DOI: 10.1126/science.1155832 , 928 (2008);
320 Science
et al.
Marty A. Condon, of Its Parts
www.sciencemag.org (this information is current as of May 16, 2008 ):
The following resources related to this article are available online at
http://www.sciencemag.org/cgi/content/full/320/5878/928
version of this article at:
including high-resolution figures, can be found in the online
Updated information and services,http://www.sciencemag.org/cgi/content/full/320/5878/928/DC1
can be found at:
Supporting Online Material
http://www.sciencemag.org/cgi/content/full/320/5878/928#otherarticles
, 5 of which can be accessed for free:
cites 24 articles
This article
http://www.sciencemag.org/cgi/collection/evolution
Evolution
:
subject collectionsThis article appears in the following
http://www.sciencemag.org/about/permissions.dtl
in whole or in part can be found at:
this article
permission to reproduce
of this article or about obtaining
reprints
Information about obtaining
is a
Science2008 by the American Association for the Advancement of Science; all rights reserved. The title
Copyright American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
Scienceon May 16, 2008 www.sciencemag.orgDownloaded from
8. J. Keske, D. A. McWhorter, B. H. Pate,Int. Rev. Phys.
Chem.19, 363 (2000).
9. P. W. Anderson,J. Phys. Soc. Jpn.9, 316 (1954).
10. R. Kubo, K. Tomita,J. Phys. Soc. Jpn.9, 888 (1954).
11. G. G. Brown, B. C. Dian, K. O. Douglass, S. M. Geyer, B. H. Pate,J. Mol. Spectrosc.238, 200 (2006).
12. J. C. McGurk, T. G. Schmalz, W. H. Flygare,J. Chem. Phys.
60, 4181 (1974).
13. K. O. Douglasset al.,J. Mol. Spectrosc.239, 29 (2006).
14. M. J. Frischet al., Gaussian 03, revision C.02 (Gaussian, Inc., Wallingford, CT, 2004).
15. K. B. Wiberg, H. Castejon,J. Org. Chem.60, 6327 (1995).
16. H. N. Volltrauer, R. H. Schwendeman,J. Chem. Phys.54, 260 (1971).
17. J. R. Durig, F. Feng, T. S. Little, A. Wang,Struct. Chem.3, 417 (1992).
18. D. J. Nesbitt, R. W. Field,J. Phys. Chem.100, 12735 (1996).
19. S. Mukamel, J. Sue, A. Pandey,Chem. Phys. Lett.105, 134 (1984).
20. T. Baer, W. L. Hase,Unimolecular Reaction Dynamics: Theory and Experiments(Oxford Univ. Press, New York, 1996).
21. D. B. Borchardt, S. H. Bauer,J. Chem. Phys.85, 4980 (1986).
22. T. Baer, A. R. Potts,J. Phys. Chem. A104, 9397 (2000).
23. B. C. Dian, A. Longarte, T. S. Zwier,Science296, 2369 (2002).
24. R. Schanz, V. Botan, P. Hamm,J. Chem. Phys.122, 044509 (2005).
25. M. J. Davis, S. K. Gray,J. Chem. Phys.84, 5389 (1986).
26. S. K. Gray, S. A. Rice,J. Chem. Phys.86, 2020 (1987).
27. D. M. Leitner, P. G. Wolynes,Chem. Phys. Lett.280, 411 (1997).
28. J. Zheng, K. Kwak, J. Xie, M. D. Fayer,Science313, 1951 (2006).
29. H. S. Yoo, M. DeWitt, B. H. Pate,J. Phys. Chem. A108, 1348 (2004).
30. C. G. Elles, M. J. Cox, F. F. Crim,J. Chem. Phys.120, 6973 (2004).
31. R. von Benten, O. Link, B. Abel, D. Schwarzer,J. Phys. Chem. A 108, 363 (2004).
32. This work was supported through the Major Research Instrumentation program of the NSF (grant no.
0215957) and the chemistry division of the NSF (grant nos. CHE-0315961 and CHE-0616660).
Supporting Online Material
www.sciencemag.org/cgi/content/full/320/5878/924/DC1 Materials and Methods
SOM Text Figs. S1 to S5 Tables S1 to S3 References Movie S1 Data Set S1
28 January 2008; accepted 3 April 2008 10.1126/science.1155736
Hidden Neotropical Diversity: Greater Than the Sum of Its Parts
Marty A. Condon,1,2*Sonja J. Scheffer,2,3,4Matthew L. Lewis,3Susan M. Swensen5
The diversity of tropical herbivorous insects has been explained as a direct function of plant species diversity. Testing that explanation, we reared 2857 flies from flowers and seeds of 24 species of plants from 34 neotropical sites. Samples yielded 52 morphologically similar species of flies and documented highly conserved patterns of specificity to host taxa and host parts.
