ArticlePDF Available

Insect herbivory close to the Oligocene–Miocene transition — A quantitative analysis

June 2010
Palaeogeography Palaeoclimatology Palaeoecology 292(3):540-550

June 2010
292(3):540-550

DOI:10.1016/j.palaeo.2010.04.029

Authors:

Hessisches Landesmuseum Darmstadt

Insects form an important part of modern terrestrial ecosystems, but while their body remains are rare in the fossil record, their trace fossils, such as feeding damage, are more common. Studies of insect herbivory on fossil leaves can provide key information for an ecological understanding of disturbance and biotic response in deep time, such as the response of insect damage frequency and diversity to changing vegetation and climate. In this contribution, I provide the first, high-resolution study documenting insect damage of fossil leaves that indicate the reaction of insect herbivores to changing regional climates and vegetation during the latest Paleogene in Europe. Insect damage censuses were conducted at six stratigraphic levels ranging in age from 27 to 23 Ma in the Siebengebirge area near Bonn, Germany. A total of 3122 fossil angiosperm leaves pertaining to 135 species were examined for the presence or absence of insect damage types (DTs); fifty-nine damage types were recorded. The most parsimonious explanation for the trends observed in this study is that the fossil insect damage represents a regional response to global environmental changes.

summaries, with total number of leaves (including monocot. and dicots.) in the collections, only dicot. leaves in the census, and percentage of damage on dicot. leaves only observed in these collections.

…

Figures - uploaded by Torsten Wappler

Content may be subject to copyright.

Content uploaded by Torsten Wappler

Content may be subject to copyright.

This article appeared in a journal published by Elsevier. The attached

copy is furnished to the author for internal non-commercial research

and education use, including for instruction at the authors 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

Author's personal copy

Insect herbivory close to the Oligocene–Miocene transition —A quantitative analysis

Torsten Wappler ⁎

Steinmann Institute for Geology, Mineralogy and Palaeontology, Division Palaeontology, Nussallee 8, D-53115 Bonn, Germany

abstractarticle info

Article history:

Received 6 November 2009

Received in revised form 20 April 2010

Accepted 25 April 2010

Available online 4 May 2010

Keywords:

Siebengebirge

Upper Oligocene

Herbivory

Plant–insect interactions

Climate change

Palaeobotany

Insects form an important part of modern terrestrial ecosystems, but while their body remains are rare in the

fossil record, their trace fossils, such as feeding damage, are more common. Studies of insect herbivory on

fossil leaves can provide key information for an ecological understanding of disturbance and biotic response

in deep time, such as the response of insect damage frequency and diversity to changing vegetation and

climate. In this contribution, I provide the ﬁrst, high-resolution study documenting insect damage of fossil

leaves that indicate the reaction of insect herbivores to changing regional climates and vegetation during the

latest Paleogene in Europe. Insect damage censuses were conducted at six stratigraphic levels ranging in age

from 27 to 23 Ma in the Siebengebirge area near Bonn, Germany. A total of 3122 fossil angiosperm leaves

pertaining to 135 species were examined for the presence or absence of insect damage types (DTs); ﬁfty-

nine damage types were recorded. The most parsimonious explanation for the trends observed in this study

is that the fossil insect damage represents a regional response to global environmental changes.

1. Introduction

Terrestrial plants and insects today account for the majority of the

Earth's biodiversity, and almost half of all insect species are herbivores

(Price, 2002). The complexity of these relationships is evident from the

sheer numbers of species of plants and insects available to enter into a

wide variety of antagonistic, neutral and mutualistic associations. The

high degree of food specialization of most herbivorous insects show

distinct and reﬁned food preferences, indicating the geologically long-

term persistence of some of these associations (Scott, 1991; Laban-

deira, 2002a,b, 2006a,b; Wilf, 2008). Palaeoecological studies of plant–

insect assemblages can provide robust evidence for palaeoclimatic

conditions, and hence potentially add major signiﬁcance to this debate

(e.g., Wilf and Labandeira, 1999; Wilf et al., 2001; Currano et al., 2008;

Wappler et al., 2009).

Fossil leaves are extremely sensitive indicators of past climates

(Wilf, 1997), and insect damage of the leaves is crucial toward

documenting the evolution of plant–insect interactions (e.g., Laban-

deira, 2006a), the development of biodiversity (e.g., Lewinsohn and

Roslin, 2008), and the effects of major environmental disturbances,

particularly climate (e.g., Labandeira, 2002a).

The Oligocene in particular is considered as a period marked by

large and abrupt climate changes, palaeogeographic changes, large

ﬂuctuations in the volume of the Antarctic Ice Sheet after the initial

formation during the earliest Oligocene, and related eustatic changes

promoted by variations in orbital frequency (e.g., Wade and Pälike,

2004; Coxall et al., 2005; Pälike et al., 2006). According to Mosbrugger

et al. (2005), cooling was especially noticeable during winter, with

marked thermal seasonality apparent at certain times during the

Oligocene. Palaeoclimatic indicators suggest ice formation during this

interval that led to a global sea level drop of nearly 70 m (Pekar et al.,

2002; Ivany et al., 2006). Major changes in upper Oligocene terrestrial

ecosystems, particularly expressed in plant physiological adaptations,

are thought to be inﬂuenced by long-term pCO

ﬂuctuations (Fletcher

et al., 2008; Pagani et al., 2009). Increased atmospheric levels of the

greenhouse gas CO

have been linked to the Paleocene–Eocene

Thermal Maximum (PETM) and the subsequent Eocene warm interval,

although Eocene pCO

has been estimated in some palaeobotanical

studies as occurring at levels comparable to today (e.g., Zachos et al.,

2001, 2008; Nicolo et al., 2007; Sluijs et al., 2009). Follow-up studies

suggest that increased insect herbivory is likely to be a net long-term

effect of increasing temperature and pCO

(e.g., Wilf and Labandeira,

1999; Wilf et al., 2001; Currano et al., 2008).

Hence, upper Oligocene environmental change is important for

understanding the context for the development and response of plant–

insect associations to climate change and, in particular, to the recovery

of associations after the pronounced decrease of temperatures during

most of the Oligocene (Utescher et al., 2000; Mosbrugger et al., 2005;

Kürschner and Kvaček, 2009). A well-preserved upper Oligocene plant

and insect fossil assemblage occurs in the Siebengebirge area, and

consists of several stratigraphically related sites that traditionally have

been designated as ‘Rott’.‘Rott’has been studied for approximately

75 years and provided the most abundant and diverse insect fauna of

this time interval in Europe. Nearly 250 plant species are known, based

on mega- and mesofossils (Weyland, 1937, 1938, 1941, 1943, 1948;

Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

⁎Tel.: +49 228734682; fax: + 49 228733509.

E-mail address: twappler@uni-bonn.de.

doi:10.1016/j.palaeo.2010.04.029

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

journal homepage: www.elsevier.com/locate/palaeo

Author's personal copy

Winterscheid, 2006). However, plant–insect interactions, a dominant

feature of terrestrial ecosystems, are not well-known from this interval

in Europe. By contrast, studies in the Western Interior of the U.S.A. for

the Paleocene and Eocene have identiﬁed important temporal trends

in insect damage of dicotyledon leaves. There is a strong correlation

between insect damage diversity and high temperature during the late

Paleocene and early Eocene (e.g., Wilf and Labandeira, 1999; Wilf et al.,

2001; Currano et al., 2008). Moreover, insect damage is very frequent

during the PETM (Currano et al., 2008).

