100 



among the tow boats. The river has not been so low in many years, 

 and the outlook for a rise this month is not promising. Two days' 

 steady rain would be necessary to bring a good boating stage, and 

 pilots do not expect this before November. A few of the prophets, 

 however, cling to the prediction that tows will be taken out before 

 Hallowe'en. The stage of the water at Davis' Island dam last night 

 was 1.3 feet; river stationary." 



As a matter of exact record the lowest stage of water at Pittsburg 

 during September, 1895, was 5.3 feet. In September, 1897, the lowest 

 stage there was 5.2 feet. 



This statement is both startling and suggestive, especially when 

 taken in connection with the following information, kindly given by 

 the National Weather Bureau, through Mr. E. R. Demain, the observer 

 at Harrisburg: 



"Deficiency of rainfall at Pittsburg from January 1, 1895, to October 

 15, 1895, was 9.47 inches. 



"Deficiency of rainfall at Pittsburg from January 1, 1897, to October 

 15, 1897, was 3.55 inches." 



Tt would seem from the above that in 1897 the water was lower at 

 Pittsburg than in 1895, though the deficiency of rainfall was in 1897 

 less than one-half (exactly 37 per cent.) what it was in 1895. 



Facts of similar import reach us from other stations. Thus, at 

 Philadelphia, the deficiency of rainfall from January 1, 1895, to Octo- 

 ber 15, 1895, was 7.42 inches. From January 1, 1897, to October 15, 

 1897, it was 0.61 inches. 



From Erie, Pa., we have these figures, thus: deficiency of rainfall 

 from January 1, 1895, to October 15, 1895, was 8.20 inches. Deficiency 

 from January 1, 1897, to October 15, 1897, was 5.16 inches. 



At Harrisburg, Pa., the deficiency of rainfall from January 1, 1895, 

 to October 15, 1895, was 18.95 inches. From January 1, 1897, to Oc- 

 tober 15, 1897, it was 11.31 inches. 



This is one of the problems to be accounted for and remedied if 

 possible. It is evident that there are too many elements of state and 

 national prosperity involved here to permit of any shortage of water 

 power if it can be avoided. 



Of course it will be noted that these rainfall data were taken in 

 Pennsylvania, though the origin of a portion of the water supply of 

 Pittsburg is quite beyond our borders. There can, however, be no 

 doubt but that over a large portion of the Allegheny and Mononga- 

 hela (basins the rainfall was greater than in 1895, yet these streams 

 seem to have been lower. There can be no doubt as to the essential 

 fact that in our periods of annual minimum waterflow, our rivers are 

 delivering less water in each successive year. Thus, the most reliable 



