Rapid recovery of Patagonian plant–insect associations after the end-Cretaceous extinction

Abstract

The Southern Hemisphere may have provided biodiversity refugia after the Cretaceous/Palaeogene (K/Pg) mass extinction. However, few extinction and recovery studies have been conducted in the terrestrial realm using well-dated macrofossil sites that span the latest Cretaceous (late Maastrichtian) and early Palaeocene (Danian) outside western interior North America (WINA). Here, we analyse insect-feeding damage on 3,646 fossil leaves from the latest Maastrichtian and three time slices of the Danian in Chubut, Patagonia, Argentina (palaeolatitude approximately 50° S). We test the southern refugial hypothesis and the broader hypothesis that the extinction and recovery of insect herbivores, a central component of terrestrial food webs, differed substantially from WINA at locations far south of the Chicxulub impact structure in Mexico. We find greater insect-damage diversity in Patagonia than in WINA during both the Maastrichtian and Danian, indicating a previously unknown insect richness. As in WINA, the total diversity of Patagonian insect damage decreased from the Cretaceous to the Palaeocene, but recovery to pre-extinction levels occurred within approximately 4 Myr compared with approximately 9 Myr in WINA. As for WINA, there is no convincing evidence for survival of any of the diverse Cretaceous leaf miners in Patagonia, indicating a severe K/Pg extinction of host-specialized insects and no refugium. However, a striking difference from WINA is that diverse, novel leaf mines are present at all Danian sites, demonstrating a considerably more rapid recovery of specialized herbivores and terrestrial food webs. Our results support the emerging idea of large-scale geographic heterogeneity in extinction and recovery from the end-Cretaceous catastrophe.

Full text

NATURE ECOLOGY & EVOLUTION 1, 0012 (2016) | DOI: 10.1038/s41559-016-0012 | www.nature.com/natecolevol 1
Articles
PUBLISHED: 7 NOVEMBER 2016 | VOLUME: 1 | ARTICLE NUMBER: 0012
Rapid recovery of Patagonian plant–insect
associations after the end-Cretaceous extinction
Michael P. Donovan1*, Ari Iglesias2, Peter Wilf1, Conrad C. Labandeira3,4,
5 and N. Rubén Cúneo6
The Southern Hemisphere may have provided biodiversity refugia after the Cretaceous/Palaeogene (K/Pg) mass extinction.
However, few extinction and recovery studies have been conducted in the terrestrial realm using well-dated macrofossil sites
that span the latest Cretaceous (late Maastrichtian) and early Palaeocene (Danian) outside western interior North America
(WINA). Here, we analyse insect-feeding damage on 3,646 fossil leaves from the latest Maastrichtian and three time slices of
the Danian in Chubut, Patagonia, Argentina (palaeolatitude approximately 50° S). We test the southern refugial hypothesis
and the broader hypothesis that the extinction and recovery of insect herbivores, a central component of terrestrial food webs,
differed substantially from WINA at locations far south of the Chicxulub impact structure in Mexico. We find greater insect-
damage diversity in Patagonia than in WINA during both the Maastrichtian and Danian, indicating a previously unknown insect
richness. As in WINA, the total diversity of Patagonian insect damage decreased from the Cretaceous to the Palaeocene, but
recovery to pre-extinction levels occurred within approximately 4 Myr compared with approximately 9 Myr in WINA. As for
WINA, there is no convincing evidence for survival of any of the diverse Cretaceous leaf miners in Patagonia, indicating a severe
K/Pg extinction of host-specialized insects and no refugium. However, a striking difference from WINA is that diverse, novel
leaf mines are present at all Danian sites, demonstrating a considerably more rapid recovery of specialized herbivores and ter-
restrial food webs. Our results support the emerging idea of large-scale geographic heterogeneity in extinction and recovery
from the end-Cretaceous catastrophe.
