286 bulletin: museum of comparative zoology. 



represented in Figures 42 and 43 are excluded, the changes in the 

 chromosomes follow the usual order. 



So far as I have been able to observe, both the direct and the indirect 

 processes have their inception in such a stage as is shown in Figure 27 

 (Plate 5). Since in the morphology of the earlier stages there is 

 nothing, as has been said, to indicate by which process the development 

 Avill take place, there must be peculiar physiological conditions which 

 determine the changes through which the egg passes; but at present 

 it is impossible to state what they are. The two centrioles which 

 accompany the second maturation spindle are already formed (Plate 5, 

 Fig. 27 a), and are surrounded by a layer of centroplasm. Stretching 

 from centriole to centriole there is at all stages (Fig. 27; Plate 7, 

 Fig. 38) a definite central spindle, which increases in length as the 

 centrioles move apart. 



Plate 7, Figure 38 (compare Plate 10, Fig. 72) shows a con- 

 dition which is typical for a fairly early stage in this indirect 

 process. The centroplasm has become limited by a definite outer 

 boundary, and from now on takes a very faint plasma stain ; the 

 primary rays ai"e conspicuous, and terminate in the outer boundary of 

 the centroplasm (c'pl.) ; the secondary rays are very faint and are grad- 

 ually becoming shorter ; the medullary zone (.s/. med.) around the deeper 

 pole has become much larger tlian in the preceding stage ; but at the 

 outer pole it is wanting, and there is nothing to indicate that cortical 

 and medullary layers have ever been differentiated there. The move- 

 ment of the maturation figure toward the surface of the egg compresses 

 the cytoplasm in this region ; this in turn reacts directly against the 

 centrosome, which thus becomes flattened. The outer centrosome never 

 becomes as large as the inner one, and its rays are less conspicuous. 



Some of the stages which I have incorporated in the indirect process 

 belong, so far as the condition of the chromatin is concerned, eaidier in 

 the description ; but since the changes in the achromatic spindle have 

 been taken as the basis for determining the sequence, and since these 

 stages have an intimate connection with the indirect process, they are 

 included here. 



The chromosomes in Figure 38 (Plate 7) are arranged in two clusters : 

 the outer one is destined to pass into the first polar cell, the inner to be 

 contained within the nuclear vesicle, the wall of Avhich has, indeed, 

 already made its appearance. Already, too, the chromosomes are con- 

 nected with the wall by linin fibres. In Figure 39 the surface of the 

 egg at the animal pole has been raised up and the abstriction of the 



