236 SIDNEY I. KORNHAUSER 



this variation, which had already occupied the attention of such 

 pioneers as Carnoy, La Valette, St. George and Sinety, should 

 be produced by parasitism. 



Brauns ('12), in his study of the oogenesis of Forficula auric- 

 ularia, believes that the haploid number in the female is 13. 

 He quotes Professor Ludwig Will of Rostock as saying that the 

 male diploid number of Forficula is 25, and the female diploid 

 number 26; but that some males show 24 chromosomes in their 

 spermatogonia. It is a striking coincidence that my counts on 

 Anisolabis are in very close agreement with the results of Pro- 

 fessor Will. 



Randolph ('08) worked on Anisolabis and described 24 chromo- 

 somes as the diploid number in both sexes. In the primary 

 spermatocytes she pictures a pair of almost equal hetero chromo- 

 somes which lag in the anaphase of the first maturation mitosis, 

 but her results are not at all in accord with those described in 

 the present paper. 



3. METHODS 



The material studied consisted of gonads and embryos. 

 Nymphs were found best for the study of the germ cells. The 

 animals were all collected at or near Cold Spring Harbor, Long 

 Island, during June, July and August of several summers since 

 1915. 



The best fixatives were Flemming's fluid (strong), Bouin's 

 fluid and Benda's modification of Flemming's fluid. Various 

 stains were employed. The Flemming material was stained in 

 Heidenhain's haematoxylin, counterstained with Orange G or 

 with safranin and lichtgrlin. The Benda material was stained 

 either with alizarin and crystal viplet, or with methyl green and 

 acid fuchsin of Bensley. These last two were especially valuable 

 in the study of the xxy-complex. 



Smears of testes were made in a moist chamber, immediately 

 exposed to osmic fumes for a few seconds, and then immersed 

 in a fixing fluid; care being taken to avoid drying. Such 

 smears were then treated and stained like sections. 