Palaeontological evidence from both continental and marine
deposits suggests that the Southern Hemisphere may have
harboured biodiversity refugia in the wake of the bolide
impact at Chicxulub, Mexico, 66.0 Myr ago (Ma)1–4. The extinc-
tion rate of Southern Hemisphere nannoplankton was lower than
that of their Northern Hemisphere counterparts, and their popu-
lations recovered nearly immediately2. Nominally Mesozoic plant
groups, including corystosperms and bennettitaleans, survived
until at least the Palaeogene in Australia1,5. Palynological data
from New Zealand revealed a sudden but short-lived disturbance,
with low overall extinction rates6,7. In Patagonia, Argentina, paly-
nomorphs from the latest Maastrichtian–early Danian Lefipán
Formation exhibited low extinction, followed by the reappear-
ances of Cretaceous pollen types3. Early Danian macrofloras from
the Salamanca Formation in Patagonia are more diverse than
comparable North American Palaeocene floras8–10. A number of sur-
viving lineages from other plant3 and vertebrate11,12 groups have also
been identified, especially in Patagonia4, although marine inver-
tebrate faunas in Antarctica underwent severe extinction13. K/Pg
boundary sections in New Zealand have provided important insights
into the response of terrestrial ecosystems6,7,14–16, but until recently
there has not been a series of well-dated, heavily sampled continen-
tal macrofloral localities anywhere in the Southern Hemisphere that
spans both the terminal Cretaceous and earliest Palaeogene.
Plant–insect interactions are fundamental components of ter-
restrial food webs, and their sensitivity to major environmental
perturbations is well known from deep time as well as the modern
world17–20. The diversity of insect-feeding damage on extant leaves
in two tropical rainforests is positively correlated with the rich-
ness of insects that caused the damage, supporting the widespread
use of insect damage on fossil leaves as a proxy for herbivorous
insect diversity when suitable insect body fossils are absent21. In
North Dakota, USA, insect-damage diversity on fossil leaves, espe-
cially specialized feeding such as mining and galling, declined
considerably across the K/Pg boundary and remained low through-
out WINA before increasing with the latest Palaeocene warming,
approximately 9 Myr after the K/Pg boundary17,18,20. The only excep-
tion to this pattern is the early Palaeocene (about 65 Ma) Mexican
Hat locality in south-eastern Montana, USA, which has typical low-
diversity flora but anomalously high insect damage diversity for the
time; this pattern is attributed to a short-lived interval of decoupled
plant and insect diversity following the K/Pg mass extinction17,20.
Much less is known about the extinction and recovery of insect
herbivores outside WINA. Late Palaeocene floras from Colombia
are associated with low richness of plants and specialized insect-
damage diversity as in Palaeocene WINA22, contrasting with high
plant and insect-damage diversity on middle Palaeocene floras from
France23 and Spitsbergen24. However, until now, no studies have
investigated changes in insect-damage diversity based on terminal
Cretaceous and early Palaeocene leaf floras from any non-WINA
study area.
Maastrichtian–Danian insect damage on Patagonian floras
Recent collections of latest Cretaceous and early Palaeocene8–10
compression macrofloras with insect-feeding damage from Chubut
Province, central Patagonia, Argentina, allow us to test the hypoth-
esis of a K/Pg refugium in the plant–insect system at approximately
50° S palaeolatitude. We analysed insect damage on fossil dicot
leaves from four coastal lowland localities (see Methods) repre-
senting four time slices: (1) Lefipán East (LefE; Lefipán Formation;
1Department of Geosciences, Pennsylvania State University, University Park, Pennsylvania 16802, USA. 2Instituto de Investigaciones en Biodiversidad y
Medioambiente, CONICET-Universidad Nacional del Comahue, San Carlos de Bariloche 8400, Argentina. 3Department of Paleobiology, National Museum
of Natural History, Smithsonian Institution, Washington DC 20013, USA. 4Department of Entomology and Behavior, Ecology, Evolution, and Systematics
Program, University of Maryland, College Park, Maryland 20742, USA. 5School of Life Sciences, Capital Normal University, Beijing 100048, China.
6CONICET−Museo Paleontológico Egidio Feruglio, Trelew 9100, Argentina. *e-mail: mpd187@psu.edu

2 NATURE ECOLOGY & EVOLUTION 1, 0012 (2016) | DOI: 10.1038/s41559-016-0012 | www.nature.com/natecolevol
Articles Nature ecology & evolutioN
67‒66 Ma; 606 specimens examined)3,25; (2) Palacio de los Loros 1
(PL1; Salamanca Formation; constrained to geomagnetic polarity
chron C29n, or 65.58–64.86 Ma; 1,089 specimens)8–10; (3) Palacio de
los Loros 2 (PL2; Salamanca Formation; chron C28n, 64.67–63.49 Ma;
1,132 specimens)8–10; and (4) Las Flores (LF; Peñas Coloradas
Formation; chron C27n, 62.52–62.22 Ma; 571 specimens)9,10.