The present study aims to provide data on the presence of insect

herbivory at six sites in the Rott deposit that span the latest Oligocene

(Magnetic Polarity Chron (MP) 26–30) of the Siebengebirge area

southeast of Bonn, Germany (Table 2;Fig. 1). In this study, temporal

trends in herbivory will be described, and investigated to ascertain if

these trends correlate with external variables, such as ﬂoral diversity

and temperature (e.g., Pearson and Palmer, 2000; Utescher et al., 2000;

Mosbrugger et al., 2005; Winterscheid, 2006; Uhl et al., 2007).

2. Geological setting

The geologic setting, stratigraphy, depositional environment, and

age of the Upper Oligocene of the Siebengebirge area is reviewed in

detail elsewhere (Mörs, 1995; Gee and Mörs, 2001; Winterscheid,

2006) and is brieﬂy summarized here. The Lower Rhine Basin is an SW

extension of the North Sea Basin bounded by pre-Paleogene, mostly

Palaeozoic rocks. Sedimentation in the Lower Rhine Basin started

during the Early Oligocene in the northwestern part, with a marine

succession of sand and clay (Zagwijn, 1989). During the Early to Upper

Oligocene, a marine transgression spread to the southeast, where

marine sediments are interspersed with ﬂuvial deltaic deposits, lignite

seams and volcanic tephra from the nearby Siebengebirge volcanic

ﬁeld. The studied area lies at the northeastern boundary of the

Siebengebirge area east of the Rhine River, approximately 12 km east

of Bonn, and includes the classic fossil deposits of Rott, Orsberg,

Stöβchen and Quegstein, that are famous for their rich ﬂora, as well as

their insect and vertebrate fauna (von Koenigswald, 1996). During the

Oligocene, clastic terrestrial sedimentary sequences were deposited as

Fig. 1. Compilation of the biostratigraphic zonation for the latest Oligocene follows Luterbacher et al. (2004), temperature trends for marine temperatures through the studied

interval, calculated from Zachos et al. (2001). Floral assemblages are as deﬁned by Winterscheid (2006). (A) Estimates of annual palaeotemperatures through the studied interval

with (B) proportions of leaves damaged at each site (black circles), and the pooled sample for each assemblage (black squares), with error bars representing one standard deviation

above and below the mean of the resamples. (C) Total damage diversity on each ﬂora standardized to 50 leaves. (D) Specialized damage diversity stardardized as in (C). (E–F) Diversity

on mine and galling morphotypes standardized as in (C).

Table 1

Collection summaries, with total number of leaves (including monocot. and dicots.) in

the collections, only dicot. leaves in the census, and percentage of damage on dicot.

leaves only observed in these collections.

Collections Leaves in

collection

Dicot. leaves

in census

% of leaves

damaged

Statz Collection 2067 1222 24.79

Kastenholz Collection 249 136 18.38

GPI Collection 1151 828 10.62

Winterscheid Collection 1132 936 17.73

541T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

lacustrine and ﬂuviatile clays, sands, and gravels that dominated the

sedimentological regime. In addition to the traditional stratigraphy

(‘Liegendschichten’, Siebengebirgs tephra, ‘Hangendschichten’), the

sequences are classiﬁed in six lithofacies complexes: (a) pelite facies,

(b) siliciclastic facies of the ‘Liegendschichten’, (c) Siebengebirge

tephra, (d) pelite and lignite facies of the ‘Hangendschichten’, with

intercalated horizons of the (e) sapropelite and diatomite facies and

(f) chert facies. Some of these facies contain exceptionally well-

preserved megaﬂoral remains in certain horizons (leaves, diaspores

and inﬂorescences). Although there have been subsequent reﬁne-

ments of the biostratigraphy (Heizmann and Mörs, 1994; Mörs, 1995),

I follow Winterscheid (2006) in retaining a more simpliﬁed ﬂoral,

assemblage-based biostratigraphy (Fig. 1). The oldest zone, charac-

terized by a prevolcanic–ﬂuviatile ﬂoral assemblage (1a), is exposed in

two localities (Allrott and Quegstein) situated at the margin of the

Lower Rhine Basin. The age of the prevolcanic assemblage is poorly

constrained. Conventional K–Ar datings of sanidines from Siebenge-

birge trachytes (zone 1b, Offenkaule exposure), that overlies directly

zone 1a, yield ages between 25.2±0.6 and 26.4 ± 0.8 Ma (Todt and

Lippolt, 1980; Mertz et al., 2007, p. 357), indicating a discrete time

mark for the termination of the prevolcanic–ﬂuviatile ﬂoral assem-

blage. The estimated mean annual palaeotemperature (MAT) for this

time and site, based on leaf-margin analysis, is 16.8 ± 1.2 °C (Win-

terscheid, 2006). Zone 2, a postvolcanic–ﬂuviatile–palustrine ﬂoral

assemblage (2), constitutes the remainder of the Siebengebirge area

exposures. Zone 2 spans most of the Upper Oligocene, from ∼26.5 Ma

to probably the Early Miocene boundary at ∼23.0 Ma (Gradstein et al.,

2004, p. 412–413). Zone 3, the postvolcanic–lacustrine ﬂoral assem-

blage (3), represented by three exposures (Orsberg, Rott, and

Stöβchen), consists of ﬂoras from lake deposits, interpreted as volcanic

crater ﬁllings (‘volcanic crater assemblage’). The estimated MAT at this

time and place, based on leaf-margin analysis, is 17.5± 1.2 °C

(Utescher et al., 2000; Winterscheid, 2006). A fauna from these

sediments consists of 20 small mammal species including didelphids,

talpids, dimylids, soricids, chiropterans, castorids, glirids, eomyids,

and cricetids (Mörs, 1995). This mammal assemblage provides for

biostratigraphic precision and places the deposits in MP 30, most

probably in the upper part of this chron. This indicates a depositional

age of latest Oligocene (Chattian) for the sediments at the southern

margin of the basin. These three ﬂoral assemblages obtained in this

study are mapped and described in Fig. 1 and Tables 4–5.

3. Material and methods

3.1. Assessment of herbivory

The fossils examined in the present work are from a number of

collections (Table 1). Fossil leaves from the upper Oligocene of Rott

are housed in the Steinmann Institute, University of Bonn (repre-

sented by the the Statz Collection, Kastenholz Collection, the

Geological–Palaeontological-Institute Collection ([GPI] at Bonn), and

in the Collection of Heinz Winterscheid, Cologne, all in Germany).

There are smaller collections of fossils from Rott in the Stadtmuseum

Siegburg, Siegburg, in the Siebengebirgsmuseum Königswinter,

Königswinter, and the Geological Museum, University of Cologne,

Germany. By far the most important collection of fossils from Rott was

accumulated between 1930 and 1940 by private collectors, in

particular Georg Statz and Anton Kastenholz. Heinz Winterscheid

and Meinolf Hellmund also collected valuable material over the last

30 years. In 2004 the Goldfuss Museum (Steinmann Institute,

University Bonn) received more than 200 fossil plant specimens

being part of the Statz Collection of the Los Angeles County Museum of

Natural History (LACM). A few months later the Steinmann Institute

acquired the Kastenholz Collection consisting of 978 specimens,

including 248 plant specimens. These two important collections are

now housed in the same repository, complementing the existing Rott

collection. These collections have become historically and palaeonto-

logically more valuable with time, as the old mine dumps at Rott

have become largely inaccessible due to dense forest cover and are

protected from private collecting by state law. If possible, the

collectors of the leaf megaﬂoras were asked to provide information

regarding collecting techniques, to identify potential biases in

collection methods that could lead to an under-representation of

damage levels.

