634. REPORT—1899. 
to the energy per c.c. of the medium. This is the same as in Maxwell’s theory, 
so that there seems very little more besides interpretation of symbols to make a 
turbulent liquid a satisfactory explanation of the structure of the ether 
I am myself satisfied, though I think Lord Kelvin is not, that the turbulency 
of a sufficiently fine-grained purely and irregularly turbulent liquid would ulti- 
mately become so slow in its diffusion from place to place that Lord Kelvin’s 
investigation would apply to it. 
3. On the Permanence of certain Gases in the Atmospheres of Planets. 
By G. H. Bryan, Se.D., PRS. 
In a paper read before the Nottingham Meeting of the Association, the author 
discussed the application of the kinetic theory of gases to explain the absence of 
an atmosphere from the moon’s surface. In the present investigation similar 
methods are applied to the atmospheres of planets, account being taken of the axial 
rotations of the planets. A test of the permanence or otherwise of different gases 
in the atmospheres of different celestial bodies at different temperatures has been 
obtained, and a superior limit has been found for the rate at which any planet 
would lose any gas by the molecules flying off from its atmosphere. To interpret 
this limit in the simplest possible form, the author has calculated the number of 
years which would have to elapse in various cases before the quantity of gas lost 
would be equivalent to that contained in a layer one centimetre thick, covering the 
surface of the earth. For simplicity absolute temperatures of 200°, 300°, 400°, 
500°, and 600° have been chosen—z.e. Centigrade temperatures of — 78°, 27°, 127°, 
and so forth. 
In the case of terrestrial hydrogen the loss in question would occupy 
84,000,000,000 years at temperature — 73°, 600,000 years at — 25°, and 222 years 
at 27°C. 
For helium on the earth’s surface, the corresponding numbers are 3'5 x 10°° 
years at — 73°, 3x 10", or 30 trillion years at 27°, 84,000,000,000 years at 127° 
600,000 years at 227°, and 222 years at 827°. This assumes the molecular weight 
of helium to be 2. 
For vapour of water on Mars, the figures are 1-2 x 10°° years at —78°, 1:9 x 10", 
or 19 thousand billion years at 27°, 2,400,000,000 years at 127°, 43,000 years 
at 227°, and 106 years at 327°. 
The removal of a layer of air 1 centimetre in thickness from the surface of the 
earth would only mean a lowering in the average barometric pressure of 73495 ofa 
millimetre, roughly. Suppose then that the afore-mentioned gases were present in 
the respective atmospheres in sufficient quantity to produce pressures comparable 
with one atmosphere, and assume that a fall of one millimetre in the average 
height of the barometer is the least secular change that could be detected; the 
above-mentioned intervals of time would have to be multiplied by 18,600 roughly, 
in order to give the numbers of years in which the escape of the respective gases 
could be detected by a barometer. 
The only possible conclusions from these results are— 
(1) That helium could exist practically permanently in our atmosphere at ordi- 
nary temperatures. 
(2) That watery vapour could exist practically permanently in the atmosphere 
of Mars at ordinary temperatures. 
(8) That if helium once existed in appreciable quantities in the earth’s atmo- 
sphere, it must have escaped when the earth was far hotter than at present. 
(4) That a similar conclusion holds good on the supposition that Mars once 
possessed, but has now lost, vapour of water as a constituent of its atmosphere, 
the temperature-limit at which the loss ceased to be appreciable being, however, 
lower than for terrestrial helium. 
(5) That hydrogen, on the other hand, may escape from the earth’s atmosphere, 
even at ordinary temperatures, to such an extent as may perhaps appreciably 
affect its permanence, 
i 
