THE SPERMATOGENESIS OF AGALENA N/KVJA. 123 



preparations than the others mentioned although all were useful 

 in some ways. 



Ripe spermatozoa from the pedipalps were smeared on a glass 

 slide and well fixed by heating at the boiling point. Smear prepa- 

 rations, after the method of Foot and Strobell, proved of value 

 for the mature sperm cells but were disappointing for the earlier 

 stages. This may be due to the fact that the chromosomes are 

 large and numerous and even when the nucleus is spread out in 

 an extremely thin film, they become so heaped up upon one 

 another that it is not possible to count them accurately. If smear 

 preparations be compared with those treated with fixing fluids, it 

 is evident that the latter cause considerable shrinkage ; and tissues 

 fixed in Flemming's fluid for a few hours show less shrinkage 

 than those fixed for fifty hours or more but the latter method 

 brings out strongly the centrosomes and spindle fibers. 



Heidenhain's iron-haematoxylin has been chiefly relied upon 

 for staining as it gave the finest results in almost every particular 

 but beautiful preparations were also obtained with Hermann's 

 triple stain. Every one grants that the different reactions to 

 staining reagents are no safe criterion in the endeavor to differen- 

 tiate nuclear elements, since the same structures do not always 

 stain in the same way during the various phases of development. 

 Nevertheless color-differentiation is often helpful and this is 

 especially true when dealing with the accessory chromosomes. 

 Their whole history can be made out clearly in iron-haematoxylin 

 preparations but when Hermann's triple stain is used, their affinity 

 for the safranin at times when the ordinary chromosomes or the 

 chromatin granules take the violet, makes them stand out in a 

 striking manner. 



OBSERVATIONS. 

 Spennatogonia. 



In young spiders, early in the summer, the testes are wholly 

 made up of spermatogonia and, unlike most forms, these cells in 

 mitosis are almost equal in size to the spermatocytes of the growth 

 period. In the metaphase the rod-shaped chromosomes are so 

 numerous, probably fifty or more, and so closely packed together, 

 that in polar views it is impossible to make even an approxi- 



