This article appeared in a journal published by Elsevier. The attached copy is 
furnished to the author for internal non-commercial research and educational 
use, including for instruction at the author’s institution and sharing with 
colleagues. 
 
 
Other uses, including reproduction and distribution, or selling or licensing 
copies, or posting to personal, institutional or third party websites are 
prohibited. 
 
 
In most cases authors are permitted to post their version of the article (e.g. in 
Word or Tex form) to their personal website or institutional repository. Authors 
requiring further information regarding Elsevier’s archiving and manuscript 
policies are encouraged to visit: 
 
http://www.elsevier.com/copyright 
A paleobiologic perspective on plant–insect interactions
Conrad C Labandeira1,2,3,4
Available online at www.sciencedirect.com
Auth r's P rsonal CopyFossil plant–insect associations (PIAs) such as herbivory and
pollination have become increasingly relevant to paleobiology
and biology. Researchers studying fossil PIAs now employ
procedures for assuring unbiased representation of field
specimens, use of varied analytical quantitative techniques, and
address ecological and evolutionarily important issues. For
herbivory, the major developments are: Late Silurian–Middle
Devonian (ca. 420–385 Maa) origin of herbivory; Late
Pennsylvanian (318–299 Ma) expansion of herbivory; Permian
(299–252 Ma) herbivore colonization of new habitats;
consequences of the end-Permian (252 Ma) global crisis; early
Mesozoic (ca. 235–215 Ma) rediversification of plants and
herbivores; end-Cretaceous (66.5 Ma) effects on extinction; and
biological effects of the Paleocene–Eocene Thermal Maximum
(PETM) (55.8 Ma). For pollination, salient issues include: Permian
pollination evidence; the plant hosts of mid-Mesozoic (ca. 160–
110 Ma) long-proboscid pollinators; and effect of the angiosperm
revolution (ca. 125–90 Ma) on earlier pollinator relationships.
Multispecies interaction studies, such as contrasting damage
types with insect diversity and establishing robust food webs,
expand the compass and relevance of past PIAs.
Addresses
1 Smithsonian Institution, National Museum of Natural History,
Department of Paleobiology, Washington, DC 20013, USA
2Rhodes University, Department of Geology, Grahamstown 6140, South
Africa
3Capital Normal University, College of Life Sciences, Beijing 100048,
China
4Department of Entomology and BEES Program, University of Maryland,
College Park, MD 20742, USA
Corresponding author: Labandeira, Conrad C (labandec@si.edu)
Current Opinion in Plant Biology 2013, 16:414–421
This review comes from a themed issue on Biotic interactions
Edited by Beverley Glover and Pradeep Kachroo
For a complete overview see the Issue and the Editorial
Available online 2nd July 2013
1369-5266/$ – see front matter, Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.pbi.2013.06.003
Introduction
The paleobiological study of plant–insect associations
(PIAs) developed later than the allied fields of paleobo-
tany and paleoentomology, or the study of modern PIAs.
It was during the latter half of the twentieth century that
an increase in the number of (largely descriptive) articlesa The designation, Ma, or mega-annum, refers to millions of years ago.
Also see Figure 1.
Current Opinion in Plant Biology 2013, 16:414–421 recorded the presence of fossil PIAs, eventually reaching
a noticeable presence within paleobiology [1]. The dis-
cipline retained a primarily idiographic bent until the late
1990s, when the testing of hypotheses regarding changes
in PIAs through time required various quantitative tech-
niques to elucidate patterns of herbivory within and
among bulk floras [2]. This transformation was spurred
by studies from the Early Permian of Texas [3,4], and the
Paleogene of the U.S. Western Interior [5], which
increased the scope of analytical techniques [6,7]. A
recent advancement is the effect of global climate change
on plants and their insect herbivores [7], especially during
the Paleocene–Eocene Thermal Maximum (PETM)
[8], and other Paleogene sites [9–14]. Current studies
involve how plants and their insect herbivores respond to
ecological crises such as mass extinctions [15,16], and the
ecological expansion of new, globally dominant clades,
including Mesozoic parasitoid insects [16,17] and angios-
perms [18,19].
Examination of fossil insect pollination and related inter-
actions has had a different history [2,7]. On the basis of
the scarcity of evidence and lack of direct data, demon-
strations of pollination have relied on the building of a
case from multiple, independent sources of evidence
[19,20,21]. Initial interest in pollination focused on Late
Carboniferous medullosan seedferns [22] (Glossary),
attributable to their very large pollen grains and distinc-
tive reproductive organs. This was followed by interest in
pollination of mid-Mesozoic seedferns such as caytonia-
leans and bennettitaleans [19,21–23], and early angios-
perms [20,24]. Recently, there has been renewed interest
in mid-Mesozoic insect–gymnosperm pollination, attribu-
table to well preserved material from eastern Eurasia
[19,20,21,25]. Because evidence for pollination is rare
in the fossil record, unlike more conspicuous herbivory,
metrics were never developed to quantify pollination
diversity or intensity.
Other fossil PAIs include very rare mimicry [26] and
various plant pathogens vectored by insects (CC Laban-
deira, R Prevec, unpublished data). These PIAs are
ancient and are documented from the mid-Mesozoic
[26,27] and earlier. Nevertheless, this review will con-
centrate on herbivory and pollination in the fossil record
in the context of recent empirical developments and
relevant theoretical issues (Figure 1).
