44 / T. CUNNINGHAM [januaky 



has no bearing on natural selection at all. Of course I admit that some 

 of the crabs in Mr. Brook's experiments grew nearly the same amount 

 in the same number of moults, but the point is that others did not. 



Another curious point brought out by Brook's records is that, when 

 a given number of moults results in greater growth, the time occupied 

 by those moults may be more or less than in the case when the growth 

 was less. Thus, in the second of the above cases the larger crab took 

 less time for the same number of moults than the smaller crab, in the 

 former case the larger crab took longer time than the smaller for the 

 same number of moults. It is evident that the relation of the growth 

 of a crab to age and ecdysis is very complicated. 



Professor Weldon has however informed me privately that he finds 

 the abnormality of frontal breadth to be very nearly independent of the 

 growth-rate during a moult, but that the difference, such as it is, shows 

 that the crabs which grow most are on the whole very slightly narrower 

 after their moult than those which grow least, in comparison with those of 

 the same size in the sea. In consequence of this it has occurred to me 

 that perhaps my first suggestion was wrong, and that the effect of con- 

 ditions on growth is exactly opposite to that described above. It is 

 certain that both the frontal breadth and the carapace length are 

 growing, and that the carapace length grows faster, since the larger 

 crab has an absolutely greater but relatively narrower frontal breadth. 

 Now, if any circumstance such as temperature or food increases the 

 growth, it may affect the carapace length more than it affects the 

 frontal breadth. In that case the crab that grew most in one moult 

 would be the narrowest. For example, suppose that a crab has a 

 carapace length of 1000 units, and a frontal breadth of 800 units. 

 Suppose that under ordinary circumstances in two moults the carapace 

 length increases to 1050 and the frontal breadth to 820, then the crab 

 is of course larger and relatively narrower. Now, suppose that the 

 growth is so hastened that it reaches the same carapace length in one 

 moult. The effect on the growth of the frontal breadth may not be so 

 great, and thus this dimension may only increase in the one moult to 

 815 ; thus the crab of the same carapace length is narrower because it 

 has grown twice as fast. Perhaps this view of the law of growth is 

 more probable than the other. 



But it will be said this new suggestion could not fit the same facts 

 as the other. It would fit the same facts of observation, though of 

 course the supposed conditions of life must be different. The con- 

 ditions on this hypothesis were more favourable to growth in 1895 

 and 1898 than in 1893. In the experiment when the narrowest 

 fronted crabs of those of the same size survived, it would have been 

 because they were younger but more vigorous than those which had not 

 grown so fast, the rapidity of growth being a sign not of weakness but 

 of strength. In the moulting experiment the amount of growth would 

 on this hypothesis be less than in crabs on the shore, and therefore the 



