444 INTRODUCTION TO CYTOLOGY 



germ and soma), or even by modifications in the surrounding soma; but 

 he denied that in the latter case the alteration would be of such a nature 

 as to cause the reappearance of the same somatic modifications in the 

 next generation. He accounted for the internally induced heritable 

 variations on the basis of his theory of germinal selection. He supposed 

 that the determinants, while multiplying in the germ-cells, are subject 

 to selection like all other organic units. Some determinants, being 

 better placed with respect to nutritive conditions, are favored thereby 

 and grow stronger and more influential, while others undergo changes in 

 the opposite direction. The parts of the organism receiving the deter- 

 minants which have had the advantage in the struggle become better 

 developed than those receiving the weaker determinants. As this process 

 continues from generation to generation, the new variation gradually 

 increases until it becomes pronounced enough to be laid hold of by 

 natural selection. 



A review of the points brought out in earlier chapters will show that 

 modern genetics, although it has shown the inadequacy of certain ele- 

 ments in the hypotheses sketched above, owes many of its leading ideas to 

 Weismann. These ideas were put to the test in the brilliant experimental 

 researches on development carried out by Roux, Boveri, R. and O. Hert- 

 wig, and others, and later in the researches of modern students of cyto- 

 genetics. Weismann's view that heredity is conditioned primarily by the 

 genetic continuity of germ-plasm has its modern expression in the 

 statement that a specific type of nuclear composition, upon which the 

 development of characters largely depends, is maintained by growth and 

 division from generation to generation. That this composition or organ- 

 ization is preserved only in germ-cells sharply distinct from somatic 

 cells is, however, a conception which no longer squares with the results of 

 experiment. It now appears that the complex of nuclear genetic units 

 is complete in both vegetative and reproductive cells, the differentiation 

 of cell types not being due to any sorting out of the units during onto- 

 genesis. Our knowledge in this field is little more than fragmentary, 

 but it seems likely that differentiation is largely a response to the diverse 

 conditions inevitably arising in a growing mass, such conditions stim- 

 ulating different groups of genes to action at different stages and in 

 different regions. The results of such action are, in turn, a part of the 

 cause of what follows. The gene, moreover, is now regarded as a con- 

 stitutional factor somehow involved with others in the metabolic proc- 

 esses and thus influencing the development of many characters, rather 

 than something standing for the characters of a complete organism or a 

 structural part thereof as did Weismann's id and determinant. 



The "germ-plasm," if this term is to be retained for the genes or 

 chromosomes, thus represents an integral part of a developing system 

 rather than an arbitrary determiner of development. For this reason 



