STÖR 
Insect Diversity in the Fossil Record
Conrad C. Labandeira; J. John Sepkoski, Jr.
Science, New Series, Vol. 261, No. 5119. (Jul. 16, 1993), pp. 310-315.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819930716%293%3A261%3A5119%3C310%3AIDITFR%3E2.0.CO%3B2-%23
Science is currently published by American Association for the Advancement of Science.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aaas.html.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.
http://www.jstor.org
Wed Jan 2 16:33:20 2008
Insect Diversity in the Fossil Record
Conrad C. Labandeira and J. John Sepkoski, Jr.
Insects possess a surprisingly extensive fossil record. Compilation of the geochronologic
ranges of insect families demonstrates that their diversity exceeds that of preserved
vertebrate tetrapods through 91 percent of their evolutionary history. The great diversity
of insects was achieved not by high origination rates but rather by low extinction rates
comparable to the low rates of slowly evolving marine invertebrate groups. The great
radiation of modern insects began 245 million years ago and was not accelerated by the
expansion of angiosperms during the Cretaceous period. The basic trophic machinery
of insects was in place nearly 100 million years before angiosperms appeared in the
fossil record.
Investigations of diversity, or taxonomic
richness, in the fossil record have provided
a wealth of information about the history of
life, identifying intervals of massive radiations,
great extinctions, and relative evolutionary
quiescence. Most of these studies
have concerned taxa above the species level
(mainly genera and families), and a variety
of fossil groups have been examined, including
marine animals (I, 2), plankton
(3), terrestrial vertebrates (4, 5), plants
(6), and even Precambrian microbes (7).
However, the most diverse class on Earth
today, the Insecta, has received much less
attention (8).
The reasons for this lack of attention
include the perception that insects are rarely
preserved as fossils and the fact that
descriptions of fossil insects, especially from
the Mesozoic era, are sparse in Englishlanguage
publications. However, the fossil
record of insects is indeed extensive, but
much of the primary study has been published
in older German and more recent
Russian literature (9). In this article, we
review this literature on fossil insects and
address (i) their fossil diversity; (ii) their
rates of evolution; (iii) their diversification
with respect to the radiation of angiosperms;
and (iv) their evolution of ecological
breadth in relation to expanding
angiosperm resources.
Our database is a compilation of the
geochronologic ranges of 1263 insect families.
These data were derived from 472
literature sources comprising several compendiums
(10) that were extensively supplemented
and updated with numerous other
sources (11-15). The rank of family was
chosen for several reasons, (i) This taxonomic
level has been analyzed in other
studies of fossil diversity (J, 2, 4, 5) and
C. C. Labandeira is in the Department of Paleobiology,
National Museum of Natural History, Washington, DC
20560. J. J. Sepkosl<i, Jr., is in the Department of the
Geophysical Sciences, University of Chicago, Chicago,
IL 60637.
seems to correlate well with underlying
species diversity (16, 17). (ii) Families are
less susceptible to irregular and biased sampling
than are fossil species and genera
(18), so that an evolutionary signal is better
maintained at this level, (iii) Families of
insects, especially extant ones, are reasonably
well established through consensus
among researchers, whereas fossil species
and genera are more idiosyncratically defined
and less frequently correspond to good
phylogenetic or phenetic units (19). (iv)
Insect families individually possess discrete,
often highly stereotyped life habits, and
their morphologies directly reflect their
trophic guilds, which are informative in
diversity studies (20). We recognize that
some of the families in our database may
not correspond precisely to monophyletic
groups as defined by cladistic methodology.
However, the inclusion of paraphyletic
families in diversity studies has not been
found to detract from general evolutionary
signals in both empirical (17) and experimental
(21) investigations.
Fossil Insect Diversity
The fossil record of insects is hardly poor
when compared to other major taxa. The
1263 families recognized for fossil specimens
exceeds the approximately 825 families
described for fossil tetrapod vertebrates
(22). Figure 1 illustrates that the preserved
diversity of insect families exceeds that of
the much more intensively studied tetrapods
at all times except for a 34-millionyear
interval during the latest Devonian
period and Early Carboniferous period.
Insect fossil diversity is not restricted to
a few, easily preserved higher taxa but is
contributed by a wide range of living and
extinct orders, as illustrated in Fig. 2. All of
the 30 commonly recognized extant orders
(15, 23) are represented as fossils. All but
one of these orders have diversities in the
late Tertiary period that are approximately
proportional to their extant familial diversities
(15, 23): The linear correlation (r)
between the numbers of living and of latest
Tertiary families for the 30 orders is 0.954
(0.932 for logarithms). The one outlier is
the Lepidoptera (moths and butterflies),
which has only about half the fossil diversity
that would be expected from their
Recent families. The Lepidoptera is characterized
by relatively large individuals with
lightly sclerotized bodies that are not well
represented in Lagerstätten (extraordinary
fossil deposits) of the Tertiary, which preserve
diverse representatives of other
groups. Elimination of the Lepidoptera increases
the linear fit between extant and
latest Tertiary familial diversities such that
r = 0.985. In summary, it appears that the
most recent part of the fossil record has
captured insect families in a largely unbiased
fashion with respect to ordinal membership.
Thus, it is not unreasonable to
assume that older parts of the record have
behaved similarly.
Few fossil insects are known from the
first 60 million years of their 390-millionyear
history (24) (Fig. 1). This paucity
probably results from a lack of appropriate
terrestrial deposits that preserved insects.
Sometime during the Early Carboniferous,
before 325 million years ago, a massive
radiation began, followed by peaks in fossil
insect diversity during the Late Carboniferous
and middle Permian. The diversity
-- Insects
Tetrapods
E400
300 200 100
Geologic time (10^ years)
Fig. 1. Family-level diversity of fossil insects
through geologic time (24), plotted at the level
of stratigraphie stages and compared to the
diversity of terrestrial tetrapod vertebrate families.
We computed diversity with the rangethrough
method, which assumes that a family
was present in all time intervals between its first
and last appearance (including the Recent),
even if not directly sampled in all intervals.
Abbreviations are D, Devonian; C, Carboniferous;
P, Permian; Tr, Triassic; J, Jurassic; K,
Cretaceous; and T, Tertiary.
310 SCIENCE ˇ VOL. 261 ˇ 16 JULY 1993
minimum between these peaks probably is
artifactual, reflecting sparse collecting sites
between the various siderite concretion faunas
of the Upper Carboniferous (for example,
Mazon Creek from Illinois and Montceau-les-Mines
from France) and the lakebed
faunas from the middle Permian (for
example, Elmo from Kansas and Chekarda
from European Russia).
