PHYLOGENY OP THE PAL^OGNATH^ AND NEOGNATH^l. 163 



the rhachis bears about 6 or 7 pairs of rami. The radii are slightly stouter, relatively. 

 As in Casuarius, fila are minute and rare. 



The nestling-down of Apteryx (PI. XLV. fig. 6) is easily distinguishable from that 

 of the remaining Palceognathce. Like that of Drommis and Casuarius, it has a 

 well-developed shaft, but it differs from these in its greater length and in the very 

 numerous rami. From Rhea and Struthio it differs entirely, in that, in these, the 

 neossoptiles are umbelliform. 



The most striking feature perhaps of the nestling-plumage of Apteryx is its great 

 length. In texture we appear to get three gradations. A. australis mantelli and 

 australis very closely resemble one another and differ from the remaining species in 

 that the feathers are coarser and slightly harsh to the touch. A. oweni and A. haasti 

 stand at the other extreme and are peculiarly soft. Some individuals of A. mistralis, 

 however, seem to stand between the two. 



A. haasti is the most markedly distinct of all. The feathers clothing the head, neck, 

 and thighs are of a peculiarly downy character, recalling, in texture, the down-feathers 

 on the breast of the adult Duck. On the back and breast they are of a peculiar 

 "woolly" texture, resembling very strongly the intermediate plumage or pseudo- 

 definitive feathers, (mesophyles of Palmer), which follow the prepenna? in the Owls. 



There is no trace of an aftershaft to these prepennse. 



The rami are closely set, very numerous (18 to 20 pairs), and moderately long. The 

 radii are relatively short, and have the appearance of being formed of a number of 

 short, laterally compressed joints, the dorsal and ventral angles of the anterior ends of 

 which are produced forwards to form a pair of minute fila. 



In the length of the shaft and the great number of the rami the prepenna of Apteryx 

 stands alone amongst the Palceognathce ; moreover it is further peculiar in that it is not 

 produced beyond the most distal radii into long filaments. 



A peculiarly interesting feature of the prepennse, and of the first definitive feathers, 

 is that the latter do not thrust out the former, as they appear to do in all other known 

 birds. Exactly how this is avoided is a point for further investigation. I hope to be 

 able to explain this mystery shortly 1 . That this process of thrusting-out is avoided 

 can be still seen in a downy nestling of A. haastii in the Eothschild Collection. Here 

 the first definitive feathers are about halfway through the skin, and show no trace 

 whatever of a prepenna surmounting the tips, as do the other Palceo- and ^Neognathm. 

 Furthermore, they are peculiar in that they are not invested by the sheath common to 

 all developing feathers. This appears to break away as soon as it reaches the surface 

 of the skin. 



1 It may possibly turn out that this nestling-down plumage may represent preplumse instead of prepennse -. 

 this would account for their non-ejectment by the teleoptiles. The fact that plumulae do not o.._.-.r in the 

 adult is of no importance. Zander [ioi] has described " pinsel-artig " down-feathers in the embryo lihea, but 

 this is probably an error. 



