ON THE EXPLOSIONS IN BRITISH COAL-MINES. 237 



such record, and a communication witli the director of the mines of Belgium 

 was also fruitless. 



The dates for the year 1859 of all the fatal explosions in the coal-fields of 

 England, Scotland, and Wales are marked in the meteorological diagram 

 (Plate v.), in which one day is represented by a horizontal space of one- 

 twentieth of an inch, and '20° Fahr. by a vertical height of one inch. 



For the meteorological data I am indebted to the kindness of Mr. Milner, 

 the surgeon of Wakefield Prison, where the instruments are read every 

 six hours, night and day. The portion of the diagram for the months of 

 October and November, showing the state of the atmosphere during the 

 passage of the ' Royal Charter ' storm, has been compared with observations 

 made at Oxford, Kevv, Stonyhurst College, Lancashire, and the Bishop's-rock 

 Lighthouse, Scilly Isles ; and the general agreement fully warrants the selec- 

 tion of the Wakefield curves as a fair type of the state of our atmosphere 

 during the year 1 859-. 



The curve of mean temperature is from results in a paper by Mr. Glaisher 

 in the 'Transactions of the Royal Society ' for 1850. 



If there were no connexion whatever between the weather and the condi- 

 tions that favour an explosion in a coal-mine, it would be found that the 70 

 or 80 vertical lines that denote fatal explosions would be scattered, as if by 

 chance, over the whole diagram, without any apparent reference to the great 

 depressions in the barometric curve, or to the great and sudden rises in the 

 thermometric curve. But this is not the case in any of the years that I have 

 examined. On the contrary, it is found that the lines of explosion have a 

 very decided tendency to group themselves about the few great atmospheric 

 perturbations of each year ; and to leave a very conspicuous and highly sig- 

 nificant blank in spaces, of a whole month's duration occasionally, where the 

 pressure has been uniformly high and the temperature moderate. 



In the 68 explosions of 1859 are found three dense groups and a number 

 of equally instructive blanks. 



The first group falls between the 11th of January and the 17th of Febru- 

 ary, during which period the diagram shows that even the nocturnal tem- 

 perature was considerably above the mean daily temperature, and the baro- 

 metric curve exhibits a succession of deep indentations marking the passage 

 of a series of storms. 



The dates and localities of the explosions forming this group are : — 

 January 11, Bewdley. 



12, Atherstone. 



15, Huddersneld. 



17, Ayr, Scotland. 



19, Wigan. 



• 25, Stevenston, Scotland. 



29, Burslem, Staffordshire. 



January 29, Aberdare, S. Wales. 

 February 2, Dudley. 



3, Coatbridge, Scotland. 



9, Willenhall. 



12, Wednesbury. 



17, Wigan. 



Two cases of death from suffocation by gas fall within this group, viz., — 

 On February 1, at St. Helen's, and 



'■ — 18, at Tiviotdale, Rowley Regis. 



An interval of a fortnight follows, with a high atmospheric pressure, and 

 no fatal accidents in mines from gas. 



During March, and the first week of April, the temperature is far above 

 the mean, and two well-characterized cyclones send the mercury in the baro- 

 meter at Wakefield down to 28-83 on the 14th, and to 28-91 on the 28th. 



The second great group of 14 explosions falls in this interval ; 8 explosions 

 happening within 8 consecutive days— exactly coinciding with one of the 



