Nov.. l'.)l()l Cytoplasm and Heredity 5 



will not weaken the evidence which other undoubted cases of 

 cyto]:)lasmic influence afford. 



Further arguments against the chromosomes as holders of 

 patent rights in heredity is found in the polarity of eggs. All 

 eggs have differentiated regions, even when these are not visibly 

 different. In most cases the animal pole becomes the aboral 

 pole of the later gastrula, while the vegetative pole becomes the 

 interior of the digestive tract. This polarity can be traced 

 back, in some cases, into the early oogonial stages, and it is not 

 improbable that it is continuous from one generation to the 

 next. It is scarcely conceivable that this polarity is due to 

 anything else than the cytoplasm. 



Symmetry in many animals is likewise apparently inde- 

 pendent of the chromosomes. This is particularly true of the 

 insects and the cephalopods. In the back swimmer Notonecta, 

 the last part of the egg to emerge in oviposition always forms 

 the same part of the larva, and there is a bilaterality of the 

 body that corresponds to a bilaterality of the egg. Since in 

 the development of the insect egg the nucleus divides repeatedly 

 before the daughter nuclei are shut off in separate cells, it is 

 scarcely conceivable that a selective distribution of chromatin 

 can occur with such minute regularity as to account for the 

 regular location of the organs. Perhaps the mere shape of the 

 egg, acting mechanically, may produce this symmetry; but in 

 any case, it is not the chromosomes. 



And finally, the case of the ascidian egg is important. These 

 eggs contain various localized metaplasmic substances which 

 can be traced into the muscles, the notochord, and the nervous 

 system of the larva. If that part of the egg which contains one 

 of these substances be removed before development begins, the 

 corresponding part of the larva is missing. The cytoplasmic 

 inclusions may not be organ-forming substances, but in that 

 case the cytoplasm itself must exert a determinative influence. 



The foregoing facts indicate a probable, in some cases 

 almost necessary, influence of cytoplasm in heredity. They are 

 not intended, however, to disprove the chromosomes hypothesis. 

 It seems to me possible to hold the view that both chromosomes 

 and cytoplasm have their influence; but they play different 

 roles. Let us examine anew the facts we have cited to show the 

 hereditary influence of cytoplasm. There was one case (varie- 

 gation in Mirabilis) which may be explained as due to a disease 