The subsequent Triassic minimum in
insect diversity also could be a consequence
of a dearth of Lower Triassic insect-bearing
deposits. However, we believe it is more
likely real, reflecting the terminal Permian
mass extinction, which severely affected
marine animals (25) and terrestrial tetrapods
(5). As Fig. 2 documents, 8 of the 27
orders of insects that have Paleozoic occurrences
evidently did not survive the Permian;
four of these orders (Palaeodictyoptera,
Permothemistida, Megasecoptera, and Diaphanopterodea)
constitute the Paleodictyopteroidea,
which is the only supraordinal
taxon within the Insecta ever to become
extinct. There are another three orders that
entered the Triassic with reduced diversities
and evidently became extinct during that
period. After the demise of Paleozoic
groups, about half of the orders that survive
to the Recent underwent modest to spectacular
diversifications during the Triassic
and Jurassic, continuing to the Recent in
most cases. This shift in phylogenetic pattern
of diversification is the most pronounced
event during the history of insects
and is largely attributable to the end-Permian
mass extinction.
For the Insecta as a whole (Fig. 1),
preserved familial diversity increases almost
steadily after the Permian-Triassic bottleneck
and rises sharply in the middle Tertiary.
This final jump is undoubtedly artifactual,
reflecting the spectacular insect faunas
preserved in the Baltic amber of Europe and
Florissant lake deposits of Colorado. As
older ambers and compression fossil deposits
receive more study, especially those from
the Cretaceous (26), many families first
represented in these Tertiary deposits probably
will have their ranges pushed back into
the Mesozoic. We expect that eventually
the marked mid-Tertiary rise in fossil diversity
will be largely extended into the late
Mesozoic.
Rates of Evolution
The rise in familial diversity among insects
after the Permian is remarkable not only for
its magnitude but also for its mechanism.
Origination rates of families were no higher
than during the Paleozoic, but extinction
rates were much lower. It is difficult to
measure rates of taxic evolution for fossil
insect families because many first and last
appearances are tied to extraordinary fossil
deposits and because nearly half of all
known families are still extant. These factors
make standard survivorship analysis
complicated (27). An alternative method
for assessing extinction rates is Lyellian
survivorship analysis (28). Lyellian survivorship
curves display the proportion of
taxa in any interval of geologic time that
are still alive today. Groups that undergo
rapid extinction are characterized by proportions
that rapidly decay backward
through time, whereas those with low extinction
rates decay only slowly.
We compared Lyellian survivorship
curves for insects (Fig. 3) with curves for two
other taxonomic groups, the tetrapod vertebrates
(22) and the marine bivalves (29).
Tetrapods are the only other major group of
terrestrial animals with a good fossil record,
and bivalves are traditionally considered to
have one of the lowest rates of extinction
(30). As Fig. 3 illustrates, insect families
exhibit low rates of familial turnover
throughout much of their recent history:
Tertiary insect faunas were composed almost
exclusively of living families, and
even 100 million years ago (middle Cretaceous)
, 84% of the fauna consisted of members
of extant families. Only during the
Jurassic and earlier did extinct families con-
^
S 2 i
o o ?ˇ
< 2 =
TO O
(O ca
E ffl
<" 5
üi ˇo IB
o <n
fr F Q.
. (D
ˇa ˇ^ <D
E <0
n t * t
Fig. 2. Spindle diagrams displaying diversities of fossil families within insect orders in stratigraphie
stages of the Phanerozoic. A scale bar is shown in the lower right. Abbreviations are Pz, Paleozoic
(Silurian through Permian); Mz, Mesozoic; and Cz, Cenozoic. Boxed illustrations (not to scale)
depict typical adult representatives of the more important orders. Angiospermous plants make their
fossil appearance approximately two-thirds of the way up the band for the Mesozoic (that is, just
above the "M" in Mz).
SCIENCE ˇ VOL. 261 ˇ 16 JULY 1993 311
stitute the majority of global insect diversity
(31). This pattern contrasts dramatically
with that of the tetrapods, in which fewer
than 20% of extant families are present in
faunas 100 million years ago and fewer than
10% are present at the Jurassic-Cretaceous
boundary. The more appropriate comparison
seems to be between insects and the slowly
evolving marine bivalves. From the JurassicCretaceous
boundary to the Recent, bivalves
experienced higher rates of extinction than
insects, although before the Jurassic-Cretaceous
boundary they had lower rates of
extinction than insects. Slightly more than
half of marine bivalve families from the Early
Triassic are still extant, but fewer than 10%
of insect families are. For insects, the lower
extinction rates after the end of the Jurassic
reflect in large part the transition from the
archaic, extinct orders of the Paleozoic Era,
which experienced high extinction rates, to
the more derived orders of the Mesozoic and
Cenozoic, whose families suffered only half
the extinction rates of their predecessors.
This kind of secular decline in extinction
seems to be general to both terrestrial and
marine animals (32).
The low extinction rates of insect families
during the Mesozoic and Cenozoic are
subject to more than one interpretation. It
is possible that some paleoentomologists
have used generalized taxonomic concepts
of extant families, shoehorning into extant
taxa fossil species that are only distantly
related to and morphologically distinct
from living species (10). Such practice
would result in modern insect families having
inordinately long geologic ranges.
Alternatively, the extinction resistance of
insect families may reflect only species richness
(33) and not species behavior. Indeed,
many extant families of insects have species
numbers in the thousands (34), whereas
tetrapod and bivalve families possess one to
at most hundreds of species. However,
when genera and species within insect families
are considered, they, too, exhibit remarkable
geologic persistence. For example,
the modern beetle genus Tetraphalerus
closely resembles 153-million-year-old Jurassic
fossils (35), and the modern crane fly
Fig. 3. Lyellian survivorship curves sliowing proportions
of extarnt families as a furnction of time
(24) for irnsects, terrestrial vertebrate tetrapods,
arid marine bivalves. Steeper curves indicate a
liigher extinction rate among families in the
group. However, tliese curves need not be
monotonie because radiations of sliort-lived
famiiies can temporariiy depress Lyeilian proportions.
Tine iow extinction rates of Insects,
especially relative to tetrapods, mean tliat even
with a relatively discontinuous fossil record. Insect
families can be traced from one Lagerstätte
to another, providing a reasonably continuous
record of insect diversity. Data are plotted for
stratigraphie stages, and abbreviations are the same as in Fig. 1.
genus Helius and leaf-mining moth genus
Stigmella have been identified from 93-million-year-old
and 89-million-year-old deposits,
respectively (36). Many insect specimens
of Eocene to Miocene age are easily
placed within modern genera (4, 37) and
even modern species (38). In the younger
part of the fossil insect record. Pliocene
aphids have been determined to be conspecific
with modern species (39), and Coope
(40) and Matthews (41) were able to place
almost all fossil beetle specimens from
Pliocene and Pleistocene lake deposits in
Britain and northern Canada, respectively,
into modern species on the basis of characters
of the genitalia; assignment was possible
despite profound changes in the biogeographic
distributions of descendant populations.
