314 H. M. BERNARD. 



cesses uniting tlieni into a network (see fig. 16). Their 

 individuality as nuclei is^ however, never quite obliterated, 

 but the union of their processes is so uniform throughout 

 the sections that there is no other conclusion possible than 

 that their chromatin substance must have run out along pre- 

 existing connecting filaments. Here, again, the filaments are 

 made visible by the presence of staining matter upon them. 



4. In some of the tadpoles killed at night all the filaments 

 are clotted over with staining matter, as shown in figs. 

 18 a — d, and are then quite visible, sometimes as a tangle, 

 even with moderate microscopic powers. Here, again, the fila- 

 ments themselves are only seen indirectly owing to the stain- 

 ing- matter which coats them. 



5. By focussing up and down with a high power rather 

 rapidly, individual nuclei can sometimes be seen to possess a 

 subtle radiation, although it is difficult to focus any particular 

 ray. They have at times reminded me of some small short- 

 rayed Sun-animalcule. At other times, as in fig. ll ,h and d, 

 the rays of some of them become clearly visible — owing, I 

 believe, to their being coated with staining matter. 



6. In fig. 17 a, the chromatin is contracted into the centres 

 of the nuclei, which, as we shall see elsewhere, is probably a 

 periodical phase passed through by the nuclei alternating 

 with a phase in which the chromatin is diffuse, as in 17 h. 

 In figs. 17 c and d, we see nuclei passing from one phase into 

 the other. ^ The filaments which run from the central chro- 

 matin mass to the nuclear " membrane " are continuations of 

 the connecting filaments, but the part within the nuclear 

 membrane is usually more distinctly visible than that without, 

 probably because more thickly coated with staining matter. 

 We see very clearly from tliese figures that the connecting 



' These nuclear pliascs have already been recorded as vital phenomena. 

 Ilertwig lias recorded I hem in Actinosplia'riuni, the contracted phase being 

 found iu starving forms (' Abh. k. Bay. Akad. Wiss.,' xix, 3 [1898], p. G37) ; 

 and de Nabias noticed contraction of the chromatin in Gastropods on the 

 application of chloroform (see ' Trav. Station Zool. Arcachon,' 1899, p. 3G). 

 Tlie nuclear phases in the retina will be discussed later ou (see below, p. 317). 



