterscience. New York. 510 p.

_. 1969 c. Selected pollen and spore types of the Suntrana Formation of
Middle Miocene age from the Alaska Range (plate of 45 grains), pl. 17- 1.
p. 403. In E. B. Leopold. Late Cenozoic Palynology. ch. 17. In R. H.
Tschudy and R. A. Scott (eds). Aspects of Palynology. Wiley-
Interscience. New York. 510 p.

138

_. 1969 (1. Selected pollen and spore types of the Middle Miocene age
from the Troublesome Formation. Granby. Colorado (plate of 44 grains).
pl. 17-2. p. 408. In E. B. Leopold. Late Cenozoic Palynology, ch. 17. In
R. H. Tschudy and R. A. Scott (eds). Aspects of Palynology. Wiley-
Interscience. New York. 510 p.

Leopold. E. B. and H. D. MacGinitie. 1972. Development and affinities of
Tertiary floras in the Rocky Mountains. ch. 12. p. 147-200. In A.
Graham (ed). Floristics and Paleofloristics of Asia and Eastern North
America. Elsevier, New York. 278 p.

Lesquereux. L. 1872. Lignitic formation and fossil flora. United States
Geological and Geographical (Hayden) Survey of the Territories. Annual
Report. 6:317-427.

___.. 1883. Contributions to the fossil flora of the Western Teritortes. Ill.
The Cretaceous and Tertiary flora. United States Geological Survey.
Annual Report 8. 283 p.

Lewis. W. H.. P. Vinay and V. E. Zenger. 1983.. Airborne and Allergenic
Pollen of North America. Johns Hopkins University Press. Baltimore.
254 p.

MacGinitie. H. D. 1953. Fossil Plants of the Florissant Beds. Colorado.
Carnegie Institution of Washington Publication 599. Contributions to
Paleontology. 198 p.

Malde. H. E. 1964. The ecological significance of some unfamiliar geologic
processes. p. 7-15. In J. J. Hester and J. Schoenwetter (eds). The
Reconstruction of Past Environments (Proceedings of the Fort Burgwin
conference on paleoecologyzl962. no. 3). Fort Burgwin Research Center.
New Mexico.

Manchester. S. R. and P. R. Crane. 1983. Attached leaves. inf lorescences. and
fruits of Eagepeis. an extinct genus of f agaceous affinity from the
Oligocene Florissant flora of Colorado. USA. American Journal of
Botany. 70:1 147-1164.

Markgraf, V. and H. L. D'Antoni. 1978. Pollen Flora of Argentina. University
of Arizona Press. Tucson. 208 p.

 

139

Martin. H. A. and G. E. Rouse. 1966. Palynology of the late Tertiary
sediments from Queen Charlotte Islands. British Columbia. Canadian
Journal of Botany. 44: 17 1 -208.

McAndrews. J. H.. A. A. Berti and G. Norris. 1973. Key to the Quaternary
Pollen and Spores of the Great Lakes Region. Royal Ontario Museum,
Publications in Life Sciences Miscellaneous Publication. 61 p.

McLeroy. C. A. and R. Y. Anderson. 1966. Laminations of the Oligocene
Florissant lake deposits. Colorado. Geological Society of America.
Bulletin. 77:605-618.

Meeuse. B. J. D. 1961. The Story of Pollination. Ronald Press. New York.
243 p.

Moore, P. D. and J. A. Webb. 1978. An Illustrated Guide to Pollen Analysis.
John Wiley and Sons. New York. 133 p.

Peale. A. C. 1874. Report of South Park Division. United States Geological
and Geographical (Hayden) Survey of the Territories. Annual Report.
7:193-273.

Piel. K. M. 1971. Palynology of Oligocene sediments from central British
Columbia. Canadian Journal of Botany. 49:1885- 1920.

Pocknall. D. '1‘. and D. C. Mildenhall. 1984. Late Oligocene-early Miocene
spores and pollen from Southland, New Zealand. New Zealand Geological
Survey. Paleontological Bulletin 51. Science Information Publishing
Center. Wellington. New Zealand. 66 p.

Root, R. R. 1983. Stratigraphic surveys for site selection of visitor and
administration facilities. Florissant Fossil Beds National Monument.
(blorado. Denver Service Center. National Park Service. United States
Department of Interior. 39 p.

Round. F. E. 1965. The Biology of the Algae. Edward Arnold Publishers. Ltd..
London. 268 p.

Sawyer, R. 1981. Pollen Identification for Beekeepers. University Cardiff
College Press, Great Britain. 111 p.

140

Schopf. J. M. 1964. Practical problems and principles in study of plant
microfossils. p. 29-57. In Cross. A. T. (ed). Palynology in Oil
Exploration--Symposium. San Francisco. 1962. Society of Economic
Paleontologists and Mineralogists Special Publication 11. Tulsa.
Oklahoma. 200 p.

_. 1969. Systematics and nomenclature in palynology. p. 49-77. In R. H.
Tschudy and R. A: Scott (eds). Aspects of Palynology. Wiley-
Interscience. New York. 510 p.

Scudder. S. H. 1882(1881). The Tertiary lake basin at Florissant, Colorado.
between South and Hayden Parks. United States Geological and
Geographical (Hayden) Survey of the Territories. Bulletin. 6:279-300.

