Content uploaded by Jiří Kvaček
Author content
All content in this area was uploaded by Jiří Kvaček
Content may be subject to copyright.
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
663
IntroductIon
Considerable attention has been devoted to the angi-
osperm radiation and their co-radiating insect herbivores
and pollinators during the mid-Cretaceous from 135 to
90 million years ago (Ma) (Crepet, 1996; Grimaldi, 1999).
However, a more prolonged, extensive, and evolutionarily
important colonization of gymnospermous seed plants,
especially by basal clades of holometabolous insects,
transpired during the mid-Triassic from about 245 to
220 Ma, continued throughout the Jurassic and into the
earlier Cretaceous at around 100 Ma (Labandeira, 2006).
The worldwide expression of this process consisted of
major assemblages of plant hosts harboring damage at-
tributable to diverse types of external feeding, galling,
leaf mining, piercing-and-sucking, seed predation, wood
boring and oviposition. Instances of this damage occur
in stereotypical and intricate patterns that are confined
to particular plant-host species but also blanket regional
floras (Krassilov & Rasnitsyn, 1983; Jarzembowski, 1990;
Reymanówna, 1991; Grauvogel-Stamm & Kelber, 1996;
Ash, 1997; Scott & al., 2004). One neglected aspect of
this insect herbivore radiation on gymnosperm hosts is
evidence for nectarivory, pollinivory, and varied damage
on ovuliferous or microsporangiate strobili, some of
which are interpretable as pollination mutualisms. These
associations represent specialized relationships that were
analogous to those occurring later on angiosperms. Thus,
these gymnosperm-based associations constituted the third
evolutionary phase of the plant-insect associational fossil
record that commenced during the Early Triassic (252 Ma),
after the end-Permian extinction, and continued through-
out the Mesozoic, albeit at significantly decreased diversity
as angiosperms assumed ecological dominance (Laban-
deira, 2000, 2006). The mid-Triassic to mid-Cretaceous
co-radiations of insects and gymnosperms are contrasted
with the fourth phase of the radiation of angiosperms and
their insect associates that commenced during the Early
Cretaceous (115 Ma), an expansion that has been increasing
in dramatic ecological ways to the present.
In this contribution, the inferred pollination-related
associations of Mesozoic gymnospermous seed plants
Pollination drops, pollen, and insect pollination of Mesozoic gymnosperms
Conrad C. Labandeira1,2, Jiří Kvaček3 & Mikhail B. Mostovski4,5,6
1 Department of Paleobiology, P.O. Box 37012 (MRC-121), National Museum of Natural History, Smithsonian
Institution, Washington, D.C. 20013-7012, U.S.A. labandec@si.edu (author for correspondence)
2 Department of Entomology, University of Maryland, College Park, Maryland 20742, U.S.A.
3 Department of Palaeontology, National Museum, Václavské nám. 68, Prague 1, Czech Republic
4
Natal Museum, Private Bag 9070, Pietermaritzburg 3200, South Africa
5
School of Biological and Conservation Sciences, University of KwaZulu-Natal, Private Bag X01,
Scottsville, 3209 South Africa
6
Paleontological Institute, Russian Academy of Sciences, 123 Profsoyuznaya Str., Moscow, 117997 Russia
Recent focus on plant-insect associations during the angiosperm radiation from the last 30 million years of the
Early Cretaceous has inadvertently de-emphasized a similar but earlier diversification that occurred among
gymnosperms. The existence of gymnosperm-insect associations during the preangiospermous Mesozoic is
evidenced by mouthparts capable of reaching and imbibing pollination drops or similar fluids, availability of
pollen types consistent with entomophily, and opportunities for related consumption of pollen, seeds, and repro-
ductively associated tissues in major seed-plant groups, namely seed ferns, conifers, cycads, bennettitaleans, and
gnetaleans. Based on stereotypical plant damage, head-adherent pollen, gut contents, wing structure, mouthpart
morphology and insect damage to plant reproductive organs, the likely nectarivores, pollinivores and pollinators
were orthopterans, phasmatodeans, webspinners, sawflies and wasps, moths, beetles, mecopteroids, and true flies.
These associations are ranked from possible to probable although the last three insect clades provide the strongest
evidence for pollinator activity. We document two mid Cretaceous examples of these associations—cycadeoidea-
ceous bennettitaleans and beetles and a cheirolepidiaceous conifer and flies—for which there are multiple lines
of evidence for insect consumption of plant reproductive tissues but also pollination mutualisms. These data
highlight the independent origin of a major phase of plant-insect pollinator-related associations during the mid
Mesozoic that served as a prelude for the separate, iterative and later colonization of angiosperms.
KEY WordS: Alvinia, Bennettitales, Cheirolepidiaceae, Coleoptera, Diptera, insects, Mesozoic, plant-
insect associations, pollination drop, seed plants
664
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
(Fig. 1A) and their insect participants (Fig. 1B) will be
documented and new data and interpretations will be pre-
sented. Previously, a near exclusive focus on angiosperms
has been predominant in the fossil insect literature, char-
acterized by skepticism with regard to evidence for pol-
lination or autecologically coupled types of feeding such
as nectarivory, pollinivory or seed predation during the
preangiospermous Mesozoic (Grimaldi & Engel, 2005).
This view persists despite considerable evidence for and
common recognition of these earlier associations (Crepet,
1974; Gottsberger, 1988; Crowson, 1991; Krassilov & Ras-
nitsyn, 1999; Labandeira, 2000; Gorelick, 2001; Klavins &
al., 2005). For example, some authors (Oberprieler, 2004)
have asserted a parallel delay in pollination and related
associations of cycads to the angiosperm dominated part
of the fossil record, even though extant major lineages of
obligately insect-pollinated cycads extend back in time
from the Early Cretaceous to Middle Triassic or possi-
bly earlier (Gao & Thomas, 1989; Klavins & al., 2003;
Anderson & al., 2007). Significantly, a predilection for
understanding angiosperms and their pollinators forms
an important backdrop for the increased number of asso-
ciations seen during the Late Cretaceous and Cenozoic
(Crepet, 1979; 1996; Crepet & Nixon, 1998; Grimaldi &
Engel, 2005), and especially the modern world (Grant &
Grant, 1965; Faegri & Pijl, 1981; Proctor & al., 1996).
An underappreciated fact is that the ensemble of
Mesozoic gymnospermous lineages were the physiog-
nomic equals of angiosperms. Mesozoic gymnosperms
included forms with herbaceous growth, lianas, shrubs,
pachycauls, mangroves, stem succulents, monocot-like
pteridospermous trees, and trees with true secondary
xylem deployed as polyaxial, branched dicot-like forms
or as monaxial conifers (Upchurch & Doyle, 1981;
Retallack & Dilcher, 1988; Tidwell & Ash, 1990; Ni-
klas, 1997; Rothwell & al., 2000; Dilcher & al., 2004).
Similarly, studies indicate that more basal Mesozoic
lineages of extant anthophilous insect lineages were
interacting with seed-plant clades in ecologically mod-
ern and specialized ways (Pellmyr, 1992; Krassilov &
Rasnitsyn, 1999; Norstog & Nicholls, 1997; Labandeira,
2000, 2005; Gorelick, 2001). For example, the obligate
associations between extant cycads and their pollina-
tors (Schneider & al., 2002) are at least as ancient and
specialized as analogous associations between figs and
fig wasps (Machado & al., 2001), and undoubtedly are
older (Farrell, 1998; Klavins & al., 2005). This increase
in specialized associations between Mesozoic plant hosts
and their herbivore- and pollinator associates forms the
ecological backdrop for the spectacular success of one
terminal, initially inconspicuous clade that displaced the
dominant, earlier Mesozoic gymnospermous clades—the
angiosperms. Given this context and evidence presented
below, it is appropriate to restate a hypothesis regarding
the origin of angiosperm pollination, namely that it was
the gymnospermous pollination drop mechanism and
its varied modifications (Baker & Hurd, 1968; Norstog
1987; Lloyd & Wells, 1992; Kato & Inoue, 1994; but see
Frame, 2003a), that served as a functional, anatomical,
and ecological prelude to early angiosperm pollination.
GYmnoSpErmouS pollInatIon
drop mEchanISm, ZooIdoGamY,
and SIphonoGamY
Seed plants originated during the Late Devonian,
at which time there was among progymnosperms a
mechanism for the reception of motile antherozooids
(Rothwell & Serbet, 1992) that evolved into an ovular
pollination-drop mechanism. The pollination drop
mechanism ancestrally was used for prepollen or pollen
capture (Fig. 2A), followed by various developmental
events that ended in fertilization. The earliest fossil evi-
dence for the pollination drop mechanism is the callisto-
phytacean seed fern Callospermarion pusillum Eggert
& Delevoryas from the Middle Pennsylvanian of the
Illinois Basin, U.S.A. (Figs. 2B, C; Rothwell, 1977; but
also see Retallack & Dilcher, 1988). Although this type
of pollen capture presumably occurred in several extinct
seed plant lineages throughout the Late Paleozoic and
Mesozoic, it currently is exclusively confined to the four
remaining gymnospermous taxa: Pinopsida (Figs. 2D,
E, G), Ginkgoopsida (Fig. 2F), Cycadopsida (Fig. 2I),
and Gnetopsida (Figs. 2H, J). There are modifications
of this mechanism in a minority of extant conifer taxa
for which pollination drops are absent, as in the case
of Araucariaceae, Saxegothaea, and some species of
Fig. 1. Phylogenetic relationships and stratigraphic ranges of gymnospermous seed-plant (A) and insect (B) clades, high-
lighting those lineages possessing significant examples of taxa engaged in pollinivory, nectarivory and pollination. Degrees
of confidence for assignment of clades are provided in the box at upper-left in (A), and based on a variety of data, principally
plant reproductive features, stereotypical plant damage, dispersed insect coprolite contents, insect gut contents, insect
mouthparts, and dietary assignments consistent with life-habit attributes of extant descendants. Data for seed plants princi-
pally are from Hilton & Bateman (2006) and Anderson & al. (2007); data for insects originate from Labandeira (1994), Rasnitsyn
& Quicke (2002), and Grimaldi & Engel (2005); time scale is that of Gradstein & al. (2004). The striped horizontal bar during the
Cretaceous represents the turnover interval from a gymnosperm- to angiosperm-dominated flora. Abbreviations: Ma, mil-
lions of years ago; Neog., Neogene; P/T, Permian-Triassic boundary; K/P, Cretaceous-Paleogene boundary; Corystosperm.,
Corystospermales; Czekanowsk., Czekanowskiales; Glossopterid., Glossopteridales; Mantophasmat., Mantophasmatodea.
665
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
200
250
150
100
50
0
400
350
300
50
200
250
150
100
0
400
350
300
cissairTnaimreP
suorefinobraC
nainoveD cissaruJ suoecaterC enegoelaP goeN
cissairTnaimreP
suorefinobraC
nainoveD cissaruJ suoecaterC enegoelaP goeN
aM doireP
aM doireP
Angiosperm
Radiation
Angiosperm
Radiation
K / P
P / T
K / P
P / T
seladacyC
selamrepsatleP
ainutuA
selaogkniG
.mrepeotsyroC
.kswonakezC
selalleitataM
selalleirteP
selatenG
selatiadroC
selaihcabeL
selaiztloV
selaniP
selatyhporuenA
seladiretpoeahcrA
aisniklE
ayecaL
aengiliB
amrepsaryL
seladiretponigyL
muignareteH
selasolludeM
arotseauQ
selatyhpotsillaC
amotnegyZ aretporemehpE
aretporeG atanodotorP atanodO
aretpoytcidoealaP aditsimehtomreP aretpocesageM aedoretponahpaiD ”aretpohtrotorP“ aretpocelP
”sdiosserpuC“
.diretpossolG
selalyxotneP
selatittenneB
selainotyaC
smrepsoigna
aretpoibmE aretparoZ
aretportyletorP aretpamreD aedottalbollyrG .tamsahpotnaM
aedoruenolaC aretponatiT aretpohtrO aedotamsahP ”aedottalbotorP“ aedottalB aretposI aedotnaM aretpomoiM aretpocosP aretparihthP adiruenoihpoL aretponasyhT ”adilrepopyH“ aretpimeH aretpoeloC aedortylessolG aretpoidihpaR aretpolageM ainnepinalP aretponemyH
”aretpocemotorP“
aretpoceM aretpanohpiS aretpiD puorg mets aretpispertS aretpohcirtaraP aretpohcirT aretpodipeL
A. Seed Plants
B. Insects
Unequivocal evidence for pollination (anthophily)
Probable evidence for palynivory, nectarivory, or
other pollination-related associations
Possible evidence for palynivory, nectarivory
or other pollination-related associations
Good evidence for clade presence
Limited evidence for clade presence
666
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
Abies and Tsuga. In a few other taxa, such as Larix and
Pseudotsuga (Doyle & O’Leary, 1935; Tomlinson & al.,
1991; Owens & al., 1998), the production of pollination
drops is delayed until after pollen lands on a dry micro-
pylar surface but occurs prior to fertilization. In extant
gymnosperms, pollination drops are timed for maximum
reception of saccate or non-saccate pollen by variously
oriented ovules (Gelbart & Aderkas, 2002).
Pollination drops are produced by nucellar and/or
adjacent secretory tissues within a chamber located at the
archegonial pole, which typically is the exposed end of the
ovule. This ovular region bears enveloping tissues that are
formed into a thickened and tubular beak, the micropyle
(Fig. 2A). The terminal end of the micropyle traps pollen
by secreting an often sugary fluid that forms an extruded
film or drop at its aperturate terminus. Soon after pollen
capture, the pollination film or drop recedes back into the
micropylar tube through evaporation or tissue resorption
until the male gametes and nuclei reach the ovular arche-
gonium that typically forms the bottom surface of a pollen
chamber (Moussel, 1980). The fluid consists principally
of carbohydrates, amino acids and lipids dissolved in
a dilute to rather concentrated solution. Among extant
gymnosperms, pollination drops typically contain dilute
to concentrated carbohydrates with ancillary amino-acid
components (Table 1; Ziegler, 1959; Chesnoy, 1993; Gel-
bart & Aderkas, 2002). The composition of sugars and
amino acids in pollination drops has been reported for
several species of conifers such as Cephalotaxus, Pinus,
and Thuja (Ziegler, 1959; Nygaard, 1977; Serdi-Benkad-
dour & Chesnoy, 1985; Chesnoy, 1993); Ginkgo (Dogra,
1964); cycadopsids (Ziegler, 1959; Baker & Baker, 1983;
Tang, 1993; 1995); and gnetopsids (Ziegler, 1959; Carafa
& al., 1992). Although the presented values for gymno-
sperms exhibit highly variable concentrations of sugars
and amino acids (Table 1), they encompass the range found
among angiosperms. It is likely that there is an imbalance
of both nutritional components given the wide range of
sugar and amino acid concentration values reported (Table
1) and a lack of correlation between concentration levels
for both nutritional classes, at least among angiosperms
(Gottsberger & al., 1984).
It is possible to reconstruct a three-step evolutionary
process by which gametes became transported to the
ovular archegonial surface by means of pollination-drop
or related processes. The first developmental type, zooi-
dogamy, was characterized by the release from prepollen
of motile antherozooids through the proximal aperture,
the antherozooids then swam to an archegonium for fer-
tilization (Poort & al., 1996). An important feature of this
mechanism was that a prepollen grain lacked a pollen
tube for providing nutrition or as a device for transporting
gamete cells or nuclei to the archegonium. This condition,
found particularly in lyginopteridalean and medullosalean
pteridosperms but also some cordaites, lasted until the late
Paleozoic. It was replaced by an intermediate developmen-
tal type in which there was continued release of motile
antherozooids proximally (zooidogamy), but with the ad-
ditional distal outgrowth of a haustorial, often branched,
pollen tube exclusively for absorbing nutrients or water
(Poort & al., 1996). This early pollen type typified late
Paleozoic conifers and seed ferns such as the callistophy-
tacean Callospermarion undulatum (Neely) (Rothwell,
1972), continued in some pteridosperm lineages into the
early Mesozoic, and is present in modern cycads and
Ginkgo. The third developmental type is siphonogamy,
in which the pollen grain abandoned motile antherozo-
oids and instead transported gametes and nuclei solely
by means of a pollen tube. Siphonogamy is found in all
modern conifers, gnetopsids, and angiosperms, although
there are significant modifications of the process, and
it probably was present in some Mesozoic gymnosperm
lineages such as bennettitaleans.
It was the pollination drop mechanism or its modi-
fication, in conjunction with true siphonogamy, which
characterized most mid-Mesozoic ovuliferous structures
attractive to insects. Although overwhelmingly a key part
of extant gymnosperm pollination, the pollination-drop
mechanism should not be construed as the only mode of
producing ovular fluids in Mesozoic gymnosperms. Like
stigmatic exudates and other reproductively associated
surface secretions present in gymnosperms and basal an-
giosperms (Gelbart & Aderkas, 2002; Frame, 2003a), the
typical pollination drop mechanism with a tubular micro-
Fig. 2. Micropyle secreted pollination drops from a Late Carboniferous seed fern (B, C) and the four major clades of extant
gymnospermous seed plants (D–I). The generalized mechanism for production of pollination drops in gymnosperms is
provided in (A), whereby an integumented unfertilized ovule bears a pollen chamber apically and is surmounted by a tubu-
lar micropyle filled with a nectar-like fluid for capture of pollen (from Gifford & Foster, 1989). This mechanism occurred in
Late Pennsylvanian seed ferns such as Calliospermarion pusillum (Rothwell, 1977), found as a permineralized substance
containing pollen grains within a micropyle in (B) (abbreviations: n, nucellus; s, sclerotesta; p, pollination drop), magnified
in (C). Pollination drops are illustrated for Sequoiadendron giganteum (Pinopsida: Cupressaceae) in (D) abbreviation: mp,
micropyle) (Takaso & Owens, 1996); Phyllocladus glaucus (Pinopsida: Podocarpaceae) in (E) (abbreviation: c, ovulate
cone) (Tomlinson & al., 1997); Ginkgo biloba (Ginkgoopsida: Ginkgoaceae) in (F); Taxus baccata (Pinopsida: Taxaceae)
in (G) (Proctor & al., 1996); an unidentified species of Ephedra (Gnetopsida: Ephedraceae) in (H) (Gifford & Foster, 1989);
Zamia pumila (Cycadopsida: Zamiaceae) in (I) (Tang, 1995); and Welwitschia mirabilis (Gnetopsida: Welwitschiaceae) in
(J) (Gifford & Foster, 1989).
667
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
668
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
pyle and subsurface chamber con-
taining secretory tissues may have
been more relevant as an adaptation
to long-proboscid nectarivores. By
contrast, other types of surface-
accessible secretions produced by
gymnosperm ovules would have
been available to insects with
shorter, non-intrusive mouthparts.
Besides insect food sources such as
ovular fluids and pollen, other (un-
fossilizable) reproductive associated
attractants and rewards are likely to
have been present such as color, heat
and fragrance, features often found
in extant cycadopsids, gnetopsids,
and angiosperms (Pellmyr & Thien,
1986; Gottsberger, 1988; Kato &
al., 1995; Stevenson & al., 1998;
Wetschnig & Depisch, 1999).
Considerable evidence exists
for extant associations between sur-
face fluid-feeding insects and gym-
nosperms that produce pollination
drops. All of the four existent gym-
nospermous clades—pinopsids, the
ginkgoopsid Ginkgo, cycadopsids,
and gnetopsids—produce pollina-
tion drops (Dogra, 1964; Kato & al.,
1995; Norstog & Nicholls, 1997);
and gnetopsids and cycadopsids,
like angiosperms, have extensive
associations with insects and over-
whelmingly are insect pollinated.
By contrast, pinopsids and Ginkgo
are fundamentally wind-pollinated
(Ackerman, 2000), although insects
do occasionally feed on their pollen
and probably pollination drops.
The dioecious Gnetopsida
consists of three, distantly related,
monogeneric families that possess a
suite of characters involved in insect
pollination. These entomophilous
features are boat-shaped saccate pol-
len that is sticky and forms clusters,
strong scents, frequent extrafloral
nectaries, and showy or otherwise
colorful bracts (van der Pijl, 1953;
Lloyd & Wells, 1992). Most impor-
tantly, these taxa produce pollination
drops on both microsporangiate and
ovuliferous cones, and include sterile
ovules positioned peripheral to the
Table 1. Extranuptial nectary, pollination drop and nectar concentration values of sugars and amino acids in major groups of vascular plants.
Sugar Amino acid Angiosperm
Insect concentration concentration nectar Overall comparative
Taxon pollinated? (%)a (μmol/ml)b site References characterization
Extranuptial nectar (ferns) Sugars: Clustered, inter-
Aspleniaceae Polybotrya osmunda N/A 46 98 Koptur & al., 1982 mediate values differing
Polypodiaceae Polypodium thyssanolepis N/A 30 391 Koptur & al., 1982 by a factor of 2.
