STUDIES ON CHROMOSOMES 75 



by the work of Stevens and myself, is widely distributed among the 

 insects (Hemiptera, Coleoptera, Diptera) and is strictly confined 

 to the male line (except when supernumeraries are present). In 

 species having an odd or accessory chromosome the Y-element 

 (small idiochromosome) is wanting, and I have urged this fact as 

 showing that this latter chromosome cannot play any essential role 

 in sex-production 10 or in the transmission of the secondary sexual 

 characters, as Castle ('09) ingeniously suggests. What I desire 

 here to point out is that by parity of reasoning we should also con- 

 clude that this chromosome is devoid of any special significance in 

 heredity of any kind, at least as far as the visible external charac- 



12 instead of 11 separate chromosomes in the first spermatocyte division, as we do 

 in the 22-chromosome individuals of Metapodius (Wilson: '09c). Throughout the 

 Hemiptera, indeed, when the accessory chromosome (or its homologue, the large 

 idiochromosome) is accompanied by a synaptic mate or Y-element, the two are 

 separate in the first division, which accordingly shows one more than the reduced 

 or haploid number i.e., *+! The photographs of Foot and Strobell show, how- 

 ever, 11 chromosomes in this division (the two m-chromosomes being of course 

 counted as one, like the other bivalents), as they should if the spermatogonial num- 

 ber be 21. 



Still, this might be a case like that of Syromastes, where no Y-element is present, 

 but the accessory is itself double though such a parallel would hardly help the 

 case, since in no form is the failure of the accessory to divide in one division more 

 indubitably shown than in Syromastes, while Foot and Strobell are persuaded that 

 it does divide in Anasa. But, secondly, my own extended additional observation, 

 like the studies of Lefevre and McGill ('08), still continues to give but one result 

 as before. The living animals (from the same locality as the material of Foot and 

 Strobell) have been kept by hundreds in the greenhouse for months at a time in 

 successive years, and have been regularly employed for class work in cytology and 

 for experimental purposes, in the course of which large numbers of additional 

 sections and smears have been prepared and examined. Others as well as myself 

 have carefully searched among these preparations for cases showing more than 21 

 spermatogonical chromosomes, without success apart from the double or multiple 

 groups that occasionally appear. The same relation continually recurs, namely 

 21 spermatogonial chromosomes of which three are larger than the others, while 

 in the dividing ovarian cells the number 22 appears with equal constancy. That 

 not even one case of 22 spermatogonial chromosomes has thus far been found is 

 indeed surprising; for plus variations in the diploid groups are known to occur in 

 some species of Hemiptera, and I have myself described such cases (e. g., Wilson: 

 '09c). 



These and other reasons lead me to believe that the conclusions of Foot and Stro- 

 bell were based on the observation either of very rare fluctuations in the normal 

 diploid number or of accidental products of the technique. 



" Loc. cit.: '06, '09a, '09d. 



