EVOLUTION IN A DETERMINATE LINE. I I I 



must accordingly perish. 1 In other words, in the first stage of 

 variation, assuming again variations I to 1,000 in egg and in 

 capsule, the chances for coincidence of even passable variations 

 would be so small that a vast proportion of the embryos would 

 fail to mature. And of the small number which did survive the 

 first selection the majority would evidently be eliminated in the 

 second stage of selection. For one egg, in short, to have run 

 the gauntlet of the favorable variations in the capsule there must 

 have been, and must still be an enormous proportion of failures. 2 

 The latter is evidently false, for it was found that the eggs of 

 Chimara are about as hardy and fruitful as the eggs of other 

 sharklike fishes. Morever, if it is necessary to show still further 

 the insufficiency of this selectionist argument, one need only re- 

 flect that nature would have to support millions of adult Ctiinuzra 

 in order that there might be laid one golden egg ! The race, in 

 short, could at the best have existed but a few generations in 

 view of its slowness of breeding and the small number of eggs 

 (possibly a dozen a year) laid by each female. But this infer- 

 ence is palpably false, for we know that the race of chim?eroids 

 has been producing specialized egg capsules from Jurassic times. 



Natural selection of fortuitous variations is, accordingly, clearly 

 valueless in explaining the evolution of the present capsule. Use 

 and disuse also are to be eliminated as factors in its development. 

 If a neo-Lamarckian suggests that the capsule is a result of use, 

 having been primitively moulded /// ntcro around the full grown 

 young, one need only reply that such a suggestion is in itself 

 anti-Lamarckian, for why would a capsule, let alone a complicated 

 one, be formed at all when the embryo is so fully grown that it 

 cannot need it, and \\hy should elaborate provision for aquatic 

 breathing be present in a uterine capsule ? One need hardly offer 



1 Thus, the proper variation for theopercular lid can only follow the favorable vari- 

 ation for the respiratory pores in the capsule, and these again can only follow if the 

 capsule is of a great size, definite shape, or durable structure. And, of course, as 

 definite an order must be followed in the stages of the embryo. And finally, the 

 accurate regulation of time between the two must coincide. Thus the breathing 

 pores of the shell must not open too early or too late. 



2 Admitting that "favorable," rather than optimum variations could have been 

 successfully selected. Even organic selection, it seems to me, reduces the difficulty 

 only in unessential regards. 



