manner in which their own money is being spent. It is quite obvious that none of the missing money finds its way into their pockets.

In reply to the suggestion that further railway statistics are desirable, the expense of compiling them has been put as an objection.

Here again we have a clear indication of the amateurishness of the railway directors. They answer as if statistics were an abstract game, and an end in themselves. The proper application of statistics is more important than their compilation, and in expert hands most valuable lessons for the undertakings with which they deal may be learned from them.

Improved railway accountancy and auditing are therefore most necessary to full reform. But, important as they are, they do not touch the most crucial reform of all.

With this remedy it is now proposed to deal.

CHAPTER III

MR. GATTIE'S SCHEME

THE inevitable conclusion to be drawn from the facts and figures which have been placed before the reader, is that the cause of nearly all this muddle and waste is the misuse of the locomotive.

When railway traffic was in its infancy, and only a few wagons had to be sorted, it was possible to do this by pushing them backwards and forwards. This was, however, a makeshift arrangement, and even to this limited extent it was a misuse of the locomotive.

The men who were originally guilty of this misuse of the locomotive were probably conscious of error, but it is not conceivable that they would have continued in the same path had they realised that they were leading their successors into the most gigantic economic blunder in the history of commerce.

The result of this bad example has been that, during the whole course of its existence, the locomotive has been used as a sorting machine. It was never intended for the purpose. It was designed by the inventor for hauling goods on a railroad, that is to say, it was designed to move in one dimension, backwards and forwards, a fact which is conclusively proved by its being fitted with flanged wheels which prevent it moving sideways. It is not designed to move laterally. It is not designed to move vertically. It is not designed to fly. If the process of sorting be carefully considered it will be seen that a sorting machine,

to be efficient, must be capable of all these three movements: upwards and downwards, backwards and forwards, and sideways. When a sorting machine which is not only capable of these three movements, but is also capable of obeying the will of the operator, has been discovered, then the perfect sorting machine has been evolved. It is the purpose of this chapter to show that such a machine is possible; nay, that it has been invented and that it exists in its full-sized dimensions. It is necessary that the fact as to its not being a model, but an actual full-sized machine, should be insisted upon, as the impression might otherwise be given that the invention is a toy, and not fitted for the heavy work required of it.

To prove the truth of what has been said about shunting, it is only necessary for the reader to take a pack of cards and spread them out face upwards in four rows on his dining-room table, the cards being arranged, of course, haphazard. Let him then imagine that his table has upon it a number of railway tracks, four of which are occupied by the four rows of cards, and that each card represents a railway wagon. Let him then imagine that his hand is a locomotive, and capable only of the movements of a locomotive. The task to be performed is to sort the fifty-two cards into four other rows, each row consisting of cards placed in a desired order. Let us say that each row must consist exclusively of one suit, and be arranged from ace to king. Let him in imagination couple his imaginary locomotive to any available wagon in the four rows, and haul any desired number of wagons over the switching points. Let him couple and uncouple the wagons, as would be done in the shunting yards, always remembering that his hand can only move backwards and forwards over the points. A longitudinal movement of 165 feet is necessary in order to deviate from one railway track to the next parallel with it. (See Fig. 13.)

Confining his hand to these movements he will be able to sort the cards into suits-hearts, diamonds, clubs, and spades.

By the time he completes his task he will find that he has moved his locomotive 300 times, and the cards 1,500 times. It will be seen that the shunting locomotive in doing this work would move on an average five cards with every one of its movements, but sometimes he would move one card and sometimes a whole row of cards.

165 feet

Fig. 13.1

Let him consult his watch, and judge of the waste of time involved in this method.

But if the reader were to manipulate 104 cards instead of fifty-two, a surprising discovery will be made. It will not take him twice as long, but four times as long, owing to increase of distance traversed in manœuvring the cards.

The number of these movements, however, will vary in accordance with the position in which the cards are originally found.

In London, every day, there are sorted over 10,000 wagons. Let the reader draw his own conclusions.

To complete the experiment, let the reader shuffle the cards and deal them out again in four rows. He again wishes to sort them into suits. Let him use both hands in sorting, as he naturally would, without

1 Molesworth's "Pocketbook of Engineering Formulas."