24 



Also on some of these points I have a few remarks: 1. A nu- 

 eleus coiild never be found. But we should consider — Brandt 

 mentions it too — that Lankester never used Brandt's staining 

 metliod but always picro-carmine only. '2. No differentiation of 

 protoplasm belonging to each chlorophyll corpuscle could be seen 

 in the intact amoebocyte. That is absolutely inexact; to me the 

 contrary proved to be the case. 



I am noiv going to show^ hy decisive jyroofs, that the chloro- 

 2)hgU corpuscles of the Spongillidae are real algae associated in 

 y^sgmhiosis^^ 'to the spo?ige^ just as Brandt declared. For that pur- 

 pose I will make use of two sets of proofs, one being morpho- 

 logical, the other physiological. 



a. Morphological proofs; description of the structure 

 of the green chlorophyll corpuscles. 



I am going to treat the living green corpuscles of Spongilla 

 only, as those of Ephydatia are quite the same. 



1. Protoplasm and chloroplast. The shape of the corpuscles is round 

 or somewhat oval ; the diameter 1.7 — 3.8 ^ci, generally 2 — 3 |U. Under 

 oil-immersion we observe that in most corpuscles the green cliloro- 

 plast takes exactly one half of the body, the other half con- 

 sisting of uncoloured protoplasm (Fig. 5). The separating line 

 between chloroplast and protoplasm, wliich in oval bodies is always 

 situated along the longest axis, may be bent a little. But there 

 are also numerous corpuscles with differently shaped chloroplasts 

 (Fig. 12 — 16, 23 — 24, 30), generally taking more than one half 

 of the corpuscle; or even two chloroplasts in one corpuscle. The 

 mutual relation between these different forms will be discussed 

 afterwards. It is a matter of course that one same chlorophyll 

 corpuscle, seen from different sides, will show different aspects; 

 so Fig. 24 and 30 give an aspect of Fig. 13, and Fig. 23 an aspect 

 of Fig. 5 and 12 — 15. The protoplasm of tlie corpuscles appears to 

 be more or less hyaline, not completely homogeneous, but usually 

 containing some diffuse, darker spots (Fig. 5). The chloroplasts, 

 on the contrary, appear to be completely homogeneous ; their 



