142 TRANSACTIONS OF THE [fEB. 1, 



boxes of the special truck ; while the strikers to depress the 

 valve when a low place is reached are carried by the journal 

 boxes of the middle pair of wheels. 



The markers are operated by compressed air, and eject the 

 paint to the rails in about l-50th of a second after the valve is 

 depressed. When they were first uSed, any deflections less than 

 •j^ of an inch were unmarked. The i)ast two or three years the 

 markers were set for the best track to -mark all deflections ex- 

 ceeding -^^ of an inch. Last j^ear they were set to pass any de- 

 flections one-thousandth of an inch less than the limit to which 

 the track was to be marked. 



The striking mechanism is now held in position by compressed 

 air instead of a spring, and has a micrometer screw which can be 

 set to one ten-thousandth of an inch. I do not wish to convey 

 the impression that we are able to work to one ten-thousandth of 

 an inch in marking, but the instrument must be made for fine 

 measurements so that we can reduce the errors to very small 

 limits. 



The summing up mechanism and the markers record elec- 

 trically, as do also those which record the time per second and 

 per ten seconds, those for the percentage of tangent and curve, and 

 those for the distance. A bell rings before the mile posts are 

 reached, and the observer touches a key to note their location, 

 as well as the location of the stations and bridges. 



The car, usually with one or two official cars in front, is run as 

 a special at a speed of twenty or twenty-five miles per hour; ac- 

 cording to the condition of the track, a constant speed being 

 maintained per mile over each road. 



The original diagrams show in detail the general surface of 

 the rails, as to permanent bends, set at the joints, deflections, 

 and irregular wear or roughness. 



When I commenced taking diagrams in 1 880 on the four and 

 four and one-half inch rails, I found the undulations in the tracks 

 much greater than had been anticipated, joints which deflected 

 a full inch being very frequent, and nearly all the rails had a 

 permanent set, the receiving ends of many of them being also 

 cut out. There was no definite standard for tracks, and in 

 making reports that year, I called all tracks " good " where the 

 joints did not deflect more than one-half inch. 



The permanent set in the rails was studied and I found I 

 could classify them under three general types, which I called 

 ^^ forms of permanent se<." 



The first form embraced all those rails which were low at the 

 joints and high at the center ; this type was found upon rails 

 with opposite joints. 



