1872.] 



between Explosions in Collieries and Weather. 



295 



been published in the Quarterly Weather Report. Stonyhurst has been 

 selected inasmuch as from its situation, near Preston, it is more central as 

 regards the coal-districts than any of the other observatories. 



In the case of certain accidents, more especially in the South- Wales and 

 Scotch districts, the records from Falmouth or Glasgow respectively 

 might possibly throw more light on the matter, but we have thought it 

 hardly worth while to complicate the diagrams by introducing them. 



The curves for barometrical pressure and for temperature have been 

 plotted from the daily maxima and minima registered for the years 1868, 

 1869, and 1870. The curves have not been actually reduced from the 

 photographic records, as the horizontal scale has been so much compressed ; 

 but they afford a fairly correct representation of the march of the pheno- 

 mena, and for the last two years they can be compared, if necessary, with 

 the Quarterly Weather Report. 



One serious disturbing cause interferes with the value of these curves, 

 arising from the weekly suspension of work in the collieries, and in many 

 instances of ventilation, too, on Sundays. The influence of this was elimi- 

 nated in the former investigation, when we were seeking for periodicity, as 

 we dealt with as many as twenty years. 



The dates of the fatal explosions are all given in the Inspector's Annual 

 Reports, those of the non-fatal explosions for 1868 have been taken from 

 Mr. Bunning's paper. For the facts relating to the years 1869 and 18/0 

 we are indebted to the kindness of the Inspectors, who have almost without 

 exception responded to our application by furnishing all the information 

 which it was in their power to give. 



The occurrences have all been entered to their proper dates, and the 

 districts in which they were reported have been noted. Fatal accidents 

 are marked by a ba.\ 



We have marked by special symbols all the explosions which appear to 

 us to be connected with sudden changes in atmospherical pressure or with 

 abnormally high temperature ; a few of the occurrences may possiblv be 

 attributable to the conjoint action of these two causes. 



Mr. Dobson, writing at a time when all our serious barometrical oscilla- 

 tions w r ere imagined to be connected with true cyclones, considered that, 

 as in the front of such an advancing storm the barometer is falling and the 

 temperature rising, he was justified in putting down the liability to ex- 

 plosion as a phenomenon of the anterior segment of such a storm. He 

 does not, however, give a continuous record of weather so that we can see 

 how many storms were unaccompanied by explosions. There can be no 

 doubt of the coincidence of certain serious explosions with severe storms ; 

 a notable instance of this will be found on the 8th of October, 1870 ; but 

 the explosions do not happen only at the commencement of a barome- 

 trical depression, but occur also two or three days after the barometer has 

 reached its lowest point and is again rising. We have taken a limit 

 of three days. 



y 2 



