272 Proceedings of the Royal Society of Victoria. 



snakes, turtles and alligators, this belief is fairly well borne out. 

 i5ut, as the unreliable nature of these figures prevents more than 

 a sort of prima facie evidence, let me pass by preference to 

 others in which there is more accuracy, though still the observa- 

 tions are often merely approximate. 



All birds and mammals, except the monotremata, and, as I 

 shall show in a future paper, the marsupiata, keep at a tempera- 

 ture which may, for the purposes of this enquiry, be considered 

 constant, so that in the following investigation we may neglect 

 temperature variations, as the figures to be dealt with are not 

 accurate enough to allow of refined adjustments. Excluding the 

 monotremata and marsupiata, the extremes of health tempera- 

 ture for birds and mammals would be 37° C and 43° C, or a 

 I'ange of only 6° C. We may therefore assume that all bii"ds 

 sitting on their eggs keep them at a tolerably definite tempera- 

 ture. Any given species, therefore, will take a certain fairly 

 <lefinite time to hatch out its eggs. Temperature, we know, 

 counts for something ; a set of hen, duck or turkey eggs placed 

 in a warm dry situation will hatch out two or three days before 

 .■mother set in a damp cold place. But, in view of the roughness 

 of the observations of naturalists, we shall assume that each 

 .species takes a tolerably definite time in hatching, the hen for 

 instance, 21 days, and the turkey, 28. 



What, then, is the reason for the difference in time, seeing 

 that in all cases the temperatures are much the .same 1 Why 

 does a humming-bird take 10 days, or a wren 10, wliile a dove 

 takes 18, a fowl 21, a turkey, 28, an ostrich about 50? St. 

 Oeorge Mivart says: "The period of incubation is much related 

 to the size of the bird." I propose in this paper to determine the 

 nature of that relation, and to show that the time of incubation 

 is directly proportional to the sixth root of the weight of the bird 

 when mature. 



The following preliminary table will .serve to illustrate tiiis 

 relation and show that / = « "/w, where / =-- time in days. 



"io = weight in 11)S. 

 n = 20. 



