CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETEROl'TERA. 149 



product of a preceding cell. And in all my work I have consistently argued for the 

 chromosomes as persisting structures, in substantiation of the idea of the individuality 

 of the chromosomes founded hy Van Beaeden, and supported by a great number of 

 students. 



Now in any consideration of the chromosomes the question presses on one : Are 

 the several chi-omosomes of a given nucleus alike in their energies, or are they dif- 

 ferent? Are they actively or potentially equivalent, or are they not? Weismann 

 and Roux were perhaps the first to take up this question, and Weismann has reasoned 

 on the basis of his determinant hypothesis, that in any cell where the chromosomes are 

 neither very small nor very numerous, each single chromosome is the bearer of all the 

 hereditable qualities i')f a whole individual of the species. Against such a valence of 

 the chromosome there is much evidence of serious weight, and it has been nowhere 

 more succinctly summed up than in the recent review by Boveri (1904). To this 

 matter of the potentiality of the chromosomes we will now turn. 



Boveri has argued very strongly (1904) that particular chromosomes have partic- 

 ular energies, that one chromosome represents certain activities not evinced by anothei-. 

 His own important empirical contribution (1902) to this idea was the analysis of the 

 abnormal development of eggs fertilized by one spermatozoon. And he concluded : 

 "that not a fixed number but a fixed combination of chromosomes is necessary for 

 normal development, and this means nothing else than that the particular chromo- 

 somes must possess different qualities." 



Another line of evidence is that afforded by the ditfei'ences in behavior of the 

 chromosomes, when the cell is not molested by experiment. Such are the allosomes, 

 of which Ave treated in the preceding section. They may behave differently from 

 the autosomes, as we have seen, either by preserving their density in the rest period of 

 the spermatogonia and the growth period of the spermatocytes, or by dividing in the 

 maturation mitoses in a different sequence from the autosomes. Therefore in nuclei 

 containing allosomes there are at least two kinds of chromosomes : the unmodified 

 autosomes, and the modified allosomes; and there can be no doubt that these have 

 different activities. 



But we may go further than this. Are we to regard the possession of chromo- 

 somes of different kinds, particularly the possession of the highly modified allosomes, as 

 simply a taxonomic peculiarity of certain forms, such as the insects, araneids, chilopods 

 and Sagittaf I think not, for if there are such great differences in the chromosomes 

 of these forms, is it not probable that there would be also chromosomal differences 

 in other forms, even if less readily demonstrable? 



For leaving the allosomes out of consideration comparative studies are proving 



