552 Mr. Horace T. Brown [March 22, 



for any plane surface, no matter where we take it within the cylinder, 

 there can be no change in the average distribution of the " green " 

 molecules throughout the cylinder — in other words, no change in any 

 part of the cylinder in the composition of the air as regards its 

 carbonic acid content. 



But now let us imagine that the bottom of the cylinder is suddenly 

 made capable of absorbing carbonic acid, say by the introduction, 

 without any disturbance of the air, of a little solution of caustic soda 

 or caustic potash. The " green " molecules which now strike the 

 bottom of the cylinder at all imaginable angles of incidence, will not 

 all rebound, as they originally did, but will be to a large extent 

 trapped in their to-and-fro excursions, so that in the first very brief 

 interval of time a very thin stratum of air, parallel to and immedi- 

 ately above the absorbing surface, will be partially freed from its 

 " green " molecules. 



Now consider the kind of exchange of " green " molecules which 

 occurs in the next very brief interval of time between this partially 

 depleted layer at the bottom and the one immediately above it. The 

 rate of exchange across the imaginary plane dividing these two con- 

 tiguous layers can no longer be equal and opposite, since the number 

 of " green " molecules in the upper stratum is greater than that in 

 the lower. A larger number of the " green " molecules must conse- 

 quently pass in a given brief interval of time from the higher to the 

 lower stratum than from the lower to the higher ; in other words, the 

 balance of exchange is in favour of the lower layer. This state of 

 affairs will rapidly propagate itself upwards until the mouth of the 

 cylinder is reached ; and provided the air outside the cylinder is kept 

 of the same composition, and the absorptive power of the bottom of 

 the cylinder is also kept constant, the uncompensated balances of ex- 

 change between the imaginary layers may be regarded as constituting 

 a steady flow or drift of the " green " molecules down the tube 

 towards the absorbent surface. 



Although within the column there is this constant flow of 

 carbonic acid molecules in the general direction of the axis of the 

 tube, the system as a whole may now be regarded as static so 

 long as all the conditions remain unchanged. The flow is then 

 strictly analogous to the flow of heat in a bar of metal which is kept 

 with its two ends at a uniform difference of temperature ; or to the 

 flow of electricity in a conductor between two regions maintained at 

 a constant difference of potential ; and static diffusion admits of pre- 

 cisely the same simple mathematical treatment as these phenomena 

 of conduction of heat and of electricity when we come to its quanti- 

 tative study. 



In such an imaginary experiment as we have been considering, it 

 is clear that the amount of carbonic acid in the air of the cylinder 

 must vary uniformly from a maximum at the top of the cylinder to a 

 vanishing jDoint at the bottom, so that if the C0 2 really had the green 

 colour, which for purposes of argument we have attributed to it, the 