Herbivory
The earliest terrestrial ecosystems with recognizably
megascopic life are from the Late Silurian to Middle
Devonian, and display the earliest occurrences of severalwww.sciencedirect.com
Fossil plant–insect interactions Labandeira 415
Author's Personal Copyland-plant tissue types associated with stems. Some of
these live tissues became consumed by small arthropod
herbivores within several million years (m.y.) (Figure 1,
cluster 1). The earliest arthropod feeding groups were
piercer-and-suckers and pith borers in stems, and con-
sumers of adjacent spores and sporangia [28,29]. Later,
during the Middle Devonian, there is evidence for con-
sumption of foliose liverwort thalli (Labandeira, Trem-
blay, Bartowski, Hernick, unpublished data), in lieu of
true vascular-plant leaves that later became ecologically
conspicuous. By the end of the Devonian, all major organs
and tissue types of land plants had evolved [28], but it was
not until various times during the Carboniferous, in
equatorial wetland environments, that the other tissue
types become consumed, after an average lag time of
91 m.y. after they first occur in the fossil record [29].
Reasons for these inordinately long lag times for herbiv-
ory of certain tissue types need to be explored, but
probably involve atmospheric O2 levels that plunged
below the threshold for sustained active respiration of
arthropod herbivores [30].
Increased herbivory ensued after a considerable rise of
atmospheric O2 levels in the Early Carboniferous, and by
the mid-Late Carboniferous (Figures 1, cluster 2 and 2a),
virtually all modern functional feeding groups, exophytic
as well as endophytic, are present in humid equatorial
habitats [1,2,16,31], with the exception of leaf miners.
This time interval represents the first trophic evidence for
the establishment of modern component communities, in
which gallers, root feeders, wood borers and ovipositing
insects were added as herbivores, providing a new base-
line for food webs that appear more driven by herbivores
than detritivores. In different environments of the Per-
mian, these feeding styles colonized varied river-associ-
ated habitats in western Euramerica (southwestern USA)
[3,4,32] (Figures 1, cluster 3 and 2b), and floristically
distinctive glossopterid habitats of south-central Gond-
wana (South Africa’s Karoo Basin) [33]. These herbivory
trends were brought to a halt at the end-Permian (P-Tr)
extinction. Preliminary evidence indicates that following
the P-Tr ecological crisis, plant, insect, and associational
diversity and intensity decreased significantly, was overly
generalized in host-specificity, and did not recover to
precrisis Permian levels until the early Late Triassic,
25 m.y. later [7,16,34,35] (Figures 1, cluster 4 and 2c).
Reasons for this delayed rediversification are being
explored [36], but equally impressive is the intensity
and diversity of the herbivore rebound [16], replete with
high host-plant specificities that rival the angiosperm
diversification event 100 m.y. later [37].
Little is known of herbivore interactions in mid-Mesozoic
ecosystems, although exploratory analyses (XL Ding, CC
Labandeira, unpublished data) of two important, well-
preserved sites in northeastern China, the Middle Jurassic
Jiulongshan Formation and the mid-Early Cretaceouswww.sciencedirect.com sites Yixian Formation, indicate substantial herbivory
in some habitats that surrounded large lake basins. Non-
herbivore PIAs also have been documented
[19,21,26,38], and will be integrated with results from
quantitative analyses on bulk, site-specific floras to pro-
vide a robust portrait of a trophically diverse, preangios-
permous food web. More is known, however, of PIAs
resulting from the end-Cretaceous (K-Pg) event, which
appears broadly similar to the P-Tr event. After the K-Pg
crisis, there is a decrease in plant–insect interactional
diversity during the Paleocene, rebounding during the
latest Paleocene when levels of interactional diversity and
intensity equaled that of the latest Cretaceous 10 m.y.
earlier. The dramatic reduction of overall diversity, feed-
ing intensity and host specialization following the K-Pg
event (Figure 1, cluster 5) was documented for the Will-
iston Basin [15,39], but not in other worldwide regions,
where the boundary is absent or remains undiscovered
[11]. Examination of the Denver Basin record would
extend the North American patterns closer to the bolide
impact site at Yucatan, but confirmation of the devastat-
ing effect of this major perturbation on PIAs awaits
examination of other stratigraphic continuous intervals
with well preserved floras outside of North America.
Herbivory patterns have been well documented for a six
m.y. interval (59–53 Ma) that includes the PETM, a
transient, 105 year long event at 55.8 Ma, during which
there was an elevation of global mean annual temperature
(MAT) by 5–78C and a doubling of atmospheric CO2
levels [8,9]. Associated with PETM physical changes
was a transformation of floral composition at mid-latitude
sites such as those examined in Wyoming and a significant
increase in herbivore intensity, as assessed by damage-
type frequency data [8] (Figures 1, cluster 5 and 2d).
These patterns persist during a six m.y. interval that
includes the post-PETM Early Eocene Climatic Opti-
mum at 53–51 Ma [5,8,9], after which there is a down-
ward but fluctuating global trajectory in temperature and
CO2 levels, occasionally with counterintuitive results
when compared to the earlier trends [12].