Tetrapods, on the other hand, experienced
major turnovers during the late
Pliocene and latest Pleistocene (42), and
few living species are more than a million
years old. Thus, the depressed extinction
rates among insect families in the fossil
record may reflect the intrinsic evolutionary
behavior of their constituent species (43).
Angiosperms and Insect Diversity
The extraordinary diversity of living insects
has been attributed by some workers to the
diversity of angiospermous plants (44),
which first appear as fossils in the Lower
Cretaceous. This notion is buttressed by
exhaustive examples of intricate ecological
coupling between the two groups, especially
with regard to insect herbivory and pollination
(45). Indeed, the great expansion of
insect families toward the Recent would
appear consistent with this proposition. But
additional evidence is needed: It must be
demonstrated that rates of diversification of
insects increased as angiosperms diversified
to dominate virtually all terrestrial plant
communities.
The null model for patterns of diversification
of radiating groups is exponential
expansion (28). Each incremental increase
in diversity augments the number of taxa
that potentially can be produced in the
next time interval, such that numbers of
100
Insects
Tetrapods
Bivalves
200 100
Geologic time (10^ years)
taxa can grow like compound interest.
Thus, any group that maintains a constant
rate of diversification will have the logarithm
of its diversity plot as a straight-line
function of geologic time (46).
As shown in Fig. 4, the familial diversity
of insects exhibits a roughly log-linear trend
from the Middle Triassic through the Early
Cretaceous. Variation around this trend
probably reflects only the uneven temporal
distribution of rich insect-bearing deposits;
for example, the spectacularly fossiliferous
lake beds at Karatau, Kazakhstan (47),
contribute a large positive outlier over the
Late Jurassic. Interestingly, the exponential
trend extrapolates to nearly the modern
diversity of insects, which we take to be 977
families (15, 23). This extrapolation might
suggest that insects have been undergoing
exponential diversification for the last 245
million years. However, we doubt this is
the case: Although the insect record is rich,
only 63% of extant insect families are represented
as fossils, and there is no reason to
believe that early Mesozoic fossil deposits
were capturing substantially larger proportions
of the insect fauna. Thus, if insects
indeed experienced exponential diversification,
there should be more extant families
than are actually observed. This result suggests
that the global insect fauna of more
recent times has been approaching saturation
such that rates of diversification have
slowed. Indeed, the deviation in preserved
diversity below the exponential trend seen
over the last 100 million years is consistent
with this inference (46). An alternative
explanation is that extinction events have
depressed insect diversity below the exponential
trend; however, other than the
terminal Permian event, no mass extinction,
including the Cretaceous-Tertiary
event, seems to have had a major effect on
insects (48).
1000
Modern diversity-
Baltic
amber
Appearance of Ascendancy of
angiosperms angiosperms
ITT
200 100
Geologic time (10^ years)
Fig. 4. insect familial diversity from the Triassic to
the Recent, plotted on semiiogarlthmic coordinates.
The dashed line Is Interpretative, Illustrating
possible exponential diversification beginning In
the Triassic and possibly continuing into the Early
Cretaceous. The temporal positions of the Baltic
amber and important events In the history of
angiosperms are labeled. Time scale and abbreviations
are the same as In Fig. 1.
312 SCIENCE VOL. 261 16 JULY 1993
ARTICLES
The more startling interpretation that
can be drawn from the data (Fig. 4) is that
the appearance and expansion of angiosperms
had no influence on insect familial
diversification. Present data on the earliest
occurrences of definitive angiosperm
pollen and macrofossils indicate that these
flowering plants may have originated during
the Hauterivian stage of the Early Cretaceous
(49). However, some cladistic analyses
and fossil identifications (50) would
place the origin as early as the Triassic,
although the absence of definitive fossil
material indicates that the group could not
have been abundant and ecologically important
before the Cretaceous. Whenever
these plants originated, the fossil data indicate
that angiosperms experienced a tremendous
radiation in all geographic regions
during the Albian and Cenomanian stages
of the middle Cretaceous (5i). However,
there is no signature of this event in the
family-level record of insects. Instead, the
data in Fig. 4 suggest that insect diversification
actually slackened as angiosperms
radiated. Even if the drop in diversity below
the exponential trend during the Cretaceous
were a result of a lack of preservation
or inadequate sampling of insect fossils
(26), the fact remains that the post-Paleozoic
radiation of insect families commenced
more than 100 million years before angiosperms
appeared in the fossil record.
Within the Insecta, orders that have radiated
strongly during the Mesozoic and Cenozoic,
such as the Coleóptera (beetles),
Diptera (true flies), and Hymenoptera
(wasps, ants, and bees), all apparently began
their expansions during the Triassic
and Jurassic (Fig. 2), long before the ascendancy
of angiosperms. Consequently, if
there were any effect on insect diversification,
angiosperms must have had an impact
at taxonomic levels below the rank of family,
where indeed many species appear highly
co-evolved with flowering plants {44,
45). Alternatively, it is possible that it was
the earlier diversification of insects that
partly propelled the rapid expansion of angiosperms
during the mid-Cretaceous {52).
Ecological Roles of Ancient Insects
Part of the presumption that angiosperms
fueled the diversification of insects involves
the inference that the diverse tissues and
organs, particularly leaves and flowers, of
these advanced land plants provided an
expanded spectrum of ecological resources
that could be exploited by various guilds of
herbivorous and pollinating insects (45).
For the fossil record, direct inferences regarding
how insects exploited plant resources
have been drawn largely from three
sources of evidence: (i) preserved host tissue
damage, (ii) insect gut contents, and
(iii) fecal pellet composition from herbivores
{53). Although this evidence is a
significant (if underappreciated) part of the
fossil record, there have been only a few
efforts in using it to assemble a comprehensive
view of the diversity of major insect
feeding guilds through time. An alternative
way to characterize how plant resources
have been exploited is to group all insects
into guilds on the basis of preservable morphological
features that function during
feeding activities. Mouthpart structure is
ideally suited for recognizing such guilds
because it is a morphological expression of
feeding type.
By means of a multivariate phenetic
analysis (34), we grouped mouthpart characters
of 1365 extant species of insects into
34 fundamental mouthpart classes (Fig.
5A), each representing a distinct adaptive
facies for manipulating, consuming, and
processing a particular food type. Each morphologically
based class can be equated
with an ecological guild, and each is composed
of a particular ensemble of taxonomic
clades. The mouthpart classes vary considerably
in constituent species diversity, and
most cross phylogenetic boundaries, encompassing
lineages that independently acquired
the same mouthpart type and thus
gained entrance into the same feeding
guild. Therefore, the mouthpart classes reflect
ecological disparity and not necessarily
taxonomic diversity.
