512 PllOCEEDINGS OF THE AMERICAN ACADEMY. 



form of the tetrads and in the attachment of the mantle fibres which 

 are interesting and instructive. A peculiarity seen in these chromo- 

 somes which I have not seen reported by any other observer is shown 

 in Figures 29-31. The chromosomes here represented were studied in 

 preparations in which the extraction of the haematoxylin had been 

 carried farther than usual. This caused the greater part of the 

 chromosome to stain not a dense black, but a dark gray. At the ends 

 of each tetrad, at the point where the mantle fibres attach, there is a 

 small, distinct body which stains much darker than the rest of the 

 chromosome, and is of about the size of the centrosome (Figure 31). 

 This spherule is embedded in the chromosome and is continuous with 

 the mantle fibre and appears to be an enlargement or knob upon the 

 end of this thread. Whether this body has any other significance than 

 that of serving as an attachment to the chromosome I am unable to say, 

 but it is a very distinct structure and is always found in the metaphase 

 chromosome. The position of this body is very useful in determining 

 the exact point of attachment of the mantle fibres and as an aid in 

 determining whether the division of the chromosome is transverse or 

 longitudinal. Usually the mantle fibres are attached at opposite ends 

 of the four-lobed chromosome, and in such cases the fact that the 

 division is longitudinal can be determined only by the shape assumed 

 by the chromosome during the gliding apart of the two component 

 dyads (McClung, :00 ; Blackman, :03, :05). Quite often, however, the 

 mantle fibres attach to two of the adjacent lobules or chromatids of 

 the tetrad (Figure 31). In such cases the chromosome lies with its 

 longer axis parallel to the equatorial plate and division occurs by the 

 separation of the chromosome along its longer axis. Such chromo- 

 somes are seen in all equatorial plates studied and are represented in 

 Figures 29-31. 



Thus the shape of the chromosomes at this time strongly indicates 

 that the division of the first spermatoc)rte is a longitudinal one, but 

 the best evidence of this is seen in the early prophase and during the 

 anaphase of the first spermatocyte division. We have seen that in 

 the prophase the first change in the chromatin segment consisted in the 

 longitudinal cleavage. It is but natural, then, to believe that this 

 longitudinal division is the one first completed in the two mitoses 

 following. In the anaphase of the first spermatoc)'te, strong evidence 

 as to the character of the chromosomic division j ust completed is fur- 

 nished by the shape of the daughter chromosomes during their passage 

 toward the poles. In Figure 32 is shown a mid-anaphase of the first 

 spermatocyte. The chromosomes are already approaching their final 

 position, but as yet are not massed together as they are at a slightly 



