19 



glass plates. One half of these vessels was exposed to daylight, 

 the other kept in darkness (at the same temperature). After a 

 week's time the sponge globules in all 6 vessels, which had been 

 exposed to light, appeared to be still intact, while those kept in 

 darkness proved to be intact in 2 vessels only, but wholly de- 

 stroyed in the other 4. These observations make us think of a 

 production of 0., by the globules in light. This is the more evi- 

 dent, when we consider that entering of O^ from the air into 

 the vessels was almost completely prevented by their glass covering. 



c. To Engelmann (18) we owe the bacteria-method, that enables 

 us to detect the production "even of traces of O2. It is based, as 

 one knows, on the fact, that many bacteria are brought to vio- 

 lent motion by the mere presence of but traces of 0^, while in 

 lack of O2 they get at rest. To prove this production of 0., by 

 the green corpuscles of the Spongillidae, I proceeded as follows: 

 On the surface of an algal-culture I found a membrane of moving 

 bacteria, from which I grafted a little on peptone-gelatine. So in 

 a few days I got a sufficient quantity of the bacteria. I then 

 ravelled a piece of green Spongilla on a glass slide, etc., etc. (see 

 Table 2). After the preparation had been kept in darkness for '/^ 

 of an hour, the bacteria were accumulating and violently moving 

 round the air bubbles, but elsewhere they had almost got at rest. 

 This proved that here we had to do with bacteria sensitive to 

 O2, as we wanted. After another period of darkness of ^/^ of 

 an hour, the movements round the air bubbles also almost had 

 stopped. When, next, the green sponge preparation was exposed 

 to light, we could observe under the microscope that, little by 

 little, the movements of the bacteria were resumed all over, till 

 the original intensity was reached. We might repeat this ex- 

 periment numerous times, always with the same result (see Table 

 2) : in darkness the movements stopped, in light they were 

 resumed. Accordingly, the preparation proved to contain many 

 amoebocytes crammed with green corpuscles, and numerous of 

 those corpuscles isolated. So these last ones produced in light 

 the 0^, necessary for the movements of the bacteria. 



I made this experiment with several preparations, also from 



