Carbon Assimilation. 129 
Table IX. 
Diameter of each hole 0-380 mm. 
Length of tube 1-0 cm. 
Distance of holes 
apart in diameters. 
Number of holes per 
sq. cm. of septum. 
Percentage area of holes 
on unit area of septum. 
Septum diffusion 
Open tube diffusion X ^ 
2-63 
100 00 
11-34 
56-1 
5-26 
25-00 
2-82 
51-7 
7-8 
11-11 
1-25 
40-6 
10-52 
6-25 
•70 
31 -4 
131 
4-00 
•45 
20-9 
15-7 
2-77 
•31 
140 
It will be observed from these numbers that the obstruction 
offered to the diffusion of gases by a multiperforate septum is 
considerably less than the actual obstruction of area. Thus when 
the area of perforation was less than 3% of the whole area of the 
septum, the actual diffusion through the perforations was 51-7% of 
the diffusion taking place through an open tube of the same area of 
cross section. That is, the diffusion through the septum is nearly 
15 times as great as it would be if it were simply proportional to 
the area of the cross section. As the distance between the holes is 
increased, the efficiency of the area of the perforations increases 
until the holes are about 10 diameters apart. In this case and in 
cases where the distance apart of the holes is increased, the diffusion 
through the perforations is about 40 times as much as it would be 
if it were proportional to the area of cross section of the tube. 
The accompanying figure (Fig. 4) illustrates what Brown and 
Escombe imagine to be the lines of equal density and the lines of flow 
of gas through such a multiperforate septum. The lines of flow of gas 
diffusing towards the septum will be approximately parallel at some 
distance from the septum, but as they pass through the perforations 
they converge, the velocity of flow increasing at the same time 
owing to the production of ellipsoidal density shells round the 
opening. After passing through the opening, the lines of flow 
diverge and as lines of flow from adjacent perforations cannot 
cross each other (for otherwise there would be shells of different 
density crossing each other, which is impossible) they must bend 
round and become once more parallel, the velocity of flow at the 
same time diminishing. From a consideration of Fig. 4, it is easy 
to understand how it is that the perforations under favourable 
conditions of distribution are so efficient for diffusion through them. 
Brown and Escombe’s results lead them to the general 
conclusion that “ the interference of the density shells of small 
