Department of Biology, California State University, Long Beach, California 90840 USA Currently, several foot types, based on the number and arrangement of the toes, are recognized in birds. One of these types, the pamprodactyl foot, is char- acterized as having all four toes ordinarily directed forward (or capable of being turned forward) (Lands- borough Thompson 1964, Van Tyne and Berger 1976, Raikow in press). As noted by Newton (1896: 972), however, "earlier ornithologists, having no better characteristics on which to rely, attached to the struc- ture of the toes a value out of all proportion to their real taxonomic importance and thus a superabun- dance of technical terms was created." Often the def- inition of terms was based on the appearance of mu- seum specimens and not living birds. Pamprodactyly is a clear example of this, and the mouse birds (Co- liidae) and some of the swifts (Apodidae) are among those prominently, but incorrectly, mentioned as ex- hibiting this type of foot. Pamprodactyly does not describe the observed action of the toes of swifts and colies in life, and it obscures the recognition of a grasping mechanism showing a remarkable conver- gence toward that found in two other classes of ver- tebrates. Although the term pamprodactyl was coined for the Coliidae (Murie 1872, 1873: 190), recent studies have shown that these birds have a very adaptable, flexible foot and can "alter the arrangement of their toes to suit the functional demands of the particular mode of locomotion employed at any time" (Bock and Miller 1959). In my observations of Colius striatus and C. macrourus in captivity, I have rarely seen them use a configuration approaching pamprodactyly; I ex- clude instances when they hung by one or two toes, with the remaining toes oriented forward but not involved in grasping. On occasion, they utilized a toe configuration similar to the lateral grasping pat- tern described below. The royology of the hindlimb and the grasping patterns in the Coliidae have re- cently been reviewed by Bertnan and Raikow (1982). All swifts of the subfamily Chaeturinae (Peters 1940), including the Cypseloidinae (Brooke 1970), have a typical anisodactyl foot (hallux directed pos- teriorly, toes II, III, and IV directed anteriorly). The Apodinae (Peters 1940), comprising the genera Apus, Cypsiurus, Tachornis, Reinarda, Micropanyptila, Aero- nautes, and Panyptila, are generally thought to be pamprodactylous. In the course of my field studies of swifts (Collins 1968, 1973a, 1980a, b), I have han- dled numerous living individuals of one to four species in each of five of these seven genera. In ad- dition, I have observed House Swifts (Apus affinis) on a daily basis from hatching to fledging (Collins 1973b). The grasping mechanism in all of these swifts is con- sistent in that toes I and II (the hallux and innermost toe) are spread medially, together or slightly apart, and oppose toes III and IV (the central and outermost toes), which are spread laterally. This arrangement forms a laterally oriented, pincer-like grasp (Fig. 1) between the two pairs of toes. Harteft (1892) noted that the toes of some swifts (Tachornis, Cypsiurus, and Reinarda) occurred in opposing pairs. Ingram (1955, 1972) and Lack (1956) also noted this condition in the newly hatched Apus apus but considered it to be the zygodactyl condition and only typical of young birds; neither of these contentions is supported by the observations presented here. The number of phalanges is reduced in toes III and IV of the Apodinae (Sclater 1865) through the fusion of elements (Zehntner 1890), making all the toes more equal in length, which would seem requisite for ef- ficient lateral grasping of the type observed. Only when the toes fail to gain purchase, as on a smooth hard surface, and the swift begins to slide downward do they tend to assume the pamprodactylous condi- tion so widely attributed to them; this toe position is also assumed in the relaxed foot and thus easily seen in museum specimens. The lateral grasping action seems particularly well adapted to holding on to the E3 ^ G Fig. 1. Convergent lateral grasping mechanism in climbing vertebrates; (A) nestling House Swift, Apus affinis (Aves: Apodidae); (B) left and right forelimbs of Jackson's chameleon, Chanaeleo jacksoni (Reptilia: Chameleonidae); (C) right forelimb of koala, Phascolarctos cinereus (Mammalia: Phascolarctidae). soft material, mostly plant floss and feathers, that makes up the nest of these swifts or to the palm- frond nest substrate of the four genera of palm swifts. Some of these swifts also roost in or on the nests in the nonbreeding season; none is known to utilize perch or roost substrates where lateral grasping would be at a disadvantage. Other birds mentioned as having a pamprodactyl foot are some nightjars, including Steatornis (Ingram 1958), and the parrots of the genus Micropsitta (Bock and Miller 1959). Recent