760 



SCIENCE 



[Vol. LVI, No. 1461 



in close proximity to the proximal one at the 

 nuclear wall; and in cells too heavily stained 

 with crystal violet they appear as a single 

 body, a fact Avhich may account for Driver's 

 view that only the firoximal centriole remains 

 near the nucleus. A full account of the oir- 

 cumstanees connected with the origin and 

 migration of the distal centriole derivative will 

 be incorporated in a later paper. 



Accounts of a distal migration of a centriole 

 or its derivatives in the spermatogenesis of 

 invertebrates are, as already indicated, rare. 

 Only among insects have such cases been re- 

 corded. Pei-ihaps the clearest and most con- 

 vincing statement is that of Otte ('07)^ for 

 Lociista viridissima. Some of his figures, nota- 

 bly No. 87 and No. 92, bear a strong super- 

 ficial resemblance to certain cells to be found 

 in the Q5«anthus preparations. Otte, like most 

 of the other observers of similar phenomena, 

 considers the migi'ating body to be an entire 

 distal centriole. My observations upon the 

 origin of the migrating body in CEcanthus 

 caused me to become skeptical of similar re- 

 ports in other cases; and through the kindness 

 of Dr. 0. L. Mohr of the University of Kris- 

 tiania, who has sent me testes of Locusta 

 viridissima fixed in Benda's solution, I was 

 enabled to make a re-examinaion of the sper- 

 matids of this insect. A preliminary survey 

 of the material, while not entirely conclusive, 

 indicates that the centrioles of the "neck- 

 region" of young spermatids are two in num- 

 ber prior to, and throughout, the period in 

 which the eentriole-like body performs its 

 migration to the posterior pole of the cell. 

 This body, therefore, appears to be a centriole 

 derivative and not the distal centriole itself. 



The discovery in CEcanthus of a centriole de- 

 rivative which migrates along the axial thread 

 to a position remote from the nucleus is of 

 interest because it presents certain transitional 

 features between two previously known and 

 common types of spermatid metamorphosis. 

 The first of these types has been reported by 

 numerous observers upon the spei-matogenesis 



1 Otte, Heinrieh : ' ' Samenreif ung und Samen- 

 bildung bei Loeusta viridissima," Zool. Jalirh., 

 Bd. 24, Heft 3, S. 431-521, 1907. 



of various invertebrates, including most of the 

 insects. In the early spermatids of these ani- 

 mals two centrioles lie together in the neck- 

 region of the young spermatid. One of these, 

 the distal centriole or blepharoplast, spins out 

 an axial thread. The two centrioles remain 

 practically unchanged in the same region 

 throughout the metamorphosis of the spenna- 

 tid, and no peripheral migration of centriole 

 derivatives occurs. The second type is charac- 

 teristic of certain vertebrates, notably the 

 mammals, and has been described by various 

 authors for man, rat, guinea-pig, Phalangista, 

 etc. In the spermatids of these forms two 

 centrioles also remain in the neck-region, but 

 the distal, after spinning out the axial thread, 

 cuts off a ring-shaped body which encircles the 

 thread and migrates distalljr, at least as far as 

 the terminus of tftie middle-piece. There is no 

 well-defined middle-piece in the insect sperm, 

 and, as already stated, the migrating body, 

 closely analogous to that of the mammal 

 sperm, passes to the posterior margin of the 

 caudal sheath. 



This study also has a bearing upon certain 

 phases of the problem of fertilization, notably 

 the origin of the first cleavage spindle. Many 

 observers hold that the first cleavage centers 

 arise from, or in the immediate neigibborhood 

 of, the neck-region of the sperm as it enters 

 the egg. It is possible that the proximal and 

 distal centrioles maintain their individuality 

 and foi-m each a pole of the spindle, but this 

 must remain a matter of conjecture. It is note- 

 worthy, however, that in practically every ani- 

 mal that has been critically studied,- por- 

 tions of both proximal and distal centrioles 

 pass into the neek-region of the mature sperm. 

 In this respect CEcanthus falls directly into 

 line with the vast majority, and probably also 

 Locusta. 



H. H. Johnson 



Columbia University, 

 New York City 



- Apparently an exception obtains in certain 

 MoUusca; vide Gatenby ('18): "The Cj'to- 

 plasmic Inclusions of the Germ- Cells. Part III — 

 The Spermatogenesis of Some Other Pulmonates. ' ' 

 Quart. Jour. Mic. ScL, N. S., Vol. 63, No. 250, 

 pp. 107-258. 



