This examination of his methods shows that the determination of space was of primary and controlling importance and that the changes in space allotment have destroyed the value of his deductions. These changes were due to his refusal to assign to the mail service the working space and temporarily unoccupied space on trains, which were necessary to the mail service and to his actually assigning much of this space to the passenger service rendered on the same trains. It is scarcely necessary to note that all kinds of traffic require "working space" in addition to the space actually occupied by the traffic itself, and that this is especially true of the mail traffic, or that where there is a preponderating movement of a certain traffic in one direction there must be some empty space on account of that traffic, sometimes called "dead" space, in trains moving in the direction of lighter traffic. Thus passenger cars must have aisles, vestibules and platforms, and postal cars must have a great deal of space in which to sort the mails while, for mail carried in baggage cars, there must be space in which to reach the pouches and to receive and deliver them through the doors. A through train must also have the full capacity required for the maximum traffic of any kind likely to seek accommodation on any part of its journey, although during much of each trip the actual traffic may be considerably below this limit. The Postmaster-General, however, refused to credit the mail service with much of the space thus required by the Department although his figures for the other passenger train services allowed fully for all such space required by them. In fact in many cases such space, actually required by the mails and so reported by the railways, was taken from the total mail space and, without reason, assigned to the passenger service. These modifications of the data correctly reported, not susceptible of justification upon any sound transportation principle, were carried so far that the tabulations of the Post Office Department, which are stated for railway mail routes having a total length of 194,977.55 miles[E] show only 926,164,459 "car-foot miles" made in the mail service, although certain railways, included therein, and having railway mail routes aggregating only 178,709.96 miles, had correctly reported mail space equivalent to 1,153,110,245 "car-foot miles." Thus, although the Department's figures cover 8.3 per cent. more mileage, its reductions of space resulted in assigning to this greater mileage about one-quarter (24.5 per cent.) less mail space. At the same time the Department actually increased the space assigned to the other passenger train services, its figures showing 12,014,065,506 car-foot miles in these services for 194,977.55 miles of mail routes which must be compared with 11,222,478,739 car-foot miles reported by the railways for 178,709.96 mail-route miles. [Footnote E: Document No. 105, p. 53.]