observations have shown that pamprodactyly is not typical of either Steatornis (Bock and Miller 1959) or Micropsitta (R. Orenstein pers. comm.). That any birds predominently or even regularly use this toe configuration in the wild should be considered doubtful. Zygodactyl and heterodacty! in the ornithological literature have more restricted definitions (Lands- borough Thompson 1964, Van Tyne and Berger 1976, Raikow in press) than simply the "yoke-toed" con- dition noted for reptiles and mammals. In the zygo- dactyl foot of birds, toes I and IV oppose 11 and III; in the heterodactyl foot, toes I and II oppose III and IV. Thus, the type of foot described here for the ^po- dinae corresponds in toe arrangement (toes [ and II opposing III and IV), but not orientation (lateral rath- er than anterior-posterior), to the heterodactyl foot of trogons (Trogonidae). A similar, laterally oriented, grasping mechanism can be seen in the "specialized 'zygodactylous' grasp- ing feet" (Romer 1956) of the chameleons (Fig. 1) (Reptilia: Chameleonidae) and the forelimbs of nu- merous species of the phalangeroid mammals (Vaughn 1972), particularly the koala (Phascolarctos cinereus; Degabriele 1980) (Fig. 1). The Chameleoni- dae and phalangeroid mammals have pentadactylous feet. In the chameleonid manus, toes I, II, and [II oppose IV and V, while in the pes toes [ and I1 op- pose II1, IV, and V. In the manus of Phascolarctos (De- gabriele 1980), toes I and II oppose III, IV, and V, while in the pes toe I opposes II-V, with II and III being syndactyl. This lateral grasping mechanism appears to be a particularly clear but overlooked case of convergence among these three classes of climbing vertebrates. It is doubtful that a single name can be usefully applied to this type of grasping foot. The further elucidation of the underlying tendon and muscle arrangements of the respective groups in which this grasping mechanism appears is a fertile area for further study. My field studies have been generously supported by the F. M. Chapman Fund, American Museum of Natural History, the California State University (Long Beach) Foundation, and Cyril K. Collins. The study of Apus affinis was made possible by a Senior Ful- bright Research Fellowship to India and the kind- ness of R. M. Naik. S. L. Warter, G. L. Callison, D. G. Huckaby, and R. J. Raikow made helpful comments on earlier versions of this paper. LITERATURE CITED BERMAN, S. L., & R. J. RAIKOW. 1982. The hindlimb musculature of the mousebirds (Coliiformes). Auk 99: 41-57. BOCK, W., & W. DEW. MILLER. 1959. The scansorial foot of the woodpeckers, with comments on the evolution of perching and climbing feet in birds. Amer. Mus. Novitates 1931: 1-45. BROOKE, R. K. 1970. Taxonomic and evolutionary notes on the subfamilies, tribes, genera and sub- genera of the swifts (Aves: Apodidae). Durban Mus. Novitates 9: 13-24. COLLINS, C. T. 1968. The comparative biology of two species of swifts in Trinidad, West Indies. Bull. Florida State Mus. 11: 257-320. 1973a. Notes on survival and band wear in White-throated Swifts. Western Bird Bander 48: 20-21. 1973b. Development of temperature regu- lation in the House Swift. Pavo 11: 1-11. 1980a. The biology of the Spot-fronted Swift in Venezuela. Amer. Birds 34: 852-855. --. 1980b. Notes on the food of the Horus Swift Apus horus in Kenya. Scopus 4: 10-13. DEGABRIEL[, R. 1980. The physiology of the koala. Sci. Amer. 243: 110-117. HARTERT, E. 1892. Catalogue of birds of the British Museum, vol. 16. London, Trustees of the Brit. Mus. INGRAM, C. 1955. The foot of the young swift Apus apus. Ibis 97: 149-150. 1958. Notes on habits and structure of the Guacharo Steatornis caripensis. Ibis 100: 113-119. 1972. The feet of young swifts Apus apus, caller and affinis. Bull. Brit. Ornithol. Club 92: 96. LACK, D. 1956. A review of the genera and nesting habits of swifts. Auk 73: 1-36. LANDSBOROUGH THOMPSON, A. 1964. A new diction- ary of birds. London, T. Nelson and Sons. Mum[, J. 1872. On the genus Colius, its structure and systematic place. Ibis 3 (3rd ser.): 262-280. 1873. On the Upupidae and their relation- ships. Ibis 4 (3rd ser.): 181-208. NEWTON, A. 1896. A dictionary of birds. London, A. & C. Black. PET[RS, J. L. 1940. Checklist of birds of the world, vol. 4. Cambridge, Harvard Univ. Press. RAIKOW, R. J. In press. The locomotor system. In Form and function in birds, vol. 3 (A. S. King and J. McClelland, Eds.). London, Academic Press. ROMER, A. S. 1956. Osteology of the reptiles. Chi- cago, Univ. Chicago Press. SCLATER, P.L. 1865. Notes on the genera and species of Cypselidae. Proc. Zool. Soc. (London) 39: 593- 617. VAN TYNE, J., & A. J. BERGER. 1976. Fundamentals of ornithology. New York, J. Wiley and Sons. VAUGHN, T.A. 1972. Mammalogy. Philadelphia, W. B. Saunders Co. ZEHNTNER, L. 1890. On the development of the feet of Cypselus melba. Ibis 1890: 196-200. Received 15 October 1982, accepted 21 March 1983.