the Feet 0/ Cypselus inelba. 197 



fourth with five phalanges. Cypselus and also, according to 

 Sclater (P. Z. S. 1865, p. 596) and Forbes (Ibis, 1882, p. 390), 

 Panyptila so far differ from this plan that the third and fourth 

 toe have each only three phalanges, while the other Cypselidae 

 display the normal condition. The genus Cypselus thus 

 lacks one phalanx in the third toe and two in the fourth. 

 This reduction has been long known, but never more closely 

 investigated. It was to be expected that embryology would 

 lead to the required explanation, and I therefore undertook 

 the research, the more readily as material was easy to be 

 obtained in Bern, where, on the tower of the cathedral, 

 Cypselus melba nests. All the extremities were investi- 

 gated, and the following is a bi'ief statement of the result : — 

 In embryos of the 5th and 6th days the skeleton of the 

 foot shows a series of continuous and diverging bars of 

 cartilage, which could only be partially followed into the toes. 

 On the 8th day the appearances are changed. Not only are 

 the three principal divisions of the limb plainly separated, 

 but also to a certain extent the phalanges (see fig. 1, p. 198). 

 Metalarsale i. corresponds in position to about the middle of 

 M. ii., and bears the two phalanges of the first toe. M. n., 

 in., and iv. are connected proximally by the distal tarsal row, 

 distally they are free. Their phalanges are plainly recogniz- 

 able ; only the boundary between the last phalanx but one 

 and the one which bears the claw has quite vanished, and is 

 represented by a broad dark band; the distal end of the 

 claw-bearing phalanx is not sharply marked. On the other 

 hand an embryo of ten days shows an appearance (fig. 2) 

 which leaves little to be desired in the way of clearness. I 

 find the first toe with two, the second with three, and the third 

 and fourth with four phalanges each. These numbers are 

 repeated in the adult Pteroclidse and in most Caprimulgidse. 

 The difference from the normal bird's foot is now reduced to 

 a single phalanx missing in the fourth toe. I believe I have 

 also found this. Fig. 3 represents a condition which is slightly 

 younger than fig. 1 . In the second toe we see one, in the third 

 two, and in the fourth three separate phalanges. After these 

 there follows in each toe (the first is unfortunately injured) 



SEK. VI. — VOL. II. p 



