Machinery occupies a very different place in the railway from that which it occupies in steam transport by sea. The engine only indirectly determines and regulates the work of the majority of railway men. Most of them are not tenders of machinery. Engine-driver, stoker, and guard are alone in close direct association with the machine. To them must be added those engaged in construction and repair within the workshops. Pointsmen and certain station officials come next in proximity to the machine; shunters and porters are also "tending" machinery, though their work is more directly dominated by general business considerations. But are we to say that the army of platelayers, navvies, etc., engaged along the line is serving machinery instead of using tools?[184] The work of ticket clerks and collectors is only governed by the locomotive in a very indirect way. Though the steam-engine is the central factor in railway work, the bulk of the labour is skilled or unskilled work in remote relation to the machine. This explains why the growth of the railway industry, after the chief work of construction has been done, is not attended by a diminishing proportion of employment. On the contrary, we find that railway employment increases faster than mileage and railway capital. The following statistics of railways in the United Kingdom illustrate this fact:-- Year. Mileage. Capital (paid up). Operatives. 1851 ... ... 25,200 1861 10,865 £362,327,338 53,400 1871 15,376 £552,661,551 84,900 1881 18,175 £745,528,162 139,500 1891 20,191 £919,425,121 186,700 But when we turn to the shipping trade, where a much larger proportion of workers is directly concerned with the tending and direction of machinery, and trace the effect upon employment of the application of steam, the result is very different. Sailing Vessels Steamers Men on Men on (Tonnage). (Tonnage). Sailing-ships. Steam-ships. 1850 3,396,359 168,474 142,730 8,700 1860 4,204,360 454,327 145,487 26,105 1870 4,577,855 1,112,934 147,207 48,755 1880 3,851,045 2,723,488 108,668 84,304 1890 2,907,405 5,037,666 84,008 129,366[185] If we take the period 1870-90, during which there is an absolute shrinkage of sailing tonnage, we find that this shrinkage is accompanied by a less than corresponding diminution of employment. On the other hand, the tonnage of steamships in this period increased more than fourfold, but brought with it an increase of employment which is less than threefold. [Illustration: TONNAGE OF SHIPS IN RELATION TO EMPLOYMENT OF SEAMEN.]