II eat and its Importance in Meteorology. 371 



early observations of the Signal Service at Cheyenne and sub- 

 sequently at Pike's Peak, as displayed on the daily weather 

 maps, we owe the knowledge, that has these twenty years been 

 public property in the United States, that warm air exists 

 above the western side of our great cold waves. Ferrel in his 

 "Recent Advances" explains this upper warm la_yer as due to 

 the compression of the descending air which latter eventually 

 flows horizontally above the lowest layer which latter is com- 

 posed of air that has cooled by conduction to the cold ground 

 near which it lies quietly resting ; he states furthermore that an 

 additional cooling is due to the evaporation of moisture from 

 the soil and also to the '' radiation of the air " but no special 

 importance is attached to the latter factor nor to the fact that 

 often during daylight sufficient solar heat does not penetrate 

 into the lowest stratum to dissipate the haze and fog that is 

 often formed during the nighttime. Hann has presented the 

 details of the continued existence of such a warm layer for 

 many days over the midst of the European area of high pres- 

 sure of ISIov., 1889; his suggested explanations, so far as they 

 go, agree with Ferrel's. 



(j.) But the slow horizontal flow, the evaporation of mois- 

 ture from the ground and the conduction of heat to the ground 

 by the air within a few meters distance, explain only a portion 

 of the observed phenomena. We need a thermal process that 

 shall do for the high area that which the condensation of vapor, 

 and its precipitation as rain or snow, do for the low area. I 

 have been . accustomed to explain this abstraction of heat in a 

 general way as due to the warming of the air by compression 

 and the loss of this evolved specific heat by direct radiation, 

 but the numerical application of this idea was hindered at first 

 by our ignorance of the coefficient of radiation and eventually 

 by the very small coefficient deduced by Maurer. The larger 

 coefficient of Hutchins and a conviction that Maurer's value is 

 too small, lead me with some confidence to ask the attention 

 of meteorologists to the present study from which we find that 

 we need not hesitate to accept Espy's conclusion that a cloud- 

 less blue sky demonstrates that descending movements are 

 present and are more efficacious than any minor ascending- 

 movements that may be present. The existence and quali- 

 tative influence of radiation from the air was discussed by me 

 with Ferrel before the publication of his "Recent Advances" 

 and admitted by him as a possibility, but in the absence of any 

 knowledge as to the coefficient of radiation Ferrel seems to 

 have decided to mention the subject only in a general way. 



(9.) If the air itself radiates its own heat then the tempera- 

 ture of an ascending or descending mass, or the general law of 

 the diminution of temperature with altitude must depend upon 