The geochronological distribution of
mouthpart classes represented among living
insects (Fig. 5A) and their diversity
through time (Fig. 5B) indicate that the
number of feeding guilds among insects
expanded well before the appearance of
fossil angiosperms. By the Middle Jurassic,
65% (low estimate) to 88% (high estimate)
of all modern insect mouthpart classes were
present (54), including several (for example,
rhynchophorate, haustoriate, and laFig.
5. The history of
mouthpart classes represented
among living
insects [updated from
(34)]. (A) Geochronologic
ranges of mouthpart
classes. The solid
pattern indicates direct
evidence for the occurrence
from fossil material
and therefore marks
the minimum age of origin;
the cross-hatched
pattern indicates indirect
evidence from the
occurrence of sistergroup
taxa; and the
hatched pattern indicates
more remote indirect
evidence such as
trace fossils and the presence of apparently conspecific developmental (that is, ontogenetic)
stages other than the stage under consideration. (B) Mouthpart class diversity as a function of time
(24), plotted at the level of stratigraphie series. The upper mouthpart diversity curve represents
inferred presence based on all evidence, whereas the lower curve is based on direct body fossil
evidence alone [solid bars in (A)]. The upper curve may be the more realistic measure of true
mouthpart diversity. Abbreviations are the same as in Fig. 1.
300 200 100
Geologic time (10^ years)
SCIENCE ˇ VOL.261 ˇ 16 JULY 1993 313
bellate) that are presently associated with
flowering plants. After the expansion of
angiosperms, only 1 (low estimate) to 7
(high estimate) of the 34 mouthpart classes
are known to have originated. However,
these have poor fossil records, and only one
(siphonomandibulate) is associated with
flowering plants. Thus, by using mouthpart
classes as proxies for ecological disparity, we
conclude that virtually all major insect
feeding types were in place considerably
before angiosperms became serious contenders
in terrestrial ecosystems. Evidence
from the fossil record of vascular plantinsect
interactions (53) also supports this
inference.
Conclusion
Insects are not an invisible part of the fossil
record. They have a rich history that must
be reconciled with evolutionary interpretations
of the taxonomic diversity and morphological
and ecological disparity of modern
insects. We have examined a few synoptic
aspects of the fossil record of insects,
and the results contradict several notions
about what macroevolutionary patterns can
be seen among fossil insects and about how
modern insect diversity can be interpreted.
At the family level, fossil insect diversity
through most of the Phanerozoic exceeds
that of tetrapod vertebrates, as might be
expected from the great difference in their
modern diversities. But what is unexpected
is that modern insect diversity results not
from high rates of origination but rather
from low rates of extinction. These low
rates were established early during the
Mesozoic era, and the great radiation of
insects began more than 100 million years
before the ascendancy of angiosperms. This
diversification was based in part on a wide
range of previously evolved trophic adaptations
that allowed insects to enter into new
food-based ecological guilds during the late
Paleozoic to mid-Mesozoic eras. The morphologies
and correlated membership in
major guilds were retained into the later
Mesozoic and Cenozoic eras when they
became co-opted by angiosperms, allowing,
perhaps, for continued increases in familylevel
diversity. Whereas seed plants in general,
and not angiosperms in particular,
provided the stage for the spectacular evolutionary
history of insects, it is also possible
that the exploitative destruction of angiosperm
communities today could reverse
245 million years of evolutionary success.
REFERENCES AND NOTES
N. D, Newell, J. Paleontol. 26, 371 (1952); M. R.
House, in The Fossil Record, W. B. Harland etal.,
Eds, (Geological Society of London, London,
1967), pp. 41-54; J. W, Valentine, Palaeontology
12, 684 (1969); J. J. SepkoskI, Jr., Paleobiology?,
36 (1981).
2. J. W. Valentine, Ed., Phanerozoic Diversity Patterns:
Profiles In Macroevolution (Princeton Univ.
Press, Princeton, NJ, 1985).
3. H. Tappan and A. R. Loebiicti, Jr., J. Paleontol.
62, 695 (1988); J. P. Bujak and G. L Williams,
Mar. Micropaleontol. 4, 1 (1979).
4. G. G. Simpson, The Major Features of Evolution
(Columbia Univ. Press, New York, 1953); J. A.
Lillegraven, Taxon2^. 261 (1972); K. Padian and
W. A. Clemens, in (2), pp. 41-96.
5. M. J. Benton, Spec. Pap. Palaeontol. 33, 185
(1985); in Evolution and the Fossil Record, K. C.
Allen and D. E. G. Briggs, Eds. (Beltiaven Press,
London, 1989), pp. 218-241.
6. K. J. Niklas, B. H. Tiffney, A. H. Knoll, Nature303,
614 (1983); in (2), pp. 97-128; A. H. Knoll and K.
J. Niklas, Rev Palaeobot Palynol. 50, 127 (1987).
7. G. Vidal and A. H. Knoll, Waiure297, 57 (1982); J.
W. Schopf, S. Afr. J. Geol. 93, 33 (1991); in The
Proterozoic Biosphere, J. W. Sctiopf and C. Klein,
Eds. (Cambridge Univ. Press, Cambridge, 1992),
pp. 529-552.
8. Alttiough see J. J. SepkoskI, Jr., and M. L. Hulver,
in (2), pp. 11-39; and K.J. Niklas, in Patterns and
Processes in the History of Life, D. M. Raup and
D. Jablonski, Eds. (Springer-Veriag, Beriin, 1986),
pp. 383-405.
9. R. W. Crowson, Entomol. Gen. 11, 99 (1985).
10. Principal sources are B. B. Rohdendorf, Ed.,
Fundamentals of Paleontology (Amerind, New
Deltii, 1991), vol. 9; and F. M. Carpenter, in
Arthropoda 4: Superclass Hexapoda, Part R of
Treatise on Invertebrate Paleontology, R. L.
Kaesler, Ed. (Geological Society of America,
Boulder, CO, 1992).
11. Examples of ottier sources are B. B. Rotidendorf
and A. P. Rasnitsyn, Eds., Trans. Paleontol. Inst.
175, 1 (1980); R. Keilbacti, Dtsch. Entomol. Z. 29,
129 (1982); Y.-C. Hong, Mesozoic Fossil Insects
of JIuquan Basin in Gansu Province (Geological
Publisfiing House, Beijing, 1982); P. E. S. Wfialley
and E. A. Jarzembowski, Bull. Br. Mus. Nat. Hist.
Geol. 83, 381 (1985); C. C. Labandeira, B. S.
