host choice by the wasps. Comparison of pre- and
postemergence samples also reveals that wasps
only oviposit into the “wrong” fly species when
their primary host is found on the same flowering
branch (table S9) (Fisher’s exact test, P = 0.0155),
which implies that wasps can identify that their
host fly is present on a plant but can less accurately discriminate among individual larvae feeding within flowers.
The Bellopius pattern contrasts sharply with
patterns of specificity shown by other parasitoids
reared from Blepharoneura. For example, the two
species of Thiemanastrepha reared from these
Blepharoneura are generalists and attack 10 species of flies in all four plant-defined niches (Fig. 3);
at least seven species of Blepharoneura are vulnerable
to the most abundant species of Thiemanastrepha
(Fig. 3). The single figitid species (Tropideucoila
blepharoneurae) found in this study also attacked
10 different fly species (Fig. 3).
Insects have a well-developed immune sys-
tem, and the defenses of larval flies against their
internal parasitoids are particularly well-studied
(12–14). Blepharoneura flies’ defenses against
Bellopius parasitoids represent a “hidden” niche
dimension revealed through analysis of preemer-
gence puparia (Figs. 3 and 4). In this tropical sys-
tem, specialized parasitoids’ offspring die in the
“wrong” species of fly. Thus, selection favors para-
sitoids able to discriminate among multiple sym-
patric species of flies infesting the same host plants.
Fidelity to host-plant parts and host-plant species
increases Bellopius parasitoids’ chances of detect-
ing the “correct” host-fly species, but such fidelity
also provides opportunities for flies to escape (Fig.
4): Flies on alternate host-plant parts (or alternate
host-plant species) escape detection by their spe-
cialized lethal parasitoid(s), which oviposit primar-
ily into a particular part and species of plant (Figs.
3 and 4). Yet, in this system, “escape” is not the
only defense against parasitoids: Inhospitable flies
are lethal to parasitoids.
Instead of representing an example of extreme
niche overlap (15), this highly diverse community of 14 fly species and 14 Bellopius parasitoid
species, all occupying flowers of just two species
of plants, represents a community with nonoverlapping niches—each distinguished by lethal interactions between parasitoids and vulnerable flies
and between parasitoids and inhospitable (lethal)
flies (Figs. 3 and 4). Highly specific virulence
(lethality) in both parasitoids and their prey adds
a previously hidden but highly dynamic dimension to analyses of tritrophic interactions, which
typically focus on unidirectional cascading patterns of escape via shifts in host plant use (10, 11, 17).
Discovery of such insect parasitoid-prey interactions in complex communities of cryptic species
depends on molecular methods to expose morphologically cryptic species and reveal interactions
(15, 19, 20). Future studies using molecular genetic
methods to identify species in other communities
throughout the Neotropics are likely to reveal a
geographic mosaic (21) of highly diverse and
dynamic interactions driven by “escape and radiate” mechanisms not only at the level of host plant
use, but also at the level of virulent interactions.
References and Notes
1. E. O. Wilson, The Diversity of Life (Norton, New York,
2. V. Novotny et al., Science 313, 1115–1118
3. Y. Basset et al., Science 338, 1481–1484 (2012).
4. L. A. Dyer et al., Nature 448, 696–699 (2007).
5. C. Darwin, On the Origin of Species by Means of
Natural Selection or the Preservation of Favored
Races in the Struggle for Life (J. Murray, London,
6. D. Schluter, The Ecology of Adaptive Radiation (Oxford
Univ. Press, Oxford, 2000).
7. D. R. Strong, J. H. Lawton, R. Southwood, Insects on
Plants: Community Patterns and Mechanisms (Harvard
Univ. Press, Cambridge, MA, 1984).
8. P. R. Ehrlich, P. H. Raven, Evolution 18, 586–608
9. P. R. Price et al., Annu. Rev. Ecol. Syst. 11, 41–65
10. J. O. Stireman III, J. D. Nason, S. B. Heard, J. M. Seehawer,
Proc. Biol. Sci. 273, 523–530 (2006).
11. A. A. Forbes, T. H. Q. Powell, L. L. Stelinski, J. J. Smith,
J. L. Feder, Science 323, 776–779 (2009).
12. B. Lemaitre, J. Hoffmann, Annu. Rev. Immunol. 25,
13. A. R. Kraaijeveld, H. C. J. Godfray, Am. Nat. 153 (s5),
Fig. 4. Tritrophic community and “hidden” lethal interactions: plant hosts (basal pie chart, N =
729 host flowers) yielding adult flies and parasitoids. Stacked bars (columns) represent numbers of
reared adults: bottom of bars, flies designated by colors; middle, generalist parasitoids (dark gray); and top
of bar (white), lethal Bellopius species (red letters). (Small letters indicate rare Bellopius.) Arrows from
Bellopius parasitoids indicate host flies from which adult wasps never emerged.