BIGELOW : NUCLEAR CYCLE OF CONIONEMUS MURBACHII. 333 



gesteil. I have not been able to trace, at any stage, a slieath of either 

 cytoplasmic or archoplasmic material surrounding the tail. 



7. Giant and Multiple Spermatids and Spermatozoa. 



While the vast majority of spermatozoa exhibit the structural condi- 

 tions just described, I have found a considerable number of abnormalities, 

 falling into two well-marked classes, which may be termed respectively 

 "giant" and "multiple." I have been able to trace some of the stages 

 in the metamorphosis of the first of these, finding that they agree in 

 their general features with those of normal cells, and strongly recall the 

 conditions found by Paulmier ('99) in the " giant " spermatids of Anasa. 

 Such spermatids are first distinguishable after the telophase of the sec- 

 ond maturation division, and probably result, as Henking ('91, page 718) 

 and Paulmier ('99, page 254) suggest, from the non-completion of the 

 first or second spermatocyte division. In early stages of metamor- 

 phosis their most striking peculiarity, the possession of several tails, is 

 not yet developed ; for this reason, as well as because they are present 

 in very small numbers, they might easily be taken for the larger cells of 

 an earlier genei'ation. But when they reach the stage shown in Figure 

 103 (Plate 5) their true nature is at once apparent from the multiple 

 condition of tails and centrosomes. They are either double, triple, or 

 even quadruple, as is shown by the size of the nucleus, and by the 

 number of tail filaments and centrosomes. The earliest stages detected, 

 one a double, the other a triple individual, are shown in Figures 103 

 and 104. In the former the nucleus, which is evidently in its period 

 of greatest size, is 3.5 p in diameter and presents an appearance similar 

 to that of the normal spermatid. The whole cell is 6.5 (j. in diameter, 

 and therefore abont twice as large as the normal spermatid. The 

 remnants of the interzonal filaments are still visible as a conical mass 

 in contact with, and near one pole of, the slightly oval nucleus. The 

 principal archoplasmic organ — probably derived here, as in normal 

 metamorphosis, from the remnants of the polar portion of the spindle 

 — consists of a single large body, lying near the opposite pole of the 

 nucleus, and containing three or four deeply staining regions. There 

 are two centrosomes, as yet undivided, lying about 30° apart, at the 

 periphery of the cell, one on either side of the archoplasmic mass. There 

 are also two distinct tails, each a simple filament arising from one of the 

 centrosomes, while extending inward from each of the centrosomes is a 

 short axial filament, not yet, however, reaching to the nucleus. 