Polypodiaceae Polypodium pyrrholepsis N/A 54 195 Koptur & al., 1982 Amino acids: Moderately
Polypodiaceae Drynaria rigidula N/A 63 98 Koptur & al., 1982 clustered, differing by ~
a factor of 4.
Pollination drops (gymnosperms) Sugars: Highly variable
Cupressaceae Thuja orientalis N — 9180 Chesnoy, 1993 values differing by ~ 1
Cycadaceae Cycas rumphii ? 5–14 — Tang, 1995 order of magnitude.
Ephedraceae Ephedra campylopoda ♀ Y 81.3 406– 606 Bino&al.,1984a Amino acids: Extremely
Ephedraceae Ephedra campylopoda ♂ Y 79.8 379–496 Bino&al.,1984a variablevalues,differingby
Gnetaceae Gnetum cuspidatum♀ Y 14–16 — Kato&al.,1995 almost2ordersofmagni-
Welwitschiaceae Welwitschia mirabilis Y 85 — Carafa&al.,1992 tude.
Zamiaceae Ceratozamia robusta Y 5–14 1200 Tang,1993
Zamiaceae Zamia pumila Y 9–12 200–1,600 Tang,1987
Nectar (basal angiosperms)b Sugars: Low to intermediate
Magnoliaceae Magnolia tamaulipana Y 69.8 — petal Dieringer & al., 1999 values differing between 1
Nymphaeaceae Nymphaea mexicana Y 3–4b 2,420 stigma Capperino & and 2 orders of magnitude.
Schneider, 1985 Amino acids: Highly vari-
Winteraceae Tasmannia insipida ♀ Y 20.3 — stigma Frame,2003a ablevaluesdiffering~by1
Winteraceae Tasmannia insipida ♂ Y 13 — stigma Frame,2003a ordersofmagnitude,but
Proteaceae Grevillea fosteri Y 23.7 284 f loralnectary Gottsberger&al.1984 fewstudies.
aSome taxa expressed as range values, others as averages; dashes indicate lack of known studies; some amino acid molar values converted from Baker scores.
bSimilar small percentages of sugars have been recorded for other congeneric species, such as 3 % for Nymphaea odorata (Meeuse & Schneider, 1980).
669
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
fertile regions of microsporangiate cones (Haycraft & Car-
michael, 1992). All three modern genera have insect visitors
feeding on the pollination drops of both ovuliferous and
microsporangiate cones, and wind pollination (anemophily)
is insignificant, although certain species of Ephedra may
have an important aerial component in the transport of pol-
len to receptive ovules (Hesse, 1984; Buchmann & al., 1989).
In monotypic Welwitschia mirabilis Hooker of
southwestern Africa, conspicuous pollination drops are
produced in ovuliferous and microsporangiate cones when
ovules are mature and receptive (Marsh 1982; Wetsch-
nig & Depisch, 1999). Pollinators are insects, consisting
mostly of a broad spectrum of cyclorrhaphan flies pos-
sessing a short but broad labella, which are adpressed to
plant surfaces for sponging fluids (Graham-Smith, 1930).
Brachyceran flies with broad labella also are pollinators,
some lineages of which extend deep into the Mesozoic.
However, the most efficient spongers are cyclorrhaphan
flies that have an origin during the Late Cretaceous
(Grimaldi & Cumming, 1999).
The genus, Gnetum, is represented by about 40 spe-
cies of lianas, shrubs and trees in tropical to subtropical
southeastern Asia, especially Indonesia (Price, 1996).
Gnetum also secretes pollination drops but is pollinated
by a wider variety of small insects, particularly small flies
and subordinately small moths (Pijl, 1953; Kato & al.,
1995). Pollen is sticky and aggregates in linear clusters
that are released synchronously with distinctive micro-
sporangiate cone odors.
The third genus, Ephedra, is represented by about
50 species throughout xeric habitats in Eurasia, northern
Africa and the Americas, and consists of scale-leaved,
woody, small, sometimes climbing, shrubs or small trees
(Price, 1996). Ephedra produces pollination drops with
levels of sugar concentration significantly higher than that
of wind-pollinated taxa of other seed plants (Porsch, 1916;
Kato & Inoue, 1994). The pollen of Ephedra is clustered
and sticky because of elevated sugar concentrations and
not because of pollenkitt, an angiosperm feature (Hesse,
1984). The micropyle for a typical species (E. distachya L.)
is 1 mm in inner minimum diameter and produces multiple
pollination drops, each following a previous pollination
episode (Moussel, 1980). Curiously, such a micropylar
diameter is narrower than the outer proboscis diameters of
many larger, actively flying insects. Pollinators typically
are small insects, particularly flies (Bino & al., 1984a;
Meeuse, 1990), but also hymenopteran parasitoid wasps
such as chalcidoids (Moussel, 1980) and small bees that
produce honey from the drops (Ordetx, 1952). A general
pattern emerges that extant gnetaleans are pollinated by
small, actively flying insects which use their proboscides to
consume surface fluids, such as small- to medium bodied
flies and moths having modestly prolonged mouthparts.
Such a pattern would be expected for fossil members of
the gnetopsids whose interval of greatest diversity is the
Early Cretaceous (Crane & Upchurch, 1987; Sun & al.,
2001), but also were present earlier during the Jurassic
(Krassilov, 1997).
The insect visitors and associates of the Cycadopsida
overwhelmingly have targeted pollen and other tissues
such as endosperm, leaf epidermis and receptacular pa-
renchyma as the principal plant rewards. As would be
expected, secreted liquids in Cycadopsida have low sugar
and amino acid content (Table 1) in contrast to the surface
fluids of Gnetopsida (Donaldson, 1997). These insects—
mostly beetles but also some thrips—enter the environs of
the pollination drop with their whole body rather than a
manipulated proboscis (Norstog & Nicholls, 1997; Terry,
2001). Thus, the associates of cycads are comparatively
small mandibulate beetles and to a lesser degree thrips,
which have “punch-and-suck” mouthparts; both groups
preferentially consume pollen (Crowson, 1991; Kirk, 1984;
Mound & Terry, 2001; Schneider & al. 2002). Limited
evidence indicates that strong musty scents and pollina-
tion drops lure pollinivorous beetles to ovules (Pearson
1906; Pellmyr & Thien, 1986; Tang 1987) even though
the latter fluids are infrequently consumed by beetles.
Rather, the dilute pollination drops likely make the mi-
cropyle terminus sticky for adherent pollen as weevils
brush across ovuliferous structures. In addition to pollen
and pollination drops, cone thermogenesis and a brood
site for oviposited eggs in the case of two Zamia species
also are major rewards (Pellmyr & Thien, 1986; Stevenson
& al., 1998; Dobson & Bergström, 2000). Major pollina-
tors of cycads are a broad array of polyphagan beetles,
including members of the Erotylidae, Boganiidae, and
especially taxa from the diverse clade Polyphaga, namely
the Belidae, Curculionidae, Brentidae, and Anthribidae
(Crowson, 1991; Schneider & al., 2002). An exception
to the dominance of beetles is the aeolothripid thrips
genus Cycadothrips that obligately pollinates the cycad
Macrozamia in Australia, an association considered to
antedate cycad-beetle associations, and possibly goes back
as early as the Middle Jurassic (Mound & Terry, 2001;
Terry & al., 2005; but see Oberprieler, 2004). Additionally,
mycetophilid midges and bees feed on pollination drops of
Zamia pumila and may play a role in pollination (Breckon
& Ortíz, 1983; Ornduff, 1991).
There is no evidence for insects playing any role in
the pollination of extant pinopsids and Ginkgo biloba L.
Typically, wind-pollinated gymnosperms are charac-
terized by round to lenticular, 20 to 50 µm in diameter,
smooth pollen grains, which lack sticky substances and
are unclumped (Wodehouse, 1935; Whitehead, 1969), and
produced in prolific amounts (Ackerman 2000; Gorelick
2001)—features typical of modern pinopsids and Ginkgo.
By contrast, the size range of entomophilous pollen (in-
cluding functional units such as tetrads and pollinia) is
670
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
much wider, varying from 10 to 150 µm or even greater.
To our knowledge, there have been no published obser-
vations of insect feeding on conifer or Ginkgo pollination
drops, although it seems likely that such surface fluids
may be sources of nutrition for insects. More notable is
the consumption of pollen on microsporangiate pinaceous
cones by holometabolous insects such as mecopteroids
and xyelid sawflies (Burdick 1961; Malyshev, 1968),
which have lineages that extend to the Middle Triassic
(Rasnitsyn, 1964; Grimaldi & Engel, 2005). Several other
insect lineages, such as hover flies, also consume coni-
fer pollen (Holloway, 1976; Stelleman, 1981; Leereveld,
1982), but are not implicated in pollination, supporting the
view that extant pinopsids are universally anemophilous.
The lack of entomophily in Ginkgo may be related to the
fact that it was only known from cultivation in eastern
Asia and hence, may have lost its ancient complement of
herbivores and pollinators.
plant provIdErS of nEctar,
pollEn and aSSocIatEd
rEproductIvE tISSuES
During the Mesozoic there was more higher-ranked
seed-plant lineages inferred to have been dominantly
insect pollinated than there are today: 5 versus 3. For
the Mesozoic interval, excluding the Pteridospermopsida
whose insect-associated members arguably are confined
to the Paleozoic, there are the Pinopsida, Cycadopsida,
Bennettitopsida, Gnetopsida, and Angiospermopsida,
compared only to the Cycadopsida, Gnetopsida, and An-
giospermopsida of today. The pollination biology of these
gymnospermous lineages will be briefly discussed with
regard to pollination drop and similar mechanisms, as well
as the role of pollen in facilitating possible entomophilous
associations of pollinating insects.
Pteridospermopsida. —
Pteridosperms (seed
ferns) are a paraphyletic assemblage of mostly early seed
plants that consist of several lineages basal to remain-
ing seed plants (Hilton & Bateman, 2006). Their ovules
were characterized by the pollination drop system and
typically bore wind-dispersed prepollen. Evidence for
insect pollination occurs among a few Late Carbonifer-
ous pteridosperm species, particularly the medullosacean
Pachytesta illinoensis, of Late Pennsylvanian age from the
Illinois Basin of north-central U.S.A. This plant geochro-
nologically is the earliest plant with a well documented
syndrome of anatomical and micromorphological features
consistent with insect pollination. Pachytesta illinoensis
had unusually large, heavy, and nonsaccate prepollen as-
signed to the form-genus Schopfipollenites (Taylor, 1978;
Dilcher, 1979); enclosure of microsynangia by glandular
trichomes and distinctive fleshy tissue that may have
provided a nutritive reward; and presence of coprolites in
the same deposit as P. illinoensis whose contents contain
monospecific populations of Schopfipollenites, indicating
pollinivore targeting of conspecific synangiate prepollen
organs (Retallack & Dilcher, 1988; Labandeira & Phil-
lips, unpubl. data). Lastly, foliage of P. illinoensis had
the most extensive herbivory of any co-occurring plant
species within the surrounding peat swamp, suggesting an
accommodationist antiherbivore strategy involving rapid
rather than delayed pollination that is more consistent with
entomophily (Eisikowitch, 1988; Labandeira & Phillips,
unpubl. data).
The heyday of medullosacean associations and pos-
sible insect pollination can be contrasted to a respectable
diversity of pteridosperm species during the Triassic and
Jurassic that generally were wind-pollinated. Accordingly,
entomophily has not adequately been demonstrated for any
well-known, Mesozoic pteridosperm species (Retallack
& Dilcher, 1988). However, taxa such as Peltaspermum
thomasii, based on a peltasperm fructification (Anderson
& Anderson, 2003), may be a candidate for insect polli-
nation. The smooth, large (up to 40 µm) lenticular pollen
was contained within pollen sacs which bore glandular
excrescences on the surface (Retallack & Dilcher, 1988).
The relatively large pollen argues against the plant being
wind-pollinated. However, it is not clear if P. thomasii
had a pollination drop mechanism and further study of
the reproductive biology of this genus, which extends to
the Late Triassic, is warranted. However, there are other
taxa related to peltaspermalean seed ferns, such as the
Ginkgoales (Hilton & Bateman, 2006), that may have
been insect pollinated. For example, seed predation on the
ginkgoalean seed, Avatia bifurcata (Anderson & Ander-
son, 2003), from the Molteno Formation of South Africa,
could be related to a pollination syndrome in which im-
matures are seed predators and adults are pollinivores or
nectarivores that pollinate the same host plant (also see
Reymanówna, 1991). Similar associational combinations
are found throughout modern cycads and angiosperms,
except that immatures, in this case larvae, are often pol-
linivores (Johnson, 1970; Norstog & al., 1992; Pellmyr &
Leebens-Mack, 1999).
Pinopsida. —
Provided that the Pinopsida does
not include the Gnetopsida as a subordinate clade or
that the two are sister taxa, the Cheirolepidiaceae may
be the only lineage of pinopsids that ever evolved insect
pollination. It should be noted that the “gnetopsids as
conifers” hypothesis (Chaw & al., 2000) conflicts with
morphological and some molecular data consistent with
the “anthophyte hypothesis” (Donoghue & Doyle, 2000;
Rydin & al., 2002), an issue that apparently remains un-
resolved. Regardless, within the several families of extant
conifers a variety of abiotic pollination mechanisms have
evolved and involve features such as varied orientations
671
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
of the ovule during pollination, presence, modification or
absence of the pollination drop mechanism, and whether
the pollen is saccate (buoyant) or nonsaccate (sinking)
(Owens & al., 1998). These varied mechanisms suggest
that other abiotic and biotic pollination mechanisms may
have been present during the earlier Mesozoic when there
was considerable proliferation of family-level clades, and
higher-rank conifer diversity was greater prior to when
global “competition” from nectar-bearing angiosperms
came into play during the Cretaceous (but see Owens &
al., 1998 for an alternative view).
Given this context, the Cheirolepidiaceae are ecologi-
cally a very anomalous taxon that colonized mostly xeric,
mesic and saline habitats (Srivastava, 1976; Upchurch &
Doyle,1981;Watson, 1988;Uličný& al,1997)during
the Middle Triassic to the Cretaceous-Paleocene bound-
ary (Scheuring, 1976) and possibly into the Paleocene
(Pocock & al., 1990; N.R. Cúneo, pers. comm.). The Chei-
rolepidiaceae represent a wide variety of growth forms
ranging from herbs, mangroves, shrubs, arborescent pole-
like trees, and succulent halophytes (Upchurch & Doyle,
1981; Alvin, 1982; Batten & MacLennan, 1984; Watson,
1988; Axsmith & Jacobs, 2005). They are morphologically
united principally by the presence of a very distinctive
pollen type that has several features consistent with insect
pollination and to a lesser extent by highly variable but
typical cone-like organization of separate microsporan-
giate and ovuliferous reproductive organs. The Cheiro-
lepidiaceae also exhibit distinctive and unique specializa-
tions of the ovuliferous scale that strongly suggest insect
pollination (Cerceau & al., 1976; Clement-Westerhof &
van Konijnenburg-van Cittert, 1991; Alvin & al., 1994;
Kvaček,2000;Axsmith&al.,2004).
Cycadopsida. —
Modern cycads are descendants
of an ancient lineage that have roots during the Permian
(Mamay, 1976; Gao & Thomas, 1989). Modern genera
and their pollinators are thought to have recently evolved
during the Cenozoic to Late Cretaceous (Oberprieler,
2004), although there is significant evidence that some
genera extend to the earliest Cretaceous and many into
the Jurassic (Pant, 1987; Artabe & Stevenson, 2004; An-
derson & al., 2007). Cycad associations with insects are
old and thought by many to extend to the earlier Mesozoic
(Crowson, 1991; Farrell, 1998; Labandeira 2000; Mound
& Terry, 2001; Brenner & al., 2003). Evidence for some
ancient associations comes principally from phylogenetic
analyses of the insect herbivores of recent cycads, which
indicate that allocorynine and antliarrhinine weevils, aul-
acoscelidine leaf beetles (Coleoptera) and cycadothripine
aeolothripids (Thysanoptera) are modern representatives
of clades that extend to the mid to latest Mesozoic (Far-
rell, 1998; Mound & Terry, 2001; Gratschev & Zherikhin,
2003). For example, there is strong biogeographic evidence
of an ancient relationship between the two closely related
host plants Encephalartos cycadifolius of southern Africa
and Macrozamia riedlei of western Australia (Zamiaceae,
Tribe Encephalarteae) and their similarly closely related
beetle pollinators, respectively Metacucujus encephalarti
and Paracucujus rostratus (Boganiidae: Tribe Paracucu-
jini). This distribution indicates a Gondwanan vicariant
separation across the Indian Ocean which began during
the Middle Jurassic for these poorly dispersing plant-host
andinsect-herbivorepairs(Endrödy-Younga&Crowson,
1986; Labandeira, 2000). In addition, several distinctive
cycad genera have host-specific pollinating erotylid
beetles that likely originated during the mid-Mesozoic,
providing circumstantial evidence of ancient associations,
at least for the more encompassing clades if not for lower-
ranked taxa as well (Oberprieler, 1995; Schneider & al.,
2002).
There is little evidence for pollinator-related cycad as-
sociations from the fossils themselves, with the exception
of galleries and coprolites in cones of a Middle Triassic
cycad from Antarctica (Klavins & al., 2005). This damage
is similar in pattern and detail to some extant beetle-cycad
associations, and attribution of the plant host to a modern
taxon of cycads was described as “remarkably similar”
(Klavins & al., 2005). Additionally, cycad foliage contains
sporadic damage from external foliage feeders during the
later Mesozoic that may be consistent with some modern
erotylid and aulacosceline leaf beetle herbivory (Laban-
deira & al., 2002; pers. observ.). Leaves and pollen are
the favored food of extant, cycad-associated adult beetles,
and pollen and associated microsporangial tissues are
typically consumed by their endophytic larvae.
Bennettitopsida. —
The Bennettitopsida, consist-
ing of the Bennettitales and tentatively, the Pentoxylales,
comprised a group, perhaps a clade, readily differentiated
from superficially similar cycads by possession in most
species of bisexual strobili and a second ovular integu-
ment (Bose & al., 1985; Crane, 1988), features which they
share with gnetopsids and angiosperms (Crepet & al.,
1991). The Bennettitales are known for their flower-like
strobili consisting of a robust, central, columnar- to dome-
shaped receptacle bearing numerous pedunculate ovules
interspersed among interseminal scales. The strobilar
axis gave rise to a few long, pinnate microsporophylls
and numerous ovules above. In those taxa that had closed
strobili, the microsporophylls were recurved toward the
receptacle, each of which housed several microsporangi-
ate synangia on lateral pinnae. These microsporophylls,
in turn, were subtended by a spiral series of enclosing and
abutting bracts that insured tight closure of the strobilus,
and effectively prevented wind-pollination because the
apices of the enclosing sterile tissue barely protruded
beyond the leaf bases of the plant trunk. In addition,
sterile tissues separately enveloped the ovuliferous
and microsporangiate reproductive structures early in
672
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
development, thus insuring their complete segregation
(Crepet, 1974).
The Bennettitales comprised the Williamsoniaceae
and Cycadeoidaceae. The former are known from the
Late Triassic to Late Cretaceous and were character-
ized by dissected, open, unisporangiate or bisporangiate
strobili on relatively delicately-branching plants (Wil-
liamsonia, Weltrichia, Williamsoniella) (Harris, 1969;
Watson & Sincock, 1992); it is likely that this group were
wind- or, more probably, insect-pollinated. By contrast,
the Cycadeoideaceae, of Late Jurassic to Late Creta-
ceous age, featured closed, bisporangiate strobili the
microsporangiate organs of which were buried among
persistent, thickened leaf bases on pachycaulous plants
(Monanthesia, Cycadeoidea) (Watson & Sincock, 1992).
It has been proposed that this group was “highly” self-
compatible but also experienced some insect-pollination
possibly by beetles (Crepet, 1974). The shift from open,
williamsoniaceous, to closed, cycadeoidaceous, strobili
probably paralleled a broad, temporal change from wind-
to insect-pollination within the Bennettitales clade. En-
tomophily in the Cycadeoidaceae may have originated
from later Jurassic williamsoniaceous descendants. The
probably related Pentoxylales, such as Pentoxylon, bore
vegetative features, particularly trunk anatomy, that was
significantly different than the Bennettitales, but the two
taxa nevertheless shared reproductive features (Bose
& al., 1985). The Bennettitales are better understood
morphologically and ecologically than the Pentoxylales
(Doyle & Donoghue, 1986; Watson, 1988); they reached
their greatest species diversity during the Middle Juras-
sic to Early Cretaceous, and inhabited xeric to mesic
habitats similar to that of the Cheirolepidiaceae and most
gnetopsids.
