75 



The deercase of the number of cohmrless algae with structure 

 in the gemmule stage will be explained lateron, when speaking 

 about the cause of the dying of the algae in sponge tissue, But 

 I want to remind that we concluded (pag. 42) that this dying 

 is closely related to the life of the sponge ; and we know that 

 the gemmules are stages of rest of the Spongillidae. So it is 

 evident that probably herein is the cause of the decrease of 

 the mortality during that period. That the. total amount of dead 

 (colourless) algae (without structure) decreases in the gemmule 

 stage, has been explained already on pag. 57. 



hl this way we have proved with the help of the data ice oh- 

 tained^ when studying tlie 6 factors (of import^ export^ reduction^ 

 groivth, multiplication and mortality) ruling together the number 

 of the symhiotic algae in sponge tissue: 1. Why in nature the 

 Spongillidae must contain such an amount of the various green 

 and colourless stages of the symhiotic algae in light and in dark- 

 ness, as we have exj)erimentally stated on pag. 46 — 48,pointl — 11. 

 2. Hoiv in nature these sponges keep up their ^colour^' (green or 

 colourless)^ and how both ,^coulour'"-types arise from each other 

 (pag. 35). 



As mentioned above, up to now I have exclusively treated the 

 cases normally occurring in nature, viz. those of green sponges 

 growing in light and colourless ones growing in darkness (pag. 46). 

 We stated, however, that sometimes green sponges might be found 

 in nature in darkness and colourless ones in light (p. 41); and 

 in Table 8, pag. 63 — 66, that as an exception to the rule green 

 sponges may become less green in light and colourless ones more 

 green in darkness. How is that to be explained? 



We saw that the amount of green algae, so the colour of the 

 sponge, is ruled in nature by the proportion (/ + r + mu) :{e + g 

 + nio) and in my experiments by mu : mo. We must consider, 

 however, that each of these factors is depending on numerous 

 circumstances. These circumstances proved in general rather con- 



