COLLIERY EXPLOSIONS AND THE WEATHER OF 1885. ANON. 251 
nection with the management of collieries. Still more does this 
become apparent when we come to deal with the very complete series 
of hourly observations carried out under the supervision of Mr. V. W. 
Corbett, in the workings of Seaham Colliery after the disaster of Sep- 
tember, 1880. The observations extend over a period of several 
months, and the diagrams show at a glance that the gas-check indi- 
cated an increase of gas with any important increase of atmospheric 
pressure. The barometer at the surface, and two others in different 
parts of the workings, agTee in their fluctuations, both as regards the 
time and the amount of the changes. The Maudlin Seam had been 
carefully sealed up, so as to have no connection with the outside 
atmosphere. The variations of pressure in this hermetically sealed 
chamber were recorded by water gauges, which showed that gas was 
escaping while the barometer was rising, and long before any decrease 
of pressure had set in. The seeming contradiction of the law of 
pressure on gasses led Mr. Corbett to the conclusion that the baro- 
meter is very tardy " in denoting the fluctuations of atmospheric 
pressures," even as much as 33, 35, 41, and 48 hours late. Were this 
a fact charts of synchronous observations of wind and barometer would 
present a strange peculiarity, viz., that the atmospheric circulation 
indicated by the winds would sometimes be more than 2,000 miles in 
front of the position shown by the barometer, a circumstance not yet 
recorded in any country, and never will be. The water-gauge and 
barometer indications were for two distinct elements, so that the 
action of the one need not accord with the action of the other in every 
particular. As, however, water-gauges are not in common use, we 
have to content ourselves with the instruments which are, and there 
seems to be no doubt that, whether we take into consideration the 
conditions of atmospheric pressure shown by the ordinary barometer 
over a tract of country at the time of explosions, or the excellent 
series of local observations by Mr. Corbett, we are forced to the con- 
clusion that gas becomes dangerous while the barometer is rising and 
long before there is any indication of a decrease of pressure. Further 
proof of this has been discovered in the investigations which have 
followed upon the great explosions in Austria within the past eighteen 
months. One of the conclusions arrived at in the report on the 
