48 KANSAS UNIVERSITY SCIENCE BULLETIN. 



possible, to secure good, clear results ; at least, neither Gilson's 

 nor Zenker's mixture did any better on cells in these stages. 



Heidenhain's iron-htematoxylin stain gave the best results, 

 although Flemming's three-color method was very satisfactory. 



Most of the observations are made from sectioned material, 

 but smear preparations were used for advanced spermatozoa. 



II. Material. 



The first observations were made in January of the present 

 year on some material prepared by Professor McClung, to whom 

 I wish here to make most grateful acknowledgment for propos- 

 ing the line of investigation, and for many helpful suggestions 

 during the progress of the same. The testes were those of adult 

 insects, and showed mostly only mature or almost mature 

 spermatozoa. Enough spermatids were seen to indicate that 

 their transformation into spermatozoa was somewhat peculiar. 



The material for the drawings was taken from specimens col- 

 lected under stones and boards on and near the University cam- 

 pus, at Lawrence, Kan. Young nymphs of Gryllus were caught 

 as early as March 1, but the cells of the testes were all in the 

 spermatogonia and spermatocyte stages. During the early part 

 of May, after the nymphs had passed their third molt, crickets 

 were secured whose testes showed all stages of development and 

 transformation. 



Gryllus assimilis, the common black field cricket, has paired 

 testes lying in the anterior dorsal part of the abdomen. These 

 have a whitish transparent appearance, which becomes duller 

 in the adult, showing sometimes a slight yellowish tinge. The 

 shape of each is that of a somewhat conically rounded body, not 

 unlike a flattened strawberry. Each testis consists of a central 

 rachis about 3 mm. long, from which extend a large number of 

 curved follicles varying in length from 1 to 3 mm. 



The follicles are larger toward the blind end and taper toward 

 the rachis. Each follicle is divided into cysts, but more often 

 transversely than longitudinally, for frequently one cyst occu- 

 pies the whole follicle in cross-section. The cysts toward the 

 rachis end of the follicle are much longer and narrower than 

 those of the blind end. The cells are not very large and con- 

 tain twenty-four chromosomes in the spermatogonial genera- 

 tions ; eleven and twelve were most frequently found after the 

 reduction. 