Beall, F. M. Hueber, Science 242, 913 (1988); J.
Kukalová-Peck, in (J5), pp, 141-179; and R. G.
Martins-Neto, J. C. K. Santos, M. V. Mesquita,
Soc. Bras. Geol. Nordeste 12, 59 (1991).
12. A. G. Ponomarenko, Ed., The Mesozoic Biocenotic
Crisis In the Evolution of Insects (in Russian)
(Izdatelstvo "Nauka," Moscow, 1988).
13. D. A. Grimaldi, Ed., Bull. Am. Mus. Nat. Hist. 195,
1 (1990).
14. G. 0. Poinar, Jr., Life In Amber (Stanford Univ.
Press, Stanford, CA, 1992).
15. Commonwealtti Scientific and Industrial Research]
Organization, Eds., The Insects of Australia: A
Textbook for Students and Researchers (Cornell
Univ. Press, Itfiaca, NY, ed. 2, 1991).
16. J. J. SepkoskI, Jr., R. K. Bambacfi, D. M. Raup, J.
W. Valentine, Nature 293, 435 (1981).
17. R. K. Bambacfi and J. J. SepkoskI, Jr., Paleontol.
Soc. Spec Publ. 6, 16 (1992).
18. D. M. Raup, Carnegie Mus. Nat. Hist. Bull. 13, 85
(1979); J. W. Valentine, in Causes of Evolution: A
Paleontological Perspective, R. M. Ross and W.
D. Alimón, Eds. (Univ. of Cfiicago Press, Cfiicago,
1990), pp. 128-150.
19. An example is tfie use of variability in modern
cockroacfi wing venation to assess Paleozoic
cockroacfi systematics [J. Scfineider, Freiberg.
Forschungsh. C 326, 87 (1977); Ibid. 391, 5
(1984)].
20. E. Mayr, Principles of Systematic Zoology (McGraw-Hill,
New York 1969); S. C. Willemstein,
Leiden Bet. Ser. 10, 1 (1987); W. P. McCafferty, in
Overview and Systematics of Ephemeroptera and
Plecoptera, J. Alba-Tercedor and A. SancfiezOretga,
Eds. (Sandfiill, Gainesville, FL, 1991).
21. J. J. SepkoskI, Jr., and D. C. Kendrick, Paleobiology
^9, 168 (1993).
22. Data are derived from M. J. Benton, Mem. Soc.
geol. Fr. Nouv. Ser. 150, 21 (1987), witfi modifications
[for example, C. M. Janis, Palaeontology
32, 463 (1989)].
23. F. W. Stefir, Ed., Immature Insects (Kendall-Hunt,
Dubuque, lA, 1987, 1991), vols. 1 and 2.
24. Geocfironology is from W. B. Hariand et ai, A
Geologic Time Scale: 1989 (Cambridge Univ.
Press, Cambridge, 1990).
25. D. H. Enwin, Annu Rev Ecol. Syst. 21, 69 (1990).
26. Major Cretaceous insect faunas from seven compression-fossil
Lagerstatten fiave been described
witfiin tfie past decade (listed fiere from oldest to
youngest); Wealden beds of England; Caracéas
Formation of Lérida, Spain; Eregskay Suite of
western Mongolia; sfiales from tfie Jiuquan Basin,
Gansu, Cfiina; Koonwarra Beds of Victoria, Australia;
Santana Formation of Ceará, Brazil; and an
unnamed crater deposit from Orapa, Botswana
(?3). In addition, insects are being described
from several Cretaceous ambers, principally from
Lebanon, France, Siberia, New Jersey, Alberta,
and Manitoba (14).
27. D. M. Raup, Paleobiology ^, 82 (1975).
28. S. M. Stanley, Macroevolution: Pattern and Process
(Freeman, San Francisco, 1979).
29. Data are from J. J. SepkoskI, Jr., Mllw. Public Mus.
Contrlb. Blol. Geol. 83, 1 (1992).
30. S. M. Stanley, Syst Zool. 22, 486 (1973); in
Patterns of Evolution, A. Hallam, Ed. (Elsevier,
Amsterdam, 1977), pp. 209-250.
31. An inverted analysis, plotting proportions of extinct
families witfiin tfie Jurassic ttirougfi Tertiary
insect fauna, was performed witfi similar conclusions
by B, B, Rodendorf and V. V. Zfierikin
[PrirodaS, 82 (1974)].
32. D. M. Raup and J. J. SepkoskI, Jr., Science 215,
1501 (1982); J. W. Valentine, B. H. Tiffney, J. J.
SepkoskI, Jr., Palalos 6, 81 (1991); J, W. Valentine,
in Molds, Molecules, and Metazoa, P. R.
Grant and H. S. Horn, Eds. (Princeton Univ. Press,
Princeton, NJ, 1992), pp. 17-32.
33. K. W. Flessa and D. Jablonski, Nature 313, 216
(1985); D. Jablonski, Scfence 231, 129 (1986).
34. C. C. Labandeira, tfiesis. University of Cfiicago
(1990).
35. R. A. Crowson, The Biology of the Coleóptera
(Academic Press, London, 1981).
36. M. V. Kozlov, in (r2), pp. 16-69; H. Donner and C.
Wilkinson, Fauna N. Z. 16, 1 (1989); R. J. Rayner
and S. B. Waters, Zool. J. Linn. Soc. 99, 309
(1990).
37. T. N. Freeman, Can. Entomol. 97, 2069 (1965); L.
J. Hickey and R. W. Hodges, Science 189, 718
(1975); S. G. Larsson, Entomonograph 1, 1
(1978); P. A. Opier, J. Lepid. Soc. 36, 145 (1982).
38. O. Bakkendorf, Entomol. Medd. 25, 213 (1948);
W. Hennig, Stuttg. Beltr. Naturk. 150, 1 (1966); L.
Masner, Proc. Entomol. Soc. Wash. 71, 397
(1969); J. G. Rozen, Jr., Univ. Calif. Berkeley Publ.
Entomol. 63, 75 (1971); E, L. Mockford, Fla. Entomol.
55, 153 (1972); P. A. OpIer, Science 179,
1321 (1973),
39. 0. E. Hele, Belh. Ber. Naturhist. Ges. Hannover e,
25 (1968); and W. L. Friedricti, Entomol.
Scand. 2, 74 (1971).
40. G. R. Coope, Symp. R. Entomol. Soc. London 9,
176 (1978); Annu. Rev. Ecol. Syst. 10, 247 (1979);
Antenna-[5, 158 (1991),
41. J. V. Mattfiews, Jr., Can. J. Zool. 48, 779 (1970);
Geol. Surv. Can. Pap. 76-1B, 217 (1976).
