brief communications However, Mongolia was characterized in and Chaoyangia fall outside Ornithurae shore birds, are placed as derived forms the Late Cretaceous by extensive lakes, in Feduccia’s own work. The explicit within Aves8,9. Thus, if the ecologies that are possibly with marine connections, and by cladistic definition of Ornithurae (most basal to the crown clade are bracketed, no the Campanian it would perhaps be most recent common ancestor of Hesperornithi- support is found for such a bottleneck. accurate to describe it as a desert, as in formes plus Aves and all descendants2) Apsaravis, because of its phylogenetic South Africa, with a swampy inland delta8. is less inclusive than Feduccia’s more sub- placement, constrains the inference of the Small, possibly volant hesperornithiforms jective definition (taxa other than those ecologies of the most recent common and Presbyornis, a widespread wader with that are not ‘modern’ enough to be ancestor of the avian crown clade. We do webbed feet, have been found at nearby ornithurine). Feduccia’s ‘Ornithurae’ is not understand how an ornithurine with sites of about the same age9. predicated on the existence of a ‘Sauriurae’ no ‘shore bird’ morphologies, from a dune- I consider Apsaravis to have little to or the paraphyletic group that contains field10, can be interpreted as compatible contribute to our understanding of avian these primitive taxa. The fact that Apsaravis with Feduccia’s idea1 of ecological restric- evolution, and its lack of a clear relationship and our analyses add to the mounting tion of these taxa to shorelines and marine with any kind of modern bird makes its evidence3,4 against sauriurine monophyly environments. If Apsaravis can simply significance ambiguous. If Apsaravis is not has been overlooked in Feduccia’s estima- be assumed, without consideration of phy- related to any modern ornithurine, how can tion of the importance of Apsaravis. logenetic tests, to have flown from an it tell us anything important about the Other specimens that we did not consid- unknown nearby lake, then we do not see evolutionary questions raised by Norell er are problematic and underscore the how Feduccia’s hypothesis is testable. and Clarke? importance of well preserved and phyloge- Julia A. Clarke*, Mark A. Norell† Alan Feduccia netically placed taxa such as Apsaravis. *Department of Geology and Geophysics, Department of Biology, University of North Carolina, Feduccia did not include the fragmentary Yale University, New Haven, Chapel Hill, North Carolina 27599-3280, USA Otogornis, Ambiortus and Gansus in his Connecticut 06520-8109, USA e-mail: [email protected] analyses5. Although he claims that Chaoyan- e-mail: [email protected] 1. Norell, M. A. & Clarke, J. A. Nature 409, 181–184 (2001). gia possesses a “toothed skull”, the holotype †Division of Paleontology, American Museum 2. Feduccia, A. The Origin and Evolution of Birds 2nd edn actually consists only of a torso and partial of Natural History, 79th Street at Central Park (Yale Univ. Press, New Haven, Connecticut, 1999). hindlimbs. The “toothed skull” belongs to a West, New York, New York 10024-5192, USA 3. Hou, L.-H., Martin, L. D., Zhou, Z. H. & Feduccia, A. Science specimen once referred to as Chaoyangia5 274, 1164–1167 (1996). 1. Feduccia, A. The Origin and Evolution of Birds 2nd edn (Yale 4. Feduccia, A. Science 267, 637–638 (1995). but later identified as the holotype of Univ. Press, New Haven, Connecticut, 1999). 6 5. Dyke, G. J. & Mayr, G. Nature 399, 317–318 (1999). Songlingornis linghensis . This specimen can- 2. Chiappe, L. M. Alcheringa 15, 333–338 (1991). 6. Bleiweiss, R. Geology 26, 323–326 (1998). not be referred to Chaoyangia (as indeed it 3. Chiappe, L. M. Nature 378, 349–355 (1995). 4. Forster, C. A., Chiappe, L. M. & Sampson, S. D. Science 382, 7. Marshall, C. R. Geology 27, 95–96 (1999). has not been6) as no element known from 8. Shuvalov, V. F. in The Cretaceous Stratigraphy and 532–534 (1996). Palaeobiogeography of Mongolia (eds Benton, M. J. et al.) the holotype is also represented in the 5. Hou L.-H., Martin, L. D., Zhou, Z.-H. & Feduccia, A. Science 256–278 (Cambridge Univ. Press, Cambridge, referred specimen. 274, 1164–1167 (1996). 2000). Although we did not comment on the 6. Hou, L.