226 Frofeasor Dewar [April 11, 



the spectrum rivals in strength that of the red and yellow rays. As 

 these gases probably include some of the gases that pervade inter- 

 planetary space, search was made for the prominent nebular, coronal 

 and auroral lines. No definite lines agreeing with the nebular spec- 

 trum could be found, but many lines occurred closely coincident with 

 the coronal and auroral spectrum. But before discussing the spectro- 

 scopic problem, it will be necessary to consider the nature and 

 condition of the upper air. 



According to the old law of Dalton, supported by the modern 

 dynamical theory of gases, each constituent of the atmosphere while 

 acted upon by the force of gravity forms a separate atmosphere, com- 

 pletely independent, except as to temperature, of the others, and the 

 relations between the common temperature and the pressure and 

 altitude for each specific atmosphere can be definitely expressed. 



If we assume the altitude and temperature known, then the pres- 

 sure can be ascertained for the same height in the case of each of the 

 gaseous constituents, and in this way the percentage composition of 

 the atmosphere at that place may be deduced. 



Suppose we start with a surface atmosphere having the compo- 

 sition of our air, only containing y^f Q-^ths of hydrogen ; then at 

 thirty-seven miles, if a sample could be procured for analysis, we 

 believe that it would be found to contain 12 per cent, of hydrogen 

 and only 10 per cent, of oxygen. The carbonic acid practically dis- 

 appears ; and by the time we reach forty-seven miles, where the 

 temperature is minus 132 degrees, assuming a gradient of 3 * 2 degrees 

 per mile, the nitrogen and oxygen have so thinned out that the only 

 constituent of the upper air which is left is hydrogen. If the gradient 

 of temperature were doubled, the elimination of the nitrogen and 

 oxygen would take place by the time thirty-seven miles was reached, 

 with a temperature of minus 220 degrees. 



The theoretical distribution of the chief components of our atmo- 

 sphere, on the assumption of steady equilibrium, is graphically 

 represented in Diagrams II. and III. In the diagrams nitrogen is 

 represented by the red colour, oxygen by the blue, hydrogen by the 

 yellow, argon by vermilion, and carbonic acid by black. A hori- 

 zontal line drawn across the diagram at any height marked in kilo- 

 metres (0 • 62 mile) shows the percentage by volume of the constituents 

 at that elevation by the respective lengths of line in the colour of the 

 individual constituents. The results of Hinrich's calculations which 

 involve no consideration of the effects of temperature, are represented 

 in Diagram II., and those of Ferrel, who assumes a temperature 

 gradient of 4° per kilometre, throughout the upper air, in Diagram III. 

 The higher the assumed temperature gradient the lower the elevation 

 at which the nitrogen and oxygen are eliminated and the true hydro- 

 gen atmosphere begins. The elevations marked A, B, C and D in the 

 diagram refer to the respective gradients of 4°, 3°, 2° and 1° per 

 kilometre, and mark the end of nitrogen or the beginning of the true 

 hydrogen atmosx^here. The jjosition A corresponds to 60 kilometres 