Gnetopsida. —
Although the Gnetopsida originated
during the Triassic (Crane, 1996), the fossil pollen record
of gnetopsids indicates a past diversity considerably
greater and more recent than the described numbers of
fossil whole-plant species would indicate. This is par-
ticularly true for the Early to mid-Cretaceous, during
which the group had the highest diversity (Crane, 1996)
and underwent a radiation that paralleled the diversifying
angiosperms (Crane & Lidgard, 1990; Wing, 2000) and
during which additional, high-ranked and extinct clades
are known. The variety of basic gnetopsid pollen types
during this interval significantly exceeded the current
level of three, disparate, extant families (Doyle & al., 1982;
Takahashi & al., 1995), and at least two, extinct, high-
ranked clades were additionally present during the Early
Cretaceous. In addition, there has been the discovery of
novel growth forms such as herbaceous taxa that no longer
are present within the extant clade (Crane & Upchurch,
1987). The ubiquity of pollination-drop formation and
insect-pollination in extant taxa strongly suggests that fossil
Mesozoic gnetopsids also were insect-pollinated (Midgely
& Bond, 1991; Lloyd & Wells, 1992). This inference is
buttressed by forms with beaked micropyles resembling
modern entomophilous Ephedra in the Lower Cretaceous
of northwestern China (Sun & al., 2001) as well as pollen
closely resembling extant insect-pollinated Welwitschia-
ceae (Rydin & al., 2003). Undescribed gnetopsid material
from the Early Cretaceous of northeastern China (Guo &
Wu, 2000) preserves ovular structures, which could have
been pollinated by long-proboscate insects.
InSEctS aS conSumErS of
nEctar, po llEn and aSSocI-
atEd rEproductIvE tISSuES
From the previous overview of Mesozoic gymno-
sperms and from an insect morphological perspective,
there are two basic ways to consume relatively inaccessible
pollination-drop or related fluids and pollen that may be
hidden in gymnosperm fructifications. The first are whole
body encounters, as in thrips, small flies, and parasitoid
wasps, and beetles with hardened elytra. The second type is
mouthpart retrieval, exemplified by proboscate Diptera and
Lepidoptera that have large bodies and robust wings. Small
insects can crawl toward the fluid of dietary choice whereas
in the second mode, considerably larger insects such as
mid-Mesozoic mecopteroid lineages and brachyceran
flies required use of elongate proboscides for food access,
assisted by hovering flight in some species. Both types of
pollinivory and nectarivory were present throughout the
Triassic and Cretaceous interval, and variously occur in
eight major orders in the Mesozoic fossil record—Orthop-
tera (katydids), Phasmatodea (stick insects), Embioptera
(webspinners), Coleoptera (beetles), Mecopteroidea (“scor-
pionflies”), Diptera (true flies), Hymenoptera (sawflies and
wasps), and Lepidoptera (moths), although weaker evidence
also exists for the Thysanoptera (thrips).
The relevant time period for the first occurrence of
these associations predates the accumulation of amber
with insect inclusions in the geologic record, taken to be
approximately 120 Ma. Thus, the relevant fossil record
is almost exclusively dependent on compression material
from the Middle Triassic to mid-Early Cretaceous. Com-
pression fossils require exceptionally good preservational
conditions of entire insect bodies, particularly heads and
mouthparts, which require rapid sedimentation in lake
and fluvial deposits.
Orthoptera. —
Katydids, comprising the suborder
Ensifera of the Order Orthoptera, have a fossil record ex-
tending to the Late Paleozoic (Sharov, 1971), and like their
descendants, presumably were herbivorous. Mesozoic En-
sifera were diverse, and include pollinivorous forms such
as the Haglidae, for which some taxa of the Prophlango-
673
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
psinae include species containing apparently fresh Clas-
sopollis pollen (the form produced by Cheirolepidiaceae)
in their guts (Figs. 3S, T; Krassilov & al., 1997). Evidence
from taxa such as Aboilus amplus Gorochov from the Late
Jurassic of Karatau, Kazakhstan, indicate that the pollen
organs of Cheirolepidiaceae were consumed by large, exter-
nally feeding insects, and thus were also available to other
large-bodied pollinators such as sawflies and stick insects.
Some taxa of extant Orthoptera are known to be nearly ex-
clusively pollinivorous (Grinfel’d, 1962). In a few instances
katydids and grasshoppers are implicated as pollinivores
and nectarivores as well as pollinators in modern ecosys-
tems, typically in understories of tropical or subtropical
habitats (Schuster, 1974; Rentz & Clyne, 1983).
Phasmatodea. —
As in the Orthoptera, the Order
Phasmatodea, or stick and leaf insects, are obligately
herbivorous, although no modern forms are known to
consume pollen. A phasmatodean, Phasmomimoides min-
utus Gorochov, was reported also from the Late Jurassic
deposits of Karatau, Kazakhstan (Krassilov & Rasnitsyn,
1999), with considerable, apparently freshly consumed
Classopollis pollen in its gut (Figs. 3Q, R). The occurrence
of pollen in the guts of phasmatodeans, orthopterans and
embiopterans (see below) may indicate the presence of a
speciose Mesozoic dietary guild of non-holometabolous
palynivores which lack modern analogs.
Embioptera. —
Embiids, or webspinners, constitute
a small group of ground dwelling insects occurring in
warm climates that construct silken galleries and typi-
cally consume a wide variety of foods, especially dead
plant tissues, moss, and fungi. Webspinners are rare in the
fossil record and are considered to be descendants of the
Grylloblattida (Rasnitsyn & Quicke, 2002) that probably
originated during the mid-Jurassic. The enigmatic, Late
Jurassic Brachyphyllophagidae are one of the earliest line-
ages of Embioptera, and share a few but important synapo-
morphies with extant webspinners (Rasnitsyn & Quicke,
2002). Within this family, the species Brachyphyllophagus
phasma Rasnitsyn and B. phantasus Rasnitsyn from the
early Late Jurassic of Karatau, Kazakhstan, notably had gut
contents that contained cheirolepidiaceous foliar fragments
accompanied by Classopollis pollen grains (Rasnitsyn &
Krassilov, 2000; Krassilov & al., 2006).
Coleoptera. —
Pollen has not been documented
in the gut contents of any presumed pollenivorous or
otherwise herbivorous beetle in a preangiospermous
deposit. Some data for possible herbivorous roles of bee-
tles on Mesozoic gymnosperms have been gleaned from
mouthpart structure (Labandeira, 1997), and from the
presence of ancestral characters of extant clades known
to consume live tissues of gymnosperms (Farrell, 1998).
Alternatively, distinctive and recurring insect damage,
such as those on bennettitalean strobili, are a significant
line of evidence for beetle-gymnosperm associations
involving reproductive organs. Modern beetles typically
pollinate seed plants by mandibulate mouthparts rather
than various types of haustellate, siphoning, sponging or
other fluid-feeding (Labandeira, 1997), and thus often
leave conspicuous damage patterns on fructifications,
microsporangia, flowers, and other reproductive organs
(Gottsberger, 1988; Proctor & al., 1996).
There is significant evidence for damage of bennet-
titalean strobili by beetles or by insects that cause beetle-
like damage. Evidence for borings, microsporangial and
seed predation and other endophytic damage by beetles on
Mesozoic gymnospermous tissues is known from cycad
pollen organs, bennettitalean strobili, pentoxylalean-like
fructifications and pinalean cone axes (Figs. 4A–F; 5A–G;
Crepet, 1974; Nishida & Hayashi, 1996; Falder & al., 1998;
Klavins & al., 2005). Likely, culprits are polyphagan bee-
tles, in many instances probably members of the subclade
Phytophaga. Other than positing extinct clades, such as the
rostrate and probable archostematan Obrieniidae which co-
occur with bennettitaleans (Gratschev & Zherikhin, 2003),
suspect members of the Phytophaga capable of producing
bennettitalean damage include the Belidae, Nemonychidae,
basal Curculionidae and Aulacoscledidae, all of which have
extant members with non-angiospermous host associations
(Zimmermann, 1994; Farrell, 1998; Santiago-Blay, 2004).
The likelihood of these potential culprits is based on: (1)
phylogenetic relationships of basal clades of polyphagan
beetles with cycad and conifer hosts that extend to the
Jurassic interval or earlier (Farrell, 1998; Zhang, 2005);
(2) stereotyped, beetle-like patterns of damage on repro-
ductive and adjacent vegetative tissues of gymnospermous
clades as varied as cycads, conifers, and bennettitaleans
(Crowson, 1981; Falder & al., 1998; Klavins & al., 2005);
and (3) at least one case of an encapsulated polyphagan
larva occurring in an extinct gymnosperm fructification
(Nishida & Hayashi, 1996). Some of these published asso-
ciations are provided in Figs. 4 (top) and 5 (A–G).
Mecopteroidea. —
Modern Mecoptera (scorpion-
flies)arelargelydetritivorous(Palmer&Yeates,2004),
but infrequently have been implicated in consuming floral
nectar and other surface plant secretions (Porsch, 1958).
Feeding in extant taxa is accomplished by an extended
hypognathous rostrum with moderately elongate man-
dibulate mouthparts that originate proximally near the
base of the head but process and comminute solid food at
the terminus, accompanied by a suction pump centered
on the frontal region of the head (Heddergott, 1938; Hep-
burn, 1969). This highly stereotyped condition is found
virtually in all extant taxa, although the basal family
Nannochoristidae has a shortened rostrum and terminally
expanded labial palps that are reminiscent of a dipteran
labellum (Hoyt, 1952). In contrast to the basic mouthpart
structure of extant species, there were three lineages of
Middle Triassic to mid-Cretaceous taxa—the Pseudo-
674
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
polycentropidae (Fig. 4I; Novokshonov, 1997; Grimaldi
& al., 2005), Aneuretopsychidae (Figs. 4M, N; Rasnitsyn
& Kozlov, 1991), and Mesopsychidae (Labandeira, pers.
observ.)—that bore significantly different mouthpart ap-
parati than that of any modern species. Members of these
three lineages, probably paraphyletic to extant Mecoptera,
exhibited conjoined maxillary galeae that were prolonged
into a tubular, uncoilable but somewhat flexible, siphonate
proboscis. These proboscides were from 0.3 to 1.4 cm long
and ventrally tipped with paired, either diminutive label-
lum-like pads or homologous larger lobes. Such mouthpart
structures were convergent with certain elements of the
brachyceran dipteran proboscis, but homologous to the
glossae of lepidopterans (Eastham & Eassa, 1955) which
evolved significantly later, during the mid-Cretaceous
(Kristensen & Nielsen, 1981; Labandeira & al., 1994). The
heads bearing proboscides in these taxa were relatively
small but bore distinctive clypeal regions, suggesting a
modest sucking pump for ingesting solutions that probably
ranged from dilute to intermediate concentrations.
The siphonate condition that occurs among these
mecopteroid lineages also is convergent with independent
originations among several major clades of brachyceran
dipterans (Nagatomi & Soroida, 1985) and in nemognathine
meloid beetles (Grinfel’d, 1975), among other extant clades.
In almost all recent lineages such proboscides are used for
uptake of nectar and other nutritious and carbohydrate rich
surface fluids of angiosperms. For the fossil taxa, the most
plausible food sources are relatively inaccessible pollination
drops secreted by micropyles and hidden strobilar necta-
ries of various seed plants and possibly ferns, examples of
which have been documented in modern, Mesozoic and
Paleozoic plants (Bonnier, 1879; Koptur & al., 1982; Power
& Skog, 1987;
Krings & al., 2002). This basal, paraphyletic
clade of “scorpionflies” evidently became extinct during
the angiosperm radiation—an interval coincident with the
diversification of glossate lepidopterans (Labandeira & al.,
1994; Grimaldi & al., 2005).
Diptera. —
Several major basal clades of brachyc-
erous dipterans were present during the Early Jurassic to
mid-Cretaceous, some of which bore long, tubular probos-
cides used for consuming fluid foods (Labandeira, 2005).
Although it is possible that some of these lineages were
hemataphagous (Grimaldi & Engel, 2005), multiple lines
of evidence indicate that these taxa were minimally nec-
tarivorous and probably engaged in pollination of various
gymnosperm seed plants. The evidence includes: (1) the
long, tubular and nonstylate structure of the proboscides;
Fig. 3. The inferred insect pollination mechanism for the mid Cretaceous whole-plant species consisting of Frenelopsis
alata (Feistmantel) Knobloch for foliage and microsporangiate cones and Alvinia bohemica Kvaček for ovuliferous cones
(Pinopsida: Cheirolepidiaceae), based on photodocumentation and reconstructions by Kvaček (2000) and additional data
on cheirolepidiaceous reproductive and vegetative material (Hluštík & Konzalová, 1976a, 1976b; Watson, 1977, 1988; Alvin
& Hluštík 1979; Pons, 1979; Alvin, 1982) (A–J). Inferred insect pollinators from earlier Cretaceous deposits are depicted in
K–P, examples of Late Jurassic insect consumption of cheirolepidiaceous Classopollis pollen are provided in (Q–T). The
ovuliferous cone, A. bohemica on F. alata foliage is shown in (A), an ovuliferous cone scale of which is enlarged in longitu-
dinal section in (B), following the reconstruction in figure 5 of Kvaček (2000). Depicted in (B) is a paramedian, longitudinal
section (cross-hatched pattern) with a distally directed and flaring funnel positioned between lateral adjacent lobes. The
funnel is magnified in (C), where it is shown three dimensionally (stippled pattern). Above and below the funnel are the
upper and lower appendages, respectively, and proximally positioned is an ovoidal anatropous ovule with an downwardly
facing micropyle and associated chamber expanding toward the ovular surface, megaspore membrane, integumentary
layer, and outer covering flap, all of which are subtended by a lower bract (B). Note a tubular structure originating from
the telescoped base of the funnel, which traverses lower ovuliferous cone scale tissues and terminates at or near the
micropyle (Kvaček, 2000). Epidermal features lining the inner sur face of the funnel, illustrated in C, include cylindrical
and apparently secretory protuberances positioned near the funnel mouth enlarged in (D), and distal from the mouth are
structurally different, slender, and acuminate trichomes shown in (E). A distal surface view of a mature ovuliferous cone
scale with the bract removed is depicted in (F). This view illustrates the position of the oblique ovular insertion ridge that
separates the two lateral upper lobes (with incised margins) from the more robust, lateral lower lobes (termed an “ovulif-
erous scale” in Kvaček, 2000). Above the dwarf shoot is the central funnel orif ice, as it may appear to an airborne insect.
This reconstruction is based on specimen F2694 in Kvaček (2000). In (G) is a view of the smaller microsporangiate cone
drawn with associated F. alata foliage (also see Hluštík & Konzalová, 1976a), of which a constituent scale is illustrated in
(H), the margin of which is rimmed with elongate trichomes in (I). Microsporangiate cones bore two (perhaps more?) pollen
sacs, each which bears numerous Classopollis pollen tetrads, one of which is enlarged in (J) (Srvistava, 1976; Courtinat,
1980; Taylor & Alvin, 1984; Pocock & al., 1990). A small, undescribed, therevid brachyceran fly (K–M), from the mid-Early
Cretaceous of Russia, was found to have Classopollis pollen on its head surface adjacent to its proboscis, illustrated as
an SEM micrograph in (N) (also see Labandeira, 2005). Miniscule flies are inferred to be one type of insect responsible for
A. bohemica pollination via entire-body entry of the funnel, although much larger bodied, long proboscate brachyceran
taxa alternatively could have thrust their proboscis deep into the funnel during hovering flight, such as Protapiocera sp.
(Mydidae) (O), which also has Classopollis pollen—though not as tetrads—on its head, as illustrated in (P). Evidence of
palynivory of Classopollis sp. pollen is found in the gut contents in Late Jurassic insects, such as the phasmatodean
(stick insect) Phasmomimoides minutus Gorochov in (Q) and (R), and aboiline haglid orthopteran (katydid) Aboilus amplus
Gorochov in (S) and (T), both from the Karatau deposits of Kazakhstan (Krassilov & al., 1997; Krassilov & Rasnitsyn, 1999).
Drawings made by the senior author. Scale bars: solid = 1cm; striped = 1 mm; stippled = 10 µm.
675
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
676
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
(2) wing shape and longitudinal veins that are upwardly
recurved on the distal wing margin venation indicating
hovering flight, especially in the Nemestrinidae; (3) prom-
inent body pubescence; (4) holoptic and dorsomedially
converging eyes; (5) presumed entomophilous pollen such
as Classopollis present on the heads of several species; and
(6) the entomophilous life-habits of their modern descend-
ants (Kneipert, 1980; Mostovski, 1998; Zaitzev, 1998;
Goldblatt & Manning, 2000; Labandeira, 2000, 2005; also
see Downes & Dahlem, 1987). Numerous brachyceran
taxa, representing multiple separate originations of the
long proboscate condition for imbibing surface fluids, are
present worldwide in Late Jurassic and Early Cretaceous
deposits. The principal relevant deposits, from which most
taxa are known, are the early Late Jurassic (Oxfordian
or Kimmeridgian) Karatau lithographic limestones from
the Karabastau Formation of eastern Kazakhstan (Dol-
udenko & al., 1990); the Early Cretaceous (Hauterivian)
shales from the Zaza Formation of Transbaikalia in Russia
(Zherikhin & al., 1999); the Early Cretaceous (Barremian)
lithographiclimestonesfromtheYixianFormation,Lia-
oning, China (Ren & al., 1995); and the Early Cretaceous
(Aptian) lithographic limestones from the Santana Forma-
tion, Ceará State, northeastern Brazil (Grimaldi, 1990).
These and other occurrences represent five lineages of
basal brachyceran Diptera that originated during the Juras-
sic and evolved long, tubular proboscides: Nemestrinidae
(Rohdendorf, 1968; Mostovski, 1998); Mydidae (Figs. 3O,
P; 4J–L); Tabanidae–Pangioninae (Ren, 1998; Labandeira,
1998a); undescribed Therevidae (Mostovski, unpublished;
Figs. 3M, N); and the enigmatic Cratomyiidae represented
by a single occurrence (Mazzarolo & Amorim, 2000; Figs.
3O, P).
The Nemestrinidae (tanglevein flies) have a compara-
tively good fossil record during the Jurassic and Early Cre-
taceous compared to other long-proboscate dipterans, and
probably were consumers of Mesozoic gymnospermous
pollination drops and perhaps pollen. Forms exhibiting
features consistent with nectarivory and pollination have
been found at Karatau in Kazakhstan, Baissa in Russia,
and Liaoning in China, spanning the early Late Jurassic to
mid-Early Cretaceous (Labandeira, 2005). Their modern
descendants are keystone species forming a nectarivore
pollination guild, together with other members such as
pangioniine tabanids (horse flies), acrocerids (small-
headed flies) and vermelionids (wormlions), which polli-
nate deep-throated flowers of Iridaceae and Geraniaceae
(Manning & Goldblatt, 1996). Morphologically similar
ancestors mentioned by Rohdendorf (1968) possessed
“head[s that are] slightly prolonged, with a prominent
proboscis,”(in Nagatomi&Yang,1998).Otherfeatures
indicating a role in pollination include wing modifications
for hovering flight, heads with holoptic eyes and extensive
body pubescence. Their occurrences in preangiospermous
deposits or in later deposits that contain flowers inappro-
priate for pollination by long-proboscid insects suggest
that gymnosperms were hosts to nemestrinid and other
long-proboscate flies. Reproductive structures reachable
by air such as the unisexual cones of Cheirolepidiaceae
or possibly ovuliferous structures of Caytoniaceae were
probably targeted by nemestrinid and other flies, rather
than the closed bisexual strobili of the Cycadeoideaceae
Fig. 4. Middle Triassic to Late Cretaceous plant-insect associations (A–H) representing cycadalean, gnetalean and prob-
able pentoxylalean hosts, and Middle Jurassic to Early Cretaceous mecopteroids and dipterans with long-proboscate
mouthparts (I–P), consistent with nectarivory and possibly pollinivory. In (A), a pollen sac of the Middle Triassic cycad
Delemaya spinulosa Klavins & al. (2003) from the Central Transantarctic Mountains is laden with numerous coprolites,
one of which is enlarged as an SEM micrograph in (B) (Klavins & al., 2005) and containing monocolpate pollen of the host
plant, an undigested example of which is provided in (C) (Klavins & al., 2003). The black arrow in (A) indicates the distinc-
tive pollen sac wall, which was avoided by palynivores. An undiagnosed gymnosperm fructification closely resembling a
pentoxylalean cone, with radially juxtaposed seeds within parenchymal ground tissue, contains an encapsulated beetle
larva (D, black arrow), enlarged in (E) and reconstructed in (F), assignable to the family Nitidulidae (Nishida & Hayashi,
1996). In (G) two food boluses are evident in the abdomen of the xyelid sawfly Ceroxyela dolichocera Rasnitsyn (Krass-
ilov & al., 2003), the larger mass occurring in abdominal segments 1–3 and the smaller mass in segments 5–7, a SEM
of this last (H), shows Cryptosacciferites pabularis Krassilov & Tekleva pollen of unknown gymnospermous affinities.
