SMALL wood: maturation OF HAMINEA SOLITARIA. 303 



that ouly one of the pronuclei is closely associated with each aster ; the 

 left aster in this case is associated with the upper pronucleus, while the 

 right aster is associated with the deeper one. There are present a few 

 more rays than in the previous stage (Fig. 88), but no distinct central 

 spindle is yet formed. In Figure 90 (Plate 13) the outline of the upper 

 nucleus is somewhat smaller than the lower one, so that the former ex- 

 tends beyond that of the latter except at one place. The central spindle 

 is partly formed, and the relation of the asters and the forming spindle to 

 the two pronuclei is clearer than in the preceding stage. The right aster 

 and its partly formed central spindle is associated with the lower nucleus, 

 while the central spindle of the left aster penetrates the upper nucleus. 

 The direction of the two half-spindles emphasizes their independence. 

 Similar conditions are further illustrated in Figure 89, where the section 

 has been made in such a direction that the two pronuclei do not cover 

 each other as in the previous cases, and therefore the relation of the 

 asters to the spindle halves is more patent. The peripheral ends of the 

 spindle fibres which ai'e in relation with the larger (right) pronucleus are 

 all bent, a condition observed in this one instance alone. 



The centre of each aster is occupied in all of these cases by a sharply 

 defined centrosome, which is enveloped in a layer of sphere-substance 

 that later becomes differentiated into cortical and medullary layers. 



That the cleavage centrosomes arise one in connection with each of 

 the pronuclei, is further indicated by the fact that in more than one hun- 

 dred eggs in which only one pronucleus occurred in the section, there was 

 never found a trace of more than one aster primarily associated with it. 

 It seems improbable that this can be due in so many cases solely to 

 the accident of cutting. This observation has been verified by a careful 

 examination of a number of whole eggs ; the earliest stages observed 

 showed two asters in the cytoplasm, lying at some distance from the 

 nuclear membrane, and no evidence was found to indicate that these 

 asters were derived from pre-existing asters. 



As the pronuclei are invaded by the central spindle, their walls break 

 down and the contents are differentiated into two substances : first, the 

 solid, or partly solid, chromatin masses that are to become the chromo- 

 somes (Plate 13, Fig. 93 ; Plate 12, Figs. 8G, 81, 82) ; secondly, a 

 rather coarsely granular substance, which takes a heavy stain in iron 

 hsematoxylin. The latter is more abundant when the nuclei first break 

 down (Fig. 93), but grailually becomes less and less until it is indis- 

 tinguishable in the cytoplasm (Fig. 82). During this period there has 

 V)een an active metabolism in the cell, as is shown by tlie condition of 