42. P. S. Martin, in Extinctions, M. H. Nitecki, Ed,
(Univ. of Cfiicago Press, Cfiicago, 1984), pp.
153-190; P. S. Martin and R. G. Klein, Eds.,
Quaternary Extinctions (Univ. of Arizona Press,
Tucson, AZ, 1984); F. G. Stetili and S. D. Webb,
Eds., The Great American Biotic Interchange (Plenum,
New York, 1985).
43. Ttiis conclusion should not be construed as vindication
of any current environmental policies regarding
insect biodiversity. Rapid deforestation in
the tropics may be quite unlike any ecological
perturbation insects have suffered during the Cenozoic
era. See E. O. Wilson, Ed., Biodiversity
(National Academy Press, Washington, DC, 1988).
44. M. Proctor and P. Yeo, The Pollination of Flowers
(Taplinger, New York, 1972); H. Zwolfer, Sonderb.
NatunNlss. Ver. Hamb. 2, 7 (1978); D. R. Strong, J.
H. Lawiton, T. R. E. Southwood, Insects on Plants:
Community Patterns and Mechanism (Harvard
Univ. Press, Cambridge, MA, 1984); M. Berenbaun
and D. Seigler, in Insect Chemical Ecology:
314 SCIENCE ˇ VOL.261 ˇ 16 JULY 1993
ARTICLES
An Evolutionary Approach, B. D. Roitberg and M.
B, Isman, Eds. (Routledge, Chapman and Hall,
New York, 1992), pp. 89-121.
45. Seminal studies of highly elaborate insect-angiosperm
associations Include P. R. Ehrlich and
P. H. Raven, Evolution 18, 586 (1964); D, H.
Janzen, ibid. 20, 249 (1966); V. F. Eastop, Symp.
R Enlomol. Soc. London 6, 157 (1972);
and M. Berenbaum, Evolution 37, 163 (1983).
Recent studies have shown that compared to their
nonherbivorous sister groups, many lineages of
herbivorous insects have preferentially diversified
with anglosperms [C. Mitter, B. Farrell, B.
Welgman, Am. Nat 132, 107 (1988); C. Mitter, B.
Farrell, D. J. Futuyma, Trends Ecol. Evol. 6, 290
(1991)] and, conversely, that relatively insectresistant
latex-bearing angiosperm lineages have
preferentially diversified compared to their nonlatex-bearing
sister groups [B. D. Farrell, D. E.
Dussourd, C. Mitter, Am. Nat 138, 881 (1991)].
46. J. J. Sepkoski, Jr., Paleobiology 5, 222 (1979);
Stiort Courses Paleontol. 4, 136 (1991).
47. B. B. Rohdendorf, Ed., Jurassic Insects ofKaratau
(In Russian) (Izdatelstvo "Nauka," Moscow,
1968); A. P. Rasnitsyn, Trans. Paleontol. Inst. 174,
1 (1980); N. S. Kaluglna and V. G. Kovalev,
Dipterous Insects from the Jurassic of Siberia (in
Russian) (Izdatelstvo "Nauka," Moscow, 1985); L.
V. ArnoI'dl, V. V. Zherlkin, L. M. NIkrItkhin, A. G.
Ponomarenko, Mesozoic Coleóptera (Smithsonian
Institution Library and National Science Foundation,
Washington, DC, 1991).
48. P. Whalley, Nature 327, 562 (1987); E. A. Jarembowski,
Mesozoic Res. 2, 25 (1989); C. C. Labandelra,
Paleontol. Soc. Spec. Publ. 6, 174 (1992);
E. M. Pike, ibid., p. 234.
49. N. F. Hughes and A. B. McDougall, Rev. Palaeobot
Palynol. 50, 255 (1987).
50. J. A. Doyle and M. J. Donaghue, Bot Rev 52, 321
(1986); Paleobiology \9. 141 (1993).
51. S. LIdgard and P. R. Crane, Nature 331, 344
(1988).
52. See also K. J. NIklas, Brittonia 30, 373 (1978); P.
R. Crane, Ann. Mo. Bol Gard. 72, 716 (1985); and
O. Pellmyr, Trends Ecol. Evol. 7, 46 (1992).
53. A. C. Rozefelds, Proc. R. Soc. Queensl. 99, 77
(1988); A. J. Boucot, Evolutionary Paleobiology of
Behavior and Coevolution (Elsevier, Amsterdam,
1990); C. C. Labandeira and B. S. Beall, Short
Courses Paleontol. 3, 214 (1990).
54. Low estimates, or minimum ages, are derived
from direct evidence of the body-fossil record.
High estimates are determined from more Indirect
lines of evidence: trace fossils, fossil presence of
a conspecific developmental (that Is, ontogenetic)
stage other than the one being analyzed, and
inferred presence due to sister-group relationships
from cladistic studies.
55. Data on Insect family ranges and mouthpart
classes are available from C.C.L. We thank D.
Chancy and F. Marsh for assistance in drafting
the figures and D. H. Enwln, D. J. Futuyma, J.
Kukalová-Peck, and A. P. Rasnitsyn for critical
commentary. Research received support from the
National Aeronautics and Space Administration
under grants NAG 2-282 and NAGW-1693 to
J.J.S. This is Evolution of Terrestrial Ecosystems
Contribution no. 12.
A Cold Suboceanic Mantle Belt
at the Earth's Equator
Enrico Bonatti, Monique Seyler, Nadia Sushevskaya
An exceptionally low degree of melting of the upper mantle In the equatorial part of the
mId-AtlantIc Ridge Is Indicated by the chemical composition of mantle-derived mid-ocean
ridge perldotltes and basalts. These data Imply that mantle temperatures below the equatorial
Atlantic are at least ~150°C cooler than those below the normal mId-AtlantIc Ridge,
suggesting that Isotherms are depressed and the mantle Is downwelling In the equatorial
Atlantic. An equatorial minimum of the zero-age crustal elevation of the East Pacific Rise
suggests a similar situation In the Pacific. If so, an oceanic upper mantle cold equatorial
belt separates hotter mantle regimes and perhaps distinct chemical and Isotopic domains
in the Northern and Southern hemispheres. Gravity data suggest the presence of high
density material in the oceanic equatorial upper mantle, which Is consistent with Its Inferred
low temperature and undepleted composition. The equatorial distribution of cold, dense
upper mantle may be ultimately an effect of the Earth's rotation.