The small, pseudopolycentropid mecopteroid, Pseudopolycentropus latipennis Novokshonov, from the Late Jurassic
(Oxfordian) Karabastau Formation of Karatau, in southern Kazakhstan, has a 2 mm long tubular proboscis (I) (black
arrow). Undetermined lower brachyceran fly Protapiocera sp. (Mydidae), from the Early Cretaceous (Hauterivian) of Zaza
Formation in Transbaikalia of Russia (J), possesses an elongate, tubular and distally labellate proboscis, the prelabellar
2.5 mm of which is enclosed by the vertical rectangle and shown in (K). Another species of Protapiocera from the same
Transbaikalian locality, is shown in (L), exhibiting a more gracile, long proboscis that terminates in a heart-shaped
labellum, of which one lobe is evident. Two species of aneuretopsychid mecopteroids, from the same Jurassic locality
of Karatau (Kazakhstan) as in (I), are illustrated in (M) and (N), with arrows indicating their proboscides (Rasnitsyn &
Kozlov, 1991). Aneuretopsyche rostrata Rasnitsyn & Kozlov is depicted in (m), and in (N) is a smaller specimen, ?A. min-
ima Rasnitsyn & Kozlov. The enigmatic brachyceran fly, Cratomyia macrorrhyncha Mazzarollo and Amorim (2000), from
the Lower Cretaceous (Aptian) Santana Formation of northwestern Brazil, has an 8 mm long proboscis (O), the labellate
terminus, indicated by the vertical rectangle, is enlarged in (P). The black arrow points to a terminal labellum. Scale bars:
solid = 1 cm; striped = 1 mm; stippled = 10 µm.
677
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
678
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
(Mostovski, 1998; Ansorge & Mostovski 2000; Mostovski
& Martínez-Delclòs, 2000). It is of note that deep-throated
angiosperm flowers evolved much later during the Cre-
taceous (Crepet & Friis, 1987; Rayner & Waters, 1991).
One question that remains to be resolved for some Meso-
zoic long-proboscate flies is whether pollination drops or
similar gymnospermous ovular fluids were sufficient for
nutritional balance, or if hematophagous feeding, espe-
cially by females supplemented such a diet with protein
and lipids. Diverse fluid diets incorporating carbohydrate,
protein and lipids are important nutritional requirements of
some descendant taxa (Wilson & Lieux, 1972; Watanbe &
Kamimura, 1975; Kneipert, 1980). If so, mouthpart differ-
ences between conspecific male taxa would be expected.
Gender-based mouthpart and behavioral dimorphisms also
have been documented for modern tabanids (Mitter, 1918;
McKeever & French, 1999). However, for lineages having
larvae that were parasitoids on other arthropods, such as
Nemestrinidae and Acroceridae, sufficient protein and
lipid reserves were likely accumulated from their larval
hosts so that gender-based nutritional and morphological
dimorphisms would be absent (Labandeira, 2002b).
Hymenoptera. —
The Xyelidae (sawflies) are Hy-
menoptera that are major consumers of pinaceous pollen
in modern ecosystems (Burdick, 1961). The occurrence
of xyelids during the Middle Triassic (Rasnitsyn, 1964) is
highly consistent with their phylogenetic position as the
earliest, basalmost clade of Hymenoptera (Rasnitsyn, 1980)
and their characterization as modern phylogenetic relicts.
Although xyelids, and presumably other pollinivorous
symphytan taxa, are not implicated in the pollination of
Mesozoic conifers, it is reasonable to assume that such a
diet could have been co-opted by gymnospermous plant
hosts as a pollination mutualism. Such an interpretation
is buttressed by several studies which have found gym-
nospermous pollen types in the guts of Mesozoic xyelids,
including pollen from cheirolepidiaceans, bennettitaleans,
gnetaleans, and angiosperms (Krassilov & Rasnitsyn, 1983;
Caldas & al., 1989; Krassilov & al., 1997, 2003). These
ingested pollen collectively lack large sacci, are relatively
large, and possess other features typical of insect dispersal.
After the mid-Triassic appearance of sawflies, an ex-
tensive radiation of parasitoid wasps occurred throughout
the Jurassic and into the Cretaceous (Rasnitsyn, 1980; La-
bandeira 2002b). Most of these wasps were small bodied and
could have played a major role in consuming gymnosperm
pollination drops from the Cheirolepidiaceae, Cycadales,
Gnetales, or even Pentoxylaceae and Caytoniaceae, much
in the way modern taxa currently do on Gnetum (Kato &
al., 1995), Ephedra (Bino & al., 1984a, b), and especially on
angiosperms (Jervis & al., 1993). Alternatively, parasitoid
wasps could have been involved in more specialized polli-
nation systems, such as the case of the cheirolepidiaceous
Frenelopsis alata/Alvinia bohemica plant discussed further
on. Dietary evidence from fossils is lacking, and inferences
come from the autecology of descendant taxa. The most
recent round of hymenopteran pollinators are bees and
their immediate sister taxa, albeit evidence for their earliest
occurrence resides in the late Early Cretaceous (Elliott &
Nations, 1998; Poinar & Danforth, 2006), a colonization
confined to angiosperm hosts.
Lepidoptera. —
Basal lepidopteran clades during
the Jurassic and Early Cretaceous may have been involved
in palynivory, nectarivory or seed predation, but fossil
mouthpart data are extremely sparse, and evidence from
modern taxa is indirect. Early lepidopteran lineages were
almost exclusively phytophagous and consist of small
moths whose modern descendants are known to feed
on nectar, including taxa with relatively short siphonate
proboscides (Downes, 1968; Brantjes & Leemans, 1976;
Kawakita & Kato, 2004). Significantly, one of the most
primitive clades of Lepidoptera, the Agathiphagidae, are
Fig. 5. Bennettitalean plant-insect associations emphasizing borings and their coprolite and other frass accumulations.
Specimens (A), (C), (D), and (E) are Cycadeoidea (Cycadeoideaceae); specimens (B) and (F) are Williamsonia and (G) is
a Bucklandia axis (both Williamsoniaceae). A Cycadeoidea dacotensis Wieland specimen in (A) shows the destruction
of an entire segment of ovular and adjacent interseminal scales to form a gallery (white arrow) that is connected with a
much narrower tunnel system (Delevoryas, 1968; Crowson, 1981). This specimen is from the Lower Cretaceous Blackhawk
locality of South Dakota, U.S.A. In (B), a Williamsonia harrisiana Bose specimen from the Upper Jurassic Rajmahal Basin
of India bears a circular gallery of frass which has replaced ovules and interseminal scales (Bose, 1968). From the Lower
Cretaceous of Poland are xylary, parenchymatic and other trunk tissues of Cycadeoidea sp. (Reymanówna, 1960) (C), which
contain a curvilinear tunnel filled with small ellipsoidal fecal pellets, probably made by a small larva. A larger gallery in (D)
cross-cuts receptacular, microsporophyll and possibly microsporangial tissues of a Cycadeoidea sp. specimen from the
Blackhawk locality (Crepet, 1972). In (E), an extensive larval tunnel occurs at the interface between ovules and interseminal
scales, and microsporophylls and associated microsporangial tissues in Cycadeoidea sp., also from the Blackhawk locality
(Crepet, 1974). A similar gallery occurs in a W. bockii Stockey & Rothwell specimen from the Upper Cretaceous of British
Columbia of Canada (F) (Stockey & Rothwell, 2003). A gallery traverses several ovules and their associated interseminal
scales. In Bucklandia kerae Saiki & Yoshida from the Upper Cretaceous of Japan (G), are tunnels within trunk cortical
tissues, representing early-instar larval or less plausibly oribatid mite activity (Saiki & Yoshida, 1999). A Monosulcites
pollen grain, typical of bennettitalean microsporangia, is illustrated in (H) (Taylor, 1973). The spatio-temporal distribution of
insect-damaged taxa illustrated in (A–G) is detailed in (I) and (J); horizontal lines represent the angiosperm radiation. The
habitus of W. sewardiana Sahni is illustrated in (K) (Taylor & Taylor, 1993). Scales: solid = 1 cm; stippled = 10 µm.
679
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
680
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
seed predators on the araucarian genus Agathis (Kris-
tensen, 1999)—an association that may extend to the
Jurassic (Powell & al., 1999). Other more derived linea-
ges are the Nepticulidae and Gracillariidae, which were
well established by the latest Early Cretaceous (Lopez-
Vaamonde & al., 2006). Unfortunately lepidopteran si-
phons are rarely preserved in the fossil record even though
their mouthpart structure in modern forms is well docu-
mented (Szucsich & Krenn, 2000).
tWo probablE mESoZoIc
pollInatIon SYndromES
on GYmnoSpErmS
We present here two examples of host plants, Fre-
nelopsis alata and Cycadeoidea dacotensis, for which
paleobotanical and paleoentomological evidence support
the existence of insect feeding on pollen, pollination drops
or other secretory fluids, or some combination thereof,
as well as the presence of pollination mutualisms. Our
reconstruction of these associations is based on from
multiple lines of evidence: the reproductive morphology
of plants, plant damage, insect gut contents and mouthpart
structure (Labandeira, 2002a).
Ovuliferous and microsporangiate cones of Fre-
nelopsis alata (Cheirolepidiaceae). —
Historically, the
pollen of Cheirolepidiaceae, Classopollis, has presented
some of the most compelling evidence for entomophily
in Mesozoic gymnosperms. Classopollis pollen is char-
acterized by comparatively large, smooth, disk-shaped to
flattened spheroidal grains resting in tetrads by adherent
exinal threads (Scheuring, 1976; Courtinat, 1980). It has
been postulated that plants bearing such pollen were self-
incompatible (Zavada & Taylor, 1986). The Classopollis
character complex was especially prevalent among taxa
occurring prior to, and during, the ecological expansion of
angiosperms; however, Late Cretaceous cheirolepidiaceous
taxa were anemophilous (Pocock & al., 1990). The presence
of a typical gymnospermous pollination drop mechanism
was probably lacking or modified because the pollen cham-
ber in these plants is atypical. Nonetheless, for most species
fluid films probably assisted pollen landing on ovuliferous
cones to germinate and directed the pollen tube toward the
ovule for a considerable distance (Taylor & Taylor, 1993).
Pollen transport to the ovuliferous cones could have been
made by taxa such as haglid orthopterans (Krassilov &
al., 1997), phasmatodeans (Krassilov & Rasnitsyn, 1999),
xyelid hymenopterans (Krassilov & Rasnitsyn, 1983),
mecopteroids (Rasnitsyn & Kozlov, 1991; Novokshonov,
1997), and brachyceran dipterans (Ren, 1998; Mostovski &
Martínez-Delclòs, 2000; Labandeira, 2005), as evidenced
by gut contents containing abundant Classopollis in the
first three examples, and siphonate mouthparts and clusters
of Classopollis pollen grains on the heads of other taxa.
These five, phylogenetically distant insect groups variously
present during the Late Triassic to Early Cretaceous, dem-
onstrate generalized pollinivory, and suggest pollination
types in which both pollen and fluid plant secretions were
important rewards. Of the five, the mecopteroids and dip-
terans may have been the only pollinators.
The relatively simple system of pollination typical of
gymnosperms (Figs. 1, 2) was present in some Mesozoic
cheirolepidiaceous taxa, whose reproductive biology
recently has become better known (Pocock & al., 1990;
Clement-Westerhof & van Konijnenburg-van Cittert, 1991;
Axsmith & al., 2004). The best example is based on the
morphology and associated pollination biology of the ov-
uliferous cone of Alvinia bohemica and its corresponding
microsporangiate cone and foliage, Frenelopsis alata ; the
whole plant of which grew in saline habitats and relatively
dry climates during the earliest Late Cretaceous in what
isnowtheCzechRepublic(Kvaček,2000).Ovuliferous
cones were broadly ovoid, surmounted on a robust axis
about 1.0 cm in width, approximately 5.0 cm long and
somewhat shorter in width, and consisted of at least 30
helically arranged imbricate, rhomboid scales (Fig. 3A).
Ovuliferous cone scales (Fig. 3B) were subtended from
the axil of a wide bract, were fleshy rather than woody,
and bore sterile basal and apical scales. The structure of
mature ovuliferous cone scales was complex and consisted
of two ovules laterally positioned next (proximal) to the
cone axis at maturity, or alternatively there was a single
functioning ovule adjacent a considerably smaller aborted
ovule. Each ovule had a 1.8 mm long micropyle projecting
proximally, and a chamber which narrowed toward the
distal ovular apex; a single strongly ribbed integument
about 1 cm in diameter surrounded the ovule (Fig. 3B).
The position, structure, and surface features of ovulifer-
ous cone-scale appendages are important for understanding
the pollination biology of Alvinia bohemica. On the upper
margin of the cone scale, proximal to its subtending axis,
was a single covering flap which overlaid and protected
each ovular region up to the level where the micropyle and
tubular “pipe-like structure” protruded (Fig. 3B; also see
below). Distal from the axis, at the upper margin of the
ovuliferous cone scale there was a medially inserted, single
appendage that was the continuation of the covering flap
and formed another upper, but outwardly directed lobe of
the ovuliferous scale. Also distally positioned, but located
centrally below the uppermost appendage were two lateral
distally directed appendages, the upper lobate margins of
which were trichome-lined and the lobes overlapped medi-
ally. Each of these lateral lobes was attached to the ovulifer-
ous scale by an oblique ridge that also was associated with
ovular insertion (Fig. 3F). These conspicuous and flattened
appendages, together with the uppermost appendage, out-
wardly and distally surrounded the deep, cone-shaped fun-
681
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
nel that narrowed proximally into a small aperture (Fig. 3C),
the continuation of which was a “pipe.” This pipe apparently
traversed lower ovuliferous scale tissues below the ovular
integument and emerged proximally adjacent to the micro-
pyle. Presumably this tube aided pollen tube growth toward
the micropyle by providing transmission tissue or perhaps
fluids, which nourished and/or directed the pollen tube (see
Frame 2003a). As a channel for directing pollen toward the
pipe, the funnel functioned as an “inverted stigma” and was
lined throughout by prominent multiseriate trichomes (Fig.
3E), laden mostly with Classopollis but also other types of
pollen. In the deeper recesses of the funnel occurred large,
cylindrical, multicellular protuberances, or papillae, that
were significantly larger than adjacent trichomes (Fig. 3D).
These cellular papillae contained secretary fluids whereas
trichomes likely were responsible for pollen entrapment.
Below the funnel and oblique ridge was the main body of
the ovuliferous scale, in whose axil was attached a single,
large, transversely elongate ellipsoidal bract (removed in
Fig. 3F), which extended as far up as the lower periphery
of the funnel orifice.
The pollen cone, Frenelopsis alata (Fig. 3G), was
the source of Classopollis pollen which pollinated A.
bohemica ovules. Male (Frenelopsis alata) cones were
considerably smaller than female (A. bohemica) ones;
they were slightly greater than 1.0 cm long and had a
more conifer-like appearance than the fleshy ovuliferous
cones. Each diamond-shaped microsporangiate cone scale
was fringed with prominent trichomes along the margin
(Fig. 3I) that were a likely source of food for fluid-feeding
insects. The microsporophylls each bore two (perhaps
more?) pollen sacs (Fig. 3H), one on each side of its base.
When cones were mature, pollen sacs released numerous
pollen tetrads (Fig. 3J), either by sac wall degradation,
insect disturbance during trichome manipulation or di-
rect consumption by insects, or a combination of these.
Inferred insect vectors include basal clades of long-pro-
boscate mecopteroids, small therevid flies (Figs. 3K–N)
and much larger long-proboscate mydid flies (Figs. 3O,
P); much larger pollinivorous phasmatodeans (Figs. 3Q,
R) and orthopterans (Figs. 3S, T) are more remote possi-
bilities. A small-bodied (3.5 mm) therevid fly is depicted
in Fig. 3 (K, L) as the pollinator of A. bohemica, although
long-proboscate flies also are candidates.
Pollination of Alvinia bohemica ovules could have been
achieved either by wind currents (anemophily), transport by
insect vectors (entomophily), or both strategies (ambophily).
Evidence for wind pollination principally is provided by
the elevated abundance of Classopollis pollen in many
Mesozoic environments occupied by cheirolepidiaceous
plants—both in palynologically macerated sedimentary
matrices (Vakhrameev, 1991; Upchurch & Doyle, 1981)
andinthereceptivetissuesofovuliferouscones(Kvaček,
2000). Such abundance, typical for anemophilous plants,
however, is known to occur in some insect-pollinated an-
giosperms, such as Papaver rhoeas L. (McNaughton &
Harper, 1960). By contrast, a diverse suite of evidence from
both plant and insect structures indicate entomophilous
pollination. First are atypical anatomical features of several
cheirolepidiaceous taxa which are understandable in the
context of insect-pollination, such as the aforementioned
stigma-like structures of A. bohemica (Figs. 3A–J; Kva-
ček,2000).SecondarefeaturesofindividualClassopollis
grains, including their comparatively large size (Figs. 3J,
N, P, R, T), thick exine, and grains occurring as tetrads
held together by sticky exinal threads (Taylor & Alvin,
1984; Clement-Westerhof & van Konijnenburg-van Cittert,
1991). Third is the presence of pollen tetrads and single
pollen grains of Classopollis on the heads of likely Late
Jurassic and Early Cretaceous pollinating insects. Fourth
is the presence of insects with elongate siphonate probosci-
des, or alternatively very small-bodied insects, particularly
dipterans (Fig. 3M), that were present in Middle Jurassic
to mid-Cretaceous floras either prior to the earliest fossil
occurrence of angiosperms or in floras contemporaneous
with the initial appearance of angiosperms but prior to the
advent of angiosperms having tubular flowers. Fifth is the
occurrence of near-monospecific Classopollis food boluses
in the guts of members of Phasmatodea, Orthoptera, and
Hymenoptera, indicating the frequent use of cheirolepidi-
aceous pollen as food. Collectively these data indicate that
entomophily was widespread among cheirolepidiaceous
plants and that pollination was carried out by several dis-
tantly related insect groups.
Bisexual strobilus of Cycadeoidea dacotensis
(Cycadeoidaceae). —
Bennettitalean strobili of the
Cycadeoideaceae were closed, bisporangiate (hermaphro-
ditic) structures not open to the dissemination of pollen by
wind, thus requiring in situ mechanisms of pollen transfer
(Crepet, 1974). Several important studies have documented
damage to receptacles, ovules, microsporophylls and other
strobilar tissues (Reymanówna, 1960; Delevoryas, 1968;
Bose, 1968; Crepet, 1972, 1974; Crowson, 1981; Saiki &
Yoshida,1999;Stockey&Rothwell,2003).Thisdamage
is analogous to known beetle damage in extant cycads
(Crowson, 1981; Norstog & Nicholls, 1997; Labandeira,
1998b). Insect-mediated bennettitalean damage, found
on taxa such as Cycadeoidea dacotensis from the Upper
Lower Cretaceous of South Dakota (Fig. 6A), but also on
several other taxa belonging to the Cycadeoideaceae and
Williamsoniaceae, provide rare autecological snapshots of
the life-history of insect larvae as they proceeded from egg
hatching, to early larval penetration of external strobilar
tissues, to consumption of various internal bennettitalean
vegetative and reproductive tissues, to eventual emergence
from the host plant and subsequent transformation to the
adult phase. In addition to the best documented species,
C. dacotensis, our reconstructed life-history is gleaned
682
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
from data of several other plant-host taxa (Figs. 5A–G),
each of which document one or more phases of tissue
consumption and formation of a tunnel or gallery systems.
These fossil data collectively are integrated to a general-
ized account of insect consumption of live tissues within
Cycadeoidea strobili and their associated pollination (Fig.
6), using the reconstruction by Crepet (1974) as a basis for
charting larval life-history. However, the insect damage
patterns do not represent any single host-plant species.
Rather, a general pattern of herbivory, ovule predation and
pollination is established that undoubtedly was common
to multiple beetle taxa engaged in similar associations
among multiple bennettitalean host plants (Reymanówna,
1991; Labandeira, 1998b).