Widespread species of plants can support 13 species of flies. Within single populations of plants, we typically found one or more fly species specific to female flowers and multiple specialists on male flowers. We suggest that neotropical herbivorous insect diversity is not simply a function of plant taxonomic and architectural diversity, but also reflects the geographic distribution of hosts and the age and area of the neotropics.
T
he diversity of neotropical herbivorous insects, ranging in number from 3 million to 30 million species (1), has been hy- pothesized to be a function of plant diversity (2, 3), but the degree to which specialization shapes that function is contentious. Plant archi- tecture (4,5) and distribution also affect patterns of insect diversity (6, 7). Diversity estimates have traditionally been generated from counts of morphologically distinguishable insect species (morphospecies) collected on plant surfaces (2,3).However, molecular evidence suggests that tallies of morphospecies underestimate both diversity and host specificity (8, 9). Temperate zone re- search has revealed diverse assemblages of host- specific cryptic species as well as recently diverged host races (10), including groups that diversified to different parts (e.g., leaves, flowers, stems) of the same host plant (11). Although
different plant tissues represent numerous niches in the tropics, few studies have assessed the diversity of concealed larvae feeding inside those parts (12,13), and even fewer have used molecular markers to reveal cryptic species (14).
To address the relationship between host and insect diversity, we focused on Blepharoneura (15), a neotropical genus of tephritid fruit flies that, as larvae, feed within the flowers or fruits of plants in the cucumber family (Cucurbitaceae).
Blepharoneuralarvae rarely cause external signs of damage, and few host records existed prior to this study. More than half of the known host species of this group ofBlepharoneurabelong to the Guraniinae, a cucurbit subtribe characterized by architectural complexity. The two largest genera within the Guraniinae (Gurania and Psiguria) have brightly colored flowers with succulent outer floral organs (calyces), typical of many hummingbird-pollinated plants (fig. S1).
Most species have male and female flowers borne on sexually dimorphic branches, which are tem- porally and spatially isolated on individual plants (16). Because only large plants produce female branches, which produce fewer flowers for a shorter period of time than male inflorescences, female flowers are rarer than male flowers.
Thus, a population of a single host species rep- resents a mosaic of morphologically distinctive
targets differing in abundance at any point in time (fig. S2).
To investigate patterns of host use and di- versity inBlepharoneura, we reared 2857 flies from 24 different cucurbit host species in nine genera and three tribes. Our sample encompassed 10 distinct biogeographic neotropical regions (table S1) spanning the geographic distribution of the subtribe Guraniinae, from Mexico to southern Bolivia (~5500 km) and from the Pacific to the Atlantic coasts of South America (~3000 km). We analyzed 419 specimens from 34 sites in 10 countries. With a conservative 4%
sequence divergence cutoff for species limits (15), a phylogenetic analysis of mitochondrial cytochrome c oxidase subunit I (mtCOI) sequence revealed 52 species of flies (figs. S3 to S8). Most of these species were morphologically indis- tinguishable (15) but had sequence differences ranging from 6 to 18% (Fig. 1). Because diver- gent groups revealed by mtCOI can incorrectly identify species (17), we examined two nuclear genes from 58 specimens from the Napo region of Ecuador. Analysis of nuclear elongation factor 1-a(EF1-a) and CAD (15) recovered the same 10 lineages (fig. S9) and corroborated the mtDNA results (Fig. 1), which suggests that these lineages indeed represent distinct species. Sub- sequent morphological analyses of a subset of these genetically defined species revealed slight but statistically significant differences in mor- phology (18).
Cryptic species of Blepharoneura showed specificity both to host part and to host taxon: Of 45 species reared from reproductive tissues, only a single species (sp. 39) fed on both flowers and seeds, whereas all other species appeared to be restricted to either flowers or seeds. Among flower-feeders, most specialized on flowers of a single gender. This is surprising because most flower-feeders feed primarily on calyx tissue, which is similar in flowers of both sexes. Fur- thermore, many species specialized on female flowers, which are rare relative to male flowers (fig. S2) (15,16), in contrast to predictions that insects are less likely to specialize on rare hosts (19). Both seed- and flower-feeders tended to
1Department of Biology, Cornell College, Mount Vernon, IA 52314, USA.2Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, DC 20560, USA.3Systematic Entomology Laboratory, ARS-USDA, Beltsville, MD 20705, USA. 4Department of Entomology, University of Maryland, College Park, MD 20742, USA.
