510 



THE AMEBIC AN NATURALIST [Vol. XLIV 



Bud variations would cerlainly arise if Mendelian segregation 

 occurred in somatic division, hut they would also occur if for 

 any reason a dominant character should become latent, It cer- 

 tainly yet remains to be demonstrated that they arise from 

 Mendelian segregation in somatic tissues. 



Kastle lias recently published a very important paper in which 

 he summarizes the results of investigations relating to oxidases 

 and related compounds. 4 The paper is too extensive to permit 

 of an adequate review at the present time. It has, however, 

 a very important bearing on many Mendelian phenomena, and 

 those who are interested in this subject will enjoy reading this 

 excellent bulletin. 



Guyer has recently given us a very thoughtful paper 5 on the 

 possible relation of the chromosomes to hereditary characters. 

 His argument is directed specifically against the YVeismannian 

 hypothesis of determinants. lie does not attempt to minimize 

 the importance of the nucleus in ontogeny or in heredity, but he 

 is inclined to regard the development of what we call hereditary 

 characters as a result of interrelations between the nucleus and 

 the cytoplasm. 



Speaking of experiments in which fragments of protozoa con- 

 taining various proportions of the nucleus regenerated the or- 



If the nucleus is an aggregate of qualitatively different morphological 



in proportion to the amount of nuclear matter removed, but the evi- 



only a longer time is required if but a small fragment of the nucleus is 

 left in the piece. 



These experiments on protozoa disprove the determinant 

 theory as held by "Woismann. de Vries and others. 



Guyer points out that the bulk of the fertilized egg is cyto- 

 plasm of maternal origin and that the developing organism must 

 therefore be more of maternal than paternal origin. 