At approximately the same time that deposits were
being lain in North America that recorded major cli-
matic events and trends during the late Paleocene to
mid-Eocene, diverse floras were deposited in southern
South America [40]. Four examined sites at Laguna del
Hunco, in southern Argentina, display a greater diver-
sity and intensity of herbivory in both finely resolved
damage types and coarser grained functional feeding
groups when compared to earlier, more depauperate
rainforest floras from northern South America [41], and
to contemporaneous floras in North America [40]
(Figure 1, cluster 5). This significant increase in herbi-
vore feeding diversity, particularly specialized inter-
actions, may explain current, elevated PIA diversity
in South America [42].Current Opinion in Plant Biology 2013, 16:414–421
416 Biotic interactions
Figure 1
200
250
150
100
50
0
400
350
300
Tr
ia
ss
ic
Pe
rm
ia
n
Ca
rb
on
ife
ro
us
D
ev
on
ia
n
Si
l
Ju
ra
ss
ic
Cr
et
ac
eo
us
P
al
eo
zo
ic
M
es
oz
oi
c
C
en
oo
zo
ic
Pa
le
og
en
e
Ne
og
en
e
Late
Lopingian
Middle
Guadalup.
Early
Middle
Early
Cisuralian
Penn.
Miss.
Late
Middle
Early
Late
Early
Late
Paleocene
Eocene
Oligocene
Miocene
Ma Pe
rio
d
Er
a
Ep
oc
h
1
1
2
2
3
3
3
4
5
6
7
1
2
6
7
8
3
5
9
4
[18]
[39] [DLU]
[SLU]
[3]
[28,29]*
[28,29]*
[28,29]*
[43]
[25]
[45]
[19-21]
[19-21]
[22]
[44]
[22] [50]
[49]
[44]
[44]
[46]
[24,47]
[51]
[19]
[19]
[19,26,39]
[LPU]
[LPU]
[LPU]
[DLU]
[11,15,39]
[5,8]
[9,10,12]
[10]
[14]
[16,37]
[35]
[16]
[33]
[34]
[16]
[4]
[13]
[12]
[5, 8 41]
[40,42]
[15,40]
[28,29]*
[19]
[22,24]
[32]
[1,2,16,37]
[1,2,16]
[LTBHU]
[28,29]
[28,29]
Paleocene-Eocene Thermal Max.
Lag and rediversification of PIAs
Lag and rediversification of PIAs
Permian-Triassic (P- Tr) crisis
Gondwana
Euramerica
early
pollination
expansion
of herbivory
Cretaceous-Paleogene (K-Pg) crisis
Post mid-Mesozoic angiosperm
pollination modes
Mid-Mesozoic gymnosperm
pollination modes
Molteno herbivore radiation
Modern-style herbivore interactions
Origin and early
evolution of herbivory
?
?
Pa
tho
ge
ns
An
gio
spe
rm
s
Gy
mn
osp
erm
s*
Qu
ant
itat
ive
ly f
ocu
sed
Qu
alit
ativ
ely
 foc
use
d
Mi
mi
cry
 / D
efe
nse
Palynophagy
& Pollination
Other
Interactions
Major herbivory (green) 
and pollination (red)
episodes
Herbivory
Current Opinion in Plant Biology
Current Opinion in Plant Biology 2013, 16:414–421 www.sciencedirect.com
Author's Personal Copy
Fossil plant–insect interactions Labandeira 417
Figure 2
(a) (b) (c) (d)
(e) (f) (f)
Current Opinion in Plant Biology
A gallery of fossil plant–insect interactions representing the perspectives of time interval, interaction type, preservational mode, and medium of
illustration. Images (a)–(d) represent herbivory from Paleozoic (a, b), Mesozoic (c), and Cenozoic (d) deposits. At (a) is Late Pennsylvanian (ca. 305 Ma)
piercing-and-sucking damage caused by mouthpart stylets of an extinct paleodictyopteroid insect extracting xylem and phloem (bracketed by the two
diagonally parallel lines at lower-left), from the marattialean tree fern, Psaronius chasei, in an anatomically preserved peat [31]. Mattoon Fm., Illinois,
USA; UIUC acetate peel 8227-Bbot, section 109. At (b) are Early Permian (ca. 296 Ma) Ovofoligallites padjetti galls in pinnular tissues of the medullosan
seedfern, Odontopteris readi, induced by a probable early hemipteroid insect [32]. Archer City Fm., north-central Texas, USA; USNM 41165a. At (c) are
Late Triassic (ca. 225 Ma) examples of dragonfly oviposition insertion scars on an equisetalean stem fragment of Equisetites kapokensis, from the
Molteno Formation, South Africa. Colorized camera lucida drawing of compression material; scale bar at bottom-right in mm increments. At (d) is a
strip of external foliage feeding showing extensive herbivory on Dicot Morphotype WW006, from the early onset of the Paleocene–Eocene Thermal
Maximum (55.8 Ma) [8]. Willwood Fm., Wyoming, USA; USNM530968. Images (e) and (f) represent Mesozoic and Cenozoic pollination. At (e) is a
Middle Jurassic (165 Ma) head and mouthparts, proboscis in yellow, of Lichnomesopsyche gloriae, an extinct scorpionfly that probably probed for
gymnosperm pollination drops [21]. Jiulongshan Fm., Inner Mongolia, China; CNU-MEC-NN-2005-024. At (f) is the Early Miocene (ca. 21 Ma)
mutualism between an agaonid fig wasp Tetrapus delclosi (at left), and (at right) an associated clump of Ficus pollen processed by specialized
mandibular appendages from the head underside [48]. La Toca Fm., Dominican Republic; AMNH DR-14-576. The reconstruction at (g) is mimicry
between a hangingfly, Juracimbrophlebia ginkgofolia (the mimic), and a coexisting, multilobed ginkgoalean leaf, Yimaia capituliformis (the model), from
the same locality as (e), based on specimens at Capital Normal University, Beijing, China [26]. Scale bars: 1 mm; except for figure (f) (right), 0.1 mm.