5Department of Biology, Ithaca College, Ithaca, NY 14850, USA.
*To whom correspondence should be addressed. E-mail:
on May 16, 2008 www.sciencemag.orgDownloaded from
B le. 55 M.P tri.
B le. 51 S .G S . B le. 251 S . G S . B le. 252 S . G S .
B le. 199 M. Geri.
B le. 195 M. Geri.
B le. 275 M. G S . B le. 95 F .G S . B le. 255 F .G S . B le. 60 F .G S . B le. 94 F .G S . B le. 254 F .G S . B le. 266 F .G S . B le. 267 M. G S . B le. 265 M. G S . B le. 256 F .G S . B le. 263 F .G S . B le. 172 M. G S . B le. 173 M. G S . B le. 264 F .G S . B le. 170 M. G S . B le. 268 M. G S . B le. 171 M. G S . B le. 272 F .G S . B le. 273 F .G S .
B le. 49 M.G eri.
B le. 200 F .Geri.
B le. 197 M. Geri.
B le. 198 M. Geri.
B le. 196 M. Geri.
B le. 253 F .G eri.
B le. 169 M. G S . B le. 93 M.G S . B le. 276 M. G S . B le. 58 M.G S . B le. 20 F .G S . B le. 277 M. G S . B le. 174 M. G S . B le. 176 M. G S . B le. 175 M. G S .
B le. 59 F .G S . B le. 19 F .G S . B le. 271 F .G S . B le. 269 F .G S . B le. 274 F .G S . B le. 262 M. G S . B le. 261 M. G S . B le. 257 M. G S . B le. 258 M. G S . B le. 280 M. G S . B le. 92 M.G S . B le. 18 M.G S . B le. 17 M.G S . B le. 278 M. G S . B le. 260 M. G S . B le. 259 M. G S . B le. 281 M. G S . B le. 270 M. G S . B le. 279 M. G S . B le. 282 M. G S .
sp. 40 sp. 38 sp. 14
sp. 30 sp. 12 sp. 11 sp. 13
sp. 8 sp. 10
sp. 4 Ecuador (Napo)
B le. 227 S . G S . B le. 323 S . G S .
B le. 225 M. caygl.
B le. 226 M. caygl.
B le. 236 M. Ga.
B le. 228 M. Ga . B le. 318 M. G a . B le. 229 M. Ga.
B le. 237 M. Ga.
B le. 247 M. G S . B le. 246 F .G S . B le. 239 M. G S . B le. 240 M. G S . B le. 316 M. G a .
B le. 232 F .Ga.
B le. 234 F .Ga.
B le. 235 F .Ga.
B le. 322 F .G a . B le. 244 F .G S .
B le. 245 F .G S . B le. 248 F .G S . B le. 233 F .Ga.
B le. 313 M. G a . B le. 320 M. G a . B le. 241 M. G S . B le. 238 M. G S . B le. 242 M. G S . B le. 243 F .G S . B le. 324 F .G S . B le. 325 F .G S . B le. 321 F .G a . B le. 315 M. G a . B le. 319 M. G a . B le. 231 M. Ga . B le. 230 M. Ga.
B le. 317 M. G a . B le. 314 M. G a .
sp. 42 sp. 28 sp. 30 sp. 21 sp. 10 sp. 2 sp. 4 sp. 3 sp. 38 Peru
B le. 381 S . E rac.
B le. 383 S . E rac.
B le. 384 S . E rac.
B le. 382 S . E rac.
B le. 386 S . C yclbr.
B le. 390 S . C yclbr.
B le. 389 S . C yclbr.
B le. 388 S . C yclbr.
B le. 391 S . C yclbr.
B le. 387 S . C yclbr.
B le. 392 S . C yclbr.
B le. 385 S . C yclbr.
B le. 395 M. P tern.
B le. 378 M. G a . B le. 361 F .G a . B le. 370 M. G a . B le. 364 M. G a.
B le. 354 F .G a . B le. 356 F .G a .
B le. 394 M. P tern.
B le. 393 F .P tern.
B le. 348 M. G S . B le. 340 M. G S . B le. 342 M. G S . B le. 346 M. G S .
B le. 362 M. G a . B le. 367 M. G a . B le. 366 M. G a . B le. 373 M. G a . B le. 365 M. G a . B le. 377 M. G a . B le. 375 M. G a . B le. 368 M. G a . B le. 345 M. G S . B le. 347 M. G S . B le. 341 M. G S . B le. 396 M. G S . B le. 343 M. G S . B le. 338 M. G S . B le. 337 M. G S . B le. 344 M. G S . B le. 399 M. G S .
