ON THE FLOAV OF WATER IN PIPES^ CONDUITS^ ETC. 149 



TABLE IV. 



Sudbury Conduit. J = .0002. 



16. The foregoing results, with many others of similar character, appear 

 to justify a close and exhaustive study of the adaptability of the general 

 form for m with the end in view of seeing whether a differentiated classifi- 

 cation can be made of pipes, conduits and canals according to degrees 

 of roughness such that one definite form of m can be utilized for the 

 purpose of constructing a formula which will give results sufficiently 

 accurate to constitute a working formula which the engineer can apply 

 with confidence under a classification so explicit as to remove indefiniteness 

 and guess work in its application. The writer made a long and careful 

 study of some fifteen hundred experiments with different solutions of the 

 general form for m with the result that finally the form of the type 



J- ^ 



— , (44) 



l + .087+V8r 



was selected as having the widest range of applicability, the most generally 

 accurate results with ease of computation. The result of the investigation 

 shows that such a formula with reasonable precision for all artificial chan- 

 nels of a fair degree of regularity and constnietion can be designed. The 

 writer is convinced from his investigations that for natural channels such 

 as rivers whose regimen departs much from the regularity of the artificial 

 channel no such formula can be designed whose service can ever be more 

 than a gross approximation which serves the purpose only of provisional 

 estimation. A study of the observations on the coastal plain rivers like 

 the Mississippi and ilissouri, and the torrential streams of the Swiss Alps, 

 will soon convince one that while a special formula may be designed for 

 an individual stream with fair approximation, such streams can not be 

 grouped into classes in such a manner that any other given stream can be- 

 definitely classified with reference to them. The United States Government, 