Abbreviation: Ma, millions of years.
(Figure 1 Legend) Geochronologic distribution of fossil plant–insect associational studies from the recent literature, including secondary sources, and
the timing of significant episodes in plant–insect associations. Herbivory is indicated in green, pollination in red and other associations in black.
References in brackets are keyed to data and events discussed in the text; circled numbers at right refer to major events mentioned in the text. Large-
numbered, elliptically shaded study-site clusters suggest time intervals of comparatively high levels of recent interest, called out in the text. For those
localities with an age-date range, the midpoint date was plotted. Some locality points for quantitatively focused data represent multiple sites, such as
Green River and Messel for 47–48 Ma [9,10,12], and several USA Western Interior sites and Cerrejo´n, Colombia, for 58–59 Ma [39,41]. Open dots with
asterisks indicate pollen feeding on pregymnospermous, free-sporing plants. Abbreviations: DLU (Ding, Labandeira, unpublished data); Guadalup.,
Guadalupian; LPU (Labandeira, Prevec, unpublished data); LTBHU, Labandeira, Tremblay (Bartowski, Hernick, unpublished data); Miss., Mississipian
Subperiod; Penn., Pennsylvanian Subperiod; PIAs, plant–insect associations; Sil., Silurian; SLU (Schachat, Labandeira, unpublished data). *Includes
Silurian and Devonian plant lineages that are not gymnospermous.
www.sciencedirect.com Current Opinion in Plant Biology 2013, 16:414–421
Author's Personal Copy
418 Biotic interactions
Author's Personal CopyPollination and related associations
In the fossil record, spores of vascular plants precede the
earliest occurrence of pollen, as does their consumption by
arthropods. Spore feeding, as demonstrated by spore-laden
coprolites in late Silurian and Early Devonian deposits
[28,29] (Figure 1, cluster 6), was a precursor to pollen
feeding, seen in later Paleozoic biotas [16,22] from insect
gut contents [22], pollen grains that display circular holes
indicating punch-and-sucking [43], and the earliest mouth-
parts with long tubular proboscides and pollen-macerating
mandibles [19,25] (Figure 1, cluster 7). This distinctive
evidence strongly indicates that during the Late Pennsyl-
vanian and Permian (306–252 Ma), pollen transfer
occurred between insects and their seed-plant hosts.
During the mid-Mesozoic, the punch-and-sucking pollina-
tion mode reappears in an association between thrips and
ginkgoaleans [44], and mandibulate insects, probably
beetles, consumed cycad pollen sacs [45] (Figures 1, cluster
8 and 2e). Prominently, there are three, independently
originating groups of long-proboscid insects — true flies
[20], scorpionflies [21], and kalligrammatid lacewings
[19] — that represent multiple mouthpart convergences.
One of the major challenges in understanding the repeated
origin of the long-proboscid pollination mode is to deter-
mine which co-occurring gymnosperm ovulate organs were
accommodated by elongate tubular mouthparts. Some long
proboscides had sponge-like, accessory terminal structures
for capillary uptake of fluids such as pollination drops, from
a variety of receptive ovulate features including integu-
mentary tubes, tubular clasping bracts, deep funnels,
channels, and siphonate micropyles [19,21]. Some tubular
structures were lined with nectary-like structures, in-
cluding resin glands, and glandular trichomes [21,22].
Various early angiosperm pollination types replaced ear-
lier modes of gymnosperm pollination (Figures 1, cluster
9 and 2f). This dramatic turnover [19] likely is attribu-
table to a more efficient system of lures, rewards, and
pollen capture by angiosperms, although such a hypoth-
esis needs testing. Most early angiosperm pollination is
generalized [24,46], although instances of specialization
are known throughout the record, including beetles [47],
fig wasps [48], bees [49], and mosquitoes and mites
caught in the act of feeding on pollen sacs [50].
Current and future discipline-wide trends
Several approaches have contributed toward understand-
ing PIAs in the fossil record (Figure 1). Most crucial are
the unbiased collection of data, proper methods of data
analysis, and importantly, inquiry based on sound evol-
utionary and ecological questions [6,7]. Historical trends
of PIAs during the past 50 years have revealed an assess-
ment of where the field has been and where it is headed
(Figure 2). From this assessment, seven issues have
garnered recent traction and are ripe for more mature
evaluation.Current Opinion in Plant Biology 2013, 16:414–421 1. A comprehensive understanding of the origin of
herbivory is needed for the earliest terrestrial
ecosystems. Currently, the data are very sparse, and
the earliest occurrences originate from only four
deposits [28]. The paucity of data may reflect actual
rarity of early herbivory, or more likely represents a lack
of searching, attributable to a major Middle Devonian to
latest Early Mississippian [30] hiatus in the documented
herbivore record (CC Labandeira, S Tremblay, KE
Bartowski, LV Hernick, unpublished data).