In the present study I compare plant–insect interactions of six

upper Oligocene megaﬂoral sites, from the top of Chron C9 (Chattian)

to the top of Chron C6C (late Chattian) and covering about 3.5 m.y. of

deposition and erosion (Table 2;Fig. 1). The dataset includes 3122

leaves. Fossil leaves and their insect damage were quantitatively and

qualitatively analysed for each site. Every morphologically identiﬁ-

able, non-monocotyledon angiosperm leaf (or leaﬂet in case of

compound leaves) of which more than half of the blade was intact

was scored for the presence or absence of 59 insect-feeding

Table 2

Summary of each quarry at the different designated ﬂoral assemblages (1a: postvolcanic/lacustrine; 1b: volcanic ash layer; 3: prevolcanic/ﬂuviatile). Errors in damage diversity are

given as one standard deviation above and below the mean of resamples, and errors on the percent of leaves damaged represent the binominal sampling error.

Quarry Floral assemblage

Coordinates # dicot leaves # dicot leaf species # damage types Damage diversity, 50 leaves % of leaves damaged

Stößchen 1a 50°36′N/7°12′E 40 18 9 n/a 22.5±6.6

Rott 1a 50°45′N/7°16′E 2476 112 55 7.5 ±2.1 19.7 ±0.8

Orsberg 1a 50°35′N/7°14′E 123 34 9 5.6 ± 1.2 17.9± 3.5

Offenkaule 1b 50°45′N/7°16′E 16 5 3 n/a 12.5±8.3

Allrott 3 50°41′N/7°13′E 62 9 6 5.2 ± 0.9 14.5± 4.5

Quegstein 3 50°40′N/7°12 E 404 36 13 4.5 ± 1.6 12.6± 1.7

Floral assemblages are on based on Winterscheid (2006).

Fig. 2. Representative insect external foliage-feeding of the Siebengebebirge ﬂora. (A) Midvein-associated hole feeding of Tremophyllum tenerrimum (Ulmaceae) (Ro-11408, DT2).

(B) Continuous cuspate margin feeding consisting of three or more adjacent cuspules (arrows) separated by a minimal ﬂap of leaf margin of T. tenerrimum (Ro-11426, DT143).

(C) Cuspate margin feeding and skeletonization of T. tenerrimum (Ro-11461, DT12, 16). (D) Primary-vein associated skeletonization of T. tenerrimum (Ro-11405, DT16). (E) Extensive

margin feeding (arrows) of T. tenerrimum (Ro-11483, DT12, 14). (F) Midvein-associated hole feeding (arrows) of Alnus menzelii (Betulaceae) (Ro-10552, DT50). (G) Cuspate

excavation along the leaf margin extending to the midrib of Salix arcinervea (Salicaceae) (Ro-10133, DT14). (H) Detail of skelezonization in D, displaying venational network.

(I) Cuspate margin feeding on leaf fragment of Pistacia septimontana (Anacardiaceae) (Ro-11637, DT14). (J) Damage by an adult leafcutter bee (Megachilidae) of Salix arcinervea

(Salicaceae) (Ro-10130, DT81). (K) Removal of interveinal tissue; reaction rim well developed of Fabales indet. —forma 1 (Ro-10746, DT17). (L) Pattern of hole feeding in intercostal

areas on unidentiﬁed dicot (Fabales) (Ro-10647, DT78). Solid scale bars =5 mm; striped scale bars= 1 mm.

542 T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

543T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

morphotypes (Labandeira et al., 2007)(Figs. 2, 3). The complete

dataset is presented in Appendix A. These damage types (DTs) can be

classiﬁed into six functional feeding groups: hole feeding (HF),

margin feeding (MF), skeletonization (S), surface feeding (SF),

galling (G), and mining (M), as described by (Labandeira et al.,

2002, 2007). MAT values for stratigraphic levels and sites are from

Utescher et al. (2000) and Winterscheid (2006). Age calibrations for

the sites were obtained with various methods, mostly biostratigra-

phy and radiometric ages, and are well constrained, at least for the

uppermost part of the sequence (Todt and Lippolt, 1980; Mörs, 1995;

Winterscheid, 2006).

3.2. Quantitative analyses of insect damage

All analyses were done using the vegan package implemented

in the R statistics environment (R Development Core Team, 2009).

The sample size for each ﬂora was standardized by selecting a random

subset of 50 leaves without replacement and calculating the damage

diversity for the subsample. This process was repeated 5000 times,

and the results were averaged to obtain the standardized damage

diversity for each ﬂora. The standard deviations (SD) for the

resampleswerecalculated.Thesameprocedurewasusedto

standardize insect diversity to 25 leaves on each of the 23 species-

site pairs with at least 20 specimens.

3.3. Estimation of leaf mass per area (LMA)

Every fossil leaf at Rott that clearly shows the attachment of an

unsplit petiole (petiolule) to the blade and had a reconstructable leaf

area was used in a leaf mass per area (LMA) analysis. The LMA

analysis uses the robust global correlation in extant leaves of petiole

width squared and leaf mass, both normalized for leaf area, to

estimate leaf mass per area in fossils (Royer et al., 2007). For the

“Rott”ﬂora I used only specimens for which both margins of the

petiole are preserved and for which leaf area could be reasonably

estimated either by digital reconstruction and measurement of

photos or by multiplying 2⁎le af length ⁎leaf width/3 (Cain and

Castro, 1959). 231 leaves, representing 21 species-site pairs, ﬁtted

these criteria. Measurements were made using Image J (http://rsb.

info.nih.gov/ij), and LMA values were calculated (Tables 4 and 5)

using the protocol of Royer et al. (2007).

4. Results

The proportions of leaves at each locality with damage, specialized

damage, mines, and galls are shown in Fig. 1 (see also electronic

supplement material in Appendix A), and the percentage of leaves at

each locality with each functional feeding group is shown in Fig. 4.

These data indicate a trend toward overall higher damage frequency

in the postvolcanic assemblages when compared to the prevolcanic

assemblages, with a signiﬁcant difference between the mean

damage frequency between the prevolcanic and the postvolcanic

assemblages (t=4.51, df=2.87, p=0.02). As shown in Fig. 4,hole

and margin feeding traces were most common. The elevated

marginal damage at Allrott and Stöβchen is attributable to a higher

frequency of common damage types (DT12, DT14, and other general

margin feeding). Both DT12 and DT14 are found on 32.1% of

Sideroxylon salicites at Allrott and on 40% of Dicot. sp. (Type 1) at

Stöβchen. Surface feeding is the next most common type of damage,

with skeletonization and galling occurring thereafter in similar

abundances. Leaf mines were the rarest damage types, and most

common at Allrott, where 3.2% of the leaves are mined, followed by

Rott (1.2%) and Orsberg (0.8%). At Allrott all mines occur on Dicot.

sp. (Type 1) and either were assigned to DT65, resembling the

serpentine mines of modern Lepidoptera, or to DT35, consisting of a

rounded blotch mine with a central chamber and coprolites. Galling

occurs most commonly at the Rott and Orsberg localities, with

frequencies of 1.9% and 1.6%, respectively, where barely 0.2% of the

Quegstein leaves exhibited the activity of gall-inducing insects. The

number of leaf specimens at the Offenkaule localilty is too low to be

statistically signiﬁcant, as is comparison of surface feeding data at

Offenkaule. Nevertheless, the percentage of leaves with specialized

damage is signiﬁcantly higher for the postvolcanic assemblages than

for the prevolcanic assemblage (t=−11.63, df =2.81, p= 0.001),

the former show a higher percentage of surface feeding, galling,

and leaf mining. Twenty-four host genera appear to be abundant

(N25 specimens), accounting for 54% of the leaves in the census.