A basic tenet of the theory of plate tectonics
states that upper mantle isotherms rise
beneath mid-ocean ridges because of the
upwelling of hot mantle material, which as
a result undergoes partial melting. The melt
fraction migrates upward, cools, and forms
the oceanic crust. The degree of melting
undergone by mantle peridotites upwelling
E. Bonatti is at the Lamont-Doherty Earth Observatory,
Columbia University, Palisades, NY 10964 and the
Istltuto Geología Marina CNR, Bologna, Italy. M. Seyler
is at the Laboratoire de Petrologle, Université Lille,
Lille, France. N. Sushevskaya Is at the Vernadsky
Institute of Geochemistry, Russian Academy of Sciences,
Moscow, Russia.
beneath mid-ocean ridges can be estimated
either from the chemical composition of
the melt fraction (mid-ocean ridge basalts
or MORBs) or the solid residue left behind
after partial melting (mid-ocean ridge peridotites
or MORPs). These estimates are
based on experimental work on the melting
of peridotites under different pressure and
temperature conditions (1, 2) and on assumptions
as to the initial composition of
the upwelling upper mantle material (], 3).
Estimates based on the composition of peridotite
samples recovered along the midAtlantic
Ridge (MAR) suggest that the
degree of melting of the upper mantle along
the MAR ranges from ~8 to ˇ25% and
varies regionally over long (ˇ1000 km) and
short (ˇ100 km) scales (4-6). Estimates
derived independently from the composition
of MORBs (7) agree in general with
those obtained from MORPs.
Recent studies of MORPs and MORBs
from the central MAR have suggested that
the upper mantle in some areas of the
equatorial Atlantic has undergone little or
no melting; thus, the mantle appears to be
relatively cold in these areas (6, 8). We
present additional data and discuss possible
causes and implications of an equatorial
belt of cold upper mantle in the Atlantic
and east Pacific.
Partial Melting of Atlantic
Mantle Peridotites
The mantle-equilibrated minerals found in
MORPs include olivine (ol), orthopyroxene
(opx), clinopyroxene (cpx), and spinel
(sp). The composition of these phases has
been determined in many peridotites from
the north and equatorial Atlantic (5, 6, 9,
10). Peridotites from the equatorial region
are of particular interest in this context.
Near the equator, the MAR axis is offset by
major transform faults, including the Romanche
fracture zone (FZ) (ˇ950-km offset)
at the equator, the St. Paul FZ just
north of the equator, and the Chain FZ just
south of it (Fig. 1).
We found that the composition of peridotites
sampled from several sites along the
Romanche FZ (Fig. 1) is different from that
of other MORPs. For instance, their mantle-equilibrated
phases have higher concentrations
of incompatible elements (elements
such as Al that partition with the
melt) and lower concentrations of refractory
elements (elements such as Mg and Cr
that stay with the solid residue during partial
melting) than MORPs from the Atlantic
and Indian oceans (Fig. 2). Representative
samples show a modal content of clinopyroxene
higher than that of other
MORPs (Fig. 3).
Laboratory experiments and theoretical
modeling show that the modal and mineral
composition of mantle peridotites changes as
a result of partial melting (1,2, 11). Assuming
an initial pyrolitic modal composition for
parental upper mantle beneath the ridge (3),
partial melting at 10 kbar causes a rapid
decrease in the content of clinopyroxene
(and eventually its disappearance when the
degree of melting is over about 20%), a
slower decrease of orthopyroxene, and an
increase of olivine (1, 2). Moreover, the
initial chemical composition of the primary
phases is modified by partial melting, so that
the ratio of refractory elements to incompatible
elements increases with an increased
degree of melting (], 2).
SCIENCE ˇ VOL. 261 ˇ 16 JULY 1993 315
http://www.jstor.org
LINKED CITATIONS
- Page 1 of5 STOR

You have printed the following article:
Insect Diversity in the Fossil Record
Conrad C. Labandeira; J. John Sepkoski, Jr.
Science, New Series, Vol. 261, No. 5119. (Jul. 16, 1993), pp. 310-315.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819930716%293%3A261%3A5119%3C310%3AIDITFR%3E2.0.CO%3B2-%23
This article references thefollowing linked citations. Ifyou are trying to access articlesfrom an
off-campus location, you may be required tofirst logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about optionsfor remote access to JSTOR.
References and Notes
' Periodicity in Invertebrate Evolution
Norman D. Newell
Journal ofPaleontology, Vol. 26, No. 3. (May, 1952), pp. 371-385.
Stable URL:
http://links.jstor.org/sici?sici=0022-3360%28195205%2926%3A3%3C371%3APIIE%3E2.0.CO%3B2-H
' A Factor Analytic Description of the Phanerozoic Marine Fossil Record
J. John Sepkoski, Jr.
Paleobiology, Vol. 7, No. 1. (Winter, 1981), pp. 36-53.
Stable URL:
http://links.jstor.org/sici?sici=0094-8373%28198124%297%3Al%3C36%3AAFADOT%3E2.0.CO%3B2-G
^ Foraminiferal Evolution, Diversification, and Extinction
Helen Tappan; Alfred R. Loeblich, Jr.
Journal ofPaleontology, Vol. 62, No. 5. (Sep., 1988), pp. 695-714.
Stable URL:
http://links.jstor.org/sici?sici=0022-3360%28198809%2962%3A5%3C695%3AFEDAE%3E2.0.CO%3B2-N
^ Ordinal and Familial Diversity of Cenozoic Mammals
Jason A. Lillegraven
Taxon, Vol. 21, No. 2/3. (May, 1972), pp. 261-274.
Stable URL:
http://links.jstor.org/sici?sici=0040-0262%28197205%2921%3A2%2F3%3C261%3AOAFDOC%3E2.0.CO%3B2-8
NOTE: The reference numberingfrom the original has been maintained in this citation list.
http://www.jstor.org
LINKED CITATIONS
- Page 2 of5 STOR

" Early Insect Diversification: Evidence from a Lower Devonian Bristletail from Québec
Conrad C. Labandeira; Bret S. Beall; Francis M. Hueber
Science, New Series, Vol. 242, No. 4880. (Nov. 11, 1988), pp. 913-916.
Stable URL:
http://links.istor.org/sici?sici=0036-8075%2819881111%293%3A242%3A4880%3C913%3AEIDEFA%3E2.0.CO%3B2-7
^^ Numerical Experiments with Model Monophyletic and Paraphyletic Taxa
J. John Sepkoski, Jr.; David C. Kendrick
Paleobiology, Vol. 19, No. 2. (Spring, 1993), pp. 168-184.
Stable URL:
http://links.jstor.org/sici?sici=0094-8373%28199321%2919%3A2%3C168%3ANEWMMA%3E2.0.CO%3B2-K
^' The End-Permian Mass Extinction
D. H. Erwin
Annual Review ofEcology and Systematics, Vol. 21. (1990), pp. 69-91.