2. The expansion of insect herbivory into new habitats
during the Permian requires documentation in greater
detail from additional sites. For the Early Permian of
southwestern Euramerica, additional studies from
north-central Texas (S Schachat, CC Labandeira,
unpublished data) provide ecological insights into the
different colonization patterns of plant hosts by insect
herbivores in various habitats. For the Middle to Late
Permian of Gondwanan South Africa, differences are
emerging in herbivory patterns among glossopterid
floras [33].
3. More recent data are needed for the immediate and
longer-term consequences of major extinction events in
terrestrial ecosystems. Work has begun to understand
the patterns of pre-event and postevent patterns of PIAs
during the P-Tr [16] and K-Pg [15] intervals. For the P-
Tr case, radioisotopic age dates need to be established
for reliable rate estimates of associational change.
4. Broad global assessments are required to record the
gradual but profound environmental change that has
affected plants and their insect herbivores. The
Paleocene–Eocene Thermal Maximum has been most
extensively examined, but the current data originate
only from one set of localities in North America. It is
unclear how universal the worldwide response is, and
whether strata of the same age can be tapped in other
terrestrial localities to elucidate a commonality of
effects.
5. Recent efforts to understand pollinator-related mouth-
part specializations of insects during the preangios-
permous Mesozoic need to be balanced by a renewed
focus on anatomical features of their likely targeted
plant hosts, particularly bennettitaleans, caytonia-
leans, cheirolepidiaceous conifers and ginkgoaleans
[19,22,23]. Structures of gymnospermous ovulate and
pollen organs that previously were enigmatic may be
explained through an understanding of coeval insect
mouthpart morphology and specializations.
6. Additional examples are required to document wing
color-patterns for mid-Mesozoic insects that indicate
mimicry and crypsis with contemporaneous plants
(Figure 2g). It is likely that mid-Mesozoic data
involving insect–ginkgoalean mimicry [26] may be
ecologically very similar to Cenozoic patterns with
angiosperms [51], particularly in the face of almost
complete turnover of important plant, insect and
vertebrate lineages.www.sciencedirect.com
Fossil plant–insect interactions Labandeira 419
Author's Personal Copy7. More integrated and intensive studies of multiple
species associations are needed that include plants,
their interactors (herbivores, pollinators, fungi, patho-
gens), and other trophic elements at local sites. A
broader, multispecies perspective [52] would supple-
ment emphases on single pairwise [53] or tritrophic
[54] interactions. One approach would examine
bilateral PIAs in a modern source community where
damage-type richness is contrasted with the linked
insect richness responsible for the damage [55]. By
contrast, construction of food webs can result in trophic
networks that have high-resolution as well as high
diversity, and are based on an accurate capture of all
evident PIAs present in a local ecosystem, recently
done for a 47 m.y. maar lake in Germany [56]. The
use of successive food webs with high trophic certainty
may be the best approach toward addressing salient
questions about ecological evolution of PIAs in deep
time.
Acknowledgements
I thank Beverley Glover and Pradeep Kachroo for an invitation to provide
this manuscript. Finnegan Marsh drafted and formatted the figures. Grant
DEB-0919071 to P. Wilf is acknowledged. This is Contribution 254 of the
Evolution of Terrestrial Ecosystem consortium at the National Museum of
Natural History, in Washington, DC.
References and recommended reading
Papers of particular interest, published within the period of review,
have been highlighted as:
 of special interest
 of outstanding interest
1. Labandeira CC: Early history of arthropod and vascular plant
associations. Annu Rev Earth Planet Sci 1998, 26:327-377.
2. Labandeira CC: The history of associations between plants and
animals. In Plant–Animal Interactions: An Evolutionary Approach.
Edited by Herrera C, Pellmyr O. Oxford, United Kingdom:
Blackwell Science; 2002. 26–74, 248–261.
3. Beck AL, Labandeira CC: Early Permian folivory on a
gigantopterid-dominated riparian flora from north-central
Texas. Palaeogeogr Palaeoclim Palaeoecol 1998,
142:139-173.
4. Labandeira CC, Allen EM: Minimal insect herbivory for the
Lower Permian Coprolite Bone Bed site of north-central
Texas, USA, and comparison to other late Paleozoic floras.
Palaeogeogr Palaeoclim Palaeoecol 2007, 247:197-219.
5. Wilf P, Labandeira CC: Response of plant–insect associations
to Paleocene–Eocene warming. Science 1999, 284:2154-2156.
6. Wilf P: Insect-damaged fossil leaves record food web
response to ancient climate change and extinction. New Phytol
2008, 178:486-502.
7. Labandeira CC, Currano ED: The fossil record of plant–insect
dynamics. Annu Rev Earth Planet Sci 2013, 41 http://dx.doi.org/
10.1146/annurev-earth-052120124139.
8.

Currano ED, Labandeira CC, Wilf P: Fossilized insect folivory
tracks temperature for six million years. Ecol Monogr 2010,
80:547-567.
This study provides the first robust analysis of the effect of the Paleo-
cene–Eocene Thermal Maximum on floras and their insect herbivory
before, during and after this exceptional elevation in atmospheric tem-
perature and CO2 increase. This study analyzes in deep-time strata a
process similar anthropogenic climate change that is occurring today
(Figure 2d).www.sciencedirect.com 9. Wilf P, Labandeira CC, Johnson KR, Coley PD, Cutter AD: Insect
herbivory, plant defense and early Cenozoic climate change.
Proc Natl Acad Sci U S A 2001, 98:6221-6226.
