SUTTON : SPERMATOGONIAL DIVISIONS IN BRACHYSTOLA MAGNA. 137 



cysts of the earlier stages are, therefore, found at the blind end of the 

 follicle, and the mature spermatozoa near the vas deferens, the inter- 

 mediate stages lying between. Ordinarily all the cells of a cyst are 

 in the same phase of development at any given time. 



Excepting the youngest secondary spermatogonia, any cyst will 

 present a later stage than its neighbor which lies farther from the vas 

 (h'fvrcns. but it does not follow that successive cysts represent con- 

 secutive stages ; for, as will be shown later, each cyst develops in- 

 dependently, but at about the same speed as the rest, so that whether 

 or not adjacent cysts show consecutive stages depends mostly upon 

 the relative time of the division of the primary spermatogonia from 

 which they arose. We may, therefore, tind several cysts in the same 

 stage ; or, between the stages of adjacent cysts, a considerable gap 

 may exist. 



The mature follicular membrane is composed of two layers, between 

 which may be seen numerous, deeply staining, flattened nuclei. 

 {". fig. 1. ) 



FIXING AND STAINING METHODS. 



Several fixing methods were used, but by far the most favorable 

 l^roved to be Flemming's chrom-osmium-acetic mixture. The testes 

 were hardened in this from four to twenty-four hours, after which 

 they were carried up through different grades of alcohol to seventy 

 per cent., where they remained until it was desired to imbed and sec- 

 tion them. 



All material was stained in section. Of the variovis stains tried, 

 Heidenhain's iron-h<i?matoxylin proved the best for nearly every pur- 

 pose. The sections were placed in a mordant of four per cent, iron 

 alum for about fourteen hours, and then, after thorough washing, 

 stained about ten hours in one-half per cent, htematoxylin. They 

 were then washed again, and finally differentiated under the micro- 

 scope in the same fluid used for the mordant. 



Flemming's three-color method proved useful in the study of the 

 behavior of the accessory chromosome. This was used as follows: 

 The sections were placed in Zwaardemaker's safranin for from forty- 

 five minutes to two hours : the excess of this was washed out in ninety- 

 five per cent, alcohol, and they were immersed for ten minutes in a 

 half-saturated aqueous solution of gentian violet. This was washed 

 out with water and the sections treated for three seconds with orange 

 G., after which the stain was differentiated with ninety-five per cent, 

 alcohol and clove oil. Kernschwarz and safranin also produce a very 

 good stain for this work. 



Sections were cut 6# microns thick and fixed to the slide with 

 Mayer's albumen water of a strength of two drojDS of the albumen to 

 an ounce of distilled water. 



