many species of specialist insect herbivores. Of course, different species may specialize by feeding on different plant organs,
or the different species may feed on the
plants at nonoverlapping times of year, but
even taking this into account, there appear
to be more consumers than distinct resource
But communities can also be structured
by the actions of higher trophic levels—
for example, the diseases, parasitoids, and
predators that afflict plant-eating insects.
Just as two consumer species cannot coexist in the long term on the same resource,
with the less efficient competitor going
extinct, so a natural enemy that attacks two
species will drive the less resilient to extinction. Similarly, just as species are predicted
to evolve to use unexploited resources, they
are also expected to evolve into “enemy-free
space”—ecological niches with few predators and parasitoids (7). This type of interaction between species mediated by shared
natural enemies has been termed “
apparent competition” (8) because its effects on
community structure are similar to those of
traditional resource competition. Experiments have shown that apparent competition
occurs in insect food webs, including in the
Condon et al. studied a type of fly
(belonging to the family Tephritidae) whose
larvae feed on the succulent parts of the
flowers of climbing squashes (cucurbits).
At their study site in Peru, two species of
squash grew together; each had morphologically distinct male and female flowers. The
fly larvae are attacked by tiny wasps, a few
millimeters in length, that lay their eggs in
young larvae. The eggs hatch and the wasp
larvae remain in a state of suspended development until their hosts become full grown,
when they resume growth and kill them.
During the developmental hiatus, the host
can fight back by mounting an immunologi-cal defense that, if successful, kills the internal parasitoid.
The first surprise about this commu-
nity was how many species were involved
and how specialized they were. The authors
reared 14 fly species, all in the same genus.
Most species of fly specialized on only one
of the four resource types—either a male
or female flower of the two species. Up to
11 species shared a single resource type. A
clue to how these species can coexist is pro-
vided by the 18 species of parasitoid wasps
that were reared. They include two species
that attack a broad range of hosts, but also a
group of closely related species—all belong-
ing to the same genus—that have very nar-
row host ranges. These wasps were each
restricted to develop in a single fly species,
even when several host types were present.
Such patterns are exactly what one expects if
resource competition and apparent competi-
tion are structuring the community.
Until recently, it has only been possible
to record the parasitoids that have over-
come host defenses, not those that have
succumbed to, or have been outcompeted
by, other parasitoid species. Using modern
molecular techniques, it is now possible to
detect the traces of unsuccessful parasitoid
attacks, allowing interaction webs rather
than trophic webs to be constructed (10).
Condon et al. do this for the first time in
a large tropical community. They find that
although most parasitoid species only suc-
cessfully develop in one host species, the
searching females lay eggs in other spe-
cies that feed on the same resource type but
are able to successfully defend themselves
against the parasitoid. This host specific-
ity is again consistent with structuring by
apparent competition, but it remains unclear
why females lay eggs in the wrong host.
Possibly, they cannot tell hosts apart; alter-
natively, laying eggs in the wrong host may
be worthwhile because the opportunity cost
of wasting an egg is low or because the off-
spring may survive in rare cases.
Condon et al. provide a wonderful example of a “society, where none intrudes.” The
web of interactions they reveal is complex
and raises fascinating hypotheses about
how different species influence each other’s
dynamics, both directly and indirectly. These
ideas need to be tested experimentally, and
this system, compared to most other tropical communities, may be relatively easy to
manipulate in the field.
1. M. A. Condon et al, Science 343, 1240 (2014).
2. A. J. Hamilton et al., Am. Nat. 176, 90 (2010).
3. J. Memmott, H. C. J. Godfray, I. D. Gauld, J. Anim. Ecol.
63, 521 (1994).
4. J. Hrcek, S. E. Miller, J. B. Whitfield, H. Shima,
V. Novotny, Oecologia 173, 521 (2013).
5. M. A. Smith et al., Proc. Natl. Acad. Sci. U. S. A. 105,
6. P. D. Colley, T. A. Kursar, Science 343, 35 (2014).
7. M. J. Jeffries, J. H. Lawton, Biol. J. Linn. Soc. London
23, 269 (1984).
8. R. D. Holt, Theor. Popul. Biol. 12, 197 (1977).
9. R. J. Morris, O. T. Lewis, H. C. J. Godfray, Nature 428, 310
10. R. Kaartinen, G. N. Stone, J. Hearn, K. Lohse, T. Roslin,
Ecol. Entomol. 35, 623 (2010).
Flowers Females Males
Flies of the genus Blepharoneura
Parasitoid wasps of the genus Bellopius
Other parasitoid wasps
Flowers of Gurania acuminata
Flowers of Gurania spinulosa
Intricate web. At their study site, Condon et al. collected 3636 flowers from two species of squash. A total of
1106 flies and 163 parasitoid wasps emerged from the flowers. In the resulting food web, the three trophic
levels are represented by rows of balls. Within each trophic level, the size of the balls and thickness of the
links are proportional to the abundance of species and the strength of feeding links, respectively. Most wasp
species can only develop in a single fly host species.