326 INTRODUCTION TO CYTOLOGY 



A strong piece of evidence supporting Boveri's conclusion was fur- 

 nished by Herbst (1909). By treating eggs of Sphcerechinus with valeri- 

 anic acid Herbst caused them to undergo cleavage artificially. While 

 the cleavage mitosis was in progress a spermatozoon of Strongylocentrotus 

 was allowed to enter the egg, where it at once gave rise to its group of 

 chromosomes (Fig. 127). These, however, arriving too late to join 

 regularly in the mitosis, were incorporated in neither of the daughter 

 nuclei of the first cleavage : they resumed the form of a nucleus, and this 

 was included in one of the blastomeres. This blastomere therefore 

 contained two nuclei, one maternal and one paternal, which combined 

 during subsequent stages, whereas the other blastomere had a maternal 

 nucleus only. Herbst regarded such nuclear behavior as responsible for 

 the frequently found larva? which are hybrid in character on one side 

 and purely maternal on the other. This experiment has been held to 

 show not only that it is the nucleus of the spermatozoon which brings in 

 the paternal characters, but also that it is the chromosomes alone that are 

 responsible. It is assumed, though perhaps without sufficient evidence, 

 that the other nuclear materials (karyolymph etc.) which may be present, 

 as well as any cytoplasmic elements, have opportunity to mix generally 

 with the egg cytoplasm, since the membrane of the male nucleus breaks 

 down and leaves the chromosomes lying free before the egg divides into 

 the two blastomeres. The paternal characters, however, appear only 

 where the chromosomes come to be located — that is, in the cells compos- 

 ing one-half of the organism. In his work on multipolar mitoses in di- 

 spermic eggs (Fig. 127 bis; see also p. 163) Boveri (1902, 1907) was able 

 to show further that abnormal chromosome distribution is associated 

 with abnormalities in development in a very definite way; and that if 

 isolated blastomeres resulting from such abnormal divisions be made to 

 develop independently, completely normal larvae result only where there 

 is statistical reason to believe that a full complement of the qualitatively 

 different chromosomes is present. 



The nuclear theory had its opponents from the beginning. Verworn, 

 Waldeyer, Rauber, and other early investigators held that the cytoplasm 

 as well as the nucleus must be concerned in the hereditary process, since 

 the spermatozoon in many cases does bring cytoplasm into the egg, and 

 also because neither nucleus nor cytoplasm can function independently 

 of the other. This view received support in certain experiments which 

 seemed to discount the power of the nucleus in controlling heredity. 

 Loeb (1903) found that when a sea urchin egg was fertilized by a starfish 

 sperm the resulting larva possessed purely maternal characters, the 

 sperm nucleus exerting no visible hereditary effect. The same thing was 

 noted by Godlewski (1906) in crosses between sea urchins and crinoids. 

 Godlewski made the further significant observation that when enucleated 

 egg fragments of Parcel) i mis (sea urchin) were fertilized by sperm of 