The host plants possess wide-ranging susceptabilities to herbivory

(Fig. 5). The maximum percentages of herbivory are found on

two deciduous species from the the postvolcanic assemblage at Rott,

Acer integrilobum (47.8%) and Tremophyllum tenerrimum (46.7%).

For the prevolcanic assemblages, high herbivory percentages are

noted for Sideroxylon salicites (32.1%). Deciduous host plants show

signiﬁcantly higher damage percentages than evergreens (F

[1,18]

4.41, p=0.05), and the frequence of herbivory in bulk samples

increases from the prevolcanic to the postvolcanic assemblages.

This rise is attributed to increasing temperature, especially since

there was a signiﬁcant warming between 25.5 and 23.0 Ma (see

Fig. 1)(Zachos et al., 2001; Mosbrugger et al., 2005; Pälike et al.,

2006; Kürschner et al., 2008). Furthermore, differences in diversity

and composition between the prevolcanic to the postvolcanic assem-

blages were also due to differing ﬂoral composition (Winterscheid,

2006).

Quantitative continental climate data, such as mean annual

temperature (MAT), mean annual precipitation (MAP), coldest and

warmest month means (CMM, WMM respectively), have been

reconstructed from European megaﬂoras by the coexistence approach

(Utescher et al., 2000; Mosbrugger et al., 2005). The upper Oligocene

warming is characterized by signiﬁcant increases in MAT by about

3 °C. The warming trends are even more pronounced (up to about

5 °C) in the coldest month mean (CMM) (e.g., Kürschner and Kvaček,

2009). The marked thermal seasonality (e.g., Mosbrugger et al., 2005;

Kürschner and Kvaček, 2009) is associated with an inﬂux of thick- and

small-leaved host plants during the latest Oligocene, whereas the

percentage of nanophyllous leaves is four times higher in the

postvolcanic assemblage (20.6%), compared to 5.4% for the prevolca-

nic assemblage. However, proportions of notophyllous and larger

leaves decrease during the observed interval (Winterscheid, 2006,

Fig. 3. Representative insect internal foliage-feeding of the Siebengebebirge ﬂora. Structurally similar, undiagnostic, circular to ellipsoidal galls occurring on primary and secondary

veins on (A) Tremophyllum tenerrimum (Ulmaceae) (Ro-11401, DT32). (B) Circular gall; surrounded by thick, dense, reaction tissue on Tetrastigmophyllum rottense (Vitaceae) (Ro-

11886, DT11), (C) Structurally similar, undiagnostic, circular to ellipsoidal galls occurring on primary and secondary veins on Carya serraefolia (Juglandaceae) (Ro-1.12, DT32), and

(D) T. tenerrimum (Ulmaceae) (Ro-11464, DT32). (E) Ellipsoid gall, attached to 2° vein; thick, woody outer rim (arrow) of radiate fusain on Daphnogene cinnamomifolia (Lauraceae)

(Ro-10930, DT120). (F) Undiagnostic gall on and secondary vein on Acer tricuspidatum (Sapindaceae) (Ro-10431, DT34). (G) Detail of galling structure in F. (H). Detail of galling

structure in B. (I) Initially threadlike, tortuous mine; undulatory trail of packed frass, smooth margins and width increases on P. septimontana (Anacardiaceae) (Ro-104.1, DT41).

(J) Leaf mine of relatively short length and with a linear margin, consisting of a solid frass trail occupying the entire mined area on T. tenerrimum (Ulmaceae) (Ro-11429, DT43).

(K) Odonatan endophytic oviposition scars on Laurophyllum presudoprinceps (Lauraceae) (Ro-10982), representing ichospecies Paleoovoidius arcuatum (vide Sarzetti et al., 2009).

(L) Enlargement of leaf showing distribution of scars over the lamina in K. (M) Serpentine mine with thin reaction rim, frass absent on Carya serraefolia (Juglandaceae) (Ro-10843,

DT94). (N) Elongate ellipsoidal blotch mine with internal frass-laden serpentine phase on Salix rottensis (Salicaceae) (Ro-11533, DT37). (O) Close-up of mine in N, displaying dark

frass trail. Solid scale bars= 5 mm; striped scale bars = 1 mm.

544 T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

545T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

p. 263). On the other hand the numbers of observed gall occurences

differ signiﬁcantly between both assemblages (χ

=44.08, df =1,

p≪0.001). Considering the correlation of leaf size and precipitation

in living vegetation (e.g., Wilf et al., 1998; Jacobs and Herendeen,

2004), the maximum in galling in the postvolcanic assemblage

(Table 3) is consistent with studies that indicate higher diversity in

drier habitats (e.g., Cuevas-Reyes et al., 2004, 2006). Thus, small leaf

size, with the possibility of rapid leaf expansion, may have been an

important defense strategy for most of the host plants (e.g., Moles and

Westoby, 2000).

Damage diversity is shown for the bulk ﬂoras as well as the pre-

and postvolcanic assemblages in Fig. 6. For the postvolcanic as-

semblage the sites are split into three groups based on their damage

diversity. At Stöβchen the diversity is higher than the pooled

sample, whereas at Orsberg the diversity is lower than the pooled

sample. Rott falls within the the error bars of the pooled post-

volcanic sites. For the prevolcanic assemblage all sites fall within the

error bars of the respective pooled samples. Like the frequency data,

the bootstrap curves for the quarries indicate greater damage

diversity in the postvolcanic assemblages, a signiﬁcant trend that

is particulaly striking in a plot that combines all the sites (Fig. 6C). A

t-test shows that the postvolcanic species assemblage has a higher

mean damage diversity than the prevolcanic assemblage (df = 18,

p≪0.001).

The differences in damage diversity between the sites and as-

semblages can probably be explained by looking at their ﬂoral

compositions and speciﬁc leaf traits (Tables 4 and 5). Higher leaf

mass per area (LMA), which is associated with thicker, tougher, better

defended, lower-nutrient, longer-lived leaves, is negatively correlated

with herbivory (Coley and Barone, 1996; Wright et al., 2004; Royer

et al., 2007). The prevolcanic assemblage, representing a lowland

riparian forest, is dominated by Fagaceae (Eotrigonobalanus furcinervis,

Trigonobalanopsis rhamnoides) and Lauraceae (Daphnogene cinnamo-

mifolia); in contrast, the postvolcanic assemblages are interpreted as

mixed mesophytic forests. However, diversities of elements from

riparian habitats were still high in the postvolcanic assemblages.