The preservation quality of the fossils varies among sites, mostly
according to grain size10. LF has the poorest preservation and PL2
the finest, including associated flowers, feathers and insect wings.
Fortuitously, the two floras closest to the age of the K/Pg boundary,
LefE and PL1, have comparable, intermediate preservation quality.
We analysed insect feeding using the standard damage type (DT)
system26,27 (see Methods). We focused on leaf mines because they
are overwhelmingly host specialized and preserve fine, informative
morphological details27. The same methods used here have been
applied in many studies in WINA17,18,20 and elsewhere, thus allow-
ing direct comparisons with changes in insect damage diversity and
potential survival of insects across the K/Pg interval. The North
American reference data used here are provided in earlier works20,26.
Diversity of insect-feeding damage
We found a greater richness of insect DTs on both the Cretaceous
(49 DTs) and the Palaeocene (60 DTs) floras from Patagonia (Fig.1)
than on comparable North American floras. This result is particu-
larly striking because the North American data20,26 come from much
larger sample sizes and cover many more time slices over a larger
spatial area (Fig.2). For example, there are 43 latest Maastrichtian
DTs from 3,155 leaf specimens and 56 early to late Palaeocene DTs
from 11,052 leaf specimens in the WINA dataset17,20. The same
pattern holds at the single-site level. Resampled total DT diversity
(Fig.2a) at single quarries is higher at each Patagonian site than at
any of the WINA localities. Both Cretaceous and Palaeocene leaf-
mine diversities (Fig.2b) are higher in Patagonia than in WINA, with
the exception of the unusual Mexican Hat site in Montana, USA20.
The noted leaf-mine richness of Mexican Hat17,20 has a higher mean
than either of the two earliest Palaeocene Patagonian localities (PL1
and PL2; uncertainties overlap). However, it is lower than at the
Cretaceous Lefipán and late Danian LF sites, even though they each
have a much lower quality of leaf preservation than Mexican Hat
(Fig.2). All of these results indicate a previously unknown excep-
tionally high richness of insect herbivores in Patagonian ecosystems
through the K/Pg interval.
Extinction and recovery
From the Cretaceous to the Palaeocene (Fig.2), we found a decrease
in plant–insect associational diversity followed by an increase to
near-Maastrichtian levels of damage diversity within approximately
4 Myr, by the late Danian (LF). From the Cretaceous (LefE) to
the Palaeocene (Fig.2), we found a decrease in plant–insect asso-
ciational diversity, followed by an increase to near-Maastrichtian
Figure 1 | Insect damage on latest Cretaceous and early Palaeocene leaves from Chubut, Argentina. a–l, Latest Cretaceous samples from the Lefipán
Formation (a–c), and early Palaeocene samples8 from the Salamanca (d–i) and Peñas Coloradas (j–l) formations. a, Multiple, overlapping blotch mines
containing centralized frass (DT299) on leaf morphotype LEF28 (LefW; MPEF-Pb 4776). b, Spheroidal galls with striated surfaces (DT303) on LEF2 (LefE;
MPEF-Pb 4259). c, Margin feeding with thickened reaction tissue (DT12) on LEF23 (LefL; MPEF-Pb 4758). d, Serpentine mine with spheroidal terminal
chamber (DT300) on Cissites patagonica (PL1; MPEF-Pb 6557). e, Elliptical gall positioned on the primary vein at the intersection with secondary
veins (DT84) on Laurophyllum piatnitzkyi (PL1; MPEF-Pb 6555). f, Row of parallel-sided holes near the leaf margin (DT64) on Dryophyllum australis (PL1;
MPEF-Pb 6560). g, Spheroidal galls with distinct outer rims positioned on the primary vein (DT117) of Cissites patagonica (PL2; MPEF-Pb 6567). h, Concentric
rings of piercing and sucking marks surrounded by dark reaction tissue (DT118) on SA19 (PL2; MPEF-Pb 4072). i, Hole feeding surrounded by a wide rim
of blotched reaction tissue (DT113) on SA43 (PL2; MPEF-Pb 6561). j, Serpentine mines that transition to blotch mines with internal, intestiniform trails
(DT301) on Fagophyllum duseni (LF; MPEF-Pb 6547). k, Elongate, curvilinear patches of skeletonized tissue (DT20) on SA70 (LF; MPEF-Pb 6549). l, Deeply
incised margin feeding damage (DT15) on Dryophyllum australis (LF; MPEF-Pb 6546). DT, damage type27 (new DTs defined in Supplementary Discussion).
