RAILWAYS 



647O 



RAILWAYS 



to the sleepers. The first is employed almost ex- 

 clusively in Great Britain, but in most other 

 countries the flat-bottomed is preferred. 



Rails are rolled from steel ingots, and stringent 

 regulations are imposed as to chemical composi- 

 tion, variation from standard sectional and 

 length measurements, and behaviour under test. 

 Rails of all sections have been carefully standard- 

 ised as regards height, width, thickness of web, 

 curves at corners, and proportionate area of parts. 

 The tendency is to increase the weight of rails, 

 and tracks designed for heavy and frequent traffic 

 are laid with rails weighing 90-100 Ib. per yard 

 run. On British railways the standard lengths 

 are 30, 36, 45, and 60 ft. Long rails give smoother 

 running by decreasing the number of joints per 

 mile ; but the gaps which must be left at joints 

 to allow for expansion are necessarily greater in 

 proportion. 



Method of Fastening Rails 



The cast-iron " chairs " in which bull-heaued 

 rails rest weigh 40 or 50 Ib. each. They are 

 secured to the sleepers by two, three, or four 

 fastenings coach-screws, bolts, spikes, and oak 

 trenails not more than two of the last being per- 

 mitted in a chair with four holes. A key of hard 

 wood is driven between the rail and the outer 

 wing of the chair to prevent movement. 



The cheapest means of fixing flat-bottomed 

 rails is the square-shanked, chisel-pointed spike 

 with a projecting head. It is quickly driven, but 

 much less reliable than the coach-screw or bolt 

 employed in the best tracks. To prevent the 

 rails from cutting into the sleepers, metal plates 

 (" tie-plates ") with holes for the spikes may be 

 interposed. They are flanged on the lower side 

 to grip the sleepers. 



A rail joint thoroughly satisfactory in all 

 respects has yet to be discovered. The com- 

 promise generally adopted is the "suspended" 

 joint, with rail ends butting between sleepers and 

 supported by fish-plates, one on either side, 

 clamped together through the rails by bolts. The 

 bolt holes are slightly larger than the bolts, to 

 give room for expansion and contraction. It has 

 been proved that the jolting at joints is due 

 rather to the bending of the rails than to the 

 " jump " across the gap, and the ordinary fish- 

 plate does not prevent bending. Various forms of 

 " bridge " joints, extending from sleeper to 

 sleeper, and supporting the rails underneath as 

 well as at the head, have been tried. 



The majority of the rlys. of the world are laid 

 on the normal or standard gauge of 4 ft. 8| ins., 

 as measured between the insides of the rails, or 

 with the 4 ft. 9 ins. gauge, which in practice is 

 identical, since rolling-stock designed for the one 

 can run on the other. But a number of other 

 gauges are to be found. For example, the 5 ft. 

 6 ins. in Spain, Portugal, India, and S. America ; 

 the 5 ft. 3 ins. in Ireland and Victoria ; the 5 ft. 

 in Russia ; and the 3 ft. 6 ins. in S. Africa, Queens- 

 land, New Zealand, the Sudan, and Japan. Other 

 gauges used on the less important lines in various 

 countries are the metre (3 ft. 3 ins.), the 2 ft. 

 6 ins., and the 2 ft. gauge. 



Both gradients and curves are undesirable be- 

 cause they increase train resistance, and, there- 

 fore, add to the cost of operation, but in practice 

 they are unavoidable. Since the interior of all 

 countries is higher than the seaboard, any line 

 that runs inland from the coast must be laid on 

 an ascending gradient, and, further, the local 

 elevations and depressions of the surface which 



