438 



NATURE 



{Sept. 8, 1887 



been made at Saarbriick and Karwin. The Austrian inquiry 

 showed that " where after a rapid rise of the barometer it con- 

 tinued to rise slightly, or remained stationary for some time at 

 its maximum, a gradual increase in the volume of gas in the air 

 would set in ; or if, after a rapid fall in the barometer, it con- 

 tinued to fall gradually, or remained stationary at its lowest 

 point, a decrease in the quantity of gas would become apparent." 



Evidently, therefore, from these researches the greatest danger 

 is not, as a rule, to be apprehended when the barometer is low 

 or falling, and this is supported by actual disasters, the majority 

 taking place under anticyclonic conditions of pressure. While 

 Mardy, Pendlebury, Penygraig, Seaham, and many others add 

 to the verdict, it will suffice to deal with some explosions of the 

 present year, and see if they do not bring home to us a new view 

 of the natural forces at work far down below the surface of the 

 earth. 



From the simultaneous observations made at 6 p.m. on Friday, 

 February 18, the Meteorological Office reported : — "The baro- 

 meter is now rising in all parts of the United Kingdom, and an 

 anticyclone is apparently advancing from the westward." An 

 hour later thirty-nine lives were lost in an explosion in the 

 Rhondda Valley. The anticyclone continued on its course to 

 the Continent, and by the morning of Wednesday, February 23, 

 when so much damage was wrought by the eai-thquake, the 

 centre had reached Southern Europe. On March i the anti- 

 cyclone was a little further north, and over the neighbourhood 

 of the Chatelus Mine, near St. Etienne, where ninety lives were 

 sacrificed. Still moving northward, the night of March 4-5 

 found the highest barometer readings over Belgium and the 

 Netherlands, when 144 miners perished at Quaregnon, near 

 Mons. 



In the last week of May another anticyclone moving from south 

 to north was marked by the loss of one life at DarcyjEever on 

 the 25th, three lives near Wigan on the 26th, and seventy lives 

 at Udston, near Glasgow, on the 28th. 



An anticyclone over Western Germany on the night of June 

 7-8 marked about sixty deaths at Gelsenkirchen. As this area 

 moved to the westward, a slight earthquake was felt near Stras- 

 burg on the iith, and a severe one in La Vendee on the 15th. 



Clearly Mr. Buddie's strong opinion is not applicable to the 

 second half of the century. The knowledge that gas is found 

 escaping with a rising barometer, and that so many explosions 

 take place as indicated, has led mining officials to blame the 

 mercury for not falling even before the gas begins to escape, 

 their idea being that pressure has actually decreased, but that 

 barometers are many hours before taking up the changes. 

 The idea may be dismissed as an erroneous one. The cause 

 must be sought for in another direction, not the direct action 

 of variations of atmospheric pressure on the gas as it leaves 

 the coal, but the effect on the earth's crust and indirectly 

 on the occluded gas. Whatever be the true cause of earth- 

 quakes, there seems to be no reason to doubt that fluctuations 

 of atmospheric pressure cause undulations of the earth's crust. 

 Prof. Darwin, taking a probable estimate for the elasticity of 

 rocks, has calculated that with a range of two inches of the 

 barometer we are at least three or four inches nearer the earth's 

 centre when the instrument stands very high than when it is 

 very low, and concludes : "It may be that the incessant strain- 

 ing and unstraining of the earth's surface is partly the cause of 

 earth-tremors, and we can at least understand that these strains 

 may well play the part of the trigger for precipitating the ex- 

 plosion of the internal seismic forces." The seismological records 

 of Japan show that earthquake shocks are twice as numerous 

 under the predominant anticyclone of the winter months, as 

 they are in the summer with lower pressure. As a result of the 

 discussion of earthquakes in Jamaica, Mr. Maxwell Hall 

 concludes that "at the time of an average earthquake shock the 

 barometer is a little above its average height. This is due to the 

 circumstance that the winter months, December, January, and 

 February, when the barometer is above its monthly average, 

 are more liable to shocks than other months of the year ; and 

 that the hours from 8 p.m. to 2 a.m., when the barometer is 

 above its diurnal average, are similarly more liable to shocks 

 than other hours of the day." Explosions of fire-damp follow 

 a similar rule ; they are most numerous in the winter 

 months, when the range of pressure is greatest, and usually 

 when the barometer is very high. Allowing for the flexure of 

 the earta's surface, we can conceive that with the downward 

 movement under increasing pressure the pent-up gases are forced 

 into the workings of our deep mines ; it may be indeed these 



movements cause infinitesimal fissures in the coal-seams through 

 which the gas passes into the workings at a time when it has 

 been customary to believe there was least danger. There is 

 some degree of probability in this from the fact, so frequently 

 noted in great explosions, that there is a suddenness in the 

 appearance of the gas which is not a common experience in 

 shallow workings. 



Taking into consideration all the recorded facts, they point 

 to the conclusion that far greater weight should be attached to a 

 period of high atmospheric pressure than has hitherto been 

 deemed necessary. In any future discussion of this important 

 subject it is to be hoped further evidence will be forthcoming, 

 and that instead of endeavouring to connect every disaster with 

 a low barometer, the distribution of pressure as a whole be 

 taken into account. 



The influence of coal-dust upon explosions has not been 

 touched upon, but it may be remarked that the dry atmosphere 

 of an anticyclone renders the dust more inflammable than the 

 dampness of a low-pressure system, so that there is a double 

 reason for giving closer attention to mines under anticyclonic 

 conditions. Hy. Harries. 



MEASUREMENT OF SPECIFIC HE A T. 



TTAVING regard to the comparatively large experi- 

 -^ ■*■ mentai error introduced by thermometers into 

 specific heat measurements, a null method appeared tc> 

 be desirable. The following metliod occurredjto me 

 about two months ago, but not having access to aTphysi- 

 cal laboratory, I have not been able to practically test its 

 accuracy. 



Two e.xactly similar calorimeters (A and B) are taken, 

 each containing a coil of thin Pt wire of resistance R, 

 so arranged as to be completely immersed in the liquid. 

 A contains a mass, M (including water equivalent), of 

 water ; B the same mass of substance the specific heat of 

 which is being measured. The wires are arranged in 

 bridge fashion, so that the ratio of the currents flowing 

 through the two wires may be made to take any value. 



A differential thermometer (not indicated in the sketch, 

 for sake of clearness) shows the least difference of tem- 

 perature between A and B. Probably the most delicate 

 and convenient arrangement is to use two thin Pt wires 

 balanced in the arms of a bridge, using a very sensitive 

 galvanometer. 



First consider the calorimeter B containing the sub- 

 stance. It receives a quantity of heat, H, from a current, 

 C, flowing through a resistance, R, for a time, /. Hence 



H = ^^A= ^MS 



(where 6 is the rise of temperature, and S the mean 

 specific heat for that interval). 

 Similarly in A, containing water, 



HI = ^'^-^ = ^M. 



