A rate per ton-mile for each class of freight is out of question, because a large part of the cost to the company consists in loading, handling, and storing the goods. Once aboard the car, it costs but little more to carry a ton of freight one hundred miles than to move it one mile. The rates per mile, therefore, are necessarily greater for short distances than for long runs. A mile-rate based on a ten-mile haul would be prohibitive to the shipper if applied to a run between Chicago and New York. On the other hand, were the charges based on the long run, the local rates would be far less than the cost of the service.[19] As a result freight rates are based very largely on the cost of the service, and this is particularly true of local freights. This practice is also modified by charging _what the traffic will bear_, and, on the whole, a combination of the two ideas gives the most reasonable and the fairest method of basing charges. Thus, a car filled with fine, crated furniture, which is light and bulky, can afford a higher rate than one filled with scrap-iron. Cars filled with grain, lumber, coal, or ore are made up in train-loads, and form a part of the daily haul; they can afford to be taken at a lower rate than the stuffs of which only an occasional car-load is hauled. In order to adjust this problem it is customary to divide freights into six general classes. [Illustration: THE PROBLEM OF FREIGHT RATES] In handling through freights the problems are many, and, if two or more roads have the same terminal points, a great deal of friction of necessity results. The longest roads must either make their through rates lower than local rates between distant points, or lose much of their through business. They cannot afford to do the latter and the statutory laws may forbid the former. As a result the laws most likely are evaded, or else openly disobeyed.[20] The difficulties in adjusting the matter of the long and the short haul, as has been shown, have caused the formation of pools and various other traffic associations, the object of which has been to prevent rate-wars. To this extent they resulted in positive good, for a rate-war in the end is apt to be as hurtful to the community as to the railway company. The attempt to settle such questions has also resulted in a great deal of legislation. Some of this has been wise and good; but not a little has been hurtful both to the railroads and to the community. The general result is seen in the great combination of competing lines and, more recently, of competing systems.