314 THOS. H. MONTGOMERY, 



were centrosomes, especially since in the following prophase we shall 

 find that a pair of centrosomes comes out of each idiozome cup. But 

 in the majority of cases such granules were not found in the rest 

 stage; and in other cases three or more granules, which may differ 

 in size (Figs. 104, 109, 121), are found in the cups; and none of 

 these granules stain very intensely. Thus such granules might re- 

 present simply portions of a cytoplasmic reticulum which had pene- 

 trated into the cavity of the idiozome cup; for these reasons I must 

 conclude that centrosomes could not be demonstrated with any cer- 

 tainty whatsoever in the rest stage, even by strong iron-haematoxyline 

 staining which rendered very clear in the same sections the centro- 

 somes in mitosis. On the other hand it is quite possible that centro- 

 somes are present in these cups in the rest stage, but may escape 

 observation by not at all or only slightly staining, for without any 

 doubt, as will shown, centrosomes afterwards come out of them 

 (Plate 21, Figs. 128 — 134, 138), The question of the persistence of 

 the centrosomes in these cells of Feripatus is further complicated by 

 the fact, that in the early anaphase the centrosomes of the last 

 spermatogonic mitosis lie at the central pole of the cell, while the 

 idiozome mass first appears at the distal pole, close to the Zwischen - 

 körperchen plate. In the early anaphase, before the mantle fibres of 

 the spindle have disappeared, a pair of exceedingly minute centrosomes 

 are present at each pole of the spindle; in no later stage have I 

 been able to detect any centrosomes at this pole of the spermatocyte. 

 In a somewhat later stage than this was the first sign of the idiozome 

 mass, but at the opposite pole of the cell. What is the explanation 

 of these facts? It must be either 1) that the centrosomes at the 

 central pole of the cell disappear, and a pair of new centrosomes 

 afterwards arises in the idiozome body; or 2) that the original pair 

 of centrosomes migrate from the central pole of the cell to the distal 

 pole, there enter the idiozome body, and possibly persist there through 

 the rest stage to finally become the centrosomes of the 1st maturation 

 mitosis. There is another possibility, namely that the idiozome mass 

 may be formed at the central pole of the cell around the centrosomes, 

 and then with the latter migrate around to the distal pole; but this 

 seems to be very improbable, for no stages were seen where the 

 idiozome body was half way between the two poles of the cell. Ac- 

 cordingly I have no means of determining whether the centrosomes 

 of the 1st maturation division are the same or direct derivatives of 

 those of the last spermatogonic division, i, e. whether there is a per- 

 sistence of the centrosomes through these generations of cells. 



