260 SUMMARY OF CURRENT RESEARCHES RELATING TO 



funnels, for it extends from slightly in front of the one to slightly behind 

 the other. The whole pronephros lies in a region which is subsequently 

 included in the skull, in stages 44-45 the niesouephric tubules appear, 

 usually about 13-15 segments behind the pronephros; their origin is. 

 somewhat uncertain. Generally, in regard to the development of the 

 pronephros, Ceratodus resembles Amphibia, especially the Urodela, and 

 the research confirms the close relationship of Amphibia and Dipnoi, 

 which is rendered probable on other grounds. 



Reproduction of Salmon in Fresh Water.* — Jousset de Bellesme has 

 succeeded in showing, after ten years' experimenting, that salmon may 

 be reared to maturity in fresh water, and may reproduce successfully. 

 He succeeded with two pairs, but he looks forward to rearing an entirely 

 fresh-water race. 



b. Histology. 



Centrosome and Centriole.f — Prof. Th. Boveri publishes an import- 

 ant paper on this subject. The paper is divided into three sections : the 

 first discusses the value of iron-haeniatoxylin staining in the study of the 

 centrosome ; the second includes minute descriptions of the changes 

 undergone by the centrosome in cell-division in four cases (spermato- 

 cytes and segmenting eggs of Ascaris, ovocytes of Diaulula, segmenting 

 eggs of Echinus) ; while the third is a general comparative discussion of 

 the significance of the centrosome. Limitations of space render it 

 necessary to confine attention to this third section. While in many 

 cases the centrosome can be distinguished by its high refracting power, 

 in others it is differentiated only by its relation to the sphere ; iron- 

 haamatoxylin cannot be relied on as a differentiating reagent. In the 

 general centroplasm a central corpuscle, or centriole, is distinguishable, 

 and has in certain cases been mistaken for the centrosome itself. The 

 size of the centrosome varies with that of the cell, and more nearly with 

 that of the spindle in a dividing cell ; the occurrence of a rhythmic series 

 of variations in size is indeed a marked characteristic of the centrosome. 

 Division of the centrosome is preceded by division of the centriole ; but 

 while the latter process appears to occur after one fashion only, there is 

 much variation in regard to the division of the centrosome. The centriole 

 appears to bear the same relation to the centrosome that the latter bears 

 to the cell in the process of division. As to the relation of centrosome 

 and centriole to sphere, the author finds that it is the centrosome itself 

 which generates the sphere ; the centriole initiates division, but neither 

 serves as a point of insertion for the radii, nor as the originator of the 

 sphere. These facts serve, with others, as means < >f distinction between 

 centrosome and centriole. As to the function of the centrosome, it is to 

 be regarded as an apparatus which automatically (rnaschinenmassig) 

 brings about cell-division, by reason of the fact that it is capable of 

 responding to stimuli received through the cell. The effect of such 

 stimuli is to permit the centrosome to resume its inhibited cycle of 

 changes, which culminate in the kinetic phase, leading to division first 

 of centrosome and then of cell. But all cell-division is not dependent 



* Comptes Rendus, cxxxii. (1901) pp. 272-4. 



t Jenaische Zeitschr. f. Naturwiss., xxxv. (1901) pp. 1-220 (8 pis. and 3 figs.). 





