1884.] NEW YORK ACADEMY OF SCIENCES. 11 



being done at all points — the power exerted, the hard places 

 of the line and the manner in which the train was rnn — 

 giving a series of facts for study which cannot be ascertamed 

 otherwise. But few railway people have seen such diagrams, 

 therefore their great value as a means of pointing out where 

 economies can be made is not understood or appreciated. Em- 

 ployees must, in many cases, understand why, as well as liow, 

 certain duties are to be performed in order to obtain the best 

 results. 



The apparatus is so arranged that experiments can be made 

 with single cars, locomotives, freight and passenger trains. Con- 

 siderable work with the present instrument has been done upon 

 passenger trains, showing that as the weight and length increases 

 tlie friction per ton decreases. For trains of 10 cars, 250 tons, 

 about 650 feet in length, speed 48 to 50 miles per hour, the 

 friction was only 10 to 11 11)8. per ton, instead of 16 to 18, as 

 given by most formulas. The expenditure of horse-power on 

 the train alone was 600 to 700 per mile. The speed of the train 

 was limited by the steam generating capacity of the boiler, in- 

 stead of the tractive and adhesive power of the locomotive. 



The ratio of adhesion of a large number of locomotives for 

 summer service I have found a1)0ve 33^ per cent, on good tracks 

 and slow speeds. 



The first dynagraph was made in 1874, while I was Chief Engi- 

 neer of the Valley Railway of Ohio, put into use in 1875, and con- 

 tinued to 1878, when the scope of the investigations required 

 more apparatus than could l)e attached to that one. The first 

 instrument used paper 11 inches wide and from 12 to 24 inches in 

 length per mile of track ; it gave a dynamometrical curve, speed 

 per second, velocity gf the wind, and had accessory apparatus to 

 show the consumption of fuel and M'ater used per trip. With 

 this instrument, in the winter of 1875-'76. many experiments 

 were made with freight trains on various roads, and a special 

 series on the Lake Shore and Michigan Southern Railroad, over 

 particular divisions, and through from Buffalo to Chicago, and 

 vice versa. On freight trains of thirty to forty eight- wheeled 

 box-cars, gross load 600 to 800 tons (2,000 pounds each), the 

 av' i-age friction per ton per foot ranged from 6 to 7^ pounds 

 for speeds from 10 to 12 miles per hour, with but slight in- 

 crease until we passed twenty miles. This refers to the long 

 run, and not the startinu' of the train. On manv miles of level 