a
f
h
g
i
j
kl
bcde
1 cm
1 mm 5 mm 5 mm 5 mm5 mm
1 cm
5 mm
1 mm
5 mm
1 cm 5 mm

NATURE ECOLOGY & EVOLUTION 1, 0012 (2016) | DOI: 10.1038/s41559-016-0012 | www.nature.com/natecolevol 3
Articles
Nature ecology & evolutioN
levels of damage diversity by the late Danian (LF), approximately
4 Myr after the K/Pg event. Resampled total insect-damage
richness (Fig.2a) was highest during the Maastrichtian (LefE), even
though preservation was comparable to that in early Danian PL1
and poorer than in PL2. Damage diversity decreased considerably
across the K/Pg (from LefE to PL1, 21.7% decrease; Fig.2a), then
increased 15.4% from PL1 to PL2 and remained high at LF, even
though that locality had the poorest preservation of the Danian sites
(Fig.2a). In addition to total DTs, the extinction and recovery pat-
tern holds for specialized damage and leaf-mine diversity (Fig.2b;
Supplementary Fig. 2). Although LefE and PL1 have similar preser-
vation, minimizing potential bias, leaf-mine richness is much lower
at PL1 (Fig.2b). PL2 has the most detailed preservation, but its leaf-
mine diversity is lower than at both LefE and LF. Our preliminary
analysis of plant extinction from these Patagonian sites shows an
approximate 30% decrease in leaf-morphotype diversity from the
Lefipán to the Salamanca Formation floras.
The temporal ranges of DTs in the Patagonian data are shown in
Fig.3. The feeding damage represents all eight of the typical func-
tional feeding groups: hole feeding, margin feeding, skeletonization,
surface feeding, piercing and sucking, mining, galling and oviposi-
tion. More specialized DTs (51) were found than generalized DTs
(18). All 17 generalized DTs from the Lefipán Formation are also
present during the Danian, which is consistent with findings from
North Dakota18 and the presumed origins of these DTs from diverse
insect clades that did not all suffer extinction. Nevertheless, 9 of
the 32 Cretaceous specialized DTs from Patagonia (28.1%) are not
found in the Danian, including 5 mining DTs, compared with 20 of
the 36 Cretaceous specialized DTs (55.6%) from North Dakota18.
This result suggests a less severe extinction of specialized herbivores
in Patagonia. However, as was done for the North American data20,
we also examined the detailed morphologies of leaf mines assigned
to the same DT for more reliable evidence of surviving herbivores.
On the basis of the fine morphological details of the six ‘surviving’
leaf-mining DTs (Fig. 3), we determined whether any mine DTs
plausibly could have been made by the same or closely related
insect species both before and after the K/Pg extinction (Methods;
Supplementary Information and Supplementary Figs 4‒10).
Our analysis suggests that none of the diverse Cretaceous
Patagonian leaf mines can be confidently linked to any of the
Palaeocene mines, even on apparently closely related plant spe-
cies. This overall pattern is the same as in WINA20 and indicates
a severe leaf-miner extinction in both hemispheres. However, the
Figure 2 | Insect-feeding damage richness for late Cretaceous and
Palaeocene floras from Patagonia and WINA. Sample sizes were
standardized to 400 leaves via random resampling; error bars indicate
± 1σ. a, Total insect-damage richness. b, Leaf-mine diversity. The age
estimates for WINA localities are detailed in Supplementary Information
and Supplementary Table 1. The Pyramid Butte (P. Butte) flora, which
appears on the K/Pg line, is located 0.2 m above the K/Pg boundary in the
Palaeocene. M. Hat, Mexican Hat.
