THE PTERYLOGEAPHY OF THE MEGAPODII. 489 



one important point. The former are true pre-pennse, the latter are somethinjr rliffor- 

 ing from, and intermediate between, pre-pennte on the one hand and definitive 

 contour feathers on the other. They represent apparently the adolescent plumage 

 which we noticed was assumed in the case of the Owls (12) during their first year; 

 but with this difference : in the Owl, these semiplumous feathers were preceded by 

 true neossoptiles, in the shape of pre-pennse ; in the young Megapode they are not : 

 for these feathers are probably only to be found ujion the embryo, as will be seen below. 



The above facts raise a most interesting question concerning the life history of 

 the Megapodes. For it may now be asked : If the plumage of the young Megapode 

 is not made up of Neossoptiles — or pre-pennse and pre-plumulte in the strict sense of 

 the word — but of a something intermediate between pre-pennsje and the definitive 

 contour feathers of the adult, is the development of these neossoptiles entirely sup- 

 pressed or rather entirely eliminated ontogenetically ? 



The material represented in this collection will I think afford an answer to this 

 question, but more specimens, and in a slightly better state of preservation, will be 

 necessaiy before it becomes absolutely convincing. For my own part, I have no 

 doubt at all, and would reply that the development of pre-pennse has been almost 

 totally eliminated, though transitory rudiments still make their appearance. 



These can well be studied in the ripe embryo, in which they appear as long 

 straw-coloured filaments depending from the tips of tiie remiges. 



Under the microscope, these filamentous processes prove to be the sheath of the 

 outer Malpighian layer, which invests the developing feather. Through this the 

 rudiments of rami can be plainly seen. These are apparently quickly disintegrated, but 

 the capsular sheath remains unbroken, until thrust out, upon the tip of the definitive 

 feather replacing it, and from which it is ultimately cast off. Normally, this sheath 

 would have been broken up and dispersed by the developing rami of the pre-penna, 

 which, later, would have been carried out on the tip of the definitive feather 

 destined to replace it. In the present case, as we have just seen, the rami never 

 attain strength — or bulk — sufficient to burst this investing sheath, which remains, for 

 a season, in the place of the pre-penna dissolved within it, attached to the tip of 

 the definitive feather, which has superseded it. 



In younger embryos, in that stage in which the feathers look not unlike much 

 elongated Porcupine quills, it will be noticed that each of these "quills" becomes 

 slightly constricted near the tip, swelling out again distad of the constriction, and 

 then tapering rapidly to a point (Fig. 7). I take it that this constriction occurs at 

 what was once the base of a pre-penna, and that the swollen, distal end of this 

 " quill " represents all that now remains of it. A comparison of Figs. 7 — 8 will make 

 this clearer. Fig. 7 represents the " quill " stage just referred to ; in this the 

 rami of the definitive feather seem clearly distinguishable from the fine rami that I 

 take to be remnants of pre-pennse. Fig. 8 is a later stage, drawn from one of 

 the "straw-coloured" filaments from the tip of a remex. Fig. 9 shows the pre-penna 

 of a Turkey attached to the distal end of the rami of a definitive feather. 



One of the most striking features of the ripe embryo, whereby it is seen to 

 differ at once from all known birds, is the great length of the remiges, and their 



