ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 397 
At present the cotton-wool is only lightly pressed into the neck 
and now the flask is dry-sterilized. This done, the flask is about half- 
filled with the liquid medium and then thrice (discontinuously) steam- 
sterilized. 
The medium is next inoculated, and the end E' of the tube is pushed 
down until it nearly touches the bottom of the vessel. Hydrogen is then 
introduced at the E end in the usual manner, and it passes through the 
medium, to escape through the cotton-wool plug. 
;Fig. 54. 
After some time the glass tube E E' is drawn up to the position in- 
dicated in fig. 54, and under the E end is placed a vessel containing 
glycerin and supported on a wooden block. 
When all the air has been driven out of the flask, the neck is filled 
with paraffin P, and the caoutchouc tube from the gas-generator re- 
moved from E. Thus air is prevented from getting into the apparatus 
by the paraffin at one end and the glycerin at the other. The bulb K 
is to prevent the glycerin (which is preferable to mercury, as this 
sometimes damages the copper parts of the incubator) from running 
up the tube. When necessary, the cotton-wool plug is easily with- 
drawn by gently heating the neck of the flask. 
Self-regulating Constant Incubator.* — Prof. L. Landois describes 
an incubator, the chief merits of which are that it is quite free from 
danger, and that it requires for its construction materials such as are to 
be obtained anywhere. The source of heat is a mineral oil lamp or a 
specially made stearin candle, so that both gas and electricity are dis- 
pensed with (fig. 55 ). 
* Centralbl. f. Bakteriol. u. Parasitenk., xiii. (1893) pp. 256-62 (1 fig.). 
