392 TRANS. ST LOUIS ACAD. OF SCIENCE. 



Unionville, Putnam Co., Mo. Lat. 40° 29' ; Lon. 93° 03'. J. G. Hart 

 obs. M. W. S., Jan. 1S78 to May 1879; B. II. Bonfoey obs. M. W. S., 

 Sept. 1879 to Jan- 1881. 



Warrensburg^, Johnsoji Co.., Mo. Lat. 38° 45' ; Lon. 93° 40' : grade ele- 

 vation 870 ft. S. L. Goodwin obs. S., July 1S68 to ilay 1875; W. L. 

 Hedges obs. M. W. S., Jan. 1878 to Nov. 187S; Pres't Geo. L. Osborne 

 obs. M. W. S., Dec. 1878 to Dec. 1881. The top of Mr. Osborne's 

 gauge is about 20 ft. above ground. The nearest object is a house 15 

 ft. high 20 ft. n.e. 



West Glaze, Camden Co., Mo. Lat. 37° 57'; Lon. 92° 35 : elevation 1450 

 ft. A. Y. Carleton obs. S., Jan. 1S72 to Oct. 1S78. 



IVyaconda Prairie, Leivis Co., Mo., near Canton. G. P. Ray obs. S., 

 Apr. 1862 to Dec. 1868. 



STATIONS IN ADJOINING STATES. 



Cairo, Ills. Lat. 37° 00' ; Lon. 89° 10' : elevation 377 ft. U. S. Signal 

 Service obs. Record from Jan. 1872 to Dec. 1^87, but only 10 years 

 record here published. The top of the gauge is 78 ft. above ground, 

 the exposure being on a building. 



Keokuk, Iowa. Lat. 40° 23'; Lon. 91° 27' : elevation 584 ft. U. S. Signal 

 Service obs. Record extends from July 1871 to Dec. 1887, but only 10 

 years published. The 'opof the gauge is 60 ft. above ground. 



Leave?iivortk, Kansas. Lat. 39° 19'; Lon. 94° 58' : elevation 842 ft. U. 

 S. Signal Service obs. The top of the gauge is 50 ft. above ground. 



Mascoutah, St. Clair Co., Ills. Lat. 38° 32' ; Lon. 89° 55'. Theo. En- 

 gelmann obs. M. W. S., Jan. 1878 to Dec. 1887. The gauge is of cop- 

 per and was made in St. Louis; the diameter is 7.9 in., and is made 

 after the Signal Service pattern, a measuring-stick being used. The 

 top of the gauge is 3 feet above ground. The nearest objects are a 

 building 100 ft. n.w., grapevines on trell ses 55 ft. high 10 ft. n., cedar- 

 trees 30 ft. high 40 ft. n.w. It is fastened between two stakes which 

 do not reach the top of the gauge. 



In the tables which follow there are first given a number of 

 stations where the record covers more than ten years. Where this 

 record covers the ten years ending December, 1887, the means for 

 that period are in all cases separately reduced, and were used in 

 the construction of the normal rain charts. 



Next follows a series of stations where observations were begun 

 with the ten-year period mentioned. Finally a series of fragmen- 

 tary records are given, some of which have been taken from Mr. 

 Schott's paper, and some being the records of our own observers 

 who have not covered the entire ten years. These series were not 