10. Smith DM: A comparison of plant–insect associations in the
middle Eocene Green River Formation and the Upper Eocene
Florissant Formation and their climatic implications. Geol Soc
Am Spec Pap 2008, 435:89-103.
11. Wappler T, Currano ED, Wilf P, Rust J, Labandeira CC: No
post-Cretaceous ecosystem depression in European forests?
Rich insect-feeding damage on diverse middle Paleocene
plants, Menat, France. Proc Roy Soc Lond B 2009, 276:
4271-4277.
12. Wappler T, Labandeira CC, Rust J, Frankenha¨user H, Wilde V:
Testing for the effects and consequences of mid Paleogene
climate change. PLoS ONE 2012, 7:e40744.
13. Currano ED, Jacobs BF, Pan AD, Tabor NJ: Inferring ecological
disturbance in the fossil record: a case study from the late
Oligocene of Ethiopia. Palaeogeogr Palaeoclim Palaeoecol 2011,
309:242-252.
14. Wappler T: Insect herbivory close to the Oligocene–Miocene
transition — a quantitative analysis. Palaeogeogr Palaeoclim
Palaeoecol 2010, 292:540-550.
15. Labandeira CC, Johnson KR, Wilf P: Impact of the terminal
Cretaceous event on plant–insect associations. Proc Natl Acad
Sci U S A 2002, 99:2061-2066.
16. Labandeira CC: Silurian to Triassic plant and insect clades and
their associations: new data, a review, and interpretations.
Arthropod Syst Phylogeny 2006, 64:53-94.
17.

Labandeira CC: Evidence for outbreaks from the fossil record
of insect herbivory. In Insect Outbreaks Revisited. Edited by
Barbosa P, Agrawal AA, Letorneau D. Chichester, United
Kingdom: Blackwell Science; 2012:269-290.
This is the first exploration of data indicating pest outbreaks in the fossil
record. It presents leaf-damage frequency and intensity data from exter-
nal foliage feeders and leaf miners for inordinately high levels of folivory.
18. Labandeira CC, Dilcher DL, Davis DR, Wagner DL: Ninety-seven
million years of angiosperm–insect association:
paleobiological insights into the meaning of coevolution. Proc
Natl Acad Sci U S A 1994, 91:12278-12282.
19.

Labandeira CC: The pollination of mid Mesozoic seed plants
and the early history of long-proboscid insects. Ann Missouri
Bot Gard 2010, 97:469-513.
This synthesis and review explores the preangiospermous, mid-Meso-
zoic, long-proboscid pollination mode by linking three lineages of long-
proboscid insects with receptive plant reproductive features of gymnos-
perm ovules (Figure 2e).
20. Ren D: Flower-associated Brachycera flies as fossil evidence
for Jurassic angiosperms. Science 1998, 280:85-88.
21. Ren D, Labandeira CC, Santiago-Blay JA, Rasnitsyn AP, Shih CK,
Bashkuev A, Logan MAV, Hotton CL, Dilcher DL: A probable
pollination mode before angiosperms: Eurasian, long-
proboscid scorpionflies. Science 2009, 326:840-847.
22. Labandeira CC, Kvacˇek J, Mostovski MB: Pollination drops,
pollen and insect pollination of Mesozoic gymnosperms.
Taxon 2007, 56:663-695.
23. Osborn JM, Taylor ML: Pollen and coprolite structure in
Cycadeoidea (Bennettitales): implications for understanding
pollination and mating systems in Mesozoic cycadeoids. In
Plants in Deep Mesozoic Time: Morphological Innovations,
Phylogeny and Ecosystems. Edited by Gee CT. Bloomington,
Indiana, USA: Indiana Univ. Press; 2010:34-49.
24. Crepet WL, Nixon KC, Gandolfo MA: An extinct calycanthoid
taxon, Jerseyanthus calycanthoides, from the Late
Cretaceous of New Jersey. Am J Bot 2005, 92:1475-1485.
25. Bashkuev AS: Nedubroviidae, a new family of Mecoptera: the
first Paleozoic long-proboscid scorpionflies. Zootaxa 2011,
2895:47-57.
26.

Wang YJ, Labandeira CC, Ding QL, Shih CK, Zhao YY, Ren D: An
extraordinary Jurassic mimicry between a hangingfly andCurrent Opinion in Plant Biology 2013, 16:414–421
420 Biotic interactions
Author's Personal Copyginkgo from China. Proc Natl Acad Sci U S A 2012, 109:
20514-20519.
This study presents an unusual occurrence of Middle Jurassic mimicry
between the whole body of scorpionfly and a species of co-occurring
ginkgoalean with leaves that are nearly identical in shape, size, silhouette
and surface texture. It extends modern styles of mimicry or crypsis to the
preangiospermous Mesozoic (Figure 2g).
27. Sung GH, Poinar GO Jr, Spatafora JW: The oldest fossil
evidence of animal parasitism by fungi supports a Cretaceous
diversification of fungal-arthropod symbioses. Mol Phylogen
Evol 2008, 49:495-502.
28. Labandeira CC: The origin of herbivory on land: the initial
pattern of live tissue consumption by arthropods. Insect Sci
2007, 14:259-274.
29.

Labandeira CC: Deep-time patterns of tissue consumption by
terrestrial insect herbivores. Naturwissenschaften 2013,
100:355-364.