58
66
64
62
60
67
63
61
59
57
20 4030 02
46
Damage richness (number of DTs) Mine richness (number of DTs)
10
Age (Ma)
65
LefE
PL1
PL2
LF
M. Hat
LefE
PL1
PL2
LF
M. Hat
Patagonia, Argentina WINA
ab
P. Butte P. Butte
K/Pg
Figure 3 | Occurrences of 69 insect damage types through four time slices spanning the latest Cretaceous to early Palaeocene in Patagonia, Argentina.
Raw presence and absence data for DTs categorized by the functional groups indicated. The symbols indicate generalized (black) and specialized (red).
Grey bars connect occurrences at more than one stratigraphic level that represent rangethroughs. MF, margin feeding; O, oviposition; PS, piercing and
sucking; SF, surface feeding. Ages are shown as midpoints of the potential age ranges.
1
66
65
64
63
62
2
3
4
7
8
6
5
17
16
26
15
14
13
12
113
64
63
57
50
10
9
48
46
82
31
30
29
79
61
56
24
22
20
19
91
69
65
60
45
42
41
40
36
101
76
138
118
52
34
33
32
1
301
300
299
298
139
109
37
304
303
302
117
112
85
84
80
62
Hole Feeding MF
Skeletonization SF PS
O
Mining Galling
Insect damage types
Age (Ma)
PL2
PL1
LF
Lefipán
K
Pg

4 NATURE ECOLOGY & EVOLUTION 1, 0012 (2016) | DOI: 10.1038/s41559-016-0012 | www.nature.com/natecolevol
Articles Nature ecology & evolutioN
subsequent recovery patterns for leaf miners are very different.
During the early Palaeocene in WINA, leaf mines are generally rare
(except at Mexican Hat) and mostly assigned to a single generalized
morphotype on a single host plant17,20. In contrast, early Palaeocene
floras in Patagonia already hosted diverse leaf mines (11 DTs;
Fig.3), although these do not indicate close relationships between
Cretaceous and Palaeocene leaf miners. Thus, the extinction of
Cretaceous leaf miners in Patagonia does not support the hypoth-
esis of a regional refugium for herbivorous insects1–4. However, the
Danian first appearances of many new mine DTs in Patagonia dem-
onstrates a faster recovery of insect herbivore diversity and terres-
trial food webs in Patagonia after the K/Pg extinction than in North
America. Unlike in the WINA dataset20, insect damage diversity
in Patagonia recovered to pre-extinction levels by the late Danian
after increasing through two time slices (Fig.2). Although damage
diversity at Mexican Hat is elevated, its richness was not sustained
and is only found at that single Palaeocene site out of at least 200
Palaeocene WINA localities studied for insect damage17,20.
Conclusions
Our results support the emerging idea that southern latitude
continental biotas suffered major extinctions but recovered much
more quickly than their North American counterparts from the
global environmental disaster after the end-Cretaceous impact.
Large-scale variation in extinction and recovery patterns indicates
that the effects of the Chicxulub impact were geographically hetero-
geneous, presumably due to varying distance from the crater28
combined with other, poorly understood factors2,29.
Methods
Localities and oras. e study is based on unbiased (censused) collections
of fossil dicot leaves from the latest Cretaceous portion of the Lepán Formation
(locality LefE; supplemental collections from localities LefL and LefW were used
for ranges and the detailed study of leaf-mine morphology), the early Danian
Salamanca Formation (localities PL1 and PL2)8–10 and the late Danian Peñas
Coloradas Formation (locality LF)9,10, all in Chubut Province, Argentina. e
Lepán localities (LefE, LefL and LefW) are in northwest Chubut, and localities
PL1, PL2 and LF are in the San Jorge Basin, near Sarmiento in southern Chubut
as recently detailed elsewhere9,10. e complete collections are curated at the
Museo Paleontológico Egidio Feruglio (MPEF-Pb) in Trelew, Chubut, Argentina
(Supplementary Information).
The LefW fossils came from seven horizons spanning 25 m of the
San Ramón section depicted in Fig.3 in Scasso et al.25; the uppermost of these
is 8.3 m below the Danian Turritella marker bed. LefE is located on the opposite
side of the same hill as LefW, approximately 1,000 m to the east, and LefL
is located in the same fossiliferous layer and 500 m to the east of LefE. Plant
fossils were collected 21.5 m and 24 m below the Turritella bed at LefE and
LefL, respectively.
