of 200,000. Generally speaking, collisions and derailments form quite a reliable standard from which to make comparisons in regard to efficiency of service. It must also be remembered that the chances for accidents are naturally increased with increase of traffic and consequent multiplication of train movements. One might reasonably expect, therefore, to find the density of conditions in Great Britain reflected in a startling list of fatalities, as compared with the United States. Yet if we take the year 1906 to illustrate our theories and anticipated conclusions, we find that there were 13,455 collisions and derailments in this country, and only 239 in Great Britain. In the same year 146 passengers were killed and 6000 injured in the United States, against 58 passengers killed and 631 injured in Great Britain. The number of employees killed and injured in train accidents was respectively 13 and 140 in Great Britain, against 879 and 7483 in this country.
It is not surprising, therefore, that figures and returns like the above, repeated from year to year with the same marked and, indeed, ever-increasing disparity, should give rise to widespread discussion and criticism, consequently leading up to a better understanding of the nature of the problem that is now submitted, with all necessary facts and illustrations, practically for the first time, to the Amer-