This paper presents fine-grained, plant histological data throughout the
fossil record, but especially from early terrestrial ecosystems indicating
that certain plant (especially photosynthetic) tissues were quickly con-
sumed by arthropod herbivores. By contrast, other (mostly structural and
meristematic) tissues had a ninefold longer lag time on average before
they were subject to arthropod herbivory.
30. Ward P, Labandeira CC, Laurin M, Berner R: Confirmation of
Romer’s Gap as a low oxygen interval constraining the timing
of initial arthropod and vertebrate terrestrialization. Proc Natl
Acad Sci U S A 2006, 103:16818-16822.
31. Labandeira CC, Phillips TL: Insect fluid-feeding on Upper
Pennsylvanian ferns (Palaeodictyoptera, Marattiales) and the
early history of the piercing-and-sucking functional feeding
group. Ann Entomol Soc Am 1996, 89:157-183 (Figure 2a).
32. Stull G, Labandeira CC, Chaney D, DiMichele WA: The ‘‘seeds’’
on Padgettia readi Mamay are insect galls: reassignment of
the plant to Odontopteris Brongniart, the gall to Ovofoligallites
gen. nov., and the evolutionary implications thereof. J
Paleontol 2013, 87:217-231 (see Figure 2b).
33. Prevec R, Labandeira CC, Neveling J, Gastaldo RA, Looy C,
Bamford MA: A portrait of a Gondwanan ecosystem: a new Late
Permian locality from Kwa-Zulu Natal, South Africa. Rev
Palaeobot Palynol 2009, 156:454-493.
34. Pott C, Labandeira CC, Krings M, Kerp H: Fossil eggs and
ovipositional damage on bennettitalean leaf cuticles
from the Carnian (Upper Triassic) of Austria. J Paleontol 2008,
82:778-789.
35. Moisan P, Labandeira CC, Matushkina NA, Wappler T, Voigt S,
Kerp H: Lycopsid–arthropod associations and odonatopteran
oviposition on herbaceous Isoetites. Palaeogeogr Palaeoclim
Palaeoecol 2012, 344/345:6-15.
36. Roopnarine PD, Angielczyk KD, Wang SC, Hertog R: Trophic
network models explain instability of Early Triassic terrestrial
communities. Proc Roy Soc B: Biol Sci 2007, 274:2077-2086.
37. Labandeira CC: The four phases of plant–arthropod
associations in deep time. Geol Acta 2006, 4:409-438.
38.

Pott C, McLoughlin S, Wu SQ, Friis EM: Trichomes on the leaves
of Anomozamites villosus sp. nov. (Bennettitales) from the
Daohugou beds (Middle Jurassic), Inner Mongolia, China:
defense against herbivorous arthropods. Rev Palaeobot Palynol
2012, 169:48-60.
This study is the best demonstration from the fossil record for the role of
trichomes as a distinct mechanism in antiherbivore defense. The tri-
chomes protect the foliage of a bennettitalean plant host which may
be linked to its suspected insect pollination.
39. Wilf P, Labandeira CC, Johnson KR, Ellis B: Decoupled plant and
insect diversity after the end-Cretaceous extinction. Science
2006, 313:1112-1115.
40. Wilf P, Labandeira CC, Johnson KR, Cu´neo NR: Richness of
plant–insect associations in Eocene Patagonia: a legacy for
South American biodiversity. Proc Natl Acad Sci U S A 2005,
102:8944-8948.
41. Wing SL, Herrera F, Jaramillo C, Go´mez C, Wilf P, Labandeira CC:
Late Paleocene fossils from the Cerrejo´n Formation,Current Opinion in Plant Biology 2013, 16:414–421 Colombia, are the earliest record of Neotropical rainforest.
Proc Natl Acad Sci U S A 2009, 106:18627-18632.
42. Price PW, Diniz IR, Morais HC, Marques ESA: The abundance of
insect herbivore species in the tropics: the high local richness
of local species. Biotropica 1995, 27:468-478.
43. Wang J, Labandeira CC, Zhang ZF, Bek J, Pfefferkorn HW:
Permian Circulipuncturites discinisporis Labandeira, Wang,
Zhang, Bek et Pfefferkorn gen. et sp. nov. (formerly
Discinispora) from China, an ichnotaxon of punch-and-
sucking insect on Noeggeranthialean spores. Rev Palaeobot
Palynol 2009, 156:277-282.
44.

Pen˜alver, Labandeira CC, Barro´n E, Delclo`s X, Nel P, Nel A,
Tafforeau P, Soriano C: Thrips pollination of Mesozoic
gymnosperms. Proc Natl Acad Sci U S A 2012, 109:8623-8628.
On the basis of exquisitely preserved amber material, this study demon-
strates an intricate pollinator relationship between an extinct lineage of
thrips and its highly probable ginkgoalean host. A highly specialized body
feature for capturing pollen, ring setae, is not found in any other extinct or
extant insect and is analogous to the branched hairs of bees.
45. Klavins SD, Kellogg DW, Krings M, Taylor EL, Taylor TN:
Coprolites in a Middle Triassic cycad pollen cone: evidence for
insect pollination in early cycads? Evol Ecol Res 2005, 7:479-
488.
46. Hu S, Dilcher DL, Jarzen DM, Taylor DW: Early steps of
angiosperm–pollination coevolution. Proc Natl Acad Sci U S A
2008, 105:240-245.
47. Gandolfo MA, Nixon KC, Crepet WL: Cretaceous flowers of
Nymphaeaceae and implications for complex insect
entrapment pollination mechanisms in early Angiosperms.