Within this assemblage toothed leaves are likely to overrepresent in

the regional vegetation and therefore underestimate MAT values

(Burnham et al., 2001; Royer et al., 2009). Sites allocated to this

vegetational type were clearly azonally arranged, with habitats

ranging from volcanic mountane slopes to the margins of crater

lakes. Consequently, I tested whether leaves from these localities had

signiﬁcant structural differences that would make them more

palatable to herbivores. Interestingly, there were no site-level dif-

ferences in LMA and damage frequency across the study interval.

An ANOVA of LMA yielded an F-value of 9.61 and p= 0.1776 (5 degrees

of freedom), indicating no signiﬁcant differences in leaf properties

between sites. Nevertheless, there is a signiﬁcant decrease in damage

at high M

[1,40]

=4.08, pb0.001) (Fig. 7,Tables 4 and 5), as

demonstrated by Royer et al. (2007). The majority of the plant species

have leaves with low M

that would be palatable to herbivorous

insects, attributable to the presence or absence of chemical or physical

defenses in the leaves.

Fig. 8 shows dicot leaf diversity rareﬁed to 50 leaves versus

damage diversity and damage frequency. In all cases there is a positive

relationship between plant diversity and insect damage, but only

ﬂoral diversity vs. gall frequency has a p-value less than 0.05 (Fig. 8D).

Fig. 4. Percent of leaves with each functional feeding group (FFG). Hole feeding (HF),

marginfeeding (MF),skeletonization(S), surface feeding (SF), galling(G), and mining(M).

Fig. 5. Percentage of leaves exhibiting feeding traces from the most abundant genera

(N25 specimens). (A) Postvolcanic assemblage, (B) Prevolcanic assemblage. Dashed

lines indicating mean of damage from the pooled assemblage; grey bars representing

genera where damage frequency is greater than the ﬂora mean.

Table 3

Galling data for the observed interval. The damage ratio is the sum of the numbers of

gall damage types found on each host species (number of damage types, host basis),

devided by the number of host species. Number of host species in the two successive

assemblages: 41, 119. Singletons are host species represented by only one specimen,

which are unevenly distributed among the four localities in two assemblages. Number

of nonsingletons hosts: 25, 97.

Galls

Prevolcanic

assemblage

Postvolcanic

assemblage

Damage types 1 10

Leaves damaged 1 47

Leaves damaged, % 0.215 1.780

Host species damaged, % 2.4 8.4

Host species damaged, no singletons, % 4.0 10.3

Damage types, host basis 1 35

Damage ratio 0.02 0.294

Damage types, host basis, no singletons 1 29

Damage ratio, no singletons 0.04 0.298

546 T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

There are signiﬁcant positive correlations between MAT and insect

herbivory; however, none have very high R

values. In general,

damage increases, as temperature increases throughout the Upper

Oligocene. However, abiotic factors, such as soil quality and water

stress, and biotic factors (e.g., life-form of host plant, plant age, plant

density and natural enemies) may also affect herbivorous insects

species richness at different scales (Fernandes and Price, 1991; Lara

and Fernandes 1996; Ribeiro et al., 1998), which, on the other hand,

could hide the correlation between temperature and herbivory.

5. Discussion

The data are highly variable for damage frequency analyses on

leaves or leaﬂets, (Lowman, 1995; Currano, 2009). This variability

strongly depends not only on the type of ﬂora and variable

taphonomic conditions (see also discussion in Wappler et al., 2009).

Previous studies indicate that the depositional environment has an

effect on apparent ﬂoral diversity (Wing and DiMichele, 1995). The

fossils used in the present study can be divided into two general

groups reﬂecting their taphonomic background, a prevolcanic–

ﬂuviatile ﬂoral assemblage vs. postvolcanic–lacustrine ﬂoral assem-

blage, whereas the lacustrine assemblages represent signiﬁcantly

more temporal averaging than the ﬂuviatile assemblages. The latter

possibly contributes also to greater beta diversity of damage. There is

a general consensus in ecology that abundance and density of insects

also had linked effects on insect herbivory over an extended period of

time (e.g., Hawkins and Porter, 2003). One further expectation is

that climate change apparently had signiﬁcant effects on herbivory

(e.g., Wilf et al., 2001; Currano et al., 2008; Wappler et al., 2009).

Variations in extant insect herbivory along climate gradients have

been observed, and greater herbivory generally occurs in the tropics

than in temperate regions (Coley and Barone, 1996) and greater

galling is apparently associated with subtropical aridity (Price et al.,

1998; Cuevas-Reyes et al., 2004, 2006). The entire Siebengebirge

dataset can be used to test whether the correlation between insect-

feeding diversity and frequency and external variables is true at a

larger geographical scale, or alternatively requires an additional

triggering mechanism for explaining the increase of plant diversity

and herbivorous insect activity between 27 and 23 Ma in the

Siebengebirge area.

Insect damage diversity and frequency on both bulk ﬂoras and

individual species are greater in postvolcanic than in prevolcanic

assemblages. There are also signiﬁcant differences in the relative

abundance of the functional feeding groups. In particular, specialized

damage like galling and some surface feeding is more diverse and

abundant in the postvolcanic assemblages. These results can be

expected because the plants from the postvolcanic assemblages grew

in a warmer climate than those from the prevolvcanic assemblages

and CO

-induced global warming was likely an important contribu-

tory factor to increasing plant consumption during the latest

Oligocene (e.g., Currano et al., 2008). It is commonly accepted that

herbivorous insect performance depends on plant nutrient supply

(Harvey et al., 2003; Wittstock et al., 2004; Johnson et al., 2006). In

general, effects of increased pCO

alone are to decrease foliar nitrogen

(N) concentration and raise the carbon:nitrogen (C:N) ratio, forcing

insects to eat more (e.g., Lau and Tifﬁn, 2009). In addition,

temperature ﬂuctuations also impact plant diversity. Several studies

show strong positive correlations between the diversity of plants and

Fig. 6. Resampling curves of insect damage diversity (DTs) for the postvolcanic

assemblages (A), the prevolcanic assemblages (B), and both assemblages on the same

plot (C). The pooled assemblage (dotted line) resamples have been truncated to 300

leaves. Error bars in (A) and (B) show 95% conﬁdence intervals (1.96 SE) for the pooled

assemblages.

Table 4

Summary statistics for the common species (N25 specimens) for the prevolcanic/ﬂuviatile ﬂoral community. Leaf mass per area was reconstructed using the petiole width analysis

(Royer et al., 2007), and errors represent the 95% prediction intervals.

Plant species % of leaves at each quarry Damage diversity

at 25 leaves

% of leaves

with damage

(g/m

)

Allrott Quegstein

Sideroxylon salicites 45.2 n/a 1.9 ± 0.05 32.1± 8.8 n/a

Daphnogene cinnamomifolia forma lanceolata 19.4 4.2 0.9 ±0.04 10.3± 5.6 n/a

Eotrigonobalanus furcinervis n/a 19.1 4.3 ±1.4 27.3 ± 5.1 52 (+117, −23)

Trigonobalanopsis rhamnoides n/a 29.5 3.8 ± 1.5 18.5± 3.5 79 (+119, −53)

547T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

that of herbivorous insects (Knops et al., 1999; Haddad et al., 2001;

Price, 2002; Asteraki et al., 2004; Procheşand Cowling, 2006;

Lewinsohn and Roslin, 2008), whereas other studies are much less

conclusive (Southwood et al., 1979; Prendergast et al., 1993; Burel

et al., 1998; Siemann, 1998; Kruess and Tscharntke, 2002). Overall,

greater plant diversity provides a greater diversity of resources for

polyphagous insects (Fernandes and Price, 1991; Lara and Fernandes

1996; Ribeiro et al., 1998).

