Spermatogenesis in spiders. 559 



4. Orig'in and behavior of the ctetosomes und 

 planosomes. 



In studying the beliavior and distribution of the peculiar 

 supernuniei'ary bodies in spiders, the observer encounters several 

 difficulties which require the exercise of a great deal of caution 

 and make the work somewhat unsatisfactory. 



Aside from the error of observation, there are two directions 

 in which oue is apt to be misled. The first of these lies in the 

 fact, that, with preservatives which contain osmic acid, the yolk 

 and fat granules tend to stain black. And while such granules 

 show an entirely different behavior from chromosomes, still in any 

 given cell, it is not always possible to say what is fat granule and 

 what chromatin; and when one wishes to study the very small 

 chromosomes, it is impossible with these artifacts present. As 

 pointed out in the beginning of this work, this difficulty has been 

 avoided by using killing fluids which did not contain osmic acid and 

 whenever possible, two fluids of a different nature were used. 



The second difficulty is of another kind. Boveri (1890) has 

 shown that with the use of iron-haematoxylin, cell structures stain 

 concentrically. Wilson has shown (Study II) that the same was 

 true of the chromosomes. This fact forces one to be very cautious 

 in a study of the behavior of these small bodies. When one 

 differentiates too far, the supernumerary chromosomes tend to disappear 

 from view. This point may be brought out more clearly in the 

 following way. In one of the slides of Ammirobius sylvestris^ the 

 distribution of the supernumerary chromosomes was studied and in 

 some cases drawings of the cells were made. In every case the 

 position of the cell was noted by a mechanical stage. Later 

 certain prophase stages were found on the same slide which were 

 not differentiated quite far enough to be studied advantageously. 

 The cover slip was removed and the sections redifferentiated for 

 these prophase cells. On returning later to an examination of the 

 other cells, some of which had been drawn, it was found that the 

 planosomes were apparently missing, but on close examination, little 

 blotches just a trifle darker than the surrounding cytoplasm were 

 seen where these bodies had been. This made it clear that one has 

 to be careful in differentiating not to carry it too far. 



After guarding against the errors described above, the observer 

 may still find it difficult to know just how small a granule may be 



37* 



