64 EXPERIMENT STATION. [Jan. 



REPORT OF THE METEOROLOGIST. 



J. E. OSTRANDER. 



At the close of the year the division will have the meteorologi- 

 cal records for this station for a period of twenty years. The 

 temperature records for the first year of the period (1889) are 

 not complete, otherwise the records are nnbroken. 



The policy of the division in general has been to make as 

 few changes as possible in methods of observation or in character 

 of instruments, in order that the comj)arisons for the different 

 years may be as reliable as is desired. 



Five years ago a change was made from tridaily to semidaily 

 observations, so as to conform with the practice of the United 

 States Weather Bnrean, — ■ this station having become a volun- 

 tary observer of that bureau at about that time. It is not thought 

 that the change has sensibly affected any of the records, so that 

 they may not safely be compared with the records existing before 

 the change was made. 



An electric sunshine recorder was installed about two years 

 later, which increased the precision of the sunshine records 

 without materially affecting the mean values. 



During the past year the work has continued along the same 

 lines as heretofore, — instrument readings taken at 8 a.m. and 

 8 P.M., self-recording instruments kept in working order, and the 

 records transcribed weekly in a jDermanent record book. 



The usual monthly bulletins, containing the more important 

 daily records and remarks on the general character of the 

 weather during the month, have been regularly issued, and the 

 annual summary will form a part of the December bulletin. 



The New England section of the United States Weather 

 Bureau has furnished us daily, except Sunday, with the fore- 

 casts for this section of the State, and the flags indicating the 

 probable weather for the following day have been regularly 



