other instances of similar increases including an increase of the key rate of the central-territory coal adjustment from Ohio No. 8 to Toledo, Ohio, from $1 prior to July 1, 1917, to $1.89 at present, or 89 percent. In Lackawanna Steel Co. v. Director General, supra, decided January 26, 1924, there were under consideration, among other things, the identical rates here considered, prior to the general reduction of 1922. One of the principal grounds for the attack was the method of making the previous general increases. Upon complete consideration of the matter it was concluded that there was no basis for a finding of unreasonableness. The present showing with respect to the percentage of increase does not warrant a different conclusion. Various parties undertook to compute the distances from and to the points here considered and, as is not unusual in coal cases, the methods employed and the results obtained were somewhat different. The Reynoldsville complainants used a distance from their district of 200.4 miles which is stated to be the average from all active mines on the Baltimore and Ohio Railroad in that district to Buffalo. The complainant in No. 27717 shows an average distance from all mines in the Reynoldsville district to Buffalo to be 199.7 miles over short tariff routes, and 204.2 miles over all tariff routes. Defendants show the average distance over all tariff routes as 208.3 miles. In arriving at their average they include distances from mines in the Butler-Mercer district, which adjoins the Reynoldsville district on the west, on the theory that it is merely a subdistrict of the Reynoldsville group and takes the same rate to Buffalo. The propriety of the inclusion of the Butler-Mercer distances in computing the average distance from the Reynoldsville group is questioned by complainants. The complainant in No. 27717 shows that the distances from the Freeport and Pittsburgh districts to Buffalo are 231.6 and 282.3 miles, respectively, over short tariff routes, and 245.7 and 289.2 miles, respectively, over all tariff. routes. The latter correspond almost exactly with defendant's figures of 245.4 and 288.1 miles respectively. These complainants show separately the average distances from that portion of the Connellsville group in Pennsylvania and that portion in West Virginia, the latter being designated as the Fairmont district. The figures are 322.3 and 363.2 miles, respectively, over short tariff routes and 342.5 and 385.2 miles, respectively, over all tariff routes. The average distance shown by defendants of 355.5 miles over all tariff routes is based on the distances from the entire Connellsville group, including the Fairmont district. In Lackawanna Steel Co. v. Director General, supra, the average short-line distances from Reynoldsville, Pittsburgh, and Connellsville to Buffalo are shown as 213, 284.33, and 330.6 miles, respectively. In Bituminous Coal to Buffalo, N. Y., supra, the average distances are shown as 205, 284, and 331 miles, respectively. All of the above distances are to Buffalo, but defendants show that Niagara Frontier points, such as Niagara Falls, Suspension Bridge, and Lockport, N. Y., are included in the Buffalo group and take the same rates. To all of these points, including Buffalo, the average distances over all tariff routes are shown by defendants as 220 miles from Reynoldsville and Butler-Mercer, 256.6 miles from Freeport, 300 miles from Pittsburgh, and 369.5 miles from Connellsville. In the circumstances it would appear that complainants may not properly criticise the use of the distances, as stated by defendants, to Buffalo, because they are less than the averages which include the distances to Niagara Frontier points north of Buffalo, which might properly be used. Based on these distances, the rates assailed would produce ton-mile earnings of 10 mills from Reynoldsville, 8.4 mills from Freeport, 7.8 mills from Pittsburgh, and 6.7 mills from Connellsville. Complainants cited by way of comparison numerous rates and earnings on coal, of which illustrative examples are shown in the appendix. Defendants criticise complainants' comparisons in detail, on various grounds, including the remoteness of some of the territories and failure to show similarity of conditions and the extent to which the rates are used; the differences in distances; that some of the rates are for proportional application; that others have been recognized by the Commission as subnormal; that others are differential group rates made higher than those from base groups by relatively small amounts for competitive reasons, and, therefore, do not afford a proper measure of reasonableness; and that others apply between extreme points in groups and do not represent the average. Without discussing