44 II. M. WOODCOCK. 



spicnous grains as centrosplieres ^ or centrosomes. Notwith- 

 standing the size to which they attain^ I am now" inclined to 

 consider them as move comparable to true centrosomes than 

 to archoplasraic structures. They are entirely homogeneous 

 in appearance^ and even when by chance less intensely 

 stained than usual^ in no case can any distinct centriole or 

 granule be perceived inside them ; moreover, they have a 

 sharply defined outline or limit. They are very different 

 from the centrospheres in Monocystis, as figured by 

 Cuenot and Prowazek (loc. cit.); nor do they appear to have 

 anything in common with the acidophile structures (probably 

 of archoplasmic nature, corresponding to attraction-spheres)^ 

 which Brasil (3) describes in connection with the nuclei of 

 U r o s p o r a . 



On tlie other hand, the centrosomes in my fig. 51 recall 

 the prominent grains at the heart of the division-centres in 

 Diplocystis (Cuenot, loc. cit., fig. 55). Again, in their 

 deeply-staining property and practically homogeneous ap- 

 pearance they agree with the centrosomes figured by Leger 

 (loc. cit.) in nuclear mitoses of Sty lorhynchus, and also 

 with the "centrioles" of Urospora (Brasil, loc. cit.), the 

 chief point of difference being that in both these cases the 

 size of the granules is less.* 



With regard to the origin of the centrosomes in C. irregu- 

 laris, they certainly arise in intimate connection with the 

 germinal nuclei, and probably with these onl^-. Each is 

 formed just at the periphery of the nucleus, and is itself most 

 likely of nuclear (chromatic) origin. Probably the ccntro- 



' I follow Wilson (39) in considering as centrospliere the inner, central 

 region of ilie aster (consisting of the so-called "medullary" and "cortical" 

 zones), restrictingthe term "centrosome" to the grain or granule, situate, when 

 present, actually at the centre of the astral sjstem (Boveri's "centriole"). 



- In my preliminary note (40) 1 termed them "centrospheres." 



3 The term " attraction-sphere " may be used in a general sense, as including 

 the whole division-centre. 



* Ilertwig (13 a), however, has described relatively large centrosomes in 

 connection with certain nuclei in Actinosj)h8eriuni (vide his figs. 2 and 3, 

 pi. 5) ; these show a general resemblance to those of Cystobia. 



