No. 2.] THE SEA-URCHIN EGG. 455 



der of the spindle-fibres undergo a similar granular degenera- 

 tion, the entire spindle breaks up into granules, and in its 

 place appears the usual cyto-reticulum. The study of a large 

 number of preparations, in every stage of the process, leads to 

 the conclusion that. the spindle-granules give rise directly to 

 the reticulum, though it is in the nature of the case difficult 

 to exclude the possibility that the granules may be absorbed 

 and the reticulum re-formed. 



Immediately after the cleavage, each aster has a horse-shoe 

 shape, and closely corresponds in general appearance to the 

 sperm-aster just before its division (Phototype 11, Text-fig. 

 X, A). The subsequent changes are nearly a repetition of those 

 involved in the division of the sperm-aster, each aster dividing 

 into two, while the rays become much shorter (Phototype 12, 

 Text-fig. X, B). In the " resting " two-cell stage each blasto- 

 mere is closely similar to the original ovum during the "pause." 

 The asters persist throughout the two-cell stage, and their 

 descendants may be traced continuously up to the 32-cell stage 

 at least. Beyond this I have not attempted to follow them. 



Conclusions regardmg the Archoplasm. — I can only interpret 

 the foregoing observations in the sense already indicated at 

 page 446, namely, that the entire system of astral rays and 

 spindle-fibres is the result of a special configuration or re- 

 arrangement of a preexisting reticulum extending through both 

 cytoplasm and nucleus. That this is the true interpretation of 

 the morphological facts observed in the sections, is almost con- 

 clusively shown by the fact that the sperm-aster moves for a 

 considerable distance tJiroiigh tJie substance of the vitelhis. If 

 with Boveri we assume the astral rays to consist of a specific 

 substance distinct from the reticulum, we must assume the 

 progress of the aster to be a movement of translation through 

 the network, which involves a mechanical problem of extreme 

 difficulty even if we assume the astral rays to end freely and to 

 be of firmer consistency than the reticular meshes. The rays, 

 however, do not end freely, but branch out into the reticulum, 

 and the movement of the aster is hardly conceivable save as a 

 progressive rearrangement of a plastic preexisting structure 

 under the influence of a center moving through it — an action 