From these data the associational pattern of insect
larvae and bennettitopsid plant hosts in general and C.
dacotensis in particular can be divided into five, more or
less characterizable larval phases. The first phase was ovi-
position and initial larval entry into accessible or otherwise
exposed strobilar tissues, most likely the tops of microspo-
rophylls, bracts, or possibly adjacent tissues such as the ra-
metum that forms the ground tissue from which the apices
of strobili project beyond the general Cycadeoidea trunk
surface (Fig. 6A, B). Insects likely oviposited in or on the
surface of these structures, the stimulus for which may have
involved a period of ovular receptivity, analogous to that
of extant cycads (Stevenson & al., 1998). The second stage
was penetration and consumption by early larval instars of
vegetative tissues initially microsporophylls and bracts, and
subsequently, trunk tissues through the creation of small-
diameter tunnels (Figs. 5C, G; 6C). As subsequent larval
instars underwent size increases, the third phase was initi-
ated by the establishment of a tunnel system that occurred
in the stem (Fig. 6D) and thence into the zone between the
inner ovule-bearing and interseminal scales originating
from the receptacular axis and the outer, more peripheral
pollen-bearing microsporophylls (Figs. 5E; 6E). The nutri-
tional lure for these larvae could have been nutritive fluids
secreted by the micropylar surfaces, nearby pollen sacs, or
perhaps the fleshy, club-shaped ends of the interseminal
scales, or any combination of these and adjacent tissues
(Haslett, 1989). The fourth phase was occasional excursions
of the tunnel system from this zone, principally consisting
of consumption of pollen sacs and their contents as well as
inner ovules and their adjacent interseminal scales. In these
gamete-associated tissues spacious galleries were formed
and filled by unconsumed plant fragments, fecal pellets,
and other frass (Figs. 5A, B, D, F; 6F, G). Larger instars
targeted the reproductive tissues, likely resulting in pollen
transfer from pollen sacs to the micropyles of unconsumed
ovules. The fifth and final phase of larval development was
when the final instar constructed a comparatively large tun-
nel that was established toward the outer, exposed tissues
and where emergence occurred (Fig. 6H, I). This phase was
terminated by pupation into the adult instar occurring either
within a terminal chamber or alternatively by the larva
dropping to the leaf litter below. Adult beetles continued
the life cycle, copulating outside the plant then ovipositing
on conspecific hosts.
Overall, the type of damage found in bennettitalean
strobili is most consistent with relatively small, compact,
robust beetles (Crepet, 1974), most likely members of the
suborder Polyphaga or its subclade, the Phytophaga, which
includes cerambycids (longhorned beetles), chrysomelids
(leaf beetles), and curculionoids (weevils) (Farrell, 1998).
The Phytophaga antedate considerably the origin of angio-
sperms (Arnol’di & al., 1977; Zhang, 2005). In particu-
lar, the plesiomorphic curculionoid families, Belidae and
Nemonychidae (Zimmermann, 1994), may be the closest
extant lineages exhibiting similar life-history attributes
to that inferred for mid-Mesozoic bennettitalean damage.
Insect damage on bennettitalean strobili is consistent with
extant coleopteran pollinators of cycads, which have life
cycles featuring adults as pollinivores or folivores and
larvae as consumers of pollen, adjacent microsporangi-
ate tissues, and receptacular pith tissues (Crowson, 1991;
Norstog & al., 1992; Oberprieler, 1995).
There is limited paleobiological data on the insect
associates of bennettitaleans. One interaction is the
presence of the bennettitalean pollen, Vitimipollis, in
the gut of a xyelid sawfly from the Early Cretaceous of
Russia (Krassilov & Rasnitsyn, 1983). However, the gut
Fig. 6. The inferred insect pollination mechanism for Cycadeoidea dacotensis Wieland (Bennettitopsida: Cycadeoidaceae).
The plant-host anatomy is based on reconstructions by Crepet (1974). For the purposes of simplicity, larval borings and
galleries are illustrated in 2-dimensions. Patterns of plant-insect damage are from sources illustrated in Fig. 5. A whole-plant
reconstruction is provided in (A) (Delevoryas, 1971), bisporangiate cones are embedded in a thick rametum of bracts and
other tissues. A three-dimensional cut-away view of C. dacotensis is provided in (B), illustrating important tissues for insect
borers involved in pollination. The initial oviposition site is given in (C), with eggs deposited on or in surface tissues. In (D)
is depicted the initial larval boring through vegetative tissues such as microsporophyll parenchyma and receptacle (and
possibly trunk?) tissues (Reymanówna, 1960). Tunnels formed at the intersection of the ovular/interseminal scale layer and
the microsporophyll layer are provided in (E), with micropylar secretions and pollen probably being the principal rewards.
Gallery formation occurs within the zone containing receptacle-borne ovules and adjacent interseminal scales, replete with
coprolites and other types of frass deposition as a consequence of extensive consumption (F) (Delevoryas, 1968; Crowson,
1981). The final phase of larval borings are provided in (H) and (I), resulting in an exit hole associated with pupation that may
occur in a terminal chamber within the host plant or on the subjacent ground as in some extant cycad pollinating beetles
(Oberprieler, 2004). Drawings made by the senior author.
683
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
684
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
contents of this specimen also contained conifer pollen
of Alisporites and Pinuspollenites, indicating a polylectic
adult feeding strategy. Also, there are three clades of basal
weevils—Belidae, Nemonychidae, and Eccoptarthridae
as well as the enigmatic Obrieniidae—that infrequently
co-occur with bennettitaleans in certain Late Jurassic and
Early Cretaceous deposits (Doludenko & Orlovskaya,
1976; Arnol’di & al., 1977; Zherikhin & Gratschev, 1997;
Sun & al., 2001; Gratschev & Zherikhin, 2003). Extant
descendants of these first two weevil clades currently are
found on cycad and conifer plant hosts (Zimmermann,
1994; Farrell, 1998). Additional features indicating insect
pollination are: (1) the relatively large size and psilate
exine sculpture of bennettitalean pollen, characters
associated with animal pollination; (2) the presence of
extranuptial nectaries in microsporangiate cones of one
species (Harris, 1973), possibly a nutritional lure for pol-
linators; and (3) decurved, rostrate mouthparts with robust
terminal mandibles characteristic of many curculionoid
taxa from Late Jurassic and Early Cretaceous deposits
indicative of ovipositing into plant tissues (Zimmerman,
1994; Anderson, 1995; Labandeira 1997).
Endophytic insect damage on bennettitalean strobili
has been found in five geographically disjunct localities
from the Late Jurassic to early Late Cretaceous (Figs. 5I,
J). The most extensive documentation of tunnel and gallery
formation is known from South Dakota, U.S.A. (Figs. 5A,
D, E; Delevoryas, 1968; Crepet, 1974), but other similar ex-
amples have been described from British Columbia, Canada
(Fig. 5F; Stockey & Rothwell, 2003), Japan (Fig. 5G; Saiki
&Yoshida,1999),Poland(Fig.5C;Reymanówna,1960),
and India (Fig. 5B; Bose, 1968). The widespread geographic
distribution, targeting of a particular plant-host clade and
highly stereotyped damage of this association collectively
suggest a distinctive pollination strategy by beetles similar
to that of extant cycads (Stevenson & al., 1998). It is possible
that there was a transfer of the bennettitalean insect pollina-
tion syndrome, though not necessarily by the same taxa, to
that of cycads during the Jurassic, long before the ecological
expansion of angiosperm pollination mechanisms. This
co-optation may have been likely since the three extant
families of cycadopsids (Cycadaceae, Stangeriaceae, Zam-
iaceae) geochronologically overlap with the bennettitalean
Williamsoniaceae (including the “Wielandiaceae” of some)
and early Cycadeoideaceae (Artabe & Stevenson, 1999,
2004; Anderson & al., 2007). It also is possible that this
syndrome of beetle pollination of relatively closed repro-
ductive structures was common during the Mesozoic, early
evidence for which is provided by Middle Triassic insect
consumption of pollen in cycad reproductive organs (Figs.
4A–C; Klavins & al., 2005), and evidence provided by a
Late Cretaceous permineralized fructification, probably
referable to the closely related Pentoxylales, which harbored
a completely preserved larva within a chamber adjacent
to seeds within the ground tissue (Fig. 4D–F; Nishida &
Hayashi, 1996). This larva was assigned to the beetle family
Nitidulidae, for which adults of modern representatives
are frequent pollen feeders (Gazit & al., 1982; Ekblom &
Borg, 1996).
thE broadEr palEobIoloGIcal
contExt
Two issues are evident from this examination of the
Mesozoic seed-plant record of insect associations involv-
ing pollination and consumption of reproductive associ-
ated structures. First, how do Mesozoic preangiospermous
pollination mechanisms fit into a larger perspective of in-
sect consumption of reproductively associated structures
through time? Second, what can be inferred in the switch
from gymnosperm- to angiosperm dominated modes of
insect pollination.
Insect consumption of reproductively-associated
plant tissues in deep time. —
The fossil record has
revealed that palynivory (consumption of spores, prepollen
and pollen) is one of the most ancient feeding strategies
of terrestrial arthropods in general and hexapods in par-
ticular (Labandeira, 2000, 2006). Coprolite assemblages
laden with spores, many plant-host specific, are known
from the latest Silurian and Early Devonian of the United
Kingdom (Edwards & al., 1995). These and other associ-
ations (Labandeira, 2000, 2006) constitute Phase 1 of the
development of palynivory, nectarivory and associated
diets, though only spore consumption occurred (Fig. 7,
middle). The subsequent, Carboniferous coprolite record
shows the consumption of fern spores and sporangia,
particularly those of marattialean tree ferns, and towards
Fig. 7. Summary of seed-plant and insect families for which there is some to considerable evidence for nectarivory, pal-
ynivory or pollination. The emphasis is on Phase 3 associations from the Middle Triassic to the mid-Cretaceous angio-
sperm radiation, as detailed in Labandeira (2000, 2006). Levels of confidence are assigned to possible (stippled), probable
(cross-hatched) and unequivocal (dark grey) associations based on available features such as plant reproductive biology,
plant damage, dispersed insect coprolites, insect gut contents, insect mouthparts, and to a lesser extent taxonomic uni-
formitarianism (see Labandeira 2002a). Note for insects, Haglidae (here the Aboilinae) are a family of Orthoptera and Aeo-
lothripidae (Cycadothripinae) a family of Thysanoptera. Geochronologic abbreviation: Dev., Devonian Period. Higher-level
plant-group abbreviations: LAG., Lagenostomopsida; PIN., Pinopsida, and BENNETT., Bennettitopsida. Geochronological
range data are from a variety of sources, including Taylor & Taylor (1993), Labandeira (1994), Grimaldi & Engel (2005), and
Anderson & al. (2007), based on the time scale of Gradstein & al. (2004).
685
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
200
250
150
100
50
0
350
300
cissairTnaimrePsuorefinobraC.veD cissaruJ suoecaterC enegoelaP enegoeN
aM doireP
angiosperm
radiation
Taxon-time envelope for Phase 3 associations
Associational
Phases
(Labandeira, 2006)
Insects:Seed
Plants:
Nectarivory, or palynivory, or pollination present
Nectarivory, or palynivory, or pollination probable
Nectarivory, or palynivory, or pollination possible
1
2
3
4
eadilihpotecyM
eadinabaT
eadinoilemreV eadirecoipA
eadiiymotarC
eadiluditiN
eadilgaH
eadiremedeO
eadipirhtoloeA eadiryleM
eadilledroM
eadinirtsemeN
eadiirugnaL eadicybmareC eadileB eadihcynomeN
eadinoilucruC eaditnerB eadirhtratpoccE
eadihcysposeM eadihcyspoteruenA eadiportnecylopoduesP eadilupiT
eadinoibiB
eadirecorcA
eadiverehT
eadidyM eadiniponecS eadidipmE eadiilybmoB eadihpryS eadicsuM eadipeloeahcrA eadigyretporciM eadilucitpeN eadixodorP eadireiP
eadilahpmyN
eadinoilipaP eadinidoiR
eadineacyL eadiutcoN eadilaryP eadileyX eadinoagA eaditelloC eaditcilaH eadinerdnA eadittileM eadilihcageM eadipA
eaecamotsonegaL eaecatyhpotsillaC eaecahtnatiadroC eaecaidipeloriehC eaecaeinotoP
eaecaimaZ
eaecasolludeM
eaecairegnatS eaecadacyC
eaecainotyaC eaecailleidnaleiW eaecainosmailliW eaecaedioedacyC eaecalyxotneP eaecahtnaoE eaecaiwerD eaecaderhpE eaecatenG eaecaihcstiwleW smrepsoigna
COLEOPTERA DIPTERAMECOPTERA LEPIDOPTERA HYMENOPTERAPIN.LAG. CYCADOPSIDA GNETOPSIDABENNETT.
686
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
the end of the period, specialist feeding on cordaite and
pteridosperm prepollen (Labandeira, 1998a; 2002a). The
deposit at Chekarda, from the Lower Permian of the
Russian central Ural Mountains, bear several species of
diverse insect lineages that have gut contents of pollen that
are affiliated with equally diverse plant hosts (Krassilov
& Rasnitsyn, 1999; Afonin, 2000). These palynivores
include early insect taxa which were consuming pollen
from peltasperm and glossopterid seed ferns, cordaites,
conifers, and gnetopsids (Rasnitsyn & Krassilov, 1996;
Krassilov & Rasnitsyn, 1999; Krassilov & al., 2006). The
prepollen and pollen form taxa occurred as both mixed and
monospecific gut accumulations, supporting earlier, well-
preserved coprolite data from coal-swamp forests from the
Late Carboniferous in which there was intense targeting of
particular host plants for consumption (Labandeira, pers.
observ.). Also at Chekarda are body fossils of insects that
display prominent, prognathous mandibulate mouthparts,
interpreted as adaptations to pollinivory (Rasnitsyn, 1977;
Novokshonov, 1997). Unlike pollinivory, Paleozoic evi-
dence for feeding on plant secretions is considerably less
and indirect, such as the existence of the pollination drop
mechanism (Rothwell, 1972, 1977) and the presence of
epidermal secretory glands adjacent to reproductive struc-
tures (Mamay, 1976; Krings & al., 2002). By the end of the
Paleozoic, insect targeting of fern spores and seed plant
pollen was well-established and were common occurrences
in at least some terrestrial habitats. The Carboniferous and
Permian interval represents Phase 2 of palynivore, nectar-
ivore and related associations (Fig. 7, middle) and provides
a prelude to Mesozoic modes of insect consumption of
gymnospermous reproductive tissues.
During the Triassic there is minimal evidence for
consumption of pollen or nectar, except for pollen-laden
coprolites within pollen organs from a taxon that is “re-
markably similar” to that of extant cycads (Figs. 4A–C;
Klavins & al., 2005). Other evidence is more indirect, such
as the earliest appearance of pollinivorous xyelid sawflies
(Rasnitsyn, 1964; Krassilov & Rasnitsyn, 1983), and early
mecopteroid lineages that are known to have had long-
proboscate, apparently nectar-feeding, mouthparts that date
from the Middle Jurassic (Fig. 4I, M, N; Rasnitsyn & Ko-
zlov, 1991; Novokshonov, 1997). During the Late Jurassic
and Early Cretaceous, there is significant documentation
of gymnosperm pollinivory from Eurasia (Figs. 3Q–T; 4G,
H; Rasnitsyn & Krassilov, 1996; Krassilov & al., 1997).
Concomitant with this, and continuing into the mid-Cre-
taceous, was the appearance of several other major insect
mouthpart types which were designed for feeding on nectar
or pollen, or for boring into ovular or seed tissues. These
forms are found among several major lineages of brachy-
ceran flies (Figs. 4J–L, N, P; Rohdendorf, 1968; Mostovski,
1998; Ren, 1998; Mazzarollo & Amorim, 2000) as well as
phytophagous curculionoid weevils (Arnol’di & al., 1977).
Thus, Phase 3 (Figs. 3–5) is marked by palynivory, nectar-
ivory and related associations during the Early Triassic to
mid-Cretaceous (Fig. 7, middle). By contrast, Phase 4 of
these associations is the most recent expansion of plant-
palynivore and related associations, which began during
the mid-Early Cretaceous expansion of angiosperms and
has continued to recent times. This co-radiation eclipsed,
but also highlighted, an earlier interval whereby currently
extinct gymnosperm host plants and their insect pollinators
contributed major pollination strategies.
The Mesozoic shift from gymnosperm to angio-
sperm entomophily. —
The disappearance of at least
some of the older, Mesozoic lineages of gymnosperms and
their insect pollinators probably was attributable to the
appearance of angiosperms during the Early Cretaceous.
Angiosperms provided a more nutritionally efficient
system for consumption of surface fluid and transfer
of pollen rewards among conspecific hosts when com-
pared to older entomophilous gymnosperm lineages, an
idea compatible with Frame’s (2003b) hypothesis of the
overall greater edibility of angiosperms. Entomophilous
cheirolepidiaceous taxa became extinct during the Late
Cretaceous, perhaps because they were at a disadvantage
in the context of more efficient angiosperm insect polli-
nation systems. For example, the presence of an intricate,
inverted stigma-like structure at considerable distance
from the micropyle but connected to it by a tubular struc-
ture that resulted in eventual pollination, may have been
inefficient when compared to early angiosperm struc-
tures, such as simple stigmatic exudates secreted prior to
anthesis (Frame, 2003a), small and perfect flowers with
nectaries (Gottsberger, 1988; Thien & al., 2000), or even
more complex floral mechanisms involving pollinator
entrapment (Thien & al., 2003; Gandolfo & al., 2004).
Similarly, but representing a different pollination pattern,
entomophilous bennettitalean taxa were supplanted by
more highly refined and host specific beetle (and thrips?)
pollination systems of extant cycad lineages (Norstog &
Nicholls, 1997; Terry, 2001). Notably, insect exploitation
of the pollination-drop was not completely transferred
to or terminated with angiosperm ecological expansion
during the Late Cretaceous. It currently survives or has
re-evolved in the form of diverse small insects, especially
flies, moths, and parasitoid wasps having abbreviated
labellate and sponging proboscides, on gnetopsid and, to
a lesser extent, cycadopsid plants providing micropylar
secretions (Kato & al., 1995; Tang, 1995).
SummarY and concluSIonS
Seven summary statements and concluding inferen-
ces can be made from this examination of pollination and
related associations from the preangiospermous Mesozoic.
687
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
Ackerman, J.D. 2000. Abiotic pollen and pollination: ecolog-
ical, functional, and evolutionary perspectives. Pl. Syst.
Evol. 222: 167–185.
Afonin, S.A. 2000. Pollen grains of the genus Cladaitina ex-
tracted from the gut of the Early Permian insect Tillyard-
embia (Grylloblattida). Paleontol. J. 34: 575–579.
Alvin, K.L. 1982. Cheirolepidiaceae: biology, structure and
paleoecology. Rev. Palaeobot. Palynol. 37: 71–98.
Alvin, K.L. & Hluštík, A. 1979. Modified axillary branching
in species of the fossil genus Frenelopsis: A new phenom-
enon among conifers. Bot. J. Linn. Soc. 79: 231–241.
Alvin, K.L., Watson, J. & Spicer, R.A. 1994. A new conifer-
ous male cone from the English Wealden and a discussion
of pollination in the Cheirolepidiaceae. Palaeontology 37:
173–180.
Anderson, J.M. & Anderson, H.M. 2003. Heyday of the gym-
There is emerging evidence for a largely extinct phase
of preangiospermous pollination syndromes involving
several major clades of seed plants and insect pollinators.
New fossil data suggest that we have only a glimpse of
these extinct pollination biologies.
(1) Extant gymnospermous seed plants provide a
source of liquid nutrition, typically in the form of mi-
cropylar pollination drops, used by a variety of insects,
especially small flies. Such associations undoubtedly were
present among gymnospermous seed plants during the
preangiospermous Mesozoic.
(2) Major Mesozoic plant-host clades or groups which
have suspected to well-documented palynivore, nectari-
vore and pollination associations include the Pteridosper-
mopsida, Pinopsida, Cycadopsida, Bennettitopsida, and
Gnetopsida. These taxa have family-level lineages that
were largely extinguished during the later Mesozoic.
(3) Major Mesozoic insect nectarivores, pollinivores
or pollinators for which there is circumstantial to well
corroborated evidence are the Orthoptera, Phasmatodea,
Embioptera, Coleoptera, Mecopteroidea, Diptera, Hymen-
optera, and Lepidoptera. The Coleoptera, Mecopteroidea
and Diptera present the most convincing evidence for an
early pollinator role, with either nectar or pollen being
the primary rewards.
(4) Many of the gymnosperm plant-host clades and
their insect nectarivore, pollinivore or pollinator clades
became extinct during the mid to Late Cretaceous for a
variety of reasons, but probably involving the more effi-
cient stigma-based angiosperm pollination system. One
reason for this replacement was that some gymnosperm
taxa possessed structurally and functionally complex pol-
lination systems that were superseded by more efficient
angiosperm systems.
(5) Two gymnosperm seed-plant clades had prob-
able pollination associations with insects. The first is
the outcrossing coniferalean Frenelopsis alata/Alvinia
bohemica whole-plant species (Cheirolepidiaceae), which
bore accessible ovuliferous scales with stigma analogs.
