NATURAL SCIENCES OF PHILADELPHIA. 207 



may be observed. la the hisrbest forms, as Anatiche, the feet are brought 

 under the centre of equilibrium, and the body is consequently supported with 

 its long axis horizontal. Some Steganopods — the next order — have much 

 the same condition ; in others, as Cormorants, the legs are much further back, 

 and the body must be held almost upright. Most Longipennes are like Anatidse 

 in this respect, but here Halodromince- offer almost the other extreme. In 

 Pygopodous families the maximum of backward position of the legs and up- 

 right position of body is attained. A few or several Alcidce stand and walk 

 tolerably well, but all Colymhidce, Podicipidce and Spheniscidi^. must rest nearly 

 upright on the rump, and progress on land with extreme difficulty. Burial of 

 the thighs and more or less of the crura in the skin of the body is ordinarily 

 in direct ratio to backward ])osition of the legs, and in inverse proportion to 

 ease of walking. Nevertheless, some good walkers, as geese, have less free 

 legs than i-ertain small petrels that scarcely walk. As a rule, utility of the 

 feet as pedestals and as paddles is directly complementary. The most adroit 

 and untiring swimmers, and those that progress under as well as on the 

 water with ease, are the poorest walkers. Compare a goose with a loon, e.g. 

 In one order, Steganopods, we have the fact illustrated in contiguous fami- 

 lies ; pelicans do not follow their prey under water; cormorants do ; there 

 is a precisely corresponding angle of inclination of the body in each case. 

 To avoid interference of the broad- webbed feet the legs are widely separated ; 

 a divarication begins in the axis of the femur, and is increased in other seg- 

 ments. Hence the characteristic "waddle'' of ducks, &c. ; their gait, com- 

 pared with that of true walkers, corresponds somewhat to the difference that 

 may be observed in the sexes of our own species, arising from difference in 

 width of pelvis. There are many other interesting items in the construction 

 of a perfect pair of paddles, too numerous to note here. A loon's legs may 

 be taken as the type. Some time ago I worked out all the details, as seemed 

 to me, in this case, and may be permitted to give the reference, in place of 

 further observations.* 



The tarsus most commonly presents no special noteworthy features. Oc- 

 casionally it is extremely compressed, as among loons and grebes. Its horny 

 covering has nothing of the importance that attaches to it among Oscinc.t, for 

 example, though the two primary types of reticulation and scutellation some- 

 times mark natural groups, as geese as distinguished from ducks. The tarsal 

 envelope is fused in one genus. 



The toes may be three or four ; if the former, it is alwaj'S the hallux that 

 aborts. All have the normal numtier of phalanges, the relative lengths of 

 which usually, if not always, conform to a general rule. In comparative 

 length as wholes the hallux is always shortest ; the others usually run 2d, 4th, 

 3d in length, but not rarely '2d, 3d, 4th. There is often only a very slight 

 inequality in the 3d and 4th, exactly compensated by complementary differ- 

 ence in length of claw, so that the claw-tips fall together, e. g. many 

 Procellariidce. The hallux, as a rule, is present; its absence or rudimentary 

 condition marks one whole family, Alcidce^ two sub-families, DlomedeincE and 

 HalodromirKB, of Procellariidce, and one species of one genus {Rissa tridactj/laf) of 



* Osteology, Ac, of Coli/mlnis torquatus, Mem. Bost. Soc. Nat. Hist.. I, pt. ii, Nov..I8CG, p. 131. 



tin examinin.^ a great many Rissfe from the North Pacific, I have found a hitherto un- 

 noticed and probablv unsuspected state of things. Tliere are in that region two perfectly- 

 distinct species of Kittivvake. One is iii'ssu brack jrluincha, Gould, {^K. brfvirnslna, Bi'andt.) 

 with coral-red or yellow legs, yellow bill, &c., &e. This has a very small hallux : but still 

 one bearing a minute claw, that looks like a little blactc speck on the extremity of the 

 knob. The other is the R. kotzebni, Bonaparte, or at any rate the bird I identified and de- 

 scribed under that name some years since. Bonaparte's diagnosis is simply ^'■halluce magis 

 explicatn." Now this form is identical with and not tiistinguished from E. tridactyla. ex- 

 cept in and by the hind toe. The hallux is almost always larger and better formed ; and 

 it ranges from a state scarcely or not different from that of H. tridactyla, up to a perfectly 

 developed conditinn. claiv and all. In some specimens the hind toe and claw are a.'* large, 

 comparatively, as those of any if rws I ever saw! The distinctive feature of the genus Rixs i 

 has disappeared. What makes the North Pacific Kittiwakes vary thus, while the Atlantic 

 birds are constant, as far as Ijuowd ? 



1869.] 15 