B le. 349 F .G S . B le. 336 M. G S . B le. 339 M. G S . B le. 398 M. G S . B le. 397 M. G S .
B le. 380 M. G a . B le. 363 M. G a . B le. 379 M. G a . B le. 376 M. G a . B le. 372 M. G a . B le. 374 M. G a . B le. 369 M. G a . B le. 359 F .G a . B le. 360 F .G a . B le. 357 F .G a . B le. 350 F .G a . B le. 352 F .G a . B le. 358 F .G a . B le. 371 M. G a . B le. 353 F .G a . B le. 351 F .G a .
0.005 substitutions/site
sp. 28 sp. 8 sp. 10 sp. 4 sp. 2 sp. 1 sp. 24 sp. 22 sp. 30 sp. 21 sp. 20 sp. 36 sp. 35 Bolivia
B le. 77 M.G M.
B le. 45 M.G C L . B le. 88 M.G C L . B le. 101 M. G C L.
B le. 4 M.G C L . B B MC .28 M.G M.
B B MC .29 M.G M.
B B MC .11 M.G M.
B le. 25 M.G M.
B B MC .16 M.G M.
B B MC .15 M.G M.
B B MC .12 M.G M.
B B MC .14 M.G M.
B le. 26 M.G M.
B le. 78 M.P tri.
B le. 312 M. P tri.
B le. 64 M.P tri.
B le. 104 M. G C L.
B le. 106 M. G C L.
B le. 87 M.G C L . B le. 85 M.G C L . B le. 3 M.G C L . B le. 112 M. G C L.
B le. 40 M.G C L . B le. 102 M. G C L.
B le. 39 M.G C L . B le. 98 M.G C L . B le. 111 M. G C L.
B le. 44 M.G C L . B le. 48 M.G C L . B le. 86 M.G C L . B le. 99 M.G C L . B le. 100 M. G C L.
B B MC .8 F . G M.
B le. 108 F .G C L . B le. 110 F .G C L . B le. 34 F .G C L . B le. 43 F .G C L . B le. 36 M.G C L . B le. 33 F .G C L . B le. 109 F .G C L . B B MC .20 F . G C L.
B B MC .17 M.G M.
B B MC .9 F . G M.
B B MC .4 F . G M.
B B MC .23 F . G C L.
B B MC .18 F . G C L.
B le. 30 F .G M.
B B MC .19 F . G C L.
B le. 32 F .G M.
B le. 1 F . G C L.
B B MC .22 F . G C L.
B le. 29 F .G M.
B B MC .7 F . G M.
B B MC .5 F . G M.
B B MC .21 F . G C L.
B B MC .26 F . G C L.
B B MC .27 F . G C L.
B le. 31 F .G M.
B B MC .24 F . G C L.
B le. 2 F . G C L.
B B MC .25 F . G C L.
sp. 16 sp. 7 sp. 5 sp. 18
sp. 25
sp. 27 Costa Rica
B le. 207 S .C ped.
B le. 203 S . G S . B le. 204 S . G S . B le. 205 S . G S . B le. 206 S . G S . B le. 210 F .G S . B le. 214 F .G S . B le. 222 M. G S . B le. 220 M. G S . B le. 217 M. G S . B le. 223 M. G S . B le. 219 M. G S . B le. 209 F .G S . B le. 212 F .G S . B le. 221 M. G S . B le. 216 M. G S . B le. 213 F .G S .
B le. 201 M. geri.
B le. 202 M. geri.
B le. 224 M. G S . B le. 215 M. G S . B le. 218 M. G S . B le. 211 F .G S .
sp. 32 sp. 37 sp. 10 sp. 27 sp. 21 sp. 9 sp. 15 sp. 6 Ecuador (Chocó)
B le. 121 S . G S . B le. 136 S . G S . B le. 151 S . G S . B le. 152 S . G S . B le. 285 S . G ret.
B le. 284 S . G ret.
B le. 286 S . G ret.
B le. 283 S . G ret.
B le. 310 S . G a.
B le. 311 S . G a.
B le. 125 S . G a.
B le. 124 S . G a.
B le. 291 F .G s ub.
B le. 292 F .G s ub.
B le. 298 M. G s ub.
B le. 299 M. G s ub.
B le. 296 M. G s ub.
B le. 113 M. G s ub.
B le. 114 M. G s ub.
B le. 293 M. G s ub.
B le. 295 M. G s ub.
B le. 294 M. G s ub.
B le. 297 M.G s ub.
B le. 168 S . G s ub.
B le. 300 M. G s ub.
