8. Cook- Spectrum of Aqueoui Vapor. '2 r> L 



The mountain selected was Moosilaukee Mountain, one of 

 the outlying peaks of the White Mountain range. The alti- 

 tude above sea level ie 4,811 feet Observations were made at 

 the summit at frequent intervals, while an assistant, Mr. E. l'>. 

 Frost, made simultaneous observations at a station well down 



toward the base. The horizontal distance between the stations 

 was about three miles. The difference of altitude was very 

 closely 8,000 feet. In addition to the spectroscope readings, 

 the ordinary meteorological observations were taken. 



On studying the varying appearance of the aqueous vapor 

 lines in the solar spectrum, one is impressed by the very wide 

 range of variation in their intensity. One day the strongest of 

 these lines may be almost invisible, while a few days later it is 

 no exaggeration to say that its intensity is forty times as great, 

 perhaps I might safely say sixty or more times greater. 

 A -sinning as we may that for a single line the light absorptive 

 power of the vapor is proportional to the amount of vapor 

 traversed, we are led to infer that on the second day there was 

 at least forty times as much vapor present as on the first da} T . 

 Does this mean that the absolute humidity was forty times as 

 much 2 The hygrometer shows nothing approaching such a 

 range. During the summer weather rarely does the amount 

 of water in a cubic foot of air on a wet day exceed by more 

 than five times that present on a dry day. Instead of assum- 

 ing that the air holds forty times as much per unit of volume, 

 is it not nearer the truth to assume rather that the vapor 

 extends upward forty times as far in the atmosphere ? I say 

 nearer the truth, with certainly no desire to deny the effect 

 due to other causes. But I do wish to emphasize, and strongly, 

 that the prevailing disposition to associate the varying vapor 

 spectrum with varying humidity, with tacit understanding that 

 other factors are acting, is all wrong. We should associate the 

 varying spectrum with varying vapor altitudes, remembering 

 always that other factors play a subordinate part. It is true 

 that a chart on which are platted together vapor line intensity 

 and absolute humidity does show a very fair, indeed I should 

 a close accordance ; but this is due in the main not directly 

 to an interdependence, but rather indirectly to the fact that 

 damp air and high vapor altitude generally go together. 



At the time the observations were made on the mountain, 

 the importance of the question of vapor altitude was not 

 appreciated. It was thought, as indeed is true, that the 

 difference between the spectra observed at the two stations 

 with identical instruments would represent the absorption of 

 the stratum of the vapor between the stations, i. e., a depth of 

 3,000 feet. It was carelessly presumed that in consequence 

 the spectrum difference would be greatest whenever this 3,000 



