528 



SCIENCE. 



[N. S. Vol. XXI. No. 536. 



mosomes would also disappear, thus con- 

 stituting a case of 'regressive mutation' in 

 the terminology of de Vries. In other 

 cases there has evidently been an increase 

 of chromosomes in some species as com- 

 pared with others; though whether this 

 increase is due to the division of originally 

 single chromosomes, or to the addition of 

 new ones through abnormalities of division 

 or distribution, or through hybridization, 

 can not now be determined. Irregularities 

 in the number and distribution of chromo- 

 somes are by no means uncommon and in 

 the case of hybrids are very frequent, as 

 has been shown by Juel, Guyer and Can- 

 non. By an increase in the number of 

 chromosomes or in the number of elements 

 of which a chromosome is composed the 

 sum of the heritable qualities would prob- 

 ably be increased, thus constituting, in the 

 language of de Vries, a 'progressive muta- 

 tion. ' There is no evidence and no proba- 

 bility that new chromatic elements are ever 

 added to the nucleus from the cytoplasm, 

 or that they ever arise de novo. Such new 

 elements must arise through new combina- 

 tions of old elements, either, as de Vries 

 considers, by an actual interchange of An- 

 lagen (material particles) between the 

 pairs of maternal and paternal chromo- 

 somes, or, as Haecker supposes, by an inter- 

 change of grandparental parts of chromo- 

 somes, or through hybridization or irregu- 

 lar mitoses. Guyer has shown that the 

 divisions of the chromosomes are frequently 

 or usually irregular in hybrids and he sug- 

 gests that such irregular mitoses may add 

 or subtract certain chromosomal elements 

 and thus constitute the basis for a muta- 

 tion. Such eases are, however, almost en- 

 tirely hypothetical and at present we are 

 compelled to admit that we do not know 

 how mutations arise or first become mani- 

 fest in the chromatin. 



As regards the cytoplasm, I have shown 

 reason for believing that it is composed in 



part of escaped nuclear material and that 

 the mechanism, therefore, exists for the 

 nuclear control of the entire cell. The 

 organization of the cytoplasm is chiefly 

 manifest in its polarity, symmetry and the 

 localization of unlike substances. There 

 are reasons for believing that many bilat- 

 eral animals are characterized by funda- 

 mental similarities in the polarity and sym- 

 metry of the unsegmented egg; the types 

 of localization of organ bases are, however, 

 very different in different phyla ; in par- 

 ticular the localizations in the eggs of 

 ctenophores, nemerteans, eehinoderms, an- 

 nelids, mollusks and ascidians are thor- 

 oughly characteristic of each phylum, and 

 except in the case of the annelids and mol- 

 lusks there are few similarities between 

 these types. Nevertheless, it is possible 

 that certain of these types may have been 

 derived from others ; in fact, such trans- 

 formations might be accomplished far more 

 ea.sily in the eg^ than in the adult. 



Despite the evident and almost insuper- 

 able difficulties involved, certain zoologists 

 have not hesitated to indicate how the adult 

 form of one phylum might have been de- 

 rived from the mature form of another; 

 thus we have the ccelenterate, the nemer- 

 tean, the echinoderm, the annelid and the 

 arthropod hypotheses as to the origin of the 

 vertebrates, and in each of these cases by 

 stupendous transformations, degenerations 

 and new formations of the adult form of 

 the invertebrate in question the vertebrate 

 is supposed to have sprung into existence 

 fully formed and panoplied, like Minerva 

 from the brain of Jove. In all these specu- 

 lations fancy occupies so prominent a place 

 and facts are so scarce that it is no wonder 

 that the whole 'phylogeny business' has 

 come into disrepute. Nevertheless, the 

 evolution idea compels us to assume that 

 there are relations more or less remote be- 

 tween all phyla and that some must have 

 come from others by natural processes. 