The second is a bennettitalean species of Cycadeoidea
(Cycadeoideaceae) possessing closed strobili. These two
plants had very different types of pollination mutualisms.
For the Cheirolepidiaceae, small-bodied or alternatively
large-bodied, long proboscid flies aerially carried pollen
from other plants, presaging the evolution of pollination
mechanisms in angiosperms. For the Cycadeoideaceae,
in situ beetle larvae consumed internal vegetative and
reproductive tissues and in the process transported pol-
len within the closed strobilus, analogous to pollination
mechanisms in cycads.
(6) Mesozoic plant-insect associations involving
gymnosperms and their nectarivores, pollinivores and
pollinators described herein (as well as other feeding
guilds of herbivores), represent the third of four distinctive
phases of plant-insect associations which characterize the
fossil record. This long-ranging third phase involved a
distinctive colonization of gymnosperm plant hosts by
insects followed by extensive diminution and replacement
by mid Cretaceous angiosperms.
(7) Anatomically and structurally well-preserved
compression and permineralized fossils provide signif-
icant evidence addressing major issues in the pollination
history of insects and seed-plants. Such fossil evidence
can supply information and insights which are otherwise
difficult (or impossible) to gain by means of phylogenet-
ical and ecological approaches. The modern perspective
is based on a highly culled, small sample of recent gym-
nospermous seed plants and their insect associates.
acKnoWlEdGEmEntS
The senior author would like to thank Dawn Frame and
Gerhard Gottsberger for an invitation to provide a presenta-
tion in the Symposium, “Generalist Flowers: Their Evolution,
Biology and Animal Associations,” at the Seventeenth Inter-
national Botanical Congress in Vienna during July of 2005.
Many thanks go to Finnegan Marsh for rendering Figures 1 to
7. Appreciation also is extended to John Anderson and Jason
Hilton for permission to use material that was in press during
the drafting of this manuscript. William Crepet, Dawn Frame
and an anonymous reviewer provided valuable comments on
an earlier draft of this contribution. Thanks also go to Carol
Hotton for identification of the pollen on the insect bodies
and to Jessica Corman for collection of data in Table 1. This
is contribution 149 of the Evolution of Terrestrial Ecosys-
tems consortium of the National Museum of Natural History
(NMNH). C.L.’s travel was funded by the IBC and NMNH,
and J.K. was supported by the Ministry of Culture of the Czech
Republic (MK 00002327201).
lItEraturE cItEd
688
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
nosperms: systematics and biodiversity of the Late Triassic
Molteno fructifications. Strelitzia 15: 1–398.
Anderson, J.M., Anderson, H.M. & Cleal, C. In press. A brief
history of the gymnosperms: classification, biodiversity
and ecology. Strelitzia 18.
Anderson, R.S. 1995. An evolutionary perspective on diversity
in Curculionoidea. Mem. Entomol. Soc. Washington 14:
103–114.
Ansorge, J. & Mostovski, M.V. 2000. Redescription of Prohir-
moneura jurassica Handlirsch 1906 (Diptera: Nemestrin-
idae) from the Lower Tithonian lithographic limestone of
Eichstätt (Bavaria). Neues Jahrb. Geol. Paläontol., Mon-
atsh. 2000: 235–243.
Arnol’di, L.B., Zherikhin, V.V., Nikritin, L.M. & Ponoma-
renko, A.G. 1977. Mesozoic Coleoptera. Trans. Paleontol.
Inst., Acad. Sci. USSR 161: 1–204 [in Russian].
Artabe, A.E. & Stevenson, D.W. 1999. Fossil Cycadales of
Argentina. Bot. Rev. 65: 219–238.
Artabe, A.E., Zamuner, A.B. & Stevenson, D.W. 2004. Two
new petrified cycad stems, Brunoa gen. nov. and Wors-
dellia gen. nov., from the Cretaceous of Patagonia (Bajo
de Santa Rosa, Río Negro Province), Argentina. Bot. Rev.
70: 121–133.
Ash, S.R. 1997. Evidence of arthropod-plant interactions in the
Upper Triassic of the southwestern United States. Lethaia
29: 237–248.
Axsmith, B.J. & Jacobs, B.F. 2005. The conifer Frenelopsis
ramosissima (Cheirolepidiaceae) in the Lower Cretaceous
of Texas: systematic, biogeographical, and paleoecological
implications. Int. J. Pl. Sci. 166: 327–337.
Axsmith, B.J., Krings, M. & Waselkov, K. 2004. Conifer pol-
len cones from the Cretaceous of Arkansas: implications
for diversity and reproduction in the Cheirolepidiaceae.
J. Paleontol. 78: 402–409.
Baker, H.G. & Baker, I. 1983. A brief historical review of the
chemistry of floral nectar. Pp. 126–152 in: Bentley, B. &
Elias, T. (eds.), The Biology of Nectaries. Columbia Univ.
Press,NewYork.
Baker, H.G. & Hurd, P.D., Jr. 1968. Intrafloral ecology. An-
nual Rev. Entomol. 13: 385–414.
Barale, G., Fernandez-Marrón, T. & Alvarez-Ramis,
C. 1988. Étude de cônes males de Frenelopsis Schenck
emend. Reymanówna & Watson, 1976 (Cheirolepidiaceae)
du Crétacé Supérieur de Torrelaguna (Procince de Mad-
rid―Espagne).Geobios 21: 187–199.
Batten, D.J. & MacLennan, A.M. 1984. The paleoenviron-
mental significance of the conifer family Cheirolepidi-
aceae in the Cretaceous of Portugal. Pp. 7–12 in: Reif,
W.-E. & Westphal, F. (eds.), Third Symposium on Mesozoic
Terrestrial Ecosystems (Tübingen, Germany). Atempto
Verlag, Tübingen.
Bino, R.J., Dafni, A. & Meeuse, A.D.J. 1984a. Entomophily
in the dioecious gymnosperm Ephedra aphylla Fork. (= E.
alte C.A. Mey.), with some notes on E. campylopoda C.A.
Mey. I. Proc. Kon. Neder. Akad. Wetensch., Ser. C, Biol.
Med. Sci. 87: 1–13.
Bino, R.J., Devente, N. & Meeuse, A.D.J. 1984b. Entomophily
in the dioecious gymnosperm Ephedra aphylla Forsk. (= E.
alte C.A. Mey.) with some notes on E. campylopoda C.A.
Mey. II. Pollination droplets, nectaries, and nectarial se-
cretion in Ephedra. Proc. Kon. Neder. Akad. Wetensch.,
Ser. C., Biol. Med. Sci. 87: 15–24.
Bonnier, G. 1879. Les Nectaires: Étude Critique, Anatomique
et Physiologique. G. Masson, Paris.
Bose, M.N. 1968. A new species of Williamsonia from the
Rajmahal Hills, India. J. Linn. Soc., Bot. 61: 121–127.
Bose, M.N., Pal, P.K. & Harris, T.M. 1985. The Pentoxylon
plant. Philos. Trans., Ser. B 310: 77–108.
Brantjes, N.B.M. & Leemans, J.A.A.M. 1976. Silene otites
(Caryophyllaceae) pollinated by nocturnal Lepidoptera and
mosquitoes. Acta Bot. Neerl. 25: 281–295.
Breckon, G. & Ortíz, V.N. 1983. Pollination of Zamia pumila
by fungus-gnats. Amer. J. Bot. 70: 106–107.
Brenner, E.D., Stevenson, D.W. & Twigg, R.W. 2003. Cycads;
evolutionary innovations and the role of plant-erived neu-
rotoxins. Trends Pl. Sci. 8: 446–452.
Buchmann, S.L., O’Rourke, M.K. & Niklas, K.J. 1989. Aero-
dynamics of Ephedra trifurca. III. Selective pollen capture
by pollination droplets. Bot. Gaz. 150: 122–131.
Burdick, D.J. 1961. A taxonomic and biological study of the
genus Xyela Dalman in North America. Univ. Calif. Publ.
Entomol. 17: 281–353.
Caldas, E.B., Martins-Neto, R.G. & Lima Filho, F.P. 1989.
Afropollis sp. (polém) no trato intestinal de vespa (Hymen-
optera: Apocrita: Xyelidae), no Cretáceo da Bacia do Ar-
aripe. Atas Simp. Geol. Nordeste, pp. 195–196. Sociedade
Brasileira de Geologia, Fortaleza.
Capperino, M.E. & Schneider, E.L. 1985. Floral biology of
Nymphaea mexicana Zucc. (Nympheaceae). Aquatic Bot.
23: 83–93.
Carafa, A.M., Carratu, G. & Pizzolongo, P. 1992. Anatomical
observations on the nucellar apex of Welwitschia mirabilis
and the chemical composition of the micropylar drop. Sex.
Pl. Reprod. 9: 330–336.
Cerceau, M.T., Hideux, M., Kachkar, G., Masure, E., Ren-
ault-Miskowsky, J., Roland, F., Taugordeau-Lantz, J.
& Ybert, J.P. 1976. A propos de la structure des spores et
pollens et de leur mode de dispersion: quelques reflexions.
Trav. Lab. Micropaléontol. 6: 231–277.
Chaw, S.-M., Parkinson, C.L., Cheng, Y., Vincent, T.M. &
Palmer, J.D. 2000. Seed plant phylogeny inferred from all
three plant genomes: monophyly of extant gymnosperms
and origin of Gnetales from conifers. Proc. Natl. Acad.
Sci. U.S.A. 97: 4086–4091.
Chesnoy, L. 1993. Les secretions dans la pollinisation des
Gymnospermes. Acta Bot. Gallica 140: 145–156.
Clement-Westerhof, J.A. & van Konijnenburg-van Cittert,
J.H.A. 1991. Hirmeriella muensteri: new data on the fertile
organs leading to a revised concept of the Cheirolepidia-
ceae. Rev. Palaeobot. Palynol. 68: 147–179.
Courtinat, B. 1980. Structure d’adhérence des grains de pollen
en tétrade du genre Classopollis Pflug, 1953, de l’Het-
tangien de Saint-Fromont, Manche (France). Géobios 13:
209–229.
Crane, P.R. 1988. Major clades and relationships in “higher”
gymnosperms. Pp. 218–272, in: Beck, C.B. (ed.), Origin
and Evolution of Gymnosperms. Columbia Univ. Press,
NewYork.
Crane, P.R. 1996. The fossil history of the Gnetales. Int. J. Pl.
Sci. 157 (Suppl. 6): S50–S57.
Crane, P.R. & Lidgard, S. 1990. Angiosperm radiation and
patterns of Cretaceous palynological diversity. Pp. 377–407
in: Taylor, P.D. & Larwood, G.P. (eds.), Major Evolutionary
Radiations. Clarendon Press, Oxford.
689
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
Crane, P.R. & Upchurch, G.R. 1987. Drewria potomacensis
gen. et sp. nov., an Early Cretaceous member of Gnetales
from the Potomac Group of Virginia. Amer. J. Bot. 74:
1722–1736.
Crepet, W.L. 1972. Investigations of North American cycad-
eoids: pollination mechanisms in Cycadeoidea. Amer. J.
Bot. 59: 1048–1056.
Crepet, W.L. 1974. Investigations of North American cycad-
eoids: the reproductive biology of Cycadeoidea. Palaeon-
tographica, Abt. B, Paläophytol 148: 144–169.
Crepet, W.L. 1979. Insect pollination: a paleontological per-
spective. Bioscience 29: 102–108.
Crepet, W.L. 1996. Timing in the evolution of derived flo-
ral characters: Upper Cretaceous (Turonian) taxa with
tricolpate and tricolpate-derived pollen. Rev. Palaeobot.
Palynol. 90: 339–359.
Crepet, W.L. & Friis, E.M. 1987. The evolution of insect
pollination in angiosperms. Pp. 181–201 in: Friis, E.M.,
Chaloner W.G. & Crane, P.R. (eds.), The Origins of Angi-
osperms and their Biological Consequences. Cambridge
Univ. Press, Cambridge.
Crepet, W.L., Friis, E.M. & Nixon, K.C. 1991. Fossil evidence
for the evolution of biotic pollination. Philos. Trans., Ser.
B 333: 187–195.
Crepet, W.L. & Nixon, K.C. 1998. Fossil Clusiaceae from the
Late Cretaceous (Turonian) of New Jersey and implications
regarding the history of bee pollination. Amer. J. Bot. 85:
1122–1133.
Crowson, R.H. 1981. Biology of the Coleoptera. Academic
Press,NewYork.
Crowson, R.H. 1991. Relationships to cycads. Pp. 13–28 in:
Zunino, M., Belles, X. & Blas, M. (eds.), Advances in Co-
leopterology. European Association of Coleopterology,
Barcelona.
Delevoryas, T. 1968. Investigations of North American cycad-
eoids: structure, ontogeny and phylogenetic considerations
of cones of Cycadeoidea. Palaeontographica, Abt. B, Pal-
äophytol. 121: 122–133.
Delevoryas, T. 1971. Biotic provinces and the Jurassic-Creta-
ceous floral transition. Pp. 1660–1674in:Yochelson,E.
(ed.), Proceedings of the North American Paleontological
Convention (September, 1969, Chicago), part 1. Allen
Press, Lawrence.
Dieringer, G., Cabrera R., L., Lara, M., Loya, L. & Reyes-
Castillo, P. 1999. Beetle pollination and floral thermo-
genicity in Magnolia tamaulipana (Magnoliaceae). Int. J.
Pl. Sci. 160: 64–71.
Dilcher, D.L. 1979. Early angiosperm reproduction: an intro-
ductory report. Rev. Palaeobot. Palynol. 27: 302–309.
Dilcher, D.L., Lott, T.A., Wang, X. & Wang, Q. 2004. A
history of tree canopies. Pp. 118–137 in: Lowman, M.D.
& Rinker, H.B. (eds.), Forest Canopies, 2nd ed. Elsevier,
San Diego.
Dobson, H.E.M. & Bergström, G. 2000. The ecology and
evolution of pollen odors. Pl. Syst. Evol. 222: 63–87.
Dogra, P.D. 1964. Pollination mechanisms in gymnosperms.
Pp. 142–175 in: Nair, P.K.K. (ed.), Advances in Palynology.
National Botanic Gardens, Lucknow, India.
Doludenko, M.P. & Orlovskaya, E.R. 1976. Jurassic floras of
the Karatau Range, southern Kazakhstan. Palaeontology
19: 627–640.
Doludenko, M.P., Sakulina, G.V. & Ponomarenko, A.G.
1990. Geological Structure of the Unique Aulie Locality
of the Late Jurassic Fauna and Flora (Karatau, Southern
Kazakhstan). Geological Institute, U.S.S.R., Academy of
Sciences, Moscow.
Donaldson, D.J. 1997. Is there a floral parasite mutualism in
cycad pollination? The pollination biology of Encephalar-
tos villosus (Zamiaceae). Amer. J. Bot. 84: 1398–1406.
Donoghue, M.J. & Doyle, J.A. 2000. Seed plant phylogeny:
demise of the anthophyte hypothesis? Curr. Biol. 10:
R106–R109.
Downes, J.A. 1968. A nepticulid moth feeding at the leaf-nec-
taries of poplar. Canad. Entomol. 100: 1078–1079.
Downes, W.L. & Dahlem, G.A. 1987. Keys to the evolution
of the Diptera: role of Homoptera. Environm. Entomol.
16: 847–854.
Doyle, J. & O’Leary, M. 1935. Pollination in Tsuga, Cedrus,
Pseudotsuga, and Larix. Sci. Proc. Roy. Dublin Soc. 21:
191–204.
Doyle, J.A., Jardiné, S. & Doerenkamp, A. 1982. Afropollis,
a new genus of early angiosperm pollen, with notes on the
Cretaceous palynostratigraphy and paleoenvironments of
northern Gondwana. Bull. Cent. Rech. Explor.-Prod. Elf-
Aquitaine 6: 39–117.
Doyle, J.A. & Donoghue, M.J. 1986. Seed plant phylogeny
and the origin of angiosperms: an experimental cladistic
approach. Bot. Rev. 52: 321–431.
Eastham, L.E.S. & Eassa, Y.E.E. 1955. The feeding mech-
anism of the butterfly Pieris brassicae L. Philos. Trans.,
Ser. B 239: 1–43.
Edwards, D., Selden, P.A., Richardson, J.B. & Axe, L.
1995. Coprolites as evidence for plant-animal interaction
in Siluro-Devonian terrestrial ecosystems. Nature 377:
329–331.
Eisikowitch, D. 1988. Flower/insect interrelations—a case of
unusual predation. Evol. Theory 8: 151–154.
Ekblom, B. & Borg, A. 1996. Pollen beetle (Meligethes aeneus)
oviposition and feeding preference on different host plant
species. Entomol. Exp. Appl. 78: 291–299.
Elliott, D.K. & Nations, J.D. 1998. Bee burrows in the Late
Cretaceous (Late Cenomanian) Dakota Formation, north-
eastern Arizona. Ichnos 4: 243–253.
Endrödy-Younga, S. & Crowson, R.A. 1986. Boganiidae,
a new beetle family for the African fauna (Coleoptera:
Cucujoidea). Ann. Transvaal Mus. 34: 253–273.
Faegri, K. & Pijl, L. van der. 1980. The Principles of Pollina-
tion Ecology, 3rd ed. Pergamon Press, Oxford.
Falder, A.B., Rothwell, G.W., Mapes, G., Mapes, R.H. &
Doguzhaeva, L.R. 1998. Pityostrobus milleri sp. nov.,
a pinaceous cone from the Lower Cretaceous (Aptian)
of southwestern Russia. Rev. Palaeobot. Palynol. 103:
253–261.
Farrell, B.D. 1998. “Inordinate fondness” explained: Why are
there so many beetles? Science 281: 555–559.
Frame, D. 2003a. The pollen tube pathway in Tasmannia in-
sipida (Winteraceae): homology of the male gametophyte
conduction tissue in angiosperms. Pl. Biol. 5: 290–296.
Frame, D. 2003b. Generalist flowers, biodiversity and flor-
ivory: implications for angiosperm origins. Taxon 52:
681–685.
Gandolfo, M.A., Nixon, K.C. & Crepet, W.L. 2004. Cre-
taceous flowers of Nymphaeaceae and implications for
complex insect entrapment pollination mechanisms in
690
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
early angiosperms. Proc. Natl. Acad. Sci. U.S.A. 101:
8056–8060.
Gao, Z. & Thomas, B.A. 1989. A review of cycad megaspo-
rophylls, with new evidence of Crossosamia pomel and
its associated leaves from the lower Permian of Taiyuan,
China. Rev. Palaeobot. Palynol. 60: 205–223.
Gazit, S., Galon, I. & Podoler, H. 1982. The role of nitidulid
beetles in natural pollination of Annona in Israel. J. Amer.
Soc. Hort. Sci. 107: 849–852.
Gelbart, G. & Aderkas, P. von. 2002. Ovular secretions as
part of pollination mechanisms in conifers. Ann. Forest
Sci. 59: 345–357.
Gifford, E.M. & Foster, A.S. 1989. Morphology and Evolution
of Vascular Plants,3rded.W.H.Freeman,NewYork.
Goldblatt, P. & Manning, J.C. 2000. The long-proboscid fly
pollination system in southern Africa. Ann. Missouri Bot.
Gard. 87: 146–170.
Gorelick, R. 2001. Did insect pollination increased seed plant
diversity? Biol. J. Linn. Soc. 74: 407–427.
Gottsberger, G. 1988. The reproductive biology of primitive
angiosperms. Taxon 37: 630–643.
Gottsberger, G., Schrausen, J. & Linskens, H.F. 1984. Amino
acids and sugars in nectar, and their putative evolutionary
significance. Pl. Syst. Evol. 145: 55–77.
Gradstein, F., Ogg, J. & Smith, A. 2004. A Geologic Time
Scale 2004. Cambridge Univ. Press, Cambridge.
Graham-Smith, G.S. 1930. Further observations on the
anatomy and function of the oral sucker of the blow-fly
(Calliphora erythrocephala). Parasitology 22: 47–115.
Grant, V. & Grant, K.A. 1965. Flower Pollination in the Phlox
Family.ColumbiaUniv.Press,NewYork.
Gratschev, V.G. & Zherikhin, V.V. 2003. The fossil record of
weevils and related beetle families (Coleoptera, Curculion-
oidea). Acta Zool. Cracovien. 46 (Suppl. “Fossil Insects”):
129–138.
Grauvogel-Stamm, L. & Kelber, K.-P. 1996. Plant-insect in-
teractions and coevolution during the Triassic in western
Europe. Paleontol. Lombarda 5: 5–23.
Grimaldi, D.A. 1990. Insects from the Santana Formation,
Lower Cretaceous, of Brazil. Bull. Amer. Mus. Nat. Hist.
195: 1–191.