Although the correlation between plant diversity and plant–insect

herbivore interactions cannot be conclusively determined from this

study, these ﬁndings suggest that climate change may have had an

important inﬂuence on plant–insect herbivore interactions, in

particular affecting highly specialized groups of herbivorous insects.

Given the data, it would be inappropriate to extrapolate the

vegetation dynamics and the responses of herbivorous insects from

one locality or region to the global in general. However, global

hypotheses for variations in insect herbivory should predict local

patterns (Wilf et al., 2001, p. 6225). Therefore, the most parsimonious

explanation for the trends observed for the Siebengebirge area is that

the fossil insect damage represents a regional response to global

environmental changes.

Acknowledgements

J. Rust (Bonn) and C.C. Labandeira (Washington, D.C.) were

extremely helpful in providing valuable feedback and advice at

many stages of the project. I thank both also for their comments and

suggestions on earlier drafts of this manuscript. I am grateful to the

anonymous reviewers for their helpful comments that signiﬁcantly

improved this manuscript and to Christen Don Shelton (Bonn) and

P. Kershaw (Melbourne) for correcting and improving the English.

I thank H. Schröder (Bonn) for support during data collection.

Furthermore, I am grateful to Heinz Winterscheid (Cologne) for

access to his Rott collection, and willing to share his knowledge about

the Siebengebirgs-ﬂora. This research was supported by funds from

the German Science Foundation (WA-1492/3-1).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in

the online version, at doi:10.1016/j.palaeo.2010.04.029.

References

Asteraki, E.J., Hart, B.J., Ings, T.C., Manley, W.J., 2004. Factors inﬂuencing the plant and

invertebrate diversity of arable ﬁeld margins. Agric. Ecosyst. Environ. 102, 219–231.

Fig. 7. Leaf mass per area (LMA) vs. insect damage frequency. Estimated M

, using the

method of Royer et al. (2007), and damage frequency for individual plant species from

the pre- and postvolcanic assemblages. All species with at least 25 leaves in the

assemblages and minimally two specimens with a reconstructable area and petiole

preserved are included in the analysis. R

and pvalues are given on the graph. For

the prevolcanic assemblage (diamonds), Taxa: Eot =Eotrigonobalanus furcinervis, and

Tri =Trigonobalanopsis rhamnoides. For the postvolcanic assemblage (circles), Taxa:

Eng =Engelhardia orsbergensis, Tet =Tetrastigmophyllum rottense, Ail =Ailanthus

ailanthifolia, Fab1, 2, 4 = Fabales forma1, 2, 4, Sal =Salix rottensis,Ac =Acerintegrilobum,

Lau =Laurophyllum pseudoprinceps, Car =Carpinus grandis, Daci =Daphnogene

cinnamomifolia forma cinnamomifolia,Ber =Berchemia parvifol ia, Zel =Zelkova

zelkovaefolia, Sid =Sideroxylon salicites, Myr =Myrica lignitum, Act =Acer tricuspidatum,

Dala =Daphnogene cinnamomifolia forma lanceolata, Tre =Tremophyllum tenerrimum,

and Cary =Carya serraefolia. Dashed line indicating a trend of lower LMA values for

tougher and thicker leaves.

Table 5

Summary statistics for the common species (N25 specimens) for the postvolcanic/lacustrineﬂoral community. Leaf mass per area was reconstructed using the petiole width analysis

(Royer et al., 2007), and errors represent the 95% prediction intervals.

Plant species % of leaves at each quarry Damage diversity

at 25 leaves

% of leaves

with damage

(g/m

)

Stößchen Rott Orsberg

Engelhardia orsbergensis n/a 0.8 5.7 5.7 ±0.5 21.6 ± 8.8 164 (+ 292, −92)

Tetrastigmophyllum rottense n/a 1.1 n/a 5.6 ±0.7 25.9 ± 8.8 97 (+ 218, −43)

Ailanthus ailanthifolia n/a 1.3 n/a 1.7 ±0.4 12.1 ± 5.7 73 (+ 165, −32)

Fabales forma 4 n/a 1.4 n/a 3.3 ±0.7 17.6 ± 6.6 143 (+ 207, −99)

Salix rottensis 2.5 1.4 0.8 3.3 ± 0.7 25.0 ± 7.2 72 (+116, −45)

Acer integrilobum 2.5 1.8 n/a 8.1± 1.3 47.8 ± 7.4 53 (+72, −39)

Laurophyllum pseudoprinceps 2.5 1.9 1.6 4.6 ± 1.6 20.4 ± 5.7 81 (+113, −58)

Carpinus grandis 2.5 2.1 0.8 6.6 ± 1.4 37.0 ± 6.6 75 (+94, −60)

Daphnogene cinnamomifolia forma cinnamomifolia 7.5 2.1 5.7 4.1 ± 1.4 19.4 ± 4.9 98 (+121, −79)

Berchemia parvifolia n/a 2.2 n/a 6.1 ±1.4 29.6 ± 5.6 118 (+ 148, −94)

Zelkova zelkovaefolia n/a 3.0 n/a 4.1± 1.4 24.3 ± 4.9 86 (+110, −67)

Sideroxylon salicites 5.0 3.0 21.1 5.7 ±1.8 29.1 ± 5.2 105 (+ 140, −79)

Fabales forma 2 n/a 3.2 3.3 3.9 ± 1.8 16.9 ± 4.1 107 (+140, −81)

Pistacia septimontana 5.0 3.4 0.8 3.5± 0.8 33.0 ± 5.0 n/a

Myrica lignitum n/a 3.6 0.8 6.1 ± 1.7 25.6± 5.1 117 (+ 160, −86)

Acer tricuspidatum 2.5 3.6 2.4 5.2 ± 1.7 25.5 ± 4.5 79 (+113, −55)

Fabales forma 1 2.5 4.0 8.1 3.5 ± 1.4 16.4 ± 3.5 115 (+145, −90)

Daphnogene cinnamomifolia forma lancelata 17.5 4.5 4.9 5.7± 1.6 24.7 ± 4.5 140 (+172, −113)

Tremophyllum tenerrimum n/a 4.9 n/a 6.9± 1.7 7.3± 2.3 101 (+ 119, −86)

Carya serraefolia n/a 5.1 n/a 6.9 ±1.7 40.8 ± 4.4 49 (+ 87, −27)

548 T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

Burel, F., Baudry, J., Butet, A., Clergeau, P., Delettre, Y., Le Coeur, D., Dubs, F., Morvan, N.,

Paillet, G., Petit, S., Thenail, C., Brunel, E., Lefeuvre, J.-C., 1998. Comparative

biodiversity along a gradient of agricultural landscapes. Acta Oecol. 19, 47–60.

Burnham, R.J., Pitman, N.C.A., Johnson, K.R., Wilf, P., 2001. Habitat-related error in

estimating temperatures from leaf margins in a humid tropical forest. Am. J. Bot. 88,

1096–1102.

Cain, S.A., Castro, G.M., 1959. Manual of Vegetation Analysis. Harper and Row, New

York.