the criticisms of the individual rate comparisons, it is enough to say that they are of sufficient substance as greatly to detract from the value of such comparisons. Complainants compared the rates assailed with lower rates on such commodities as sand, gravel, and crushed stone between various points, and on iron ore moving from Buffalo and other lake ports to inland consuming points. Obviously, the rates on such low-grade commodities as sand, gravel, and crushed stone are not appropriate for comparison with the rates on coal, and in Lackawanna Steel Co. v. Pennsylvania R. Co., supra, the rates on coal from Reynoldsville, Pittsburgh, Connellsville, and related districts to Buffalo were alleged to be unreasonable and unduly prejudicial in comparison with the rates on ore from Lake Erie ports to inland consuming points. The complaint was dismissed. Complainants presented comparisons based on class rates. Chambersville, Pa., 200 miles from Buffalo, is selected as representative of the Reynoldsville group. The first-class rate from this point to Buffalo is $14.60 and the coal rate of $2.09 is 14.3 percent thereof. A number of coal rates are shown from western origins to points in Missouri, Nebraska, and the Dakotas which range from 7.3 to 11.8 percent of the first-class rates between the same points, but in eastern territory, such for instance as from the Clearfield district in Pennsylvania, to New York, New Jersey, and Pennsylvania, it appears that the percentage relation between the coal rates and the first-class rates is substantially identical with that shown from Reynoldsville to Buffalo. Complainants point out that in 1935 the Pennsylvania Railroad Company reduced the rate on coal from $2.47 to $2.24 from the Clearfield district to Buffalo and that this was followed by corresponding reductions from points in the Clearfield district on the New York Central Railroad Company lines and from points in the Meyersdale district on the Baltimore and Ohio Railroad Company lines in connection with fourth-section relief granted in Coal from Clearfield District to Niagara Frontier, 215 I. C. C. 611, and Coal from Meyersdale District to Niagara Frontier, 220 I. C. C. 256. Defendants say that the Clearfield rate of $2.47 was out of line with the general adjustment and moved no traffic, that it was reduced to the Pittsburgh basis for competitive reasons, and that even under the reduced rates there has been very little movement to Buffalo. Defendants cited a large number of coal rates within the territory here considered and surrounding territories which have been prescribed or approved by us and with which, in general, the assailed rates do not compare unfavorably. Substantionally all of these rates are considered and discussed in cases previously cited herein, and particularly in Intrastate Rates on Bituminous Coal Within Ohio, supra, and Showalter, Receiver, v. Akron, C. & Y. Ry. Co., supra, and only a few of those more closely associated with the assailed rates need be mentioned. Reference has already been made to the finding in Lackawanna Steel Co. v. Director General, supra, that the identical rates here considered, prior to the general reduction of 1922, were not unreasonable. In that same proceeding coal rates from the same origin districts to Lockport, N. Y., 29 miles northeast of Buffalo, which were 14 cents higher than the rates to Buffalo, were found not unreasonable, but unduly prejudicial to the extent they exceeded the rates to Black Rock, Tonawanda, and Lancaster, N. Y., to which the Buffalo rates applied. This decision resulted in the extension of the Buffalo group to include Lockport. To the Rochester, N. Y., group, east of the Buffalo group, the rates from the origins here considered are 13 cents higher than to the Buffalo group. The average distances from Reynoldsville and Pitts burgh to the Rochester group are about 248 and 362 miles respectively. In East Rochester Assn. v. Baltimore & O. R. Co., supra, the Rochester group rates were found reasonable for application to East Rochester, N. Y., a few miles east of Rochester. In International Paper Co. v. Buffalo, R. & P. Ry. Co., supra, coal rates from the origin territory here involved to destinations in northern New York east of the Rochester group were under consideration. From Clearfield, the base group in this adjustment, to the Syracuse, N. Y., group, immediately east of the Rochester group, a rate of $2.55 was prescribed for an average haul of about 306 miles. To Olean, N. Y., in the extreme southern part of the Buffalo group, a rate of $1.90 was prescribed on coal from mines in the Reynoldsville district in Kistler Leather Co. v. Pittsburg, S. & N. R. Co., supra. The average distance from the Reynoldsville district to Olean