Proc Natl Acad Sci U S A 2004, 101:8056-8060.
48. Pen˜alver E, Engel MS, Grimaldi DA: Fig wasps in Dominican
amber (Hymenoptera: Agaonidae). Am Mus Novit 2006, 3541:1-
16 (Figure 2f).
49.

Danforth BN, Poinar GO Jr: Morphology, classification, and
antiquity of Melittosphex burmensis (Apoidea:
Mellitosphecidae) and implications for early bee evolution. J
Paleontol 2011, 85:882-891.
This follow up study from an earlier study indicates that the earliest
described bee was present during the late Early Cretaceous, soon after
the origin of eudicots. The size and body modifications present in this
specimen indicate that a specialized mode of pollination occurred early in
bee history.
50.

Hartkopf-Fro¨der C, Rust J, Wappler T, Friis EM, Viehofen A: Mid-
Cretaceous charred fossil flowers reveal direct observation of
arthropod feeding strategies. Biol Lett 2011 http://dx.doi.org/
10.1098/rsb.2011.0696.
This exception discovery records a very rare instance in the fossil record
where the mouthparts of an arthropod (a mite) was inserted into an
angiosperm pollen sac. An association of a mosquito with the same
chloranthaceous flower also indicates feeding on nectar.
51. Wedmann S, Bradler S, Rust J: The first fossil leaf insect: 47
million years of specialized cryptic morphology and behavior.
Proc Natl Acad Sci U S A 2007, 104:565-569.
52. Kaplan I, Denno RF: Interspecific interactions in phytophagous
insects revisited: a quantitative assessment of competition
theory. Ecol Lett 2007, 10:977-994.
53. Wilf P, Labandeira CC, Kress JW, Staines CL, Windsor DM,
Allen AL, Johnson KR: Timing the radiations of leaf beetles:
hispines on gingers from latest Cretaceous to Recent. Science
2000, 289:291-294.
54.

Hughes D, Wappler T, Labandeira CC: Life after death: ancient
death-grip leaf-scars reveal ant–fungal parasitism. Biol Lett
2011, 7:67-70.
This study is the first use of distinctive fossil plant damage to establish a
tritrophic interaction known from present biological research. The study
links a clade of zombie ants that produce a characteristic death-grip leaf
scar induced by a parasitoid fungus in control of the ant’s brain.
55. Carvalho MR, Wilf P, Barrios H, Currano ED, Windsor DM,
Jaramillo CA, Labandeira CC: Tropical canopy insects link leaf
damage in fossil and living forests. Geol Soc Am Abstr Prog
2011, 43:381.www.sciencedirect.com
Fossil plant–insect interactions Labandeira 421
Author's Personal Copy56.

Labandeira CC, Labandeira CC, Williams RJ: The Messel food
web. In The World at the Time of Messel: Puzzles in
Palaeobiology, Palaeoenvironment, and the History of Early
Primates. Edited by Lehmann T, Mossbrugger V. Frankfurt-am-
Main, Germany: Schaal Senckenberg Gesellschaft fu¨r
Naturforschung; 2011:95-97.
This preliminary study is the first time that a high-resolution and high
certainty food web has been produced for any ancient ecosystem and
spotlights the crucial importance of detailed plant–insect interactional
data for the establishment of trophic links. It is the largest food web
published and employs new network modeling and analytical
approaches.
57. Anderson JM, Anderson HM, Cleal CJ: Brief History of
Gymnosperms: Classification, Biodiversity, Phytogeography and
Ecology. Pretoria, South Africa: South Africa National Biodiversity
Institute; 2008, .
Glossary of extinct, major seed-plant groups (after [57])
Bennettitales: Gymnosperms with compact, many-ovular ‘gynoecia’
composed of a honeycomb-like aggregate of ovule bearing and (in some
lineages) sterile cells. Middle Triassic–Oligocene.
Caytoniales: Gymnosperms consisting of single, bilaterally symmetrical,
reflexed cupules fully enclosing 6 to >30 ovules. Late Triassic–Early
Cretaceous.www.sciencedirect.com Cheirolepidiaceae: Conifers bearing elliptical cones with
megasporophyll units comprising large, free bracts; ovule-bearing scales
complex, with 6–10 lobes and usually two ovules enclosed in cutinized
sacs. Late Triassic–Paleocene.
Ginkgoales: Gymnosperms with short shoots bearing fascicles of leaves
and reproductive organs; ovulate strobili lax, compact, and spicate, with
many megasporophylls reduced to single or a pair of ovulate heads, each
containing one to five ovules. Early Triassic? Eocene, one species
Recent.
Glossopteridales: Gymnosperms with megasporophylls of bract/scale
complexes in which megasporangia are variously palmate, many-ovuled
structures attached to midrib of sterile bracts ranging from unmodified to
reduced glossopterid leaves. Early to Late Permian.
Medullosales: Gymnosperms with radiospermic ovules borne singly on
fronds or in loose clusters on dichotomously branched axes; vascularized
nucellus free from integument. Middle Mississippian–Early Permian.
seedfern: A mostly late Paleozoic to late Mesozoic polyphyletic
assemblage of seed plants with fern-like foliage and naked seeds.
Including the Medullosales, Glossopteridales and Caytoniales. Synonym:
pteridosperm.Current Opinion in Plant Biology 2013, 16:414–421