302 UNITED ST A TES DA IL Y A TMOSPHERIC S UR VE Y 



the bend of their parabolic course, at about latitude thirty. They 

 have for years furnished a fruitful theme for the thoughts of the 

 investigator. 



For twenty-seven years the forecasters of the Weather Bureau 

 have studied the inception, development, and progression of these 

 different classes of atmospheric disturbances. From a knowledge 

 personally gained by many years" service as an official forecaster, 

 I do not hesitate to express the opinion that we long since reached 

 the highest degree of accuracy in the making of forecasts possible 

 to be attained with surface readings. It is patent that we are 

 extremely ignorant of the mechanics of the storm, of the opera- 

 tions of those vast yet subtle forces in free air which give incep- 

 tion to the disturbance and which supply the energy necessar}' 

 to continue the same. 



Having long realized this, I determined at once, on coming to 

 the control of the United States Weather Bureau, to systemat- 

 ically attack the problem of upper-air exploration, with the hope 

 ultimately of being able to construct a daily synoptic weather 

 chart from simultaneous readings taken in free air at an altitude 

 of not less than one mile above the earth, as it appeared to me 

 that previous plans for investigating the upper air by means of 

 free and uncontrollable, balloons, by observers in balloons, or by 

 independent kite stations were of little value in getting the in- 

 formation absolutely necessary to improve our methods of fore- 

 casting. Simultaneous observations at a uniformly high level 

 from many kite stations was the plan I inaugurated for the pros- 

 ecution of this important investigation. Professor Marvin was 

 assigned to the difficult task of devising appliances and making 

 instruments, and I am pleased to say that we have improved on 

 kite-flying to such an extent that apparatus is now easily sent 

 up to a height of one mile in only a moderate wind. We have 

 made an automatic instrument that, while weighing less than 

 two pounds, will record temperature, pressure, humidity, and 

 wind velocity. Before next spring we expect to have not less 

 than twenty stations placed between the Rocky mountains and 

 the Atlantic ocean taking daily readings at an elevation of one 

 mile or more. 



We shall then construct a chart from the high-level readings 

 obtained at these twenty stations and study the same in connec- 

 tion with the surface chart made at the same moment. Being 

 thus able to map out not only the vertical gradients of tempera- 

 ture, humidity, pressure, and wind velocity, but the horizontal 