B le. 167 S . G s ub.
B le. 165 M. G S . B le. 153 M. G S .
B le. 307 M. G a.
B le. 308 M. G a.
B le. 117 F .G a.
B le. 118 F .G a.
B le. 304 F .G a.
B le. 288 M. G ret.
B le. 146 M. G ret.
B le. 290 M. G ret.
B le. 289 M. G ret.
B le. 287 M. G ret.
B le. 145 M. G ret.
B le. 250 M. G a.
B le. 115 M. G a.
B le. 249 M. G a.
B le. 309 M. G a.
B le. 128 F .G S . B le. 164 M. G S . B le. 154 M. G S . B le. 120 M. G a.
B le. 134 F .G S . B le. 133 F .G S .
B le. 156 F .G S . B le. 166 M. G S . B le. 161 M. G S . B le. 127 M. G S . B le. 158 M. G S . B le. 162 M. G S . B le. 159 M. G S .
B le. 303 F .G a.
B le. 132 F .G S . B le. 131 F .G S . B le. 135 F .G S . B le. 126 F .G S . B le. 130 F .G S . B le. 155 F .G S . B le. 129 F .G S . B le. 119 M. G a.
B le. 122 M. G a.
B le. 123 M. G a.
B le. 306 M. G a.
B le. 305 M. G a.
B le. 116 M. G a.
B le. 163 M. G S . B le. 160 M. G S . B le. 301 M. G a.
B le. 302 M. G a.
B le. 157 M. G S .
sp. 10 sp. 1 sp. 4 sp. 39
sp. 21 sp. 29
sp. 30 sp. 11 sp. 8 sp. 37 b
c a French Guiana
sp. 11
Echinopepon racemosus Gurania acuminata
G. costaricensis
G. reticulata
G. spinulosa
Psiguria ternata P. triphylla G. eriantha
G. makoyana Cayaponia glandulosa
Hosts
Cyclanthera spp. G. subumbellata
Fig. 1. Multiple sympatric species, often specific to flower gender, infest single species of hosts in diverse geographic areas. Neighbor-joining trees were constructed from mtCOI sequence data from specimens collected in 6 of the 10 biogeographic regions we sampled (15). Branch colors indicate host species (see key). Terminal labels indicate host part infested: red, female flowers; blue, male flowers; tan, seeds. Species are identified with numbers as in figs. S3 to S8; underlined identification numbers indicate species reared from more than one host species; letters (a, b, c) highlight
lineages within species 37 that would be recognized as distinct species if less conservative criteria were used to delineate species (15). See table S1 for locality data for transects (identified by italic letters and numbers): Trees denoted Costa Rica (C14), the Chocó region of northwestern Ecuador (E20), the Napo of eastern Ecuador (E19), and Peru (P31) were constructed from samples collected along single transects (each <20 km long with elevation variation of <400 m); trees denoted French Guiana (F22toF25) and Bolivia (B1toB7) were from samples collected along more than one transect.
on May 16, 2008 www.sciencemag.orgDownloaded from
specialize on a single species; 80% were reared from a single host plant species (15). The proportion of host taxon specialists did not
significantly change when we used less conserv- ative criteria for species delimitation (15). We also found, contrary to predictions (19), that the
rarer host plants in this study supported only host-taxon specialists and were not hosts to generalist species (Fig. 1).
Maximum likelihood analysis of the genetic data revealed distinct groups of seed- and flower- feeding species restricted to specific plant parts and plant taxa (Fig. 2). Two clades of flower- feeders associated with groups characterized by relatively large flowers and succulent tissues.
Species 1 to 30 were flower-feeders found on plants within the subtribe Guraniinae (98% boot- strap support). Alternatively, species 41 to 43 were restricted to the tribe Cucurbiteae (100%
bootstrap support). Within the seed-feeding clade (spp. 31 to 39), two strongly supported lineages appeared to be restricted to the subtribe Sicyeae (100% bootstrap support) and the Guraniinae (100% bootstrap support), respectively (Fig. 2).
Thus, the number ofBlepharoneuraspecies re- flects the diversity of different plant parts and host taxa but exceeds the sum of plant part types for some plant taxa, because multiple species of flies infest a single type of plant part. We observed that some host species were infested by at least 13 species of Blepharoneura. We typically found four to six species infesting flowers in a population of a single species of host (Fig. 1 and table S1).
