BIGELOW: NUCLEAR CYCLE OF GONIONEMUS MURBACHII. 321 



we must pause to consider the most important question connected with 

 the first maturation division ; that is, whether it is a reducing or an 

 equation division. It must he admitted that the internal evidence bear- 

 ing on this point is but slight, and not very convincing. Since there is 

 no apparent formation of tetrads, nor any trace of longitudinal splitting 

 of the spireme strands, it is vain to turn for light on the sequence of the 

 maturation divisions to the prophase, as McCIung (:05.) and Black- 

 m in (: 05) have so successfully done in arthropods and Montgomery ( : 00) 

 in Amphibia. The manner of cleavage can seldom be ascertained with 

 certainty — as is now well understood — by studying the behavior of 

 the chromosomes during the metaphase, since they are then so compacted 

 that the planes of cleavage, if any exist during the late prophase, are 

 wholly obliterated. This stage does, however, afford some little evi- 

 dence, although it is not conclusive. Taking, in the first place, the 

 events of the late prophase, we have seen a series of phenomena, which, 

 except that we are here dealing with only one-half the somatic number 

 of chromatin bodies, are exactly comparable to those seen in the sper- 

 matogonia, in which the division is, beyond any doubt, equational. The 

 separation of the chromosomes, now probably of a viscid consistency, 

 through a gradual elongation and drawing apart of their thickened ends, 

 suggests a very firm coherence of their halves and argues against their 

 separation along a recent plane of synapsis. I believe that the weight 

 of the little internal evidence which exists is on the side of the view 

 that the first maturation division in Gonionemus is longitudinal, and 

 that its daughter chromosomes are to be regarded as dyads. Yet I 

 must at the same time admit that in this animal, as in Lumbricus 

 (Calkins, '95), it is impossible from the history of the chromatin struc- 

 tures, to state definitely which of the two maturation divisions results in 

 reduction and which in equation. Hydra is so nearly related to Gonion- 

 emus that Downing's (:05) results on that form are here of especial in- 

 terest ; I come to the same conclusion as Downing, who bases his view on 

 the evidence that the daughter chromosomes of this division contain the 

 same number of chromomeres as their parents. 



Telophase. — The telophase of the first maturation division presents 

 no features of especial interest. The chromosomes are closely massed 

 near the poles of the spindle (Fig. 68), but later separate again, where- 

 upon it is seen that they have become connected with one another so as 

 to form an irregular network (Fig. 69), exactly as occurs in the preceding 

 mitosis. The new nuclear membrane is now formed and the remnants of 

 both centrosome and spindle disappear. At the same time the chro- 



VOL. XLVIII. — NO. 4 21 



