66 



March 20 inclusive, only 0.94 inches of rain fell. P'rom Fig. 3 it is apparent 

 that according to the government rain gage at Purdue, and a ])rivate gage, 

 about 1 inch of water fell pi^ceding the 23d, enough to thoroughly saturate 

 the soil. On the 23d, 1.75 inches of rain fell; another inch on the 24th; 1.35 

 on the 25th-26th and snow on the 26th, which did not immediately melt. 

 While there are no other rain gages on the watershed above this point from 

 which records were obtainable, it is ciuite ]irobable that the diagram repre- 

 sents average conditions. (See Table in Engineering News, April 3,1913, 

 p. 381.) 



The daily maximum temperature during the flood period is also shown 

 on Fig. 3. While there had been no snow the saturated condition of the ground, 

 which was free from frost, the temperature and the distribution of rainfall 

 caused tlie highest known stage of the river. 



Gagings of the Wabash River here have been made by students at Purdue 

 University for several jears and by the Weather Bureau and U. S. Geo- 

 logical Survey. 



From these we find the following greatest annual discharge: 



Date. Max. for Year in Cu. Ft. pel' Sec. 



1904, March 27 70,000 (estinuited.) 



1907, March 15 41,500 



1908, March 7 .■)7,()00 



1909, February 25 44,000 



1910, January 19 49,000 



1911, January 29 31,000 



1912, March 20 15,900 



1913, March 20* 95,400 (inclu<liiig (low over levee.) 



From the above data it is evident that the Hood of 1913 was greater than 

 any other recent one. 



The maximum flood rate at Lafayette was less than 20 cubic feet per 

 second per square mile. For Logansport, the flood of 1904 gave less than 20 

 cubic feet per second per square mile. 



These are low rates and as the rainfall did not average as great as has 

 been recorded for equal areas otherwheres it was not a flood which would 

 occur only once in a hundred years, but may be expected nioic freciucntly 

 than that. 



*NoTE. — .\ more extended investigation of the flood gaging.s indicates that the iiiaxiiiium dis- 

 charge may have reached 1.30,000 cubic feet per second. 



