430 
Baker . — Quantitative Experiments on 
In the earlier experiments the method of timed weighings was 
employed. Allowance was made for the amount of water lost during 
weighing, by weighing the loss in the next interval of the same time. The 
weighings were done as rapidly as possible, but could not be very accurate, 
because of the continually changing values. 
The second method used was to weigh the wet cultures, or ‘ house ’ 
inside a small paper hood which completely protected them from the out- 
side air, and which could be weighed before or after the weighings. Even 
by this method a small loss, due to evaporation or condensation on the 
hood, occurred, but its amount was not great and its estimation could not 
lead to serious errors. 
Besides these losses, a small loss always occurs in transferring the wet 
cultures to or from the desiccators. This was reduced as much as possible 
by speed in manipulation. 
Precautions to render the Apparatus Air-tight. 
It is of the utmost importance, in these experiments, that the whole 
apparatus should be absolutely air-tight. A considerable number of 
experiments were made before this result could be attained. The first 
essential is to have all rubber connexions irreproachable ; absolutely sound, 
well-fitting corks, and thick pressure tubing, closed with double screw-clips, 
for all connexions. The second essential is to have air-tight ground-glass 
joints round the stoppers and lids of the desiccators. This last was not 
easy to secure. It was found that, though every joint was carefully greased, 
and then sealed over with paraffin wax, after about a week little bubbles of 
air worked their way through the grease round the desiccator lids, collected 
under the wax sheath, and finally cracked the wax and sprung a leak. 
After several experiments had been spoiled in this way, it was found that 
the desiccator lids could be kept air-tight by applying a little pressure from 
the outside to counteract the internal air-pressure. For this purpose my 
brother — Mr. G. R. Baker — designed and made a wooden clip to hold the 
lids tightly together. This consisted of a wooden frame fitting under the 
flange of the desiccator, with four wooden clips screwed down from it on to 
the upper edge of the desiccator lid, by means of metal thumb-screws. Two 
of these are shown in section in Text-fig. 4. When these were screwed down, 
and the lids greased and waxed round, no sign of leakage could be detected 
in any part of the apparatus during the whole time. 
Finally, to prevent the movement of air from one part of the apparatus 
to another when the air-current was not passing, the screw-clips at each 
side of the weighed apparatus and the desiccator were closed, immediately 
after stopping the pump. Only one of these was kept open, that between 
the desiccator and the calcium chloride U-tube, in order to prevent the air 
exit- and inlet-tubes from condensing moisture. 
