The Ovogenesis of Hydra. 319 



33. I have been unable to demonstrate a centrosome; spindle 

 fibres are evidently present and faint asters. 



34. Tetrads are not formed but the first polar spindle has twelve 

 rod-shaped chromosomes. These divide longitudinally as is seen in 

 Fig-. 14 and 15, Plate 12, drawings of the early anaphase of the 

 first polar spindle, seen from the pole. In the same section, by 

 focussing" a little deeper there may be seen the twelve corresponding- 

 chromosomes going to the other pole. 



35. Reduction is foreshadowed by a union of these twelve 

 chromosomes, two by two, as they approach the pole, as shown in 

 Fig. 16. In Fig. 17 I show a drawing of the late anaphase of the 

 first polar spindle. In the section next to this in the series the 

 first polar body is found and its chromatin is already undergoing 

 degeneration: the nucleus of the polar body seems to disintegrate 

 very rapidly. 



36. Beauer explicitly states that there is no rest between the 

 first and second polar spindles: with this statement I can not agree. 

 A rest stage follows the first spindle a stage of short duration but 

 with rapid and extensive changes. During the telophase of the 

 first maturation division the nucleus again forms a coarse net (Fig. 18, 

 Plate 12). This net becomes finer as the nucleus grows rapidly; the 

 nucleoplasm is now very lable, often with amoeboid borders (Fig. 19, 

 Plate 12). Both the ovogenesis and the spermatogenesis of hydra 

 illustrate the converse of Boveei's law that no rest occurs between 

 the first and second maturation divisions when tetrads are formed, 

 i. e. when tetrads are not formed a rest stage does follow the first 

 maturation division. The changes resulting in the formation of the 

 second polar spindle are entirely homologous to those preceding the 

 first. A decrease in size occurs again with the expulsion of more 

 chromatin; a spireme is formed which segments into the six chromo- 

 somes (Fig. 20, Plate 12; Fig. 21, Plate 12). These have nearly the 

 same size as the chromosomes of the first spindle and they have 

 much the same structure; there are four chromomeres in each as 

 may be seen in favorable spindles (Fig. 12, Plate 12). 



Though I have studied manj^ hundreds of sections of the ovaries 

 of H. dioecia my material is still insufficient to provide as complete 

 a series of stages as I was able to find for the spermatogenesis. 

 The desired %g^ stages are, of course, much less numerous in an 

 animal than the corresponding stages of the spermatocytes since 

 sperm are produced in immense numbers. Then too the chromosomes 