Smiley. C. J. and W. C. Rember. 1981. Paleoecology of the Miocene Clarkia
Lake (Northern Idaho) and its environs. ch. 17. p. 551-590. In J. Gray.
A. J. Boucot and W. B. N. Berry (eds). Communities Of The Past. Dowden.
Hutchinson and Ross, Stroudsburg, Pennsylvania.

 

Taggart. R. E. 1971. Palynology and paleoecology of the Miocene Sucker
Creek flora from the Oregon-Idaho boundary. Ph.D. Dissertation
(Botany). Michigan State University. East Lansing. Michigan. 173 p.

__ and A. T. Cross. 1980. Vegetation change in the Miocene Succor Creek
flora of Oregon and Idaho: a case study in paleosuccession, ch. 7. p. 185-
210. In D. L. Dilcher and T. N. Taylor (eds). Biostratigraphy of Fossil
Plants. Dowden. Hutchinson and Ross. Stroudsburg, Pennsylvania.

Thompson. G. G. 1969. Paleoecology of palynomorphs in the Mancos Shale.
southwestern Colorado. Ph. D. Dissertation (Geology). Michigan State
University. East Lansing. Michigan. 199 p.

Traverse. A. 1955. Pollen analysis of the Brandon lignite of Vermont.
United States Department of Interior. Report of Investigations 5151.
107 p.

Tschudy. R. H. 1957. Pollen and spore formulae--a suggestion.
Micropaleontolosy. 32277-280.

_. 1969. Applied Palynology. Ch. 7. pp.103-126. In R. H. Tschudy and R.
A. Sc0tt (eds). Aspects of Palynology. Wiley-Interscience. New York.
510 p.

 

141

Wadsworth. M. E. 1883. Tufa from Florissant (Colorado). United States

Geological and Geographical (Hayden) Survey of the Territories. Annual
Report. 12:276-278.

Wing. S. L. 1987. Eocene and Oligocene floras and vegetation of the Rocky
Mountains. Annals of the Missouri Botanical Garden. 74: 748-784.

Wobus. R. A. and R. C. Epis. 1978. Geologic map of the Florissant 15-minute
quadrangle, Park and Teller Counties. Colorado. United States Geological
Survey. Miscellaneous Investigations Series. Map I-1044.

Wodehouse. R. P. 1959. Pollen Grains: Their structure. identification and
significance in science and medicine. Hafner Publishing Co.. New York.
574 p.

142

PLATE 1

Illustration

Abies Type 3
Fungi Type 1
Mai Tyne 2
Monolete Type 2
Enhedra Type 1 a
Ephedra Type 1 b
Abies Type 2
£12th Type 2

9 Monolete Type 3
10 Abies Type 1
1 1 Monolete Type 4
12 Monolete Type 1

 

IDOVO‘MAWN—d

All illustrations 666K.

 

144

PLATE 2
Illustration

1 Bicea Type 1

2 Taxodiaceae/Cupressaceae/Taxaceae Type 1 c
3 Taxodiaceae/Cupressaceae/Taxaceae Type 1 a
4 Taxodiaceae/Cupressaceae/Taxaceae Type 1 b
5 Taxodiaceae/Cupressaceae/Taxaceae Type 1 d
6 Trilete Type 3

7 Trilete Type 5

8 Trilete Type 4

9 Eigea WW 2

10 Trilete Type 6
1 l Trilete Type 7
12 Trilete Type 2
13 Bseufintsuza/Larix
l4 Trilete Type 8
15 Trilete Type 1
16 Trilete Type 9

All illustrations 666x.

 

 

 

 

146

PLATE 3

Illustration

1 Eodocarpus

2 Hans Type 2

3 Finns Type 4

4 Plans Type 5

5 Einus Type 3

6 Plans Type 1

7 Gramineae Type 2

8 UM 8

9 Populus
10 Gramineae Type 1
11 CM 5
12 Elma/Wm Type 1
13 [knits/Mam Tyne 2
14 o a to
15 Celtis Type 2
16 game Type 1
17 Cam Type 3
l8 Caryn Type 2
19 UM 2
20 Enzelhandla

All illustrations 666x.

 

i

20

 

148

PLATE 4
Illustration
1 Proteaceae/Sapindaceae/Elaeagnaceae
2 Nymphaeaceae .
3 Ulmus/Zelkova Type 3
4 Ulmns/Zelkoxa Tyne 2
S lllmns/Zelkoya Type 1
6 Am/Di'pteronia Type 2
7 _A_ce£/Diptergnia Type 1
8 Aces/1219121121112 Type 3
9 Aeer/Dipteronia Type 4
10 Sui} Type 2
11 Sell; Type 1
12 llex
13 UM l
14. . 1311a
15 UM 3
16 Compositae
17 Install;
18 Emisltznslfolia
19 Qnem Type 4
20 Quencns Type 2
21 99111: Type 1
22 Quernns Type 1
23 Castanea
24 UM 7
25 UM 6
26 9111219115 Type 3
27 Linuldamhax
28 UM 4
29 1149112912113
30 Chenopodiaceae/ A maranthaceae Type 2
31 Chenopodiaceae/Amaranthaceae Type 3
32 Chenopodiaceae/Amaranthaceae Type 1
33 Algae Type 1

All illustrations 666x.