EXPLOSIONS IN COAL-MINES AND REVOLVING STORMS. 3 



to the shaft in the Bensham ; and having made its appearance in the Bensham 

 engine chimney, it was found necessary to extinguish the fire. The waste- 

 man says that the glass does not fall two degrees without a change being 

 perceptible below." 



Notwithstanding the absence of an explosion in each of these cases, it is 

 manifest that the readings 28-81 and 27'4'8 ought to have appeared in the 

 tables. 



Sndly. By estimating the importance of the explosion by the number of 

 persons killed, the great explosions are omitted which have occurred at times 

 when few persons were in the mine. 



Srdly. By taking a// the ^rrca^ explosions, some cases are included which have 

 arisen from known accidental causes, unconnected with atmospherical changes. 



These tables are, therefore, defective with respect to a large and important 

 class of cases, and redundant with respect to others which have no relation to 

 meteorological agency. But even if they had been perfect, the results would 

 still have been illusory, so long as the attention was confined to the action of 

 the barometer and thermometer at the time of explosion ; for the transit of 

 a great atmospheric storm generally occupies several days, during which a 

 mine may continue in a "foul" and dangerous state, ready to explode at any 

 stage of the storm's progress. The mercurial column, therefore, at the time 

 of explosion, may have any length comprised within the extreme limits of the 

 range of the barometer. The condition of ignition, and therefore the explo- 

 sion, may even be deferred until the storm has entirely passed over, and the 

 mercury has resumed the height and stability peculiar to settled weather. 



Thus, on the 3rd and 4th of November 1850, "a most violent storm of 

 wind" caused great loss of life and property in Great Britain ; blowing down 

 walls, chimneys, trees, &c. on land, and destroying many vessels along the 

 coasts. At the Royal Observatory, Greenwich, the passage of the storm is 

 recognized by a sudden and considerable depression of the mercury on the 

 3rd and 4th of November, but the readings range above 30 inches on the 

 9th, 10th, and 11th (see Plate v.). 



On the 11th of November, twenty-six persons perished by an explosiou 

 in the Houghton pit, Newbottle, county of Durham. 



The remark that " the workmen had been apprehensive of an explosion for 

 more than a week" connects this accident with the storm of the preceding 

 week. 



That such cases of delayed danger are not uncommon, appears from the 

 following statement of Mr. Mather to the Parliamentary Committee in 1854 

 (Second Report, &c., Qu. 1564). " The Killingworth explosion was pre- 

 viously indicated for eight days by three separate explosions ; the Washing- 

 ton explosion gave notice for five weeks of the coming catastrophe; and 

 Wallsend, that killed 102 people, showed its state for three days in red-hot 

 Davy-lamps. All of them gave large and decided indications of gas being 

 present for days before they happened ; and these are some of the chief acci- 

 dents that have occurred. In one instance there was, for a period of six 

 weeks, carburetted hydrogen to be found in a most positive manner." 



The opinion that explosions in coal-mines are, in some manner, dependent 

 upon certain changes in the ordinary conditions of the atmosphere, seems to 

 have been long entertained by the colliers of the various mining districts of 

 Great Britain and France ; and is repeatedly expressed in the minutes of evi- 

 dence taken by the Select Committee of the House of Lords on " Accidents 

 in Coal-mines," in 1849; and by the several Committees of the House of 

 Commons, on the same subject, in 1835, 1852, 1853, end 1854. 



It appears to have been satisfactorily established by observation, that the 



B 2 