Grimaldi, D.A. 1999. The co-radiations of pollinating insects
and angiosperms in the Cretaceous. Ann. Missouri Bot.
Gard. 86: 373–406.
Grimaldi, D.A. & Cumming, J. 1999. Brachyceran Diptera
in Cretaceous ambers and Mesozoic diversification of the
Eremoneura. Bull. Am. Mus. Nat. Hist. 195: 1–191.
Grimaldi, D.A. & Engel, M.S. 2005. Evolution of the Insects.
CambridgeUniv.Press,NewYork.
Grimaldi, D.A., Zhang, J.-F., Fraser, N.C. & Rasnitsyn, A.P.
2005. New and revised scorpion flies of the extinct family
Pseudopolycentropodidae (Mecopteroidea), including two
mosquito-like species in Cretaceous amber. Insect Syst.
Evol. 36: 443–458.
Grinfel’d, E.K. 1962. The Origin of Anthophily in Insects. St.
Petersburg Univ., St. Petersburg. [In Russian]
Grinfel’d, E.K. 1975. Anthophily in beetles (Coleoptera) and
a critical evaluation of the cantharophilous hypothesis.
Entomol. Rev. 54: 18–22.
Guo, S.-X. & Wu, X.-W. 2000. Ephedrites from the latest
JurassicYixianFormationinwesternLiaoning,northeast
China. Acta Paleontol. Sin. 39: 81–91.
Harris, T.M. 1969. The Yorkshire Jurassic Flora. III. Bennet-
titales. British Museum (Natural History), London.
Harris, T.M. 1973. The strange Bennettitales. 19th Sir Albert
Charles Seward Mem. Lect. (1970), Birbal Sahni Inst.,
Lucknow, India.
Haslett, J.R. 1989. Adult feeding by holometabolous insects:
pollen and nectar as complementary nutrient sources for
Rhingia campestris (Diptera Syrphidae). Oecologia 81:
361–363.
Haycraft, C.J. & Carmichael, J.S. 2001. Development of
sterile ovules on bisexual cones of Gnetum gnemon (Gne-
taceae). Amer. J. Bot. 88: 1326–1330.
Heddergott, H. 1938. Kopf und Vorderdarm von Panorpa
communis L. Zool. Jahrb. (Anat.) 64: 229–294.
Hepburn, H.R. 1969. The skeleton-muscular system of Mecop-
tera: The head. Univ. Kansas Sci. Bull. 48:721–765.
Hesse, M. 1984. Pollenkitt is lacking in Gnetatae—Ephedra
and Welwitschia—further proof of its restriction to the
angiosperms. Pl. Syst. Evol. 144: 9–16.
Hilton, J. & Bateman, R.M. 2006. Pteridosperms are the
backbone of seed-plant phylogeny. J. Torrey Bot. Soc.
133: 119–168.
Hluštík, A. & Konzalová, M. 1976a. Polliniferous cones of Fre-
nelopsis alata (K. Feistm.) Knobloch from the Cenomanian
of Czechoslovakia. Vĕst. Ústřed. Ust. Geol. 51: 37–44.
Hluštik, A. & Konzalová, M. 1976b. Frenelopsis alata (K.
Feistm.) Knobloch (Cupressaceae) from the Cenomanian
of Bohemia, a new plant producing Classopollis pollen.
Pp. 125–131 in: Novák, V.J.A. & Pacltová, B. (eds.), Pro-
ceedings of the International Conference on Evolutionary
Biology, (1975, Liblice, Czechoslovakia). Czechosolovak
Biological Society, Prague.
Holloway, B.A. 1976. Pollen-feeding in hover-flies (Diptera:
Syrphidae). New Zealand J. Zool. 3: 339–350.
Hoyt, C.P. 1952. The evolution of the mouth parts of adult
Diptera. Microentomology 17: 61–125.
Jarzembowski, E.A. 1990. A boring beetle from the Wealden
of the Weald. Pp. 373–376 in: Boucot, A.J. (ed.), Evolution-
ary Paleobiology of Behavior and Coevolution. Elsevier,
Amsterdam.
Jervis, M.A., Kidd, N.A.C., Fitton, M.G., Huddleston, T.
& Dawah, H.A. 1993. Flower-visiting by hymenopteran
parasitoids. J. Nat. Hist. 27: 67–105.
Johnson, C.D. 1970. Biosystematics of the Arizona, California,
and Oregon species of the seed beetle genus Acanthoscel-
ides Schilsky (Coleoptera: Bruchidae). Univ. Calif. Publ.
Entomol. 59: 1–116.
Kato, M. & Inoue, T. 1994. Origin of insect pollination. Nature
368: 195.
Kato, M., Inoue,
T. & Nagamitsu, T. 1995. Pollination biology
of Gnetum (Gnetaceae) in a lowland mixed dipterocarp
forest in Sarawak. Amer. J. Bot. 82: 862–868.
Kawakita, A. & Kato, M. 2004. Obligate pollination mutual-
ism in Breynia (Phyllanthaceae): further documentation
of pollination mutualism involving Epicephala moths
(Gracillariidae). Amer. J. Bot. 91: 1319–1325.
Kirk, W.D.J. 1984. Pollen-feeding in thrips (Insecta: Thysan-
optera). J. Zool. 204: 107–117.
Klavins, S.D., Kellogg, D.W., Krings, M., Taylor, E.L. &
Taylor, T.N. 2005. Coprolites in a Middle Triassic cycad
pollen cone: evidence for insect pollination in early cy-
cads? Evol. Ecol. Res. 7: 479–488.
691
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
Klavins, S.D., Taylor, E.L., Krings, M. & Taylor, T.N. 2003.
Gymnosperms from the Middle Triassic of Antarctica: the
first structurally preserved cycad pollen cone. Int. J. Pl.
Sci. 164: 1007–1020.
Kneipert, F.-W. 1980. Blood-feeding and nectar-feeding in
adult Tabanidae (Diptera). Oecologia 46: 125–129.
Koptur, S., Smith, A.R. & Baker, I. 1982. Nectaries in some
Neotropical species of Polypodium (Polypodiaceae): pre-
liminary observations and analyses. Biotropica 14: 108–
113.
Krassilov, V.A. 1997. Angiosperm Origins. Pensoft Publishers,
Sofia.
Krassilov, V.A. & Rasnitsyn, A.P. 1983. A unique find: pollen
in the intestine of Early Cretaceous sawflies. Paleontol.
J. 4: 80–95.
Krassilov, V.A. & Rasnitsyn, A.P. 1999. Plant remains from
the guts of fossil insects: Evolutionary and paleoecological
inferences.Pp.65–72in:Vršanský,P.(ed.),Proceedings
of the First Palaeoentomological Conference. Moscow,
1998. AMBA Projects; Bratislava.
Krassilov, V.A., Rasnitsyn, A.P. & Afonin, S.A. 2007. Pollen
eaters and pollen morphology: co-evolution through the
Permian and Mesozoic. African Invertebr. 48: 3–11.
Krassilov, V.A., Tekleva, M., Meyer-Melikyan, N. & Ras-
nitsyn, A.P. 2003. New pollen morphotype from gut
compression of a Cretaceous insect, and its bearing on
palynomorphological evolution and palaeoecology. Cre-
taceous Res. 24: 149–156.
Krassilov, V.A., Zherikhin, V.V. & Rasnitsyn, A.P. 1997.
Classopollis in the guts of Jurassic insects. Palaeontology
40: 1095–1101.
Krings, M., Taylor, T.N. & Kellogg, D.W. 2002. Touch-
sensitive glandular trichomes: a mode of defence against
herbivorous arthropods in the Carboniferous. Evol. Ecol.
Res. 4: 779–786.
Kristensen, N.P. 1999. The non-glossatan moths. Pp. 41–49 in:
Kristensen, N.P. (ed.), Handbuch der Zoologie. Band IV
(Arthropoda: Insecta), Teilband 35 (Lepidoptera, Moths
and Butterflies), Volume 1 (Evolution, Systematics, and
Biogeography). De Gruyter, Berlin.
Kristensen, N.P. & Nielsen, E.S. 1981. Intrinsic proboscis
musculature in non-ditrysian Lepidoptera–Glossata:
structure and phylogenetic significance. Entomol. Scand.
(Suppl.) 15: 299–304.
Kvaček, J. 2000. Frenelopsis alata and its microsporangiate
and ovuliferous reproductive structures from the Cenoma-
nian of Bohemia (Czech Republic, Central Europe). Rev.
Palaeobot. Palynol. 112: 51–78.
Labandeira, C.C. 1994. A compendium of fossil insect families.
Milwaukee Public Mus. Contrib. Biol. Geol. 88: 1–71.
Labandeira, C.C. 1997. Insect mouthparts: Ascertaining the
paleobiology of insect feeding strategies. Annual Rev. Ecol.
Syst. 28: 153–193.
Labandeira, C.C. 1998a. How old is the flower and the fly?
Science 280: 57–59.
Labandeira, C.C. 1998b. The role of insects in Late Jurassic
to Middle Cretaceous ecosystems. New Mexico Mus. Nat.
Hist. Sci. Bull. 14: 105–124
Labandeira, C.C. 2000. The paleobiology of pollination and
its precursors. Paleontol. Soc. Pap. 6: 233–269.
Labandeira, C.C. 2002a. The history of associations between
plants and animals. Pp. 26–74, 248–261 in: Herrera, C.M.
& Pellmyr, O. (eds.), Plant-Animal Interactions: An Evo-
lutionary Approach. Blackwell, London.
Labandeira, C.C. 2002b. Paleobiology of predators, parasi-
toids, and parasites: accommodation and death in the fossil
record of terrestrial invertebrates. Paleontol. Soc. Special
Pap. 8: 211–250.
Labandeira, C.C. 2005. Fossil history and evolutionary
ecology of Diptera and their associations with plants. Pp.
217–272in:Yeates,D.K.&Wiegmann,B.M.(eds.),The
Evolutionary Biology of Flies. Columbia Univ. Press, New
York.
Labandeira, C.C. 2006. The four phases of plant-arthropod
associations in deep time. Geol. Acta 4: 409–438.
Labandeira, C.C., Dilcher, D.L., Davis, D.R. & Wagner, D.L.
1994. Ninety-seven million years of angiosperm-insect asso-
ciation: paleobiological insights into the meaning of coevo-
lution. Proc. Natl. Acad. Sci. U.S.A. 91: 12278–12282.
Labandeira, C.C., Johnson, K.R. & Lang, P. 2002. Prelim-
inary assessment of insect herbivory across the Creta-
ceous-Tertiary boundary: Major extinction and minimum
rebound. Geol. Soc. Amer. Special Pap. 361: 297–327.
Leereveld, H. 1982. Anthecological relations between reputedly
anemophilous flowers and syrphid flies. III. Worldwide
survey of crop and intestine contents of certain anthophi-
lous syrphid flies. Tijdschr. Entomol. 125: 25–35.
Lloyd, D.G. & Wells, M.S. 1992. Reproductive biology of
a primitive angiosperm, Pseudowintera colorata (Win-
teraceae), and the evolution of pollination systems in the
Anthophyta. Pl. Syst. Evol. 181: 77–95.
Lopez-Vaamonde, C., Wikström, N., Labandeira, C.C.,
Godfray, C.H.J., Goodman, S.J. & Cook, J.M. 2006.
Fossil-calibrated molecular phylogenies reveal that leaf-
mining moths radiated several million years after their
host plants. J. Evol. Biol. 19: 1314–1326.
Machado, C.A., Jousselin, E., Kjellberg, F., Compton, S.C.
& Herre, E.A. 2001. Phylogenetic relationships, historical
biogeography and character evolution of fig-pollinating
wasps. Proc. Roy. Soc. London, Ser. B, Biol. Sci. 268:
685–694.
Malyshev, S.I. 1968. Genesis of the Hymenoptera and the
Phases of their Evolution. Meuthen, London.
Mamay, S.H. 1976. Paleozoic origin of the cycads. U.S. Geol.
Survey Prof. Pap. 934: 1–48.
Manning, J.C. & Goldblatt, P. 1996. The Prosoeca peringueyi
(Diptera: Nemestrinidae) pollination guild in southern
Africa: long-tongued flies and their tubular flowers. Ann.
Missouri Bot. Gard. 83: 67–86.
Marsh, B. 1982. An ecological study of Welwitschia mirabilis
and its satellite fauna. Transvaal Mus. Bull. 4 (Suppl.):
3–4.
Mazzarolo, L.A. & Amorim, D.S. 2000. Cratomyia macror-
rhyncha, a Lower Cretaceous brachyceran fossil from the
Santana Formation, Brazil, representing a new species,
genus and family of the Stratiomyomorpha (Diptera).
Insect Syst. Evol. 31: 91–102.
McKeever, S. & French, F.E. 1999. Comparative study of adult
mouthparts of fifty-two species representing thirty genera
of Tabanidae (Diptera). Mem. Entomol. Int. 14: 327–353.
McNaughton, I.H. & Harper, J.L. 1960. The comparative
biology of closely related species living in the same area.
I. External breeding-barriers between Papaver species.
New Phytol. 59: 15–26.
692
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
Meeuse, A.D.J. 1990. Flowers and Fossils. Eburon Delft, The
Hague, Netherlands.
Meeuse, B.J.D. & Schneider, E.L. 1980. Nymphaea revisited:
a preliminary communication. Israel J. Bot. 28: 65–79.
Midgely, J.J. & Bond, W.J. 1991. How important is biotic pol-
lination and dispersal to the success of the angiosperms?
Philos. Trans., Ser. B 333: 209–215.
Mitter, J.L. 1918. Note on the method of feeding of Corizoneura
(Pangonia) longirostris Hardwick, with a description of the
mouth parts. Indian J. Med. Res. 5: 523–528.
Mostovski, M.B. 1998. A revision of the nemestrinid flies
(Diptera, Nemestrinidae) described by Rohdendorf, and
a description of new taxa of the Nemestrinidae from the
Upper Jurassic of Kazakhstan. Paleontol. J. 32: 369–
375.
Mostovski, M.B. & Martínez-Delclòs, X. 2000. New Neme-
strinoidea (Diptera: Brachycera) from the Upper Jurassic-
Lower Cretaceous of Eurasia, taxonomy and palaeobiology.
Entomol. Probl. 31: 137–148.
Mound, L.A. & Terry, I. 2001. Thrips pollination of the central
Australian cycad, Macrozamia macdonnellii (Cycadales).
Int. J. Pl. Sci. 162: 147–154.
Moussel, B. 1980. Gouttelette réceptrice du pollen et pollinisat-
ion chez l’Ephedra distachya L. Observations sur le vivant
et en microscopies photonique et életronique. Rev. Cytol.
Biol. Veg. Botaniste 3: 65–89.
Nagatomi, A. & Soroida, K. 1985. The structure of the mouth-
parts of the orthorrhaphous Brachycera (Diptera) with a
special reference to blood-sucking. Beitr. Entomol. 35:
263–368.
Nagatomi, A. & Yang, D. 1998. A review of extinct Mesozoic
genera and families of Brachycera (Insecta, Diptera, Or-
thorrhapha). Entomol. Monthly Mag. 134: 95–192.
Niklas, K.J. 1997. The Evolutionary Biology of Plants. Univ.
Chicago Press, Chicago.
Nishida, H. & Hayashi, N. 1996. Cretaceous coleopteran larva
fed on a female fructification of extinct gymnosperm. J.
Pl. Res. 109: 327–330.
Norstog, K.J. 1987. Cycads and the origin of insect pollination.
Amer. Sci. 75: 270–279.
Norstog, K.J. & Nicholls, T.J. 1997. The Biology of the Cy-
cads.CornellUniv.Press,Ithaca,NewYork.
Norstog, K.J., Fawcett, P.K.S. & Vovides, A.P. 1992. Beetle
pollination of two species of Zamia: Evolutionary and
ecological considerations. Palaeobotanist 41: 149–158.
Novokshonov, V.G. 1997. Some Mesozoic scorpionflies
(Insecta: Panorpida = Mecoptera) of the families Mes-
opsychidae, Pseudopolycentropodidae, Bittacidae, and
Permochoristidae. Paleontol. J. 31: 65–71.
Nygaard, P. 1977. Utilization of exogenous carbohydrates for
tube growth and starch synthesis in pine pollen suspension
cultures. Physiol. Pl. 39: 206–210.
Oberprieler, R.G. 1995. The weevils (Coleoptera: Curculionoi-
dea) associated with cycads. I. Classification, relationships,
and biology. Pp. 295–324 in: Vorster, P. (ed.), Proceedings
of the Third International Conference on Cycad Biology.
The Cycad Society of South Africa, Stellenbosch.
Oberprieler, R.G. 2004. “Evil weevils”—the key to cycad
survival and diversification? Pp. 170–194 in: Lindstrom,
J.A. (ed.), Proceedings of the Sixth Internatinal Conference
on Cycad Biology. Nong Nooch Tropical Botanical Garden,
Chonburi, Thailand.
Ordetx, G.S.R. 1952. Flora Apicola de America Tropical.
Editorial Lex, Havana.
Ornduff, R. 1991. Size classes, reproductive behavior, and in-
sect associates of Cycas media (Cycadaceae) in Australia.
Bot. Gaz. 152: 203–207.
Owens, J.N., Takaso, T. & Runions, C.J. 1998. Pollination in
conifers. Trends Pl. Sci. 3: 479–485.
Palmer, C.M. & Yeates, D.K. 2004. Diet and feeding behavior
in adults of the Apteropanorpidae (Mecoptera). J. Insect
Behav. 18: 209–231.
Pant, D.D. 1987. The fossil history of the Cycadales. Geophy-
tology 17: 125–162.
Pearson, H.H.W. 1906. Notes on South African cycads. I.
Trans. S. African Philos. Soc. 16: 341–354.
Pellmyr, O. 1992. Evolution of insect pollination and angio-
sperm diversification. Trends Ecol. Evol. 7: 46–49.
Pellmyr, O. & Leebens-Mack, J. 1999. Forty million years
of mutualism: evidence for Eocene origin of yucca-yucca
moth association. Proc. Natl. Acad. Sci. U.S.A. 96: 9178–
9183.
Pellmyr, O. & Thien, L.B. 1986. Insect reproduction and floral
fragrances: key to the evolution of the angiosperms. Taxon
35: 76–85.
Pocock, S.A.J., Vasanthy, G. & Venkatachala, B.S. 1990.
Pollen of Circumpolles—an enigma of monotetrads show-
ing evolutionary adaptation. Rev. Palaeobot. Palynol. 65:
179–193.
Poinar, G.O., Jr. & Danforth, B.N. 2006. A fossil bee from
Early Cretaceous Burmese amber. Science 314: 614.
Pons, D. 1979. Les organs reproducteurs de Frenelopsis alata
(K. Feistm) Knobloch, Cheirolepidiaceae du Cénomanien
de l’Anjou, France. Compt. Rend. Congr. Natl. Soc. Savan-
tes, Sec. Sci. 103: 209–231.
Poort, R.J., Visscher, H. & Dilcher, D.L. 1996. Zoidogamy
in fossil gymnosperms: the centenary of a concept, with
special reference to prepollen of late Paleozoic conifers.
Proc. Natl. Acad. Sci. U.S.A. 93: 11713–11717.
Porsch, O. 1916. Die Nektartropfen von Ephedra campylopoda.
Ber. Deutsch. Bot. Ges. 34: 202–212.
Porsch, O. 1958. Alte Insektentypen als Blumenausbeuter.
Oesterr. Bot. Z. 104: 115–163.
Powell, J.A., Mitter, C. & Farrell, B. 1999. Evolution of
larval food preferences in Lepidoptera. Pp. 403–422 in:
Kristensen, N.P. (ed.), Handbuch der Zoologie. Band IV
(Arthropoda: Insecta), Teilband 35 (Lepidoptera, Moths
and Butterflies), Vol. 1 (Evolution, Systematics, and Bio-
geography). De Gruyter, Berlin.
Power, M.S. & Skog, J.E. 1987. Ultrastructure of the extra-
floral nectaries of Pteridium aquilinum. Amer. Fern J. 17:
1–15.
Price, R.A. 1996. Systematics of the Gnetales: a review of
morphological and molecular evidence. Int. J. Pl. Sci. 157
(Suppl.): S40–S49.
Proctor, M., Yeo, P. & Lack, A. 1996. The Natural History of
Pollination. Timber Press, Portland, Oregon.
Rasnitsyn, A.P. 1964. New Triassic Hymenoptera from Central
Asia. Paleontol. Zhurn. 1: 88–96. [In Russian]
Rasnitsyn, A.P. 1977. New Paleozoic and Mesozoic insects.
Paleontol. Zhurn. 1: 60–72.