Our data show that geographically wide- spread host plants supported more species, both locally and regionally, than did hosts with more limited distributions (Fig. 3), reflecting patterns observed in the temperate zone (6,7). The most abundant and widespread species ofGurania (G. spinulosa) was infested by 13 species of Blepharoneura(Figs. 1 to 3, table S1, and ap- pendix S1).Gurania acuminata, another wide- spread but locally rare species, hosted nine species ofBlepharoneura, four of which were reared from flowers collected on a single day from the same tangle of vines in Peru (spp. 2, 3, 21, and 28; Fig. 1). The close ecological asso- ciations of sympatric species feeding on the same host taxon and tissue were not unusual; even on less common hosts (e.g.,G. eriantha), we reared multiple species from single inflorescences (spp.
13 and 14; see Napo, Fig. 1).
More than half of theBlepharoneuraspecies we discovered were found at only one site and may have very limited ranges (Figs. 1 and 2, figs.
S3 to S8, and table S1). Species endemic to particular regions were not restricted to rare hosts or to hosts with limited geographic distributions.
Three of the 13 species infesting the widespread G. spinulosaappeared to be local endemics: One species appeared to be restricted to the Napo region of eastern Ecuador (sp. 12; Fig. 2 and fig.
S5), one to northwest Ecuador (sp. 9; Fig. 2 and fig. S5), and one to the southernmost limit of G. spinulosa’s distribution in Bolivia (sp. 24;
Fig. 2 and fig. S3). Most widespread species of Blepharoneurashowed a high degree of fidelity to host species and tissue, but we detected some geographic variation in host use patterns (e.g., spp. 27 and 30; Fig. 1 and figs. S6 and S7). For example, species 10 was distributed Fig. 2.Maximum likeli-
hood tree of 45 species of Blepharoneura. Maxi- mum likelihood analysis used a combined data set of the nuclear CAD and EF1-agenes and the mito- chondrial COI gene. Fly species, each represented by a single individual, are identified by numbers (figs.
S3 to S8; spp. 44 and 45 lack nuclear sequences and are not included here). Collection localities for each species are indi- cated by letters: A, Pan- ama; B, Bolivia; C, Costa Rica; E, eastern Ecuador;
F, French Guiana; G, Guy- ana; M, Mexico; P, Peru;
V, Venezuela; W, western Ecuador (table S1). Colors of branches and letters indicate most commonly used hosts in an area; col- ors of rectangles around letters indicate additional hosts at some sites. Flower color indicates host taxon:
red, Guraniinae; white, Cucurbiteae. Host nomen- clature is as in appen- dix S1.
Ble.372 B, F Ble.380 B, F, P Ble.321 P
Ble.25 A,C
Ble.45 C Ble.347 B, E, F
Ble.209 W Ble.269 B, E, F, P, V, W Ble.273 E, F, P
Ble.170 E Ble.200 E Ble.199 E Ble.202 W Ble.77 C
Ble.137 C Ble.64 C, M Ble.56 V Ble.395 B, P
Ble.370 B, F, P, V, W Ble.394 B, P Ble.89 C Ble.340 B
Ble.104 C
Ble.145 F
Ble.148 C Ble.207 W Ble.69 C Ble.6 C Ble.381 B Ble.387 B Ble.51 E, P, V Ble.168 F, G Ble.55 E
Ble.139 C Ble.225 P Ble.63 C, M Ble.141 stem.Sech.CR
Ble.416 B.furcifer 10 changes
Ble.373 B. P 100 56
98 99 89
83 100100
96 88 71 98
94 59
93
88
100 81 75 100
100
50 100
95
1 23 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Out1Out2 53
G. subumbellata G. tubulosa
Sicyeae
Cyclanthera Echinopepon G. acuminata G. costaricensis Gurania
HOST TAXA
G. reticulata G. spinulosa
P. racemosa P. ternata P. triphylla G. eriantha G. makoyana
Guraniinae
Cucurbita Polyclathra Psiguria
Cucurbiteae
Rytidostylis
Cayaponia
HOST PARTS
seeds flowers
FLY SPECIES ID
Fig. 3.Diversity ofBlepharoneura species per host species increases with number of transects. Host plants found in multiple transects host more species ofBlepharoneura than do rarer species and species with more limited distribution (r = 0.9337).
Colors (see key) identify hosts; tran- sect identification numbers are listed by country (table S1). All transects are <20 km long with elevation var- iation of <400 m.
4 2 0 0 2 4 6 8 10 12 14 16
6 8 10 12
Number of transects
Number ofBlepharoneura species per host species
(12 spp)
(B1, B6) (E19-21) (F22-25) (P31) (V34)
(B1-3, 8) (F25) (P31)
(C11,13,14)
(C10,14)
(B5-7) (B1,3) (P32)
(E19) (M28) (E19,20) (C14) (PA3)
(C18) (M29)
HOST COLOR CODE Gurania spinulosa G. acuminata G. costaricensis Psiguria triphylla G. makoyana G. eriantha Cyclanthera langei G. reticulata P. ternata Cucurbita sp.
