262 C. S. Cook — Spectrum of Aqueous Vapor. 



feet of air contained the most water — when its humidity was 

 greatest. After returning home, the figures obtained were 

 tabulated, and the results platted in curves. To my surprise 

 no such relation appeared t Indeed, there was rather more than 

 a suggestion of the contrary — the paradoxical appearance that 

 the more vapor present in the 3,000 feet of air, the less the 

 light absorptive effect produced by it. The effect of altitude 

 of vapor was then more closely considered, and it was at once 

 apparent that if there were great variation of altitude its effect 

 might be as observed. Let us illustrate this point by the facts 

 observed. On many days, so little vapor was there above the 

 level of the summit of the mountain that the aqueous vapor 

 spectrum contained no line of measurable strength ; indeed, it 

 was sometimes invisible. At the same time at the lower station 

 the line was very easily measured. Indeed, once only did the 

 line utterly fail to appear at the lower station. This was after 

 a long storm, and about twenty -four hours after the rain 

 ceased, and about twelve hours after the clouds broke away. 



On another occasion, the day following the close of a heavy 

 storm, about thirty hours after the rain ceased, and about seven 

 hours after the clouds broke away, a very faint spectrum was 

 observed at the summit. Here it could be directly seen that 

 there was much vapor at a low level, with exceedingly little at 

 a high level. The spectroscope acted just like a telescope. 

 When pointed at the horizon, a very distinct line was seen 

 measuring 25 units. When elevated 5 degrees, a very faint 

 line of only 6 units was seen. A most striking and unusual 

 difference for this small difference of altitude. The lower 

 stratum of air in direct contact with the earth was well 

 charged with moisture by evaporation from the wet earth, 

 while the upper air was completely exhausted of moisture by 

 the storm which had passed. The air was exceptionally clear 

 at this time — as clear as I have ever seen it. 



During several equally clear days in September, when the 

 weather was cool, and the air very dry, the line at the horizon 

 ranged between and 10 units. It was only in fair weather 

 and long in advance of rain that the spectroscope revealed no 

 vapor above the mountain top. 



Returning now to the point under consideration, we have 

 noted a condition of the atmosphere when there is no vapor 

 line to be seen at the top, with a moderately strong line of say 

 25 units at the base. The spectroscope difference is then 25 — 

 or the whole 25 units. The maximum difference between the 

 two stations is then obtained when the vapor line is at the 

 summit. This difference will be a minimum when there is a 

 very great amount of vapor above the summit. It is easily 

 seen that if the vapor extends to a height of several miles in 