Rasnitsyn, A.P. 1980. Origin and evolution of the Hymenoptera
(Insecta). Trans. Paleontol. Inst. Acad. Sci. U.S.S.R. 174:
1–191. [In Russian]
693
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
Rasnitsyn, A.P. & Kozlov, M.V. 1991. A new group of fossil in-
sects: scorpions [sic] with cicad and butterfly adaptations.
Trans. U.S.S.R. Acad. Sci., Earth Sci. 310: 233–236.
Rasnitsyn, A.P. & Krassilov, V.A. 1996. Pollen in the gut
contents of fossil insects as evidence of coevolution. Pa-
leontol. J. 30: 716–722.
Rasnitsyn, A.P. & Krassilov, V.A. 2000. First documented
occurrence of phyllophagy in pre-Cretaceous insects: leaf
tissues in the gut of Upper Jurassic insects from southern
Kazakhstan. Paleontol. J. 34: 301–309.
Rasnitsyn, A.P. & Quicke, D.L.J. (eds.) 2002. History of
Insects. Kluwer Academic Publ., Dordrecht.
Rayner, R.J. & Waters, S.B. 1991. Floral sex and the fossil
insect. Naturwissenschaften 78: 280–282.
Ren, D. 1998. Flower-associated Brachycera flies as fossil
evidence for Jurassic angiosperm origins. Science 280:
85–88.
Ren, D., Guo, Z.-G., Lu, L.-W., Ji, S.-A. & Han, Y.-G. 1995.
Faunae and Stratigraphy of the Jurassic-Cretaceous in
Beijing and the Adjacent Areas. Seismic Publ. House,
Beijing.
Rentz, D.C.F. & Clyne, D. 1983. A new genus and species of
pollen- and nectar-feeding katydids from eastern Australia
(Orthoptera: Tettigoniidae: Zaprochilinae). J. Austral. En-
tomol. Soc. 22: 155–160.
Retallack, G.J. & Dilcher, D.L. 1988. Reconstructions of
selected seed ferns. Ann. Missouri Bot. Gard. 75: 1010–
1057.
Reymanówna, M. 1960. A cycadeoidean stem from the western
Carpathians. Acta Palaeobot. 1: 3–28.
Reymanówna, M. 1991. Are developing ovules and seeds of
Mesozoic gymnosperms protected against the environ-
ment? Contr. Paleontol. Mus. Univ. Oslo 364: 53–54.
Rohdendorf, B.B. 1968. New Mesozoic nemestrinids. Pp.
180–189 in: Rohdendorf, B.B. (ed.), Jurassic Insects of
Karatau. Nauka, Moscow.
Rothwell, G.W. 1972. Evidence of pollen tubes in Paleozoic
pteridosperms. Science 175: 772–774.
Rothwell, G.W. 1977. Evidence for a pollination-drop mech-
anism in Paleozoic pteridosperms. Science 198: 1251–
1252.
Rothwell, G.W., Grauvogel-Stamm, L. & Mapes, G. 2000.
An herbaceous fossil conifer: gymnospermous ruderals
in the evolution of Mesozoic vegetation. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 156: 139–145.
Rothwell, G.W. & Serbet, R. 1992 Pollination biology of
Elkinsia polymorpha, implications for the origin of
gymnosperms. Pp. 225–231 in: Schaarschmidt, F. (ed.),
Anatomical Investigations of Fossil Plants. Courier For-
schungsinstitut, Senckenberg.
Rydin, C., Källersjö, M. & Friis, E.M. 2002. Seed plant re-
lationships and the systematic position of Gnetales based
on nuclear and chloroplast DNA: conflicting data, rooting
problems, and the monophyly of conifers. Int. J. Pl. Sci.
163: 197–214.
Rydin, C., Mohr, B. & Friis, E.M. 2003. Cratonia cotyledon
gen. et sp. nov.: a unique Cretaceous seedling related to
Welwitschia. Proc. Roy. Soc. London, Ser. B., Biol. Sci.
270 (Suppl.): S1–S4.
Saiki, K. & Yoshida, Y. 1999. A new bennettitalean trunk with
unilacunar five-track nodal structure from the Upper Cre-
taceous of Hokkaido, Japan. Amer. J. Bot. 86: 326–332.
Santiago-Blay, J.A. 2004. Some aspects of the biology of the
Aulacoscelinae Chapuis 1874 (Orsodacnidae or Chrys-
omelidae), sensu lato, with a description of seven new
species of Janbechynea. Pp. 1–66 (CD-ROM) in: Jolivet,
P., Santiago-Blay, J.A. & Schmitt, M. (eds.), New Devel-
opments in the Biology of Chrysomelidae. SPB Academic
Publishers, The Hague.
Scheuring, B.W. 1976. Proximal exine filaments, a widespread
feature among Triassic Protosaccites and Circumpolles
to secure the dispersal of entire tetrads. Pollen & Spores
18: 611–639.
Schneider, D., Wink, M., Sporer, F. & Lounibos, P. 2002.
Cycads: their evolution, toxins, herbivores and insect pol-
linators. Naturwissenschaften 89: 281–295.
Schuster, J.C. 1974. Saltatorial Orthoptera as common visitors
to tropical flowers. Biotropica 6: 138–140.
Scott, A.C., Anderson, J.M. & Anderson, H.M. 2004. Ev-
idence of plant-insect interactions in the Upper Triassic
Molteno Formation of South Africa. J. Geol. Soc. London
161: 401–410.
Serdi-Benkaddour, R. & Chesnoy, L. 1985. Secretion and
composition of the pollination drop in the Cephalotaxus
drupacea (gymnosperm, Cephalotaxaceae). Pp. 345–350
in: Cristi, M., Gori, P. & Pacini, E. (eds.), Sexual Reproduc-
tion in Higher Plants. Springer-Verlag, Amsterdam.
Sharov, A.G. 1971. Phylogeny of the Orthopteroidea. Israel
Program for Scientific Translations, Jerusalem.
Srivastava, S.K. 1976. The fossil pollen genus Classopollis.
Lethaia 9: 437–457.
Stelleman, P. 1981. Anthecological relations between reputedly
anemophilous flowers and syrphid flies. V. Some special
aspects of the visiting of Plantago media and P. lanceolata
by insects. Beitr. Biol. Pflanzen 55: 157–167.
Stevenson, D.W., Norstog, K.J. & Fawcett, P.K.S. 1998. Pol-
lination biology of cycads. Pp. 277–294 in: Owens, S.J. &
Rudall, P.J. (eds.), Reproductive Biology in Systematics,
Conservation and Economic Botany. Royal Botanic Gar-
dens, Kew.
Stockey, R.A. & Rothwell, G.W. 2003. Anatomically preserved
Williamsonia (Williamsoniaceae): evidence for bennet-
titalean reproduction in the Late Cretaceous of western
North America. Int. J. Pl. Sci. 164: 251–262.
Sun, G., Zheng, S.-L., Dilcher, D.L., Wang, Y.-D. & Mei, S.-
W. 2001. Early Angiosperms and their Associated Plants
from Western Liaoning, China. Shanghai Scientific and
Technological Education Publishing House, Shanghai.
Szucsich, N.U. & Krenn, H.W. 2000. Morphology and func-
tion of the proboscis in Bombyliidae (Diptera, Brachycera)
and implications for proboscis evolution in Brachycera.
Zoomorphology 120: 79–90.
Takahashi, M., Takai, K. & Saiki, K. 1995. Ephedroid fossil
pollen from the Lower Cretaceous (Upper Albian) of Hok-
kaido, Japan. J. Pl. Res. 108: 11–15.
Takaso, T. & Owens, J.N. 1996. Ovulate cone, pollination
drop, and pollen capture in Sequoiadendron (Taxodiaceae).
Amer. J. Bot. 83: 1175–1180.
Tang, W. 1987. Insect pollination in the cycad Zamia pumila
(Zamiaceae). Amer. J. Bot. 74: 90–99.
Tang, W. 1993. Nectar-like secretions in female cones of cy-
cads. Cycad Newslett. (special issue) 16: 10–13.
Tang, W. 1995. Pollination drops in female cycad cones. Palms
Cycads 48: 20–22.
694
TAXON 56 (3) • August 2007: 663–695Labandeira & al. • Pollination and Mesozoic gymnosperms
Taylor, T.N. 1973. A consideration of the morphology, ultra-
structure and multicellular microgametophyte of Cycad-
eoidea dacotensis pollen. Rev. Palaeobot. Palynol. 16:
157–164.
Taylor, T.N. 1978. The ultrastructure and reproductive signif-
icance of Monoletes (Pteridospermales) pollen. Canad. J.
Bot. 56: 3105–3118.
Taylor, T.N. & Alvin, K.L. 1984. Ultrastructure and devel-
opment of Mesozoic pollen: Classopollis. Amer. J. Bot.
71: 547–587.
Taylor, T.N. & Millay, M.A. 1979. Pollination biology and
reproduction in early seed plants. Rev. Palaeobot. Palynol.
27: 329–355.
Taylor, T.N. & Taylor, E.L. 1993. The Biology and Evolution
of Fossil Plants. Prentice Hall, Englewood Cliffs, New
Jersey.
Terry, I. 2001. Thrips and weevils as dual, specialist pollinators
of the Australian cycad Macrozamia communis (Zamia-
ceae). Int. J. Pl. Sci. 162: 1293–1305.
Terry, L.I., Walter, G.H., Donaldson, J.S., Snow, E., For-
ster, P.I. & Machin, P.J. 2005. Pollination of Australian
Macrozamia cycads (Zamiaceae): effectiveness and behav-
ior of specialist vectors in a dependent mutualism. Amer.
J. Bot. 92: 931–940.
Thien, L.B., Azuma, H. & Kawano, S. 2000. New perspec-
tives on the pollination biology of basal angiosperms. Int.
J. Pl. Sci. 161 (Suppl.): S225–S235.
Thien, L.B., Sage, T.L., Jaffré, T., Bernhardt, P., Pontieri,
V., Weston, P.H., Malloch, D., Azuma, H., Graham,
S.W., McPherson, M.A., Rai, H.S., Sage, R.F. & Dupre,
J.-L. 2003. The population structure and floral biology
of Amborella trichopoda (Amborellaceae). Ann. Missouri
Bot. Gard. 90: 466–490.
Tidwell, W.D. & Ash, S.R. 1990. On the Upper Jurassic stem
Hermanophyton and its species from Colorado and Utah,
USA. Palaeontographica, Abt. B, Paläophytol 218: 77–
92.
Tomlinson, P.B., Braggins, J.E. & Rattenbury, J.A. 1991.
Pollination drop in relation to cone morphology in Podo-
carpaceae: a novel reproductive mechanism. Amer. J. Bot.
78: 1289–1303.
Tomlinson, P.B., Braggins, J.E. & Rattenbury, J.A. 1997.
Contrasted pollen capture mechanisms in Phyllocladaceae
and certain Podocarpaceae (Coniferales). Amer. J. Bot.
84: 214–223.
Uličný, D., Kvaček, J., Svobodová, M. & Špičáková, L. 1997.
High-frequency sea-level fluctuations and plant habitats
in Cenomanian fluvial to estuarine succession: Pecínov
quarry, Bohemia. Palaeogeogr. Palaeoclimatol. Palaeo-
ecol. 136: 165–197.
Upchurch, G.R. & Doyle, J.A. 1981. Paleoecology of the co-
nifers Frenelopsis and Pseudofrenelopsis (Cheirolepidia-
ceae) from the Cretaceous Potomac Group of Maryland and
Virginia. Pp. 167–202 in: Romans, R.C. (ed.), Geobotany,
vol.2.PlenumPress,NewYork.
Vakhrameev, V.A. 1991. Jurassic and Cretaceous Floras
and Climates of the Earth. Cambridge Univ. Press, New
York.
Van der Pijl, L. 1953. On the flower biology of some plants
from Java, with general remarks on fly-traps (species of
Annona, Artocarpus, Typhonium, Gnetum, Arisaema and
Abroma). Ann. Bogor. 1: 77–99.
Watanabe, M. & Kamimura, K. 1975. Nectar sucking be-
havior of Tabanus iyoensis (Diptera: Tabanidae). Jap. J.
Sanit. Zool. 26: 41–47.
Watson, J. 1977. Some Lower Cretaceous conifers of the Chei-
rolepidiaceae from the U.S.A. and England. Palaeontology
20: 715–749.
Watson, J. 1988. The Cheirolepidiaceae. Pp. 382–447 in: Beck,
C.B. (ed.), Origin and Evolution of Gymnosperms. Colum-
biaUniv.Press,NewYork.
Watson, J. & Sincock, C.A. 1992. Bennettitales of the Eng-
lish Wealden. Monogr. Palaeontogr. Soc. 145: 1–228, pls.
1–23.
Wetschnig, W. & Depisch, B. 1999. Pollination biology of Wel-
witschia mirabilis Hook. (Welwitschiaceae, Gnetopsida).
Phyton 39: 167–183.
Whitehead, D.R. 1969. Wind pollination in the angiosperms:
evolutionary and environmental considerations. Evolution
23: 28–35.
Wilson, B.H. & Lieux, M. 1972. Pollen grains in the guts of
field collected tabanids in Louisiana. Ann. Entomol. Soc.
Amer. 65: 1264–1266.
Wing, S.L. 2000. Evolution and expansion of flowering plants.
Paleontol. Soc. Pap. 6: 209–231.
Wodehouse, R.P. 1935. Pollen Grains. McGraw-Hill, New
York.
Zaitzev, V.F. 1998. Emergence and evolution of anthophily in
Diptera. Pp. 154–155 in: Ismay, J.S. (ed.), Abstracts Vol-
ume, Fourth International Congress of Dipterology. The
Congress, Oxford. [Abstract]
Zavada, M.S. & Taylor, T.N. 1986. The role of self-incom-
patibility and sexual selection in the gymnosperm-angi-
osperm transition: a hypothesis. Amer. Naturalist 128:
538–550.
Zhang, J.-F. 2005. The first find of chrysomelids (Insecta:
Coleoptera: Chrysomeloidea) from Callovian-Oxfordian
Daohugou biota of China. Geobios 38: 865–871.
Zherikhin, V.V. & Gratschev, V.G. 1997. The Early Cretaceous
weevils from Sierra del Montsec, Spain (Insecta: Coleop-
tera: Curculionoidea). Cretaceous Res. 18: 625–632.
Zherikhin, V.V., Mostovski, M.B., Vršanský, P., Blagoderov,
V.A. & Lukashevich, E.D. 1999. The unique Lower Cre-
taceous locality Baissa and other contemporaneous fossil
insect sites in North and West Transbaikalia. Pp. 185–191
in: Vrsansky, P. (ed.), Proceedings of the First Palaeoen-
tomological Conference (Moscow, 1998). AMBA Projects,
Bratislava.
Ziegler, H. 1959. Über die Zusammensetzung des Ëbestaubung-
stropfensi und den Mecanismus seiner Sekretion. Planta
52: 587–599.
Zimmermann, E.C. 1994. Australian Weevils (Coleoptera:
Curculionoidea), vol. 1. Orthoceri: Anthribidae to Attel-
abidae (The Primitive Weevils). CSIRO Australia, Melb-
ourne.
695
Labandeira & al. • Pollination and Mesozoic gymnospermsTAXON 56 (3) • August 2007: 663–695
fIGurE crEdItS
Subfigures not listed herein are either our own or are extracted
from journals that are no longer published.
Fig. 2A. Reproduced with permission from W.H. Freeman and
Company: Gifford, E.M. & Foster, A.S., 1989. Morphology
and Evolution of Vascular Plants, fig. 15-7, p. 352.
Figs. 2B, 2C. Reproduced with permission by the American As-
sociation for the Advancement of Science: Rothwell, G.W.
1972. Evidence of pollen tubes in Paleozoic pteridosperms.
Science 175: figs. 1A and 1B on p. 1251.
Fig. 2D. Used with permission by the Botanical Society of
America: Takaso, T. & Owens, J.N. 1996. Ovulate cone,
pollination drop, and pollen capture in Sequoiadendron
(Taxodiaceae). Amer. J. Bot. 83: fig. 5, p. 1176.
Fig. 2E. Used with permission from the Botanical Society of
America: Tomlinson, P.B., Braggins, J.E. & Rattenbury,
J.A. 1991. Pollination drop in relation to cone morphology
in Podocarpaceae: a novel reproductive system. Amer. J.
Bot. 78: fig. 3, p. 216.
Fig. 2F. Bioimages (http://www.cas.vanderbilt.edu/bioimages/
species/frame/gibi2.htm).
Fig. 2G. Reproduced with permission from Timber Press, Port-
land,Oregon:Proctor,M.,Yeo,P.&Lack,A.1996.The
Natural History of Pollination, fig. 9.4, p. 274.
Figs. 2H, 2J. Reproduced, with permission by W.H. Freeman
and Company: Gifford, E.M. & Foster, A.S. 1989. Mor-
phology and Evolution of Vascular Plants, fig. 18-16, p.
469; fig. 18-23, p. 476.
Fig. 2I. No response from the Palm and Cycad Societies of
Australia after 3 weeks of repeated e-mailings.
Fig. 3D, 3E. Reproduced with permission by Elsevier, Ltd.:
Kvaček,J.2000. Frenelopsis alata and its microsporangi-
ate and ovuliferous reproductive structures from the Ceno-
manian of Bohemia (Czech Republic, Central Europe). Rev.
Palaeobot. Palynol. 112: fig. 4, p. 72, and fig. 3, p. 74.
Fig. 3I. Reproduced with permission by Elsevier, Ltd.: Barale,
G., Marrón, T.F. & Alvarez-Ramis, C. 1988. Étude de cônes
males de Frenelopsis Schenck emend. Reymanówna &
Watson, 1976 (Cheirolepidiaceae) du Crétacé Supérieur
deTorrelaguna(ProcincedeMadrid―Espagne).Geobios
21: fig. 4, p. 197.
Figs. 3N, 3O.Yeates,D.K.&Wiegmann,B.M. (eds.).The
Evolutionary Biology of Flies, copyright © 2005, Colum-
bia University Press. Reprinted with permission of the
publisher.
Figs. 3Q, 3R, 3S, 3T. Reproduced with permission, courtesy
of Pensoft Publishers.
Figs. 4A, 4B. Used with permission by Dr. Sharon Klavins.
Fig. 4C. Used with permission by The University of Chicago
Press: Klavins, S.D., Taylor, E.L., Krings, M. & Taylor,
T.N. 2003. Coprolites in a Middle Triassic cycad pollen
cone: evidence for insect pollination in early cycads? Evol.
Ecol. Res. 7: figs. 1d and 1e, p. 482.
Figs. 4D, 4E, 4F. Reproduced, with permission, from Springer
Science and Business Media: Nishida, H. & Hayashi, N.
1996. Cretaceous coleopteran larva fed on a female fruc-
tification of extinct gymnosperm. J. Pl. Res. 109: fig. 1, p.
328, and figs. 5 & 11, p. 329.
Figs. 4G, 4H. Reproduced with permission from Elsevier,
Ltd.: Krassilov, V., Tekleva, M., Meyer-Melikyan, N. &
Rasnitsyn, A.P. 2003. New pollen morphotype from gut
compression of a Cretaceous insect, and its bearing on
palynomorphological evolution and palaeoecology. Cre-
taceous Res. 24: fig. 2, p. 150.
Fig. 4I. Reproduced with permission by the Russian Academy
of Sciences.
Figs. 4O, 4P. Photographs kindly provided by Dr. D.S. Amorim,
Universidade de São Paulo.
Figs. 4M, 4N. Reproduced with permission by the Russian
Academy of Sciences.
Fig. 5A. Reproduced with permission by Schweizerbart’sche
Verlagsbuchhandlung (www.schweizerbart.de): Delev-
oryas, T. 1968. Investigations of North American cycade-
oids: structure, ontogeny and phylogenetic considerations
of cones of Cycadeoidea. Palaeontographica, Abt. B,
Paläophytol. 121: fig. 10, plate 36.
Fig. 5B. Reproduced with permission by Blackwell Publishing:
Bose, M.N. 1968. A new species of Williamsonia from the
Rajmahal Hills, India. J. Linn. Soc., Bot. 61: fig. 1, plate
1, following p. 122.
Fig. 5C: No response from the Polish Academy of Sciences,
after 3 weeks of repeated e-mailings.
Fig. 5H. Reproduced with permission from Elsevier, Ltd.:
Taylor, T.N. 1973. A consideration of the morphology,
ultrastructure and multicellular microgametophyte of
Cycadeoidea dacotensis pollen. Rev. Palaeobot. Palynol.
16: fig. 1, p. 160.
Fig. 6A: Reproduced with permission by Allen Press, Law-
rence, from Delevoryas, T. 1971. Biotic provinces and the
Jurassic-Cretaceous floral transition. Pp. 1660–1674 in:
Yochelson, E. (ed.),Proceedings of the North American
Paleontological Convention (September, 1969, Chicago),
part 1, fig. 15, p. 1672.