G. subumbellata Echinopepon racemosus
(F22,25) (G26)
on May 16, 2008 www.sciencemag.orgDownloaded from
throughout tropical South America, fed exclu- sively on female flowers, and, in all but one lo- cality, fed on a single species of host (Fig. 1 and fig. S4). Other species (e.g., sp. 27) fed almost exclusively on female flowers (30 of 32 speci- mens) of at least two host species in Central America, but commonly fed on male (N= 4) and female (N= 4) flowers in areas west of the Andes in Ecuador (Fig. 1 and fig. S6). These variable patterns of host use form a mosaic that varies from community to community across large geographic areas (20) and complicates attempts to extrapolate local samples to global estimates of tropical diversity (21).
Although we report diversity exceeding the original morphological estimates by an order of magnitude (15), this must underrepresent the actual diversity of this group because our cri- terion for species delimitation is highly conserv- ative (15). This is because we used a 4% mtCOI divergence, whereas other studies recognize species differing by less than 1% (8). As a result of this conservative criterion, we may be lumping biologically distinct species together, and single generalist species may actually represent multiple host-specific species (e.g., sympatric mono- phyletic lineages feeding on separate hosts; see sp. 37 in French Guiana, Fig. 1 and fig. S8). Also, our samples are limited; most of our collections were made during single trips, and our samples were restricted to species in fruit or flower at that time (table S1). Finally, the number of fly species recorded for a particular host plant species was most likely limited because the number of insect species detected rose as the number of collection localities increased (Fig. 3).
We also found that the distribution of hosts may also predict herbivore diversity at both local and regional scales (6,7). The neotropics include a mosaic of biogeographic zones reflecting a long history of repeated habitat fragmentation (22).
During periods of habitat fragmentation, insect populations may be more likely than these plant populations to diverge, as insects have shorter generation times and can evolve more quickly than plants with long generation times (15). Fur- thermore, sexual selection accelerates rates of evolution in insects, particularly in groups with complex courtship displays such asBlepharoneura (9, 18, 23). When these new species come together, as habitats expand and host popula- tions rejoin, assemblages of highly host-specific cryptic species result. In local assemblages of Blepharoneura(Fig. 1), the minimum pairwise divergence among sympatric species is ~6%, which suggests that they diverged at least 2.6 million years ago (24). During the past 2.6 million years, even seemingly uniform habitats experienced multiple cycles of fragmentation and expansion (22). If host plants represent “hard boundaries”
(25) for ranges of host-specific insects, simple neutral models incorporating changes in habitat area (25) as well as time (26,27) could help ac- count for patterns of diversity. Conflicting as- sessments of host specificity and diversity in the
tropics (2,3,28) may reflect differences in geo- graphic scale rather than differences in evolu- tionary or ecological processes.
References and Notes
1. R. M. May,Philos. Trans. R. Soc. London Ser. B330, 293 (1990).
2. V. Novotnyet al.,Science313, 1115 (2006); published online 13 July 2006 (10.1126/science.1129237).
3. L. A. Dyeret al.,Nature448, 696 (2007).
4. I. S. Winkler, C. Mitter, inSpecialization, Speciation, and Radiation: The Evolutionary Biology of Herbivorous Insects, K. J. Tilmon, Ed. (Univ. of California Press, Berkeley, CA, 2008), pp. 240–263.
