when the present data are viewed from several different 

 points. Some come promptly to mind ; according to modern 

 taxonomists, the megapodes are near relatives of the do- 

 mestic hen, and they should have, by this token, a length of 

 incubation somewhat similar to that of the hen, yet it is in 

 reality much longer, and far longer than one would predict, 

 from my point of view, with birds of the comparatively high 

 standing of the megapodes. Hence their incubation length 

 does not seem to fit in with this "ascent theory;" how it fits 

 in with the temperature side of the present explanation I do 

 not know, as I have no temperature records of these queer 

 birds. Wliat are the possibilities in the premise? These 

 birds may upset the whole argument, or they may be one of 

 a few exceptions, or there are two other possibilities. It has 

 been maintained in this discussion that the true length of 

 incubation is fixed by the parent's temperature as applied to 

 the eggs; thus this method of fixing cannot apply to the 

 megapodes because their eggs are hatched, not by heat sup- 

 plied by the parent bird, but by that arising in the decaying 

 mound, its temperature being given as 95.0 to 9G.0 F. (131), 

 which are the lowest successful incubation temperatures, for 

 birds' eggs, known to me. It might be assumed, in the first 

 place, that the mound incubation habit is very ancient, dat- 

 ing from the time when the megapodial ancestors were much 

 less removed from the pro-bird than at present, and that 

 the ancestral species had at that time, a comparatively low 

 temperature, approximating that of the present incubation 

 mound. If the mound building habit arose at that time, 

 and continued to the present, it would explain the inappro- 

 priately long incubation because, on this last assumption, it 

 has not varied during all the past ages, notwithstanding that 

 in the interim there have been great change and variation 

 in the morphology of the group. The eggs have, during all 

 this time, been hatched by a temperature which has varied 

 little, if at all, resulting in an unchanging incubation length. 

 On the contrary, it is possible that this mound incubation 

 habit is of recent origin, and that the present recorded meg- 

 apod length of incubation is an apparent one only, being the 

 true length plus X days of slowing down of embryonic de- 

 velopment caused by the low temperature of the mound; it 

 is difficult to picture just how, under this suggestion, the 

 eggs successfully became adapted to such abnormal pro- 

 longation of incubation. This might be tested experiment- 

 ally by subjecting megapode eggs to varying temperatures 

 in artificial incubators ; it also would give an added light if 

 one knew wdiat is the true length of incubation of the genus 

 Maleo, since eggs of this genus are said to be incubated, not 

 in mounds, but by the sun's heat or by hot springs. The 

 case of the megapodes falls without the confines of my 



71 



