322 N. M. Stevens. 



torial plate oi the first maturation spindle. All are connected by 

 linin threads, and every possible arrangement of the 5 chromo- 

 somes is found in different cells. In c the longest chromosome is 

 seen to be double and as a side view of the spindle in metaphase 

 alw^ays show^s the chromosomes double, it is probable that they 

 come into the mitotic figure in that condition from the preceding 

 conjugation stage. A second longitudinal split to form a tetrad 

 cannot be detected at this stage. In Fig. 36, a side view and an 

 oblique view of the equatorial plate show the double chromosomes. 

 Two stages of the anaphase appear in Figs. 37 and 38. There 

 is always one "lagging" chromosome in this division, and it cer- 

 tainly is not usually the longest, as I at first thought likely. In 

 fact it appears in most cases to be either the second or third in size. 

 After the two spermatocytes of the second order are fully formed, 

 as shown in Fig. 39, the two daughter elements of this "lagging" 

 chromosome are still connected by a thread extending through the 

 cytoplasm of each cell. This phenomenon seems to be a peculiar 

 characteristic of one of the chromosomes in this particular division. 

 I have never seen an exception, nor have I ever seen anything 

 similar in the second spermatocyte division or in fact in any other 

 mitosis. 



If, as I have supposed, this first maturation division simply 

 separates paired chromosomes, it is possible that the pair that 

 shows this peculiarity has a different linin connection from the 

 others. In Fig. 35, it will be seen that the 5 chromosomes are 

 united by linin threads into a chain with free ends. A side 

 view (Fig. 36) shows two pairs connected by two parallel linin 

 threads. Now if one of the end pairs always has its two elements 

 connected by a single thread, we might expect such figures as 37, 

 38 and 39. 



Fig. 40 is the equatorial plate of the second maturation mitosis, 

 showing again the 5 chromosomes of characteristic form and size. 

 Fig. 41 shows one of a few fortunate sections showing the daughter 

 chromosomes in anaphase, and removing all doubt as to the kind 

 of division, longitudinal or transverse. 



Judging from analogy in other insects, I fully expected to find 

 one longitudinal and one transverse division, but was soon con- 

 vinced that both are longitudinal, and from such figures as 23 and 

 26 that the same is true in the maturation division of the winter 

 egg. This point will be more fully discussed later. 