5. T. Huntet al.,Science318, 1913 (2007).
6. T. R. E. Southwood,J. Anim. Ecol.30, 1 (1961).
7. H. V. Cornell,Ecology66, 1247 (1985).
8. P. D. Hebert, E. H. Penton, J. H. Burns, D. H. Janzen, W. Hallwachs,Proc. Natl. Acad. Sci. U.S.A.101, 14812 (2004).
9. D. Bickfordet al.,Trends Ecol. Evol.22, 148 (2006).
10. J. O. Stireman, J. D. Nason, S. B. Heard,Evolution59, 2573 (2005).
11. J. B. Joy, B. J. Crespi,Evolution61, 784 (2007).
12. H. C. T. Godfray, O. T. Lewis, J. Memmott,Philos. Trans.
R. Soc. London Ser. B354, 1811 (1999).
13. V. Novotny, A. R. Clarke, R. A. I. Drew, S. Balagawi, B. Clifford,J. Trop. Ecol.21, 67 (2005).
14. M. A. Condon, G. J. Steck,Biol. J. Linn. Soc.60, 443 (1997).
15. See supporting material onScienceOnline.
16. M. A. Condon, L. E. Gilbert,Am. J. Bot.75, 875 (1988).
17. D. J. Funk, K. E. Omland,Annu. Rev. Ecol. Evol. Syst.34, 397 (2003).
18. M. A. Condonet al.,Biol. J. Linn. Soc.93, 779 (2008).
19. D. A. Norton, R. K. Didham,Science315, 1666b (2007).
20. J. N. Thompson,The Geographic Mosaic of Coevolution (Univ. of Chicago Press, Chicago, 2005).
21. C. D. Thomas,Nature347, 237 (1990).
22. J. Cracraft, R. O. Prum,Evolution42, 603 (1988).
23. T. C. Mendelson, K. L. Shaw,Nature433, 375 (2005).
24. We used Brower’s (29) calibration of 2.3% (pairwise divergence) MY-1, which is used as the standard mitochondrial molecular clock estimate (27).
25. R. K. Colwell, D. C. Lees,Trends Ecol. Evol.15, 70 (2000).
26. S. P. Hubbell,The Unified Neutral Theory of Biodiversity and Biogeography(Princeton Univ. Press, Princeton, NJ, 2001).
27. M. A. McPeek, J. M. Brown,Am. Nat.169, E97 (2007).
28. V. Novotnyet al.,Nature448, 692 (2007).
29. A. V. Z. Brower,Proc. Natl. Acad. Sci. U.S.A.91, 6491 (1994).
30. We thank many colleagues and students for assistance;
museums and governmental agencies in Bolivia, Costa Rica, Ecuador, French Guiana, Peru, the United States, and Venezuela for assistance and permission to carry out the study; and D. C. Adams, D. J. Futuyma, D. H. Feener, L. E. Gilbert, S. H. McKamey, C. Mitter, A. L. Norrbom, M. C. Singer, and I. S. Winkler for helpful comments on the manuscript. Supported by the Smithsonian Institution, NSF, Hofstra University, Ithaca College, and Cornell College. Sequences ofBlepharoneuraspecimens have been deposited in GenBank (accession numbers EF531751 to EF531769, EF531789 to EF531828, EF531890, EF531891, EU601764 to EU60230, and EU623470).
Supporting Online Material
www.sciencemag.org/cgi/content/full/320/5878/928/DC1 Materials and Methods
Figs. S1 to S10 Tables S1 and S2 Appendices S1 and S2
29 January 2008; accepted 9 April 2008 10.1126/science.1155832
Surface Tension Transport of Prey by Feeding Shorebirds:
The Capillary Ratchet
Manu Prakash,1David Quéré,2John W. M. Bush3
The variability of bird beak morphology reflects diverse foraging strategies. One such feeding mechanism in shorebirds involves surface tension–induced transport of prey in millimetric droplets: By repeatedly opening and closing its beak in a tweezering motion, the bird moves the drop from the tip of its beak to its mouth in a stepwise ratcheting fashion. We have analyzed the subtle physical mechanism responsible for drop transport and demonstrated experimentally that the beak geometry and the dynamics of tweezering may be tuned to optimize transport efficiency. We also highlight the critical dependence of the capillary ratchet on the beak's wetting properties, thus making clear the vulnerability of capillary feeders to surface pollutants.
P
halaropes (Fig. 1A) and several other shorebirds with long thin beaks feed primarily on small crustaceans and other invertebrates (1). By swimming in a tight circle on the water surface, they generate a vortex that draws underlying fluid and suspended prey toward the surface (2). By pecking on the water surface at a rate of ~1.5 Hz (1,3–6), the birds capture water droplets with a characteristic scale of ~2 mm between their upper and lower man- dibles (movie S1). Suction cannot be used to raise the drops mouthward because of the geom- etry of the open beak; gravity acts to oppose thedrop motion. Nevertheless, the birds succeed in raising the drops mouthward by opening and closing their beaks successively (1,5,7,8). Al- though the importance of surface tension in this process was inferred (1), the physical mechanism responsible for the droplet transport, specifically
1Center for Bits and Atoms, Massachusetts Institute of Technology (MIT), 20 Ames Street, Cambridge, MA 02139, USA. 2Physique et Mécanique des Milieux Hétérogènes, UMR 7636 du CNRS, Ecole Supérieure de Physique et de Chimie Industrielles, 10 rue Vauquelin, 75005 Paris, France.
3Department of Mathematics, MIT, 77 Massachusetts Avenue, Cambridge, MA 02139, USA.
on May 16, 2008 www.sciencemag.orgDownloaded from