WEATHER — WOOLARD 225 



been initiated (largely through the work of William Ferrel in the 

 United States about 1855, pi. 2), progress was relatively slow in the 

 attempt to interpret the current weather phenomena shown on the 

 synoptic charts in terms of the physical processes in operation. 



Weather forecasts were based fundamentally on the fact of obser- 

 vation (demonstrated by Benjamin Franklin, Leverrier, Fitzroy, and 

 Buchan) that existing weather conditions travel over the globe in a 

 fairly regular manner from day to day, altering more or less as they 

 move (fig. 1 and pi. 3, figs. 1 and 2) ; and from a detailed study of 

 the daily movements and transformations recorded on series of past 

 synoptic charts, it is possible to estimate from a given current map 

 the weather that will occur in the near future. For the most part, 

 weather phenomena were empirically associated with the barometric 

 pressure distribution, particularly with the cyclones or "lows" (re- 

 gions of relatively low pressure), and anticyclones or "highs" (re- 

 gions of relatively high pressure) , which are continually moving over 

 the temperate regions; and the emphasis was mainly on the weather 

 conditions that had in the past been observed usually to accompany 

 the particular movements and developments of these pressure forma- 

 tions that were to be expected on the basis of past experience, with 

 little consideration of physical explanations and almost no attempt 

 to make direct use of any physical principles. 



The obstacles arise partly from the complexity of the phenomena, 

 and partly from the difficulty of securing adequate observational 

 data and effectively utilizing them. A complete investigation of the 

 weatlier phenomena over a given region requires regular observations 

 at fairly frequent intervals, both from the surface of the earth and 

 aloft to moderately great heights in the free air, over a territory that 

 extends far beyond the limits of the area under immediate consider- 

 ation; a satisfactory understanding of the weather over the entire 

 continental United States, therefore, would necessitate adequate data 

 for at least the region from the mid-Pacific Ocean to Bermuda and 

 from northern Alaska, Canada, and Greenland to Mexico and the 

 West Indies, and it would be desirable to extend this area over a still 

 larger part of the northern hemisphere. This ideal is far from hav- 

 ing yet been attained, and much remains to be learned as to how to 

 interpret and apply even all the data that now are available; but 

 during recent years, notable increases have been effected in the niun- 

 ber of reporting stations, including ships at sea and stations for 

 upper air soundings, as well as in the extent of territory covered by 

 the observing network, and in the completeness and frequency of the 

 observations; and steady progress has been made in utilizing the 

 data effectively, from both the empirical and the physical viewpoints. 



The empirical association of weather conditions with pressure dis- 



