It is indisputable that the great dynamic force in railway operation inheres in the value of service idea. The traffic manager who is always considering how much it will cost to handle business, will seldom adventure into new territory. The United States, as a rapidly growing country, is consequently the field in which charging what the traffic will bear, has been most ardently upheld as the only practicable basis for rate making. A few detailed illustrations will serve to show the results of its application in practice. Not infrequently does it happen that rates are different over the same line for shipments between two given points in opposite directions. Where this is due to a preponderance of traffic in one direction, and a consequent movement of "empties" which invite a back loading at very low rates, the difference of charges according to direction may actually be due to differences in the cost of carriage.[153] An empty train, which must be returned from New York to Chicago for another loading of grain, or to Georgia or Oregon for shipments of lumber, if loaded with merchandise, can be moved with no allowance for dead weight of cars or locomotives; inasmuch as the train must move in any event, whether loaded or empty. But even where this defence of difference in the cost of service fails, the practice may be entirely proper from every point of view. By increasing the total tonnage a special rate may ultimately contribute to lower charges all along the line. Raisin culture began in California in 1876. Prior to that time the Spanish product had supplied the American market. The first thing to do was to find a market for the California raisins in the East. They would not bear the freight rate which had previously been charged for Spanish raisins moving over the transcontinental lines westward. A very low rate was all that the new traffic would bear. During the year 1876 therefore 70,000 lbs. of California raisins were carried east at one and three-fourths cents per 100 lbs., while simultaneously 1,000,000 lbs. of Spanish raisins were carried west over the same lines at a rate of three cents. No such difference in the cost of service in opposite directions existed, although a preponderance of empties moving eastward undoubtedly cheapened the service from California. The aim of the commodity rate was to upbuild a new industry. How far this succeeded appears from the fact that in 1891, no Spanish raisins were carried west at all; while the eastbound shipments amounted to 37,600,000 lbs.[154] The preceding illustration leads us then to this further conclusion. The cost of service principle might most conceivably be applied to a railway in a purely static state. But, dynamically considered, as involving the growth and development of business, it fails utterly by itself to meet the necessities of the case.