THE IRRIGA T10N A GE. 417 



is now universally adopted by American builders, were tried at the 

 same tests. Their average efficiency was 12.5 percent, less than those 

 of the Jonval type, and in selecting water-wheels for the city the com- 

 mittee would not consider them because their efficiency was so low. 

 This shows how completely in its infancy wa^s the American type of 

 water- wheel in 1859. 



In 1876, seventeen years later, another 'competitive test of water- 

 wheels was carried on in Philadelphia at Fairmount Park. This test 

 was under the auspices of the Centennial Exposition. Sixteen water- 

 wheels were tried. Fifteen were of the American type. The only 

 Jonval wheel, or wheel of foreign type, was entered by the same man- 

 ufacturers who secured the order which depended on the tests of 1859. 

 Their wheel was designed by the same French engineer, who had 

 been trained in the best French schools, where he had studied the best 

 French models. Its chutes and buckets were set in finished grooves, 

 cut with perfect accuracy. They were made of rolled brass, so as to 

 allow the water the smoothest and best course. 



The American wheel which led the test in 1876 was a Risdon 

 wheel, a plain iron casting. It was from*5 to 10 per cent, better at 

 full gate and gave 33 per cent, more power out of the water at part 

 gate, so much had the American type of wheel been developed at 

 that time. 



From 1859 to 1876 a corresponding change had been going on 

 among our large New England factories. There, in 1859, the French 

 Fourneyron turbine had crowded out the old pitch-back breast wheels 

 and had obtained high records. 



After 1876 the American type of wheel almost completely con- 

 trolled the New England trade. The Fourneyron wheel was expensive 

 to build, it clogged easily, it was not efficient at the part gate, and it 

 ran too slowly. Now as to the origin and history of this great change. 

 Until 1850 there was lacking a simple water-wheel which could be 

 used on large streams where the water in the tail-race rises in every 

 storm, and destroys the efficiency of a breast or overshot wheel. This 

 was especially true of saw-mills. They were used only in winter and 

 spring, when the supply of water was ample and it was not necessary 

 to economize it. At first the nutter wheel was used, consisting only 

 of a wheel with a set of square blades, against which a sheet of water 

 impinged and gave force. They only gave an efficiency of about 30 

 per cent., but that was enough. The flutter- wheel would not run in 

 back-water and every storm stopped the mill. This led to the adop- 

 tion of the reaction wheel 



The model has two such wheels placed on opposite ends of the 

 same shaft. The water there is let into the flume between them in 

 such a way that it can only escape through the water-wheels. 



Passing through the curved vanes o such a wheel, even if it had 

 no stationary chutes or guides, would give the water a whirling motion' 