The Lefipán Formation is a tide-dominated delta deposit that ranges in
age from the Maastrichtian to early Palaeocene25. The K/Pg boundary impact
layer is not preserved, apparently due to bioturbation25; the generic diversity
of marine molluscs in this formation decreased from the Maastrichtian to the
Danian30. The fossil plant localities are dated to approximately 67–66 Ma on
the basis of age-diagnostic Maastrichtian marine invertebrates25, dinoflagellates
and palynomorphs3. We tabulated insect damage from LefE (606 specimens) for
the quantitative analyses because it is the best-sampled Lefipán locality with an
unbiased collection. We supplemented the LefE data for the analyses of leaf-mine
morphology and ranges (see Morphological Analyses section; Fig.3) with fossil
leaves from LefW (140 specimens) and LefL (107 specimens).
The ages of the Danian floras are well constrained by a suite of recent
integrated data from the absolute dating of associated volcanic strata;
palaeomagnetic stratigraphy; biostratigraphy of calcareous nannoplankton,
foraminifera and dinoflagellates; lithostratigraphy; and sequence stratigraphy9,10.
The Salamanca Formation was deposited in a suite of estuarine environments10.
Plant fossils were collected from two localities8 representing two early
Danian time slices: Palacio de los Loros 1 (PL1; 1,089 specimens) in chron
C29n (65.58–64.86 Ma, all ages follow the international 2012 geologic time
scale31) and Palacio de los Loros 2 (PL2; 1,133 specimens) in chron
C28n (64.67–63.49 Ma31)9,10. Plant fossils from a third Danian time slice are found
in the basal Peñas Coloradas Formation, which overlies the Salamanca Formation
and is a volcanic–fluvial unit deposited during chrons C27n (62.52–62.22 Ma) and
C26r (62.22–59.24 Ma)9. Fossil leaves were collected from LF9 (571 specimens),
which is located in the basal portion of the formation, correlating to chron
C27n. Despite the change in palaeoenvironment, the floral composition
at LF is very similar to that at PL1 and PL2 (Supplementary Information).
Analyses of insect damage. We recorded the presence or absence of
insect-feeding DTs on the four fossil leaf samples using the standard reference27
(new DTs are described in the Supplementary Information). As in the
compared WINA studies17,20, all fossil leaves included in the analyses were dicots
(non-monocot angiosperms) identifiable to morphotype with at least 50% of the
leaf blade intact. Insect damage can be described as specialized, made by insects
that feed on one or a few closely related host plants, or generalized, made by
insects that feed on a variety of taxonomically unrelated plants27. DTs are assigned
to increasing specialization levels on a scale26,27 of 1–3. We considered all DTs
assigned to groups 2 and 3 as specialized for the quantitative analyses19,27.
Sampling intensity varied across the sites, so we compared insect damage
diversity among sites via random resampling of fossil leaves26 (Supplementary Fig. 1),
in which 400 leaves were randomly sampled from a flora without replacement and
the mean damage diversity was calculated from 5,000 iterations for total, specialized
and leaf-mine diversity, respectively (Fig.2, Supplementary Fig. 2). We compared the
outputs with previous resampling analyses of DT diversity on floras from WINA17,18,20
(Fig.2; Supplementary Information). Damage diversity on single-leaf morphotypes
with more than 20 specimens each was resampled to 20 leaves using the same
methods (Supplementary Fig. 3). All analyses were conducted with R version 3.2.232.
We described all leaf-mine damage from the four Patagonian samples and
compared their morphologies to estimate the survival of leaf-mining insects after
the K/Pg extinction (Supplementary Information and Supplementary Table 2).
We focused on leaf mines because they preserve a wide variety of informative
characters (Supplementary Information). We also focused on leaf morphotypes
that crossed the K/Pg boundary to test whether they provided a refugium for
herbivorous insects after the bolide impact.
Data availability. The Patagonian insect damage data generated and
analysed during this study are included in this Article and its Supplementary
Information files. The WINA insect damage data used during this study are
available in references17,18,20.
Received 6 June 2016; accepted 6 September 2016;
published 7 November 2016
References
1. McLoughlin, S., Carpenter, R. J., Jordan, G. J. & Hill, R. S. Seed ferns survived
the end-Cretaceous mass extinction in Tasmania. Am. J. Bot. 95, 465–471 (2008).