Coley, P.D., Barone, J.A., 1996. Herbivory and plant defense in tropical forests. Annu. Rev.

Ecol. Syst. 27, 305–335.

Coxall, H.K., Wilson, P.A., Pälike, H., Lear, C.H., Backman, J., 2005. Rapid stepwise onset of

Antarctic glaciation and deeper calcite compensation in the Paciﬁc Ocean. Nature

433, 53–57.

Cuevas-Reyes, P., Quesada, M., Hanson, P., Dirzo, R., Oyama, K., 2004. Diversity of gall-

inducing insects in a Mexican tropical dry forest, the importance of plant species

richness, life-forms. Host plant age and plant density. J. Ecol. 92, 707–716.

Cuevas-Reyes, P., Quesada, M., Oyama, K., 2006. Abundance and leaf damage caused by

gall-inducing insects in a Mexican tropical dry forest. Biotropica 38 (1), 107–115.

Currano, E.D., 2009. Patchiness and long-term change in early Eocene insect feeding

damage. Paleobiology 35, 484–498.

Currano, E.D., Wilf, P., Wing, S.L., Labandeira, C.C., Lovelock, E.C., Royer, D.L., 2008.

Sharply increased insect herbivory during the Paleocene–Eocene Thermal

Maximum. Proc. Natl. Acad. Sci. U. S. A. 105, 1960–1964.

Fernandes, G.W., Price, P.W., 1991. Comparisons of tropical and ternperate galling

species richness, the roles of environmental harshness and plant nutrient status. In:

Price, P.W., Price, Lewinsohn, T.M., Fernandes, G.W., Benson, B.B. (Eds.), Plant–

animal Interactions, Evolutionary Ecology in Tropical and Temperate Regions.

Wiley, New York, pp. 91–115.

Fletcher, B.J., Brentnall, S.J., Anderson, C.W., Berner, R.A., Beerling, D.J., 2008.

Atmospheric carbon dioxide linked with Mesozoic and early Cenozoic climate

change. Nat. Geosc. 1 (1), 43–48.

Gee, C.T., Mörs, T., 2001. Aquatic macrophytes from the upper Oligocene fossillager-

stätte of Rott (Rhineland, Germany). Part I, seeds. Palaeontographica B259, 33–45.

Gradstein, F.M., Ogg, J.G., Smith, A.G. (Eds.), 2004. A Geologic Time Scale 2004.

Cambridge University Press, Cambridge.

Haddad, N.M., Tilman, D., Haarstad, J., Ritchie, M., Knops, J.M.H., 2001. Contrasting

effects of plant richness and composition on insect communities, a ﬁeld

experiment. Am. Nat. 158, 17–35.

Harvey, J.A., VanDam, N.M., Gols, R., 2003. Interactions over four trophic levels,

foodplant quality affects development of a hyperparasitoid as mediated through a

herbivore and its primary parasitoid. J. Anim. Ecol. 72, 520–531.

Hawkins, B.A., Porter, E.E., 2003. Does herbivory diversity depend on plant diversity?

The case of California butterﬂies. Am. Nat. 161, 40–49.

Heizmann, E.P.J., Mörs, T., 1994. Neue Wirbeltierfunde aus dem Oberoligozän der

Tongrube Kärlich und ihre Bedetung für die Tertiär-Stratigraphie des Neuwieder

Beckens (Rheinland-Pfalz, Deutschland). Neues Jb. Geol. Paläontol. Abh. 192,

17–36.

Ivany, L.C., van Simaeys, S., Domack, E.W., Samson, S.D., 2006. Evidence for an earliest

Oligocene ice sheet on the Antarctic Peninsula. Geology 34, 377–380.

Jacobs, B.F., Herendeen, P.S., 2004. Eocene dry climate and woodland vegetation in

tropical Africa reconstructed from fossil leaves from northern Tanzania. Palaeo-

geogr. Palaeoclimatol. Palaeoecol. 213, 115–123.

Johnson, S.N., Birch, A.N.E., Gregory, P.J., Murray, P.J., 2006. The ‘mother knows best’

principle, should soil insects be included in the preference-performance debate?

Ecol. Entomol. 31, 395–401.

Knops, J.M.H., Tilman, D., Haddad, N.M., Naeem, S.,Mitchell, C.E., Haarstad, J., Ritchie, M.E.,

Howe, K.M., Reich, P.B., Siemann, E., Groth, J., 1999. Effects of plant species richness

on invasiondynamics,disease outbreaks,insect abundancesand diversity.Ecol. Lett.2,

286–293.

Kruess, A., Tscharntke, T., 2002. Contrasting responses of plant and insect diversity to

variation in grazing intensity. Biol. Conserv. 106, 293–302.

Kürschner, W.M., Kvaček, Z., 2009. Oligocene–Miocene CO

ﬂuctuations, climatic and

palaeoﬂoristic trends inferred from fossil plant assemblages in central Europe. Bull.

Geosci. 84 (2), 189–202.

Kürschner, W.M., Kvaček, Z., Dilcher, D.L., 2008. The impact of Miocene atmospheric

carbon dioxide ﬂuctuations on climate and the evolution of terrestrial ecosystems.

Proc. Natl. Acad. Sci. U. S. A. 105, 449–453.

Labandeira, C.C., 2002a. The history of associations between plants and animals. In:

Herrera, C.M., Pellmyr, O. (Eds.), Plant–animal Interactions, An Evolutionary

Approach. Blackwell Science, London. 26–74, 248–261.

Labandeira, C.C., 2002b. Paleobiology of middle Eocene plant–insect associations from

the Paciﬁc Northwest. A preliminary report. Rocky Mt. Geol. 37 (1), 31–59.

Labandeira, C.C., 2006a. The four phases of plant–arthropod associations in deep time.

Geol. Acta 4 (4), 409–438.

Labandeira, C.C., 2006b. Silurian to Triassic plant and hexapod clades and their

associations, new data, a review, and interpretations. Arth. Syst. Phyl. 64 (1), 53–94.

Labandeira, C.C., Johnson, K.R., Lang, P.J., 2002. Preliminary assessment of insect

herbivory across the Cretaceous–Tertiary boundary: major extinction and mini-

mum rebound. In: Hartman, J.H., Johnson, K.R., Nichols, D.J. (Eds.), The Hell Creek

Formation of the Northern Great Plains. Geol. Soc. Am. Spec. Pap, pp. 297–327.

Boulder, Colorado.

Labandeira, C.C., Wilf, P.,Johnson, K.R., Marsh,F., 2007. Guideto Insect (and Other)Damage

Types on Compressed Plant Fossils, Version 3.0. Washington, D.C., Smithsonian

Institution. [http://paleobiology.si.edu/pdfs/insectDamageGuide3.01.pdf].

Lara, A.C.F., Fernandes, G.W., 1996. The highest diversity of galling insects, Serra do

Cipó. Braz. Biodivers. Lett. 3, 111–114.

Lau, J.A., Tifﬁn, P., 2009. Elevated carbon dioxide concentrations indirectly affect plant

ﬁtness by altering plant tolerance to herbivory. Oecologia 161 (2), 401–410.

Lewinsohn, T.M., Roslin, T., 2008. Four ways towards tropical herbivore megadiversity.

Ecol. Lett. 11 (4), 398–416.

