ON THE CARBONIC ACID, ORGANIC MATTER, ETC., IN AIR. 
G3 
placed in a horizontal position and turned round until the jelly is just beginning to 
set; the whole of the inner surface of the tube is by this means coated with jelly, and 
a flat surface is formed at the bottom, where most of the jelly collects before setting. 
When the tube is to be used it is set horizontally on a tripod, or other convenient 
stand, with the flat surface of jelly on the lower side of the tube. The outer cap is 
then removed, and a measured quantity of air drawn slowly through the aperture in 
the inner cap and along the tube by means of an aspirator, the india-rubber tube of 
which is connected with the end of the small tube passing through the cork of the 
large tube. For aspirating, two graduated bottles are used, one being hung to the 
upper part of the tripod, and the other standing on the ground. The rate of flow is 
regulated by the size of the glass nozzle through which the water flows into the lower 
bottle, or by means of a small screw clip. 
The whole of the micro-organisms in the aspirated air always settle on the lower 
half of the tube; and when a proper rate of flow is chosen the resulting colonies 
(which develop in the course of a few days) appear closest near the cap, gradually 
becoming more and more sparse as the farther end of the tube is approached, so that 
the last part of the tube is entirely free from colonies (see Plate 6, figs. 2-5). For 
further information we must refer to Hesse’s paper; but it is desirable to mention 
more specially some details of the method as it was employed by us. As regards 
length and width of tubes, we did not consider it necessary in our experiments to 
adhere strictly to one size. It is exceedingly difficult and expensive to procure tubing 
of the same width and in such lengths as will cut up without great waste into 
pieces of say 70 cm. long; and for our purpose the advantages would have been 
scarcely appreciable, 
Hesse recommends that the tubes should have edges turned out at one end, in order 
that the caps may fit more tightly. This, however, increases the risk of cracking during 
sterilisation, and makes it difficult to remove or replace the cap without a good deal 
of inconvenient manipulation, which increases the danger of artificial infection of the 
air. We found it better to employ square pieces of india-rubber sheeting, which were 
fastened by means of stout umbrella rings. The inner cap is first fixed on, the part 
of the cap next the end of the tube being freed from creases by slightly stretching 
the cap. It is then easy to fix the outer cap by means of a ring applied quite close to 
the end of the tube. If any creases are left under the ring the tube is sure to leak. 
An additional ring applied higher up keeps the edges of the square in proper position. 
W r hen the tube is used the last-mentioned two rings are removed, and the outer cap 
can then be taken off without any manipulation likely to infect the air. During the 
experiment the cap is placed in a solution of corrosive sublimate and replaced still 
moist, the superfluous drops having, however, been first carefully shaken off. 
The apparatus used for sterilisation was essentially the same as that employed by 
Hesse, and consisted of a sufficiently large cylinder of iron plate, about 3^ feet long, 
fitted on the top of an ordinary potato-steamer and provided with a perforated lid. 
