1885.] MR. F. E. BRDOAKD ON THE CUCKOOS. 183 



branches; just before the point of separation of these two branches, 

 a row of feathers, at first wider, then narrowing to a single row, is 

 given off to the hypopteron ; the outer branch of the ventral tract 

 is short and only consists of a single row of feathers ; the inner 

 branch is at first three feathers wide, then diminishes to two, and 

 finally ends in a single row a little way in front of the cloaca. 



The spinal tract is, as usual, closely feathered and bifurcates 

 between the shoulder-blades ; it is not connected by any feathering 

 with the humeral tract, and is also separated by a more or less 

 complete break from the remainder of the spinal tract. 



The latter at its commencement is somewhat narrow on either 

 side, and is not in any way connected with the ventral feathering ; 

 posteriorly it widens out and is continuous with the femoral tract. 

 The femoral tract consists of about six rows of feathers parallel with 

 each other, and almost at right angles with the long axis of the 

 limb, separated by considerable intervals ; rather behind the point 

 of articulation of the femora with the pelvis, the two halves of the 

 spinal tract unite. 



It does not seem to me that Nitzsch has any grounds for separating, 

 as he does, this genus of Cuckoos to form a separate group. Phosni- 

 cophaes is evidently closely allied to Geococcyx, Guira, &c., in its 

 pterylosis. 



The arrangement of the feather- tracts in an example of a species 

 of Centropus from Celebes (see fig. 7, p. 18-1) seems to me to be 

 rather different from Nitzsch's figure of C. philippensis \ 



The ventral tract is particularly closely and strongly feathered in 

 the region of the throat, where the rows of feathers form a series of 

 chevrons, quite in a similar fashion to that which I have already 

 described in Cuculus ; the feathering, moreover, entirely fills up 

 the space between the two rami of the mandible. On a level 

 with the attachment of the humerus is given off the humeral tract, 

 which is also strongly and closely feathered ; it is about three 

 feathers wide, and at its commencement gives off a branch which 

 runs along the anterior margin of the patagium ; the latter is not 

 completely covered v/ith feathers, which are distributed much as 

 shown in the accompanying figure. Opposite the axilla the ventral 

 tract of either side divides into two. The inner branch is at first 

 two feathers wide, but just before its termination, a little way off 

 the cloacal aperture, is reduced to a single row. The outer branch 

 rapidly dwindles to a single row of feathers, and terminates some way 

 in front of the inner branch, without, however, showing any signs of 

 becoming fused with it. The ventral tract is therefore very different 

 from that of Centropus philippensis as figured by Nitzsch (tab. 

 et fig. cit.), where the two branches of either half of the tract are 

 of equal breadth and reunite just in front of the cloaca. 



The feathering on the head is continuous but rather scanty on 



the upper surface of the hesid. The spinal tract is very closely and 



strongly feathered, and is of a pyramidal form, gradually diminishing 



in breadth from before backwards ; at about the level of the junction 



' I/^c cit. pi. iv. fig8. 1,", 14. 



