828 



SCIENCE 



[N. S. Vol. II. No. 51. 



For illustrations of this proposition I am 

 indebted to an officer of one of the great 

 railway systems of the country, who has 

 kindly furnished, during the past three or 

 four years, most interesting statistical in- 

 formation relating to accidents, collected by 

 him with the object of studying the results 

 in the interests of the corporation with 

 which he is connected. I have put the 

 principal results in graphic form, but for 

 those who like to see the numbers from 

 which the diagrams were constructed I 

 have included tables showing the classifica- 

 tion of accidents, as to occupation, results, 

 etc. Four years are included in this in- 

 vestigation, and there is shown in the tables 

 and diagrams the average for the whole 

 period. 



on the track or other property of the cor- 

 poration. The meaning of the diagrams 

 will be readily seen, the circle, in each case, 

 being divided into sectors proportional to 

 the number of accidents in the several 

 classes, the whole area representing the 

 total in every instance, regardless of the 

 numerical magnitude of that total, approximate 

 constancy of ratio of distribution being 

 the point under consideration. The per- 

 sistency of this ratio is certainly very strik- 

 ing. Naturally the railway corporation 

 collects this information with the view of 

 being benefited by it, and therefore it may 

 be expected that, as its character is de- 

 veloped from year to year, the operation of 

 what has been called ' chance ' in control- 

 ling the distribution of accidents will be 



Fig. 2. 



Fig. 3. 

 occupation. 



1. Passengers. 



2. Travelers on highway. 



Fig. 4. Fi( 



3. Employ(^»s. 4. Trespassers. 



The first five diagrams show the distribu- 

 tion of accidents among the various occu- 

 pations of the injured, at the time of the 

 injury. The division is into the four gen- 

 eral classes of passengers, employes, travel- 

 ers on the public highways and trespassers 



interfered with by new influences which will 

 tend strongly to diminish the number of 

 casualties, especially in those classes in which 

 accident is most costly to the corporation. 

 The results exhibited herewith furnish evi- 

 dence that this influence is already felt. 



