474 ON INCREASE IN SIZE [pt. iii 



is then plotted on double-log. paper against the weight of the adult 

 animal in kilos. Evidently the relation is also linear in this case, but 

 the interesting thing is that the smallest animals take far the longest 

 time to construct unit weight. Thus the mouse performs the feat of 

 producing a kilo of mice in 1 790 days while the elephant produces 

 a kilo of elephant in 0-16 day. This must be due to the fact that 

 contained in i kilo of mouse there is a great deal more organisation 

 and differentiation than in i kilo of elephant, in other words that 

 the degree of heterogeneity is greater. Whether all mammals can 

 make unit quantity of differentiation in the same time is a question 

 one would like to have answered, but which seems to be at the 

 present time unanswerable. 



So far, we have only considered the relation between adult weight 

 and gestation time. It would obviously be better to use birth- 

 weights for this purpose, but unfortunately only a few are known 

 (see Table 60 taken from Przibram). Nevertheless when the birth- 

 weight is plotted against the gestation time on double-log. paper, 

 a straight-line relation is found, except for a slight deviation in the 

 case of the heaviest animals ; this is shown in Fig. 68. The scattering 

 of the points is evidently considerable, but we may say that there is 

 some law which ensures that certain limits shall be held to. Thus if 

 an animal proposes to weigh 100 gm. at birth it must resign itself to 

 an incubation period of between 40 and 1 50 days, while if it is to 

 weigh I gm. it may be between 10 and 30 days in utero. Within 

 these wide limits individual species evidently have the power of 

 making drastic shortenings or lengthenings. 



Thus gestation time alone may not be a very fundamental constant. 

 In the first place there are great differences in degree of development at 

 birth between such animals as the pig on the one hand and the rat on 

 the other, the former being born almost ready to assume complete 

 motor control of its musculature, the latter by no means ready to 

 do so ; the former able to see, the latter blind ; the former covered 

 with hair, the latter hairless. Any relation between weight and 

 gestation time can therefore only be approximate, and the law 

 governing it must be, as it were, elastic. Again the difference 

 between polytocous and monotocous animals will make itself felt, 

 and the large differences between the relative weights of new- 

 born and mother. The following table (Table 61), which has 

 been constructed from Franck's information, shows how large 

 these are: 