2. Jiang, S., Braklower, T. J., Patzkowsky, M. E., Kump, L. R. & Schueth, J. D.
Geographic controls on nannoplankton extinction across the Cretaceous/
Palaeogene boundary. Nat. Geosci. 3, 280–285 (2010).
3. Barreda, V. D. et al. Cretaceous/Paleogene oral turnover in Patagonia: drop in
diversity, low extinction, and a Classopollis spike. PLoS ONE 7, e52455 (2012).
4. Wilf, P., Cúneo, N. R., Escapa, I. H., Pol, D. & Woodburne, M. O. Splendid
and seldom isolated: the paleobiogeography of Patagonia. Annu. Rev. Earth
Planet. Sci. 41, 561–603 (2013).
5. McLoughlin, S., Carpenter, R. J. & Pott, C. Ptilophyllum muelleri (Ettingsh.)
comb. nov. from the Oligocene of Australia: last of the Bennettitales?
Int. J. Plant Sci. 172, 574–585 (2011).
6. Vajda, V., Raine, J. I. & Hollis, C. J. Indication of global deforestation at the
Cretaceous-Tertiary boundary by New Zealand fern spike. Science 294,
1700–1702 (2001).
7. Vajda, V. & McLoughlin, S. Fungal proliferation at the Cretaceous-Tertiary
boundary. Science 303, 1489 (2004).
8. Iglesias, A. et al. A Paleocene lowland macroora from Patagonia reveals
signicantly greater richness than North American analogs. Geology 35,
947–950 (2007).
9. Clyde, W. C. et al. New age constraints for the Salamanca Formation and
lower Río Chico Group in the western San Jorge Basin, Patagonia, Argentina:
implications for Cretaceous-Paleogene extinction recovery and land mammal
age correlations. Geol. Soc. Am. Bull. 126, 289–306 (2014).
10. Comer, E. E. et al. Sedimentary facies and depositional environments of
diverse early Paleocene oras, north-central San Jose Basin, Patagonia,
Argentina. Palaios 30, 553–573 (2015).
11. Case, J. A. & Woodburne, M. O. South American marsupials: a successful
crossing of the Cretaceous-Tertiary boundary. Palaios 1, 413–416 (1986).
12. Apesteguía, S., Gómez, R. O. & Rougier, G. W. e youngest South American
rhynchocephalian, a survivor of the K/Pg extinction. Proc. R. Soc. B. 281,
20140811 (2014).
13. Witts, J. D. et al. Macrofossil evidence for a rapid and severe Cretaceous–
Paleogene mass extinction in Antarctica. Nat. Commun. 7, 11738 (2016).
14. Kennedy, E. M. Late Cretaceous and Paleocene terrestrial climates of
New Zealand: leaf fossil evidence from South Island assemblages.
N. Z. J. Geol. Geophys. 46, 295–306 (2003).

NATURE ECOLOGY & EVOLUTION 1, 0012 (2016) | DOI: 10.1038/s41559-016-0012 | www.nature.com/natecolevol 5
Articles
Nature ecology & evolutioN
15. Pole, M. & Vajda, V. A new terrestrial Cretaceous-Paleogene site in
New Zealand—turnover in macroora conrmed by palynology. Cretac. Res.
30, 917–938 (2009).
16. Steinthorsdottir, M., Vajda, V. & Pole, M. Global trends of pCO2
across the Cretaceous–Paleogene boundary supported by the rst Southern
Hemisphere stomatal proxy-based pCO2 reconstruction. Palaeogeogr.
Palaeoclimatol. Palaeoecol. http://dx.doi.org/10.1016/j.palaeo.2016.04.033
(2016).
17. Wilf, P., Labandeira, C. C., Johnson, K. R. & Ellis, B. Decoupled plant
and insect diversity aer the end-Cretaceous extinction. Science 313,
1112–1115 (2006).
18. Labandeira, C. C., Johnson, K. R. & Wilf, P. Impact of the terminal
Cretaceous event on plant-insect associations. Proc. Natl Acad. Sci. USA 99,
2061–2066 (2002).
19. Currano, E. D. et al. Sharply increased insect herbivory during the
Paleocene-Eocene ermal Maximum. Proc. Natl Acad. Sci. USA 105,
1960–1964 (2008).
