BIGELOW: NUCLEAR CYCLE OF GONIOXEMUS MURBACIIII. 349 



scure nature, are the differences seen in the ground substance of different 

 examples. This is, as a rule, homogeneous ; in many cases, however, 

 (Figs. 132, 133), the vacuoles appear to be enclosed by walls of more 

 deeply staining substance. Occasionally, also, the material enclosed by 

 these denser walls is apparently of the same composition and density as 

 the ordinary ground substance, instead of presenting the pale appearance 

 of vacuoles (compare Fig. 137 with Fig. 138). In many cases the nu- 

 cleoli enclose deeply staining strands and granules (Figs. 134, 135, 130), 

 and in a few instances I have found specimens with no vacuoles at all, 

 but traversed by a series of rather evident strands, variously thickened 

 and connected with one another by deeper stained centres (Figs. 134, 

 135). These differences are not due to different degrees of development, 

 for they occur in the smaller as well as in the more mature nuclei ; and 

 there is no evidence that such strands are stored chromatin, an inter- 

 pretation placed upon similar appearances in echinoderms by Hart- 

 maim ( : 02) and Guenther ( : 04). It seems to me more probable that we 

 have in Gonionemus such a condition as Montgomery ('98 b , p. 509) found 

 in Polydora, where, "owing to the gradual confluence of the vacuoles, 

 which thus produce anastomosing cauals of vacuolar substance in the 

 ground substance of the nucleolus, it is the true ground substance which 

 repi'esents a skein-like appearance." 



In iron-haematoxylin preparations the chief nucleoli show great indi- 

 vidual variation, being sometimes intensely stained, in other cases very 

 pale, and occasionally almost transparent. But these differences are no 

 more dependent on the degree of development of the nucleus as a whole, 

 or on the condition of the chromatin structures, than are the variations 

 in the nucleolar ground substance. To test this question further, I dis- 

 mounted some of the preparations, washed out the haematoxylin, and 

 restained in the Auerbach mixture, with the result that the same nucleoli 

 which previously were stained hardly at all, were now stained as deeply 

 as others of t'he same size ; the only noticeable variation being that there 

 seems to be a very slight loss of staining capacity corresponding to the 

 increase in bulk. This fact is of great importance, for it shows that 

 there is no loss of substance on the part of any of the nucleoli ; and it 

 emphasizes the danger of basing conclusions as to the relations of nucleoli 

 to chromatin structures on iron-haematoxylin staining unchecked by 

 other methods. 



The accessory nucleoli, though varying considerably in size and out- 

 line, show no important differences in structure ; all stain very deeply 

 with ordinary nuclear dyes, being either of spongy or multiple nature, 



