CHROMOSOMES IN THE SPERMATOGENESIS OF THE HEMIPTERA HETEROPTERA. 161 



We can decide this much about numerical relations of the chromosomes, that cor- 

 respondence in number by no means implies community of race ; one has simply to 

 list the different animals with the number 24 to be sure of this. On the other hand 

 there is often constancy through smaller groups such as genera or species. The ques- 

 tion is then : when we find a genus like Ascaris, with chromosomal number ranging 

 from 2 to 48, are we to judge from this variation that chromosomal number has no 

 taxonomic significance, or are we to decide that the forms combined in the genus 

 Ascaris are really not generically related ? 



This is an exceedingly difficult question to decide. If our present relegation of 

 the species of Ascaris be justified, then clearly chromosomal numbers have not even 

 generic worth. But our whole classification of somatic individuals is at present merely 

 tentative, and the grouping of the various species of the Nematodes in particular 

 seems to be very artificial. There is uncertainty at both ends of the argument. We 

 must commence with the premise, that seems to me fully justified, that the species is 

 one and the same from the egg up to the adult condition. Therefore it is per- 

 missible to classify germ cells as well as adults, and, c. g., to compare chromosomal 

 relations through a series of germ cells as we would conditions of the nervous system 

 through a series of somatic individuals. The chromosomes as portions of the very 

 conservative nuclear element should surely offer as good a basis for genetic compari- 

 sons as any set of somatic structures. That is to say, an entirely rational phylogeny 

 of organisms might be founded in part upon relations of the germ cells ; therefoi'e 

 nuclear constituents be used as characters quite as much as any other sets of structures. 

 The only reason to prefer comparisons of adult individuals is because they exhibit 

 differentiation more than germ cells do, and not because they are really more differ- 

 entiated. 



Therefore when germ cells show differences in chromosomal numbers, these can 

 signify only differences of the individuals that contain them. And while numerical 

 differences are among the least important of the anatomical characters, yet when they 

 are differences of so important an organ as the nuclear element they should be granted 

 some degree of importance in a rational taxonomy. Consequently, it would be 

 incorrect to place different species, some with 4 and others with 48 chromosomes, in 

 the same genus, for such differences of the chromosomal number must constitute at 

 least genetic and much more than specific difference. Were this not so, we could not 

 explain why in so many cases there is constancy of chromosomal number in groups 

 much higher than genera. Therefore chromosomal number is a character that should 

 be considered in taxonomy. 



At the same time number is only one of the properties of chromosomes, they have 



A. P. S.— XXl. R. 27, 8, '06. 



