596 liEroiiT— 1884. 



Pennsylvania Railroad Company is to receive rails, and of which an 

 official record is kept at each mill, is to be exhibited to the rail inspector. 

 ' 3. A test bar, three quarters of an inch wide, and about ten inches 

 long, is to be taken from a web of rail made from each charge. 



' 4. The number of the charge and place and year of manufacture shall 



be marked in plain figures and letters on the side of the web of each rail. 



' 5. The sections of the rails rolled shall correspond with the respective 



templates issued by the Pennsylvania Railroad Company, showing the shape 



and dimensions of the different rails adopted as their standard. 



' 6. The space between the web of the rails and template representing 

 the splice-bar shall not be less than one quarter of an inch, nor more than 

 three-eighths of an inch. 



' 7. The weight of rails shall be kept as near to the standard weights 

 as can be demanded after complying with section No. 5. 



'8. Circular holes one inch in diameter shall be drilled through the 

 web in the centre thereof, at equal distances from the upper surface of the 

 flange and lower surface of the head, and three and fifteen-sixteenths 

 inches from the end of the rail to the centre of the first hole, and of five 

 inches from the centre of the first hole to the centre of the second hole. 



' 9. The lengths of rails at sixty degrees Fahrenheit shall be kept 

 within one quarter of an inch of the standai'd lengths, which are thirty 

 feet, twenty-seven and one-half and twenty-five feet. That not more 

 than ten per cent, of the shorter lengths, nor more than five per cent, of 

 No. 2 rails, will be accepted on any one contract. 



' 10. The rough edges produced at the ends of the rails by the saw 

 shall be well trimmed off and filed. 



'11. All rails are to be straightened in order to insure a perfectly 

 straight track. 



' 12. The causes for temporary rejection of the rails are — 

 ' (1) Crooked rails. 



' (2) Imperfect ends (which, after being cut off, would give a per- 

 fect rail of one of the standai'd short lengths). 

 ' (3) Missing test reports. 



' (4) A variation of more than one quarter of an inch from the 

 standard lengths. 

 ' 13. The causes for the permanent rejection of a rail, as a No. 1 rail, 

 are — 



' (1) A bad test report, showing a deficiency or excess of carbon. 

 ' (2) The presence of a flaw of one quarter of an inch in depth 



in any part of the rail. 

 ' (3) A greater variation between the rail and splice-bar than is 



allowed in paragraph No. G. 



' (4) The presence of such other imperfections as may involve a 



possibility of the rail breaking in the track.' 



In the construction of a railroad, the rails should be accurately laid to 



line and level stakes as given by the engineer. On straight lines the two 



rails of a track must be laid to the same level, but on curves the outer 



rail is elevated according to the degree of curvature, the elevation com- 



mencing at each end back of the point of curvature, by a distance also 



depending upon the sharpness of the curve, and increasing to the curve 



itself, around which the full elevation is carried uniformly. The amount 



of elevation varies on different roads, and indeed on the branches and 



main stem of the same road, depending upon the velocity at which trains 