Fig. 8. Correlation between ﬂoral diversity and insect damage at the quarries. Results from R's linear model function with ﬂoral diversity (rareﬁed to 50 leaves) as the independent

variable and damage metrics as the dependent variable. Total, specialized, mine and gall diversity are at 50 leaves.

549T. Wappler / Palaeogeography, Palaeoclimatology, Palaeoecology 292 (2010) 540–550

Author's personal copy

Lowman, M.D., 1995. Herbivory in Australian forests —a comparision of dry sclerophyll

and rainforest canopies. Proc. Linn. Soc. N. S. W. 115, 77–88.

Luterbacher, H.-P., Ali, J.R., Brinkhuis, H.,Gradstein, F.M., Hooker, J.J., Monechi,S., Ogg, J.G.,

Powell,J., Röhl, U., Sanﬁlippo, A., Schmitz,B., 2004. The Paleogeneperiod. In:Gradstein,

F.M., Ogg, J.G., Smith, A.G. (Eds.), A Geologic Time Scale. Cambridge University Press,

Cambridge, pp. 384–408.

Mertz, D.F., Renne, P.R., Wuttke, M., Mödden, C., 2007. A numerically calibrated

reference level (MP28) for the terrestrial mammal-based biozonation of the

European Upper Oligocene. Int. J. Earth Sci. 96 (2), 353–361.

Moles, A.T., Westoby, M., 2000. Do small leaves expand faster than large leaves, and do

shorter expansion times reduce herbivore damage? Oikos 90 (3), 517–524.

Mörs, T., 1995. Die Sedimentationsgeschichte der Fossillagerstätte Rott und ihre

Altereinstufung anhand neuer Säugetierfunde (Oberoligozän, Rheinland). Cour.

Forsch.-Inst. Senckenb. 187, 1–129.

Mosbrugger, V., Utescher, T., Dilcher, D.L., 2005. Cenozoic continental climatic evolution

of Central Europe. Proc. Natl. Acad. Sci. U. S. A. 102, 14964–14969.

Nicolo, M.J., Dickens, G.R., Hollis, C.J., Zachos, J.C., 2007. Multiple early Eocene

hyperthermals, their sedimentary expression on the New Zealand continental

margin and in the deep sea. Geology 35, 699–702.

Pagani, M., Caldeira, K., Berner, R., Beerling, D.J., 2009. The role of terrestrial plants in

limiting atmospheric CO

decline over the past 24 million years. Nature 460, 85–88.

Pälike, H., Norris, R.D., Herrle, J.O., Wilson, P.A., Coxall, H.K., Lear, C.H., Shackleton, N.J.,

Tripati, A.K., Wade, B.S., 2006. The heartbeat of the Oligocene climate system.

Science 314, 1894–1898.

Pearson, P.N., Palmer, M.R., 2000. Atmospheric carbon dioxide concentrations over the

past 60 million years. Nature 406, 695–699.

Pekar, S.F., Christie-Blick, N., Kominz, M.A., Miller, K.G., 2002. Calibrating eustasy to

oxygen isotopes for the early icehouse world of the Oligocene. Geology 30, 903–906.

Prendergast, J.R., Quinn, R.M., Lawton, J.H., Eversham, B.C., Gibbons, D.W., 1993. Rare

species, the coincidence of diversity hotspots and conservation strategies. Nature

365, 335–337.

Price, P.W., 2002. Resource-driven terrestrial interaction webs. Ecol. Res. 17, 241–247.

Price, P.W., Fernandes, G.W., Lara, A.C.F., Brawn, J., Barrios, H., Wright, M.G., Ribeiro, S.P.,

Rothcliff,N., 1998. Global patternsin local number of insectgalling species. J. Biogeogr.

25, 581–591.

Procheş, S., Cowling, R.M., 2006. Insect diversity in Cape fynbos and neighbouring South

African vegetation. Glob. Ecol. Biogeogr. 15, 445–451.

R Development Core Team, 2009. R: A Language and Environment for Statistical

Computing. R Foundation for Statistical Computing, Vienna, Austria.

Ribeiro, K.T., Madeira, J.A., Monteiro, R.F., 1998. Does ﬂooding favour galling insects?

Ecol. Entomol. 23, 491–494.

Royer, D.L., Sack, L., Wilf, P., Lusk, C.H., Jordan, G.J., Niinemets, U., Wright, I.J., Westoby,

M., Cariglino, B., Coley, P.D., Cutter, A.D., Johnson, K.R., Labandeira, C.C., Moles, A.T.,

Palmer, M.B., Valladares, F., 2007. Fossil leaf economics quantiﬁed, calibration,

Eocene case study, and implications. Paleobiology 33 (4), 574–589.

Royer, D.L., Kooyman, R.M., Little, S.A., Wilf, P., 2009. Ecology of leaf teeth, a multi-site

analysis from Australian subtropical rainforest. Am. J. Bot. 96, 738–750.

Sarzetti, L.C., Labandeira, C.C., Muzón, J., Wilf, P., Cúneo, N.R., Johnson, K.R., Genise, J.F.,

2009.Odonatan endophyticoviposition fromthe Eocene of Patagonia.The ichnogenus

Paleoovoidus and implications for behavioral stasis. J. Paleont. 83 (3), 431–447.

Scott, A.C., 1991. Evidence for plant–arthropod interactions in the fossil record. Geol.

Today 7, 58–61.

Siemann, E., 1998. Experimental tests of the effects of plant productivity and plant

diversity on grassland arthropod diversity. Ecology 79, 2057–2070.

Sluijs, A., Schouten, S., Donders, T.H., Schoon, P.L., Röhl, U., Reichart, G.-J., Sangiorgi, F.,

Kim, J.-H., Sinninghe Damste, J.S., Brinkhuis, H., 2009. Warm and wet conditions in

the Arctic region during Eocene Thermal Maximum 2. Nat. Geosci. 2, 777–780.

Southwood, T.R.E., Brown, V.K., Reader, P.M., 1979. The relationship of plant and insect

diversities in succession. Biol. J. Linn. Soc. 12, 327–348.

Todt, W., Lippolt, H.J., 1980. K–Ar age determination on Tertiary volcanic rocks. V.

Siebengebirge-Graben. J. Geophys. 48, 18–27.

Uhl, D., Klotz, S., Traiser, C., Thiel, C., Utescher, T., Kowalski, E., Dilcher, D.L., 2007.

Cenozoic paleotemperatures and leaf physiognomy —a European perspective.

Palaeogeogr. Palaeoclimatol. Palaeoecol. 248 (1–2), 24–31.

Utescher, T., Mosbrugger, V., Ashraf, A.R., 2000. Terrestrial climate evolution in

northwest Germany over the last 25 million years. Palaios 15 (5), 430–449.

von Koenigswald, W. (Ed.), 1996. Fossillagerstätte Rott bei Hennef am Siebengebirge.

Rheinlandia, Siegburg.

Wade, B.S., Pälike, H., 2004. Oligocene climate dynamics. Paleoceanography 19, PA2019.

doi:10.2029/2004PA001042.

Wappler, T., Currano, E.D., Wilf, P., Rust, J., Labandeira, C.C., 2009. No post-Cretaceous

ecosystem depression in european forests? Rich insect-feeding damage on diverse

middle Palaeocene plants, Menat France. Proc. R. Soc. Lond. B276, 4271–4277.

Weyland, H., 1937. Beit räge zur Kenntnis der rheinischen Tertiärﬂora II. Erste