Stable URL:
http://links.jstor.org/sici?sici=0066-4162%281990%2921%3C69%3ATEME%3E2.0.CO%3B2-%23
^^ Taxonomic Survivorship Curves and Van Valen's Law
David M. Raup
Paleobiology, Vol. 1, No. 1. (Winter, 1975), pp. 82-96.
Stable URL:
http://links.jstor.org/sici?sici=0094-8373%28197524%291%3Al%3C82%3ATSCAVV%3E2.0.CO%3B2-M
^** Effects of Competition on Rates of Evolution, With Special Reference to Bivalve MoUusks
and Mammals
Steven M. Stanley
Systematic Zoology, Vol. 22, No. 4. (Dec., 1973), pp. 486-506.
Stable URL:
http://links.istor.org/sici?sici=0039-7989%28197312%2922%3A4%3C486%3AEOCORO%3E2.0.CO%3B2-E
^^ Mass Extinctions in the Marine Fossil Record
David M. Raup; J. John Sepkoski
Science, New Series, Vol. 215, No. 4539. (Mar. 19, 1982), pp. 1501-1503.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819820319%293%3A215%3A4539%3C1501%3AMEITMF%3E2.0.CO%3B2-Y
NOTE: The reference numberingfrom the original has been maintained in this citation list.
http://www.jstor.org
LINKED CITATIONS
- Page 3 of5 STOR

^^ Evolutionary Dynamics of Plants and Animals: A Comparative Approach
J. W. Valentine; B. H. Tiffney; J. J. Sepkoski, Jr.
PALAIOS, Vol. 6, No. 1. (Feb., 1991), pp. 81-88.
Stable URL:
http://links.jstor.org/sici?sici=0883-1351%28199102%296%3Al%3C81%3AEDOPAA%3E2.0.CO%3B2-B
^^ Background and Mass Extinctions: The Alternation of Macroevolutionary Regimes
David Jablonski
Science, New Series, Vol. 231, No. 4734. (Jan. 10, 1986), pp. 129-133.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819860110%293%3A231%3A4734%3C129%3ABAMETA%3E2.0.CO%3B2-%23
ˇ^^ Lepidopteran Leaf Mine from the Early Eocene Wind River Formation of Northwestern
Wyoming
Leo J. Hickey; Ronald W. Hodges
Science, New Series, Vol. 189, No. 4204. (Aug. 29, 1975), pp. 718-720.
Stable URL:
http://links.istor.org/sici?sici=0036-8075%2819750829%293%3A189%3A4204%3C718%3ALLMFTE%3E2.0.CO%3B2-U
^^New Species, Records, and Synonymy of Florida Belaphotroctes (Psocoptera: Liposcelidae)
Edward L. Mockford
The Florida Entomologist, Vol. 55, No. 3. (Sep., 1972), pp. 153-163.
Stable URL:
http://links.jstor.org/sici?sici=0015-4040%28197209%2955%3A3%3C153%3ANSRASO%3E2.0.CO%3B2-M
^**Late Cenozoic Fossil Coleóptera: Evolution, Biogeography, and Ecology
G. R. Coope
Annual Review ofEcology and Systematics, Vol. 10. (1979), pp. 247-267.
Stable URL:
http://links.jstor.org/sici?sici=0066-4162%281979%2910%3C247%3ALCFCEB%3E2.0.CO%3B2-O
"^ Butterflies and Plants: A Study in Coevolution
Paul R. Ehrlich; Peter H. Raven
Evolution, Vol. 18, No. 4. (Dec., 1964), pp. 586-608.
Stable URL:
http://links.jstor.org/sici?sici=0014-3820%28196412%2918%3A4%3C586%3ABAPASI%3E2.0.CO%3B2-8
NOTE: The reference numberingfrom the original has been maintained in this citation list.
http://www.jstor.org
LINKED CITATIONS
- Page 4 of5 STOR

^^ Coevolution of Mutualism Between Ants and Acacias in Central America
Daniel H. Janzen
Evolution, Vol. 20, No. 3. (Sep., 1966), pp. 249-275.
Stable URL:
http://links.istor.org/sici?sici=0014-3820%28196609%2920%3A3%3C249%3ACOMBAA%3E2.0.CO%3B2-%23
'*' The Phylogenetic Study of Adaptive Zones: Has Phytophagy Promoted Insect
Diversification?
Charles Mitter; Brian Farrell; Brian Wiegmann
The American Naturalist, Yol 132, No. 1. (JuL, 1988), pp. 107-128.
Stable URL:
http://links.jstor.org/sici?sici=0003-0147%28198807%29132%3Al%3C107%3ATPSOAZ%3E2.0.CO%3B2-G
"' Escalation of Plant Defense: Do Latex and Resin Canals Spur Plant Diversification?
Brian D. Farrell; David E. Dussourd; Charles Mitter
The American Naturalist, Vol. 138, No. 4. (Oct., 1991), pp. 881-900.
Stable URL:
http://links.istor.org/sici?sici=0003-0147%28199110%29138%3A4%3C881%3AEOPDDL%3E2.0.CO%3B2-9
""^A Kinetic Model of Phanerozoic Taxonomic Diversity II. Early Phanerozoic Families and
Multiple Equilibria
J. John Sepkoski, Jr.
Paleobiology, Vol. 5, No. 3. (Summer, 1979), pp. 222-251.
Stable URL:
http://links.jstor.org/sici?sici=0094-8373%28197922%295%3A3%3C222%3AAKMOPT%3E2.0.CO%3B2-U
'** Phylogenies and Angiosperm Diversification
James A. Doyle; Michael J. Donoghue
Paleobiology, Vol. 19, No. 2. (Spring, 1993), pp. 141-167.
Stable URL:
http://links.istor.org/sici?sici=0094-8373%28199321%2919%3A2%3C141%3APAAD%3E2.0.CO%3B2-X
^^ Coupled Evolutionary Rates and the Fossil Record
Kari J. Niklas
Brittonia, Vol. 30, No. 3. (Jul. - Sep., 1978), pp. 373-394.
Stable URL:
http://links.jstor.org/sici?sici=0007-196X%28197807%2F09%2930%3A3%3C373%3ACERATF%3E2.0.CO%3B2-0
NOTE: The reference numberingfrom the original has been maintained in this citation list.
http://www.jstor.org
LINKED CITATIONS
- Page 5 of5 STOR

'^ Phylogenetic Analysis of Seed Plants and the Origin of Angiosperms
Peter R. Crane
Annals ofthe Missouri Botanical Garden, Vol. 72, No. 4. (1985), pp. 716-793.
Stable URL:
http://links.jstor.org/sici?sici=0026-6493%281985%2972%3A4%3C716%3APAOSPA%3E2.0.CO%3B2-F
NOTE: The reference numberingfrom the original has been maintained in this citation list.