20. Donovan, M. P., Wilf, P., Labandeira, C. C., Johnson, K. R. & Peppe, D. J.
Novel insect leaf-mining aer the end-Cretaceous extinction and the
demise of Cretaceous leaf miners, Great Plains, USA. PLoS ONE 9,
e103542 (2014).
21. Carvalho, M. R. et al. Insect leaf-chewing damage tracks herbivore richness
in modern and ancient forests. PLoS ONE 9, e94950 (2014).
22. Wing, S. L. et al. Late Paleocene fossils from the Cerrejón Formation,
Colombia, are the earliest record of Neotropical rainforest. Proc. Natl Acad.
Sci. USA 106, 18627–18632 (2009).
23. Wappler, T., Currano, E. D., Wilf, P., Rust, J. & Labandeira, C. C.
No post-Cretaceous ecosystem depression in European forests? Rich
insect-feeding damage on diverse middle Palaeocene plants, Menat, France.
Proc. R. Soc. B 276, 4271–4277 (2009).
24. Wappler, T. & Denk, T. Herbivory in early Tertiary Arctic forests. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 310, 283–295 (2011).
25. Scasso, R. A. et al. Integrated bio- and lithofacies analysis of coarse-grained,
tide-dominated deltaic environments across the Cretaceous/Paleogene
boundary in Patagonia, Argentina. Cretac. Res. 36, 37–57 (2012).
26. Wilf, P. & Labandeira, C. C. Response of plant-insect associations to
Paleocene-Eocene warming. Science 284, 2153–2156 (1999).
27. Labandeira, C. C., Wilf, P., Johnson, K. R. & Marsh, F. Guide to Insect (and
Other) Damage Types on Compressed Plant Fossils Version 3.0. (Smithsonian
Institution, 2007); http://paleobiology.si.edu/insects/index.html
28. Schulte, P. et al. e Chicxulub asteroid impact and mass extinction at the
Cretaceous-Paleogene boundary. Science 327, 1214–1218 (2010).
29. Vajda, V., Ocampo, A., Ferrow, E. & Koch, C. B. Nano particles as the
primary cause for long-term sunlight suppression at high southern
latitudes following the Chicxulub impact—evidence from ejecta deposits
in Belize and Mexico. Gondwana Res. 27, 1079–1088 (2015).
30. Aberhan, M. & Kiessling, W. Rebuilding biodiversity of Patagonian
marine molluscs aer the end-Cretaceous mass extinction. PLoS ONE 9,
e102629 (2014).
31. Gradstein, F. M., Ogg, J. G., Schmitz, M. & Ogg, G. e Geologic Time Scale
2012 (Elsevier, 2012).
32. R Development Core Team R: A Language and Environment for Statistical
Computing (R Foundation for Statistical Computing, 2013).
Acknowledgements
The authors thank E. Currano, K. Johnson, P. Puerta, L. Reiner and E. Ruigómez
for field and laboratory assistance and T. Bralower, D. Hughes and M. Patzkowsky for
discussion. This study was supported by grants to M.P.D. from the Evolving Earth
Foundation, the Geological Society of America, Sigma Xi, the Paleontological Society
and the P. D. Krynine Memorial Fund of Penn State Department of Geosciences;
and to P.W., A.I. and N.R.C. from NSF awards DEB-0919071 and DEB-1556666.
Author contributions
M.P.D., A.I., P.W. and C.C.L. designed the research. M.P.D., A.I., P.W. and N.R.C. did
the fieldwork. M.P.D., A.I. and C.C.L. collected the DT data. N.R.C. led research on the
Lefipán flora. M.P.D. performed the analyses and wrote the manuscript. All authors
commented on and substantially contributed to the manuscript.
Additional information
Supplementary information is available for this paper.
Reprints and permissions information is available at www.nature.com/reprints.
Correspondence and requests for materials should be addressed to M.P.D.
How to cite this article: Donovan, M. P., Iglesias, A., Wilf, P., Labandeira, C. C. &
Cúneo, N. R. Rapid recovery of Patagonian plant–insect associations aer the
end-Cretaceous extinction. Nat. Ecol. Evol. 1, 0012 (2016).
Competing interests
The authors declare no competing financial interests.