illCUllllCU WUU 11, Ul 1UIC1CDUU IU 113successYet it sbould be said that it is very doubtful whether the brakes on the rear third of a 50 car train, in either of the two systems, exert much effective retarding force. Although the gauges on the last car always indicated nearly as as.much air or vacuum pressure as those at the front end, and showed that the brakes were on in from 14 to 19 seconds, other facts lead to the belief that the stops were made before the rear brakes were effectively applied. It was certainly shown that 50 car trains could be well handled if the brakes were cut out of the last twenty cars.The general facts established by the tests, as regards the systems taking part in them, may be summarized as follows:1. Continuous brakes, operated by air, can be applied to and released from the wheels of trains of fifty cars with all the promptness required in the service.2. Brakes operated by compression of the draw-bar may be used with a good degree of success upon trains not exceeding 25 cars, if handled intelligently. To be used on long trains, some device must be contrived by which the brakes will be steadily held to the wheels during the whole stop.3. In running down grades, the buffer brakes do not show retarding power relatively proportioned to that which they exhibit on level track, or to that shown by the continuous brakes on down grades.4. The comparatively gradual putting on of the full power of continuous brakes in what are called “service” stops is sufficient to meet all or uearly all the exigencies of service. The records show that these stops, or, at least, many of them, are made in just about the same distance as the corresponding “emergency” stops, while the injurious shocks which accompany the latter are avoided if the brakes are applied with a fair degree of skill.5. The proper handling of long freight trains with either continuous or buffer brakes, requires a coupling device which shall largely decrease the amount of slack caused by the use of the ordi-narn link and pin. This proposition is, perhaps, not absolutely established bythe tests, but its truth .is ce.tainly indicated.