WEATHER BUREAU. 91 



number of those reporting temperature conditions has resulted in 

 keeping that class of stations down to a number thought sufficient 

 for all needs, at the same time arranging for a few new stations at 

 points where temperature observations are considered of value. 



On the other hand, effort is being made to increase the niunber of 

 rainfall measuring stations in sections where local topography 

 causes marked variations in the precipitation at near-by points. It 

 is thought no difficulty will be experienced in securing cooperation 

 along that line at various additional points, and recommendation 

 has been made that appropriation for the purchase of an increased 

 number of gauges be provided for. 



Inspection of cooperative stations. — The withholding of travel ex- 

 penditures for economical reasons and because officials could not 

 absent themselves from their stations due to a general shortage of 

 force has prevented the needed inspection of substations to a con- 

 siderable extent. As a result some deterioration in the character of 

 reports rendered has undoubtedly occurred, due to agencies not 

 readily apparent to the cooperative observers. The most important 

 of these is the tendency of maximum thermometers to become " re- 

 treaters," a condition of the thermometers not readily apparent to 

 the observer, but easily detected by the inspector. The increasing 

 number of instruments developing this defect has been commented 

 on by several inspectors. The ultimate cause of this is a- falling off 

 in the high quality of maximum thermometers as now produced by 

 manufacturers. 



Another serious condition to be guarded against is the constant 

 tendency to an improper exposure of the rain gauge. Faults of this 

 kind are not readily noticed by the observer and are most surely 

 remedied by an inspection. 



Xo expenditure of Government funds will render more valuable 

 returns than those providing for the reasonably frequent inspection 

 of cooperative stations, which furnish such a great volume of most 

 valuable data at no cost except those of forms, instruments, and in- 

 spections. 



Cliinatological publications. — The material for the Annual Report 

 of the Chief of Bureau and for the Monthly Weather Review has 

 remained practically as during previous years. 



The Snow and Ice Bulletin, due to congressional action, was not 

 issued during the first few weeks of the season, but later on it was 

 incorporated in the new department publication. Weather, Crops, and 

 Markets. This soon proved unsatisfactory on account of late issue, 

 and the Snow and Ice Bulletin was therefore issued as an individual 

 publication in nearly the form adopted prior to 1919. 



The 45 sections of the Climatological Service are issuing their 

 monthh' and annual summaries substantially as last year, save that 

 a few sections have requested permission to increase slightly the 

 number of pages per issue which, for reasons of economy, had been 

 reduced to the lowest possible extent. 



The somewhat smaller size of these publications has enabled their 

 printing and distribution at a distinctly earlier date than had been 

 possible for several years past. 



The publication of the Alaska Section Summary, which had fallen 

 greatly in arrears, due to certain printing difficulties, was finally 