is about 155 miles. In National Petroleum Assn. v. Baltimore & O. R. Co., supra, a rate of $2.05 was prescribed on coal from Pittsburgh to the Warren, Pa., group, south of the Buffalo group, for an average haul of about 210 miles. In Intrastate Rates on Bituminous Coal Within Ohio, supra, coal rates from Pittsburgh to Cleveland and Mansfield, Ohio, for instance, of $1.84 and $2.03 respectively, were found reasonable for hauls of about 162 and 201 miles respectively. As stated in Coal from West Virginia to Northeastern Ohio, 220 I. C. C. 333, all, or practically all, of the coal-rate adjustments of the official-classification territory carriers have been before us in recent years, in a broad way at least, and the present rate levels in general have been prescribed by us or received our sanction. These rates, as a whole, constitute a homogeneous and highly competitive rate structure in which distance commonly has been subordinated to other elements affecting the level and relation of the rates. A reduction in an important key rate in the adjustment, such as that from Reynoldsville to Buffalo, should be required only upon the most clear and convincing showing that it is unreasonable. Complainants' comparisons do not constitute such a showing. Complainants contend that changes in their operations over period of years have resulted in greater efficiency and economy, in the operations of the railroads serving them, which should be reflected in reduced rates. For instance, complainant in No. 27690 shows that in 1920 it operated 28 mines whose average production was 245,634 tons, and that in 1928 it operated 19 mines whose average production was 278,250 tons, and loaded an average of 5,540 cars per mine to an average weight of 50.22 tons per car, whereas in 1936 it operated only 8 mines whose average production was in ex cess of 600,000 tons, and loaded an average of 11,201 cars per mine to an average weight of 55.09 tons per car. Complainant in No. 27717 stated that the experience in the Freeport, Pittsburgh, and Connellsville districts was similar but no details were given. In connection with the subject of changed conditions, defendants showed, as found in International Paper Co. v. Buffalo R. & P. Ry. Co., supra, that when the $1.10 rate was first established from Reynoldsville to Buffalo it applied from only five mines, for an average haul of 166 miles, and produced earnings of 6.63 mills per ton-mile. The present rate in connection with the average haul, as computed by defendants, of 208 miles, earns 10 mills per ton-mile. If the shortest average distance over short tariff routes which is shown of record, namely, 199.7 miles, were used, the earnings would be 10.46 mills. Therefore, while the rate has been increased 90 percent, even the higher of the ton-mile earnings mentioned represents an increase in earnings of only 57.7 percent. In addition to matters of defense already alluded to, defendants also refer to various others, such, for instance, as the fact that coal cars usually return empty from Buffalo, whereas from other lake ports they obtain a substantial return loading of ore, and the difficult and expensive character of the terminal service on coal at Buffalo and the relatively large switching charges which must be absorbed by line-haul carriers in order to make deliveries at that point. The parties are in complete disagreement as to the revenue effect of the rate reductions sought by complainants. Complainants say that the reduced rates would cause traffic, now moving over rail-andlake routes at the lake-cargo rates to the ports of transshipment, to move by rail, and that the resulting increase in the gross revenues of the rail carriers would more than offset the loss in revenue on traffic which is now moving all rail. Defendants emphatically deny that this is so. Furthermore, they contend that, if the reductions sought were made, it would be difficult, if not impossible, to avoid corresponding reductions in the rates on a very large movement from and to other origins and destinations which, although not made in direct relation to the rates assailed, are parts of the same general competitive adjustment. They say, therefore, that the instant proceedings constitute a grave potential menace to their revenue from coal traffic, which is one of their most important sources of revenue. The revenue showing by both sides is based upon assumptions as to what the situation will be in the future, under conditions different from those which have existed in the past, and which therefore cannot be accurately checked. We have given careful consideration to the evidence with respect to this matter, but it is unnecessary to set out a detailed analysis thereof. A few general observations with respect |