THE PROPAGATION OF EXPLOSIONS IN MIXTURES 

 OF PETROLEUM VAPOR WITH AIR IN TUBES. 



By CHARLES E. MUNROE, 



Professor of Chemistry, George Washington Unr'ersity, 

 Washington, D. C. 



(Read April 22, 1910.) 



On !May 12, 1902, a series of accidents from fire and explosions 

 occurred at or contiguous to the Sheraden Yard of the P. R. R. 

 at Pittsburg, Pa., as the result of the collision of a tank car con- 

 taining " stove naphtha " with another car, or cars, whereby the 

 tank car was perforated and naphtha escaped. As the collision 

 occurred in the late afternoon at a time when the yard switch lamps 

 and the lamps of other safety devices had been lighted, the vapors 

 of the naphtha which was spilled from the perforated tank car, 

 became ignited and through the conflagration thus initiated other 

 tank cars containing naphtha were heated to such a point as to force 

 their contents out where they could contribute to the general con- 

 flagration, or, where, as in one instance certainly, the tank valve 

 was too firmly set to thus yield to this effect, the tank itself was 

 ruptured and its heated contents were discharged into the atmos- 

 phere, producing most disastrous results. 



The Sheraden Yard was artificially constructed by filling a 

 ravine between steep hills extending from the high ridge along the 

 of level surface required for so extensive a railroad yard. As 

 Ohio River, near the confluence of the Alleghany and Monongo- 

 hela rivers, to the northwest of Pittsburg, so as to obtain the area 

 might have been expected, the bottom of the ravine was occupied 

 by a stream, known as Cork Run, and this was fed by several lateral 

 streams which entered the ravine through fissures or ravines in 

 the surrounding hills. Good engineering demanded the preserva- 

 tion of these water courses after the filling of the ravine, and this 

 was done by means of sewers, the main one occupying, in general. 



203 