-H. Mesozoic Birds of China (Taiwan Provincial Feng 9. Kurochkin, E. N. in The Cretaceous Stratigraphy Huang Ku Bird Park, Nan Tou, 1997). and Paleobiology of Mongolia (eds Benton, M. J. et al.) implications of Apsaravis for the timing of 7. Marshall, C. R. Geology 27, 95–96 (1999). 533–559 (Cambridge Univ. Press, Cambridge, 2000). the origin of Aves, Feduccia’s conjecture 8. Groth, J. G. & Barrowclough, G. F. Mol. Phylog. Evol. 12, that it cannot inform our understanding of 115–123 (1999). 9. Cracraft, J. & Clarke, J. A. in Perspectives on the Origin and this origin (because it is not part of an Early Evolution of Birds: Proceedings of the International extant lineage) is incompatible with his Symposium in Honor of John H. Ostrom (Peabody Mus. Nat. Norell and Clarke reply — Given that own arguments. In recounting the origin of Hist., New Haven, Connecticut, in the press). Feduccia has explicitly stated that there is Aves, he invokes taxa such as ichthyornithi- 10.Loope, D. B., Dingus, L., Swisher, C. C. & Minjin, C. Geology 26, 27–30 (1998). a near absence of ornithurine birds in forms and hesperornithiforms, which are Late Cretaceous continental deposits1 and not parts of extant lineages. Furthermore, it has speculated that ornithurines may have has been argued7 that gap analyses may be been more or less restricted to shoreline and consistent with Cretaceous or Tertiary retraction marine deposits during this time1, we do diversification of avian lineages, depending Furtive mating in female chimpanzees not believe that we misrepresented Feduc- on what model of diversification rate and Pascal Gagneux, David S. Woodruff & Christophe Boesch cia’s hypothesis. We reported the finding recovery potential is considered realistic. Nature 387, 358–359 (1997). of an almost complete skeleton of an Reasoning derived from phylogenetic In this genetic analysis of a community of chimpanzees in ornithurine from Late Cretaceous conti- analysis is a powerful way to test hypoth- the Taï forest, Côte d’Ivoire (carried out in 1994), we nental deposits, and do not see how this eses of relationships or the evolution of concluded that 7 out of 13 offspring were sired by males specimen could have no bearing on Feduc- morphology (for example, enantiornithine not found in the mother’s social group. Now a study of cia’s previous arguments. monophyly and novelties in the flight paternity using quantified and automated methods shows Feduccia comments that other Mesozoic apparatus). We used a phylogenetic test to that the incidence of extra-group paternity is much lower bird specimens are more, or just as, useful assess the idea that transitional ‘shore birds’ (1 out of 14 offspring; ref. 1). Direct comparison at the for tackling questions concerning the origin gave rise to all extant birds through an only satellite locus re-examined reveals that 10 out of 66 of extant bird lineages. Although all speci- ecological bottleneck1. alleles (15%) and 9 out of 33 individuals (27%) were mens contain some information, we If such a bottleneck occurred, then when inaccurately genotyped. Possible sources of error in the disagree with Feduccia’s current assertion ecology is bracketed phylogenetically for first study include allelic dropout in the amplification of that Apsaravis is simply one of a group living birds, ‘shore bird’ morphology and degraded DNA from field-collected samples of shed hair, of “abundant” ornithurine fossils. The ecology should be basal to the crown clade, inconsistent visual autoradiograph interpretation (stutter specimens he mentions are either not as well as in its nearest sister taxa. However, bands), contamination and sample mix-up. The new ornithurine or are so poorly preserved that virtually all molecular and morphological analysis confirms that extra-group paternity can occur in they have not shed much light on their own evidence places ‘land birds’ (tinamous, nature, but shows that the social community probably phylogenetic positions, let alone on broader ratites, galliforms and anseriforms, for corresponds to the reproductive unit in chimpanzees. 8,9 patterns of avian evolution. example) at the base of Aves . Charadri- 1. Vigilant, L., Hofreiter, M., Siedel, H. & Boesch, C. Proc. Natl The two “ornithurine” birds Liaoningornis iformes, the extant lineage referred to as Acad. Sci. USA 98, 12890–12895 (2001). 508 © 2001 Macmillan Magazines Ltd NATURE | VOL 414 | 29 NOVEMBER 2001 | www.nature.com.