DEVELOPMENT AND HEREDITY 1357 



at a fairly early stage. If in the two-celled stage of the frog's egg one 

 cell be destroyed by means of a hot wire, the other cell develops to 

 form half an embryo, thus suggesting that each cell of the two-celled 

 embryo could give rise only to the corresponding half of the body. 

 This limitation of development, however, only occurs if the intact cell 

 be left in connection with the cell that has been injured. If, in 

 echinoderm larvae, the cells be entirely separated, even as late as the 

 eight-celled stage of division, each cell will give rise to a whole embryo, 

 differing from a normal embryo only in respect of size. This difference 

 suggests that the number of divisions that each cell can undergo is 

 predetermined in the egg-cell itself, but shows also that the cells into 

 which the egg divides are, at first at any rate, equipotential. We must 

 assume therefore that the reason why one cell under ordinary circum- 

 stances only forms one half of the embryo is that its development is 

 regulated and determined by the presence of the other cell in connec- 

 tion with it forming the other half of the embryo. That is to say, the 

 development of the egg involves the reaction to environment of a 

 protoplasm of certain properties and powers of reaction. The final 

 product of development depends (1) on the nature of the protoplasm 

 (including the nucleus) of which the egg is composed, and (2) on the 

 environmental conditions to which the egg is subjected during the rapid 

 growth and multiplication attending its development. We could 

 therefore speak of a morphology of inheritance, but the morphology 

 would be ultra-microscopic and have relation to the molecular structure 

 of the protoplasm of which the egg was composed. 



In the transmission of the potentialities of development from 

 parent to fertilised egg we must regard the nucleus as the essential 

 structure. In ordinary development the spermatozoon furnishes 

 only a nucleus and centrosome, the ovum supplying the whole of the 

 cytoplasm. There seems, however, no grounds for assigning any 

 directive power to the latter structure. In echinoderm ova it is 

 possible to get rid of the nucleus and then by the introduction of 

 spermatozoa to have an individual entirely paternal in origin, which, 

 on development, produces a larva of the paternal character. In division 

 of the egg the only part of the cell which divides so that each daughter 

 cell shall include an equal part of both parental germs is the nucleus. 

 The constant number of the chromosomes in each species, and their 

 accurate division on mitosis, suggest that the hereditary transmission 

 of the potentialities of the cell is bound up with the chromosomes. 

 It has been suggested that every character is located in a chromosome 

 or part of a chromosome. If this be the case one might regard the 

 differentiation into various tissues, which occurs in the process of 

 development, as occasioned by an actual loss or degeneration of the 

 constituent parts of one or more chromosomes. There is no doubt 



