April 2, 18SS.] 



♦ KNOW^LEDGE ♦ 



125 



attraction at the top and bottom of the mine is a matter of 

 sufficient simplicity. 



Though it is tolerably obvious that this method of deter- 

 mining the earth's attraction must be even less trustworthy 

 than experiments of the Schehallien type, the method com- 

 mended itself to the mind of Sir George (then Professor) 

 Airy, and after two failures in the Dolcoath mines, he had 

 the method tried in the Harton Colliery, near South 

 Shields, where a depth of 1,260 feet was available. It is 

 not necessary to describe the contrivances by which two 

 pendulums were compared, one swinging at the top the 

 other at the bottom of the mine, the pendulums being inter- 

 changed after intervals of 10-t hours and 60 hours — that is, 

 each working 104 hours above and below, and then each 

 working 60 hours above and below. The work was con- 

 ducted with great care and skill by Mr. Dunkin, of the 

 Greenwich Observatory, to whom, however, no portion of 

 the credit has hitherto been given in treatises on astronomy. 



The observations showed that at the foot of the mine 

 either pendulum gained two seconds and a quarter per day, 

 bhow-ing that gravity was increased by xaJ^jTy part. This 

 result cannot be sensibly in eiTor, so carefully were Mr. 

 Dunkin's operations conducted. But the inferred increase 

 of density towards the earth's centre — that is, the deduced 

 mean density of the earth — can by no means be regarded as 

 ascertained with corresponding accuracy. We do not make 

 an ordinary foot rule more precise for measuring purposes 

 by careful and complicated experiments on the changes it 

 undergoes under the varying influences of temperature, 

 moisture, and so forth — it remains a foot-rule still, with all 

 a foot-rule's shortcomings as a measurer. The complicated 

 calculations and corrections effected by Mr. Dunkin did not 

 even touch the real defects of the mine method of weighing 

 the earth ; it remained rough and untrustworthy. A com- 

 parison of the final i-esult with that obtained bj- experiments 

 of the Schehallien class — which, though certainly not trust- 

 worthj', were at least as trustworthj' as the mine experi- 

 ments — suffices to show how little either method can be 

 trusted. The earth's mean density, as calculated by Jlr. 

 Airy from the Harton experiments, was 6-565 times that of 

 water, and the results of the mountain and mine methods 

 combined appear as follows : — • 



Mean DeDsity 

 (Water"s=l). 

 M.askelyne'3 Schehallien experiment (corrected 



by Piayfair) 4 713 



Carlini's pendulum experiments on Mont Cenis 



(corrected by Giolio) 4 950 



Colonel H. James, from attraction of Arthur's 



Seat 5316 



Dnnkin and Airy, from experiments in Harton 



Colliery 6.JGJ 



Mean 



5-386 



The mean value is probably, as we shall see, near the 

 truth ; but as the greatest value differe from the least by 

 1852, or more than a third of the mean, the results — 

 except Colonel James's — are discredited by the very circum- 

 stance that the mean value is nearly right ; and all three 



attraction at Q is greater than, equal to, or less than the attraction 

 at P. Putting (■)■) equal to zero, we get 



3 ■jYa'^^b- = 3 J -f 2a 

 or 



36 a= + 9 J= = 9 J= ^ 12 o J ^- 4 a^ 

 that is 



.32 a = V2h. 

 Hence, if i, the radius of the cylindrical mine, is equal to tour- 

 thirds its total depth, the attraction at P will be equal to that at 

 Q. If the mine be wider, the attraction will be greater at P than 

 acQ. 



All other cases may be similarly treated, though, of course, the 

 problem will not in all cases be so simple as the above. 



methods are discredited, the correctness of Colonel James's 

 result being thus shown to be merely accidental. If the 

 length of a piece of ground had to be measured, and one 

 workman said that measuring it in a certain way he found 

 it to be 471 feet long, while another using a different 

 method made it 49.5 feet long, and a third, using yet another 

 method, found it to be 657 feet long, their employer would 

 not regard their work or their methods as satisfactory if he 

 presently found that a thoroughly trustworthy measurement 

 showed the piece of gi-ound to be 550 feet long, even though 

 that is not far from the mean of the results obtained by 

 three unsatisfactory methods. And if, later, a fourth work- 

 man, employing one of those methods, deduced as a result 

 539 feet, trust in that method would not te greatly increased. 

 It would be felt that either the approach of the result 

 to the truth was accidental or it was more or less consciously 

 forced. 



Airy expressed the opinion that the result of the Harton 

 Colliery experiment was comparable on at least equal terms 

 with those obtained by other methods, though it differs by 

 20 per cent, from the mean of the results obtained by all 

 methods, and the results presently to be considered do" not 

 differ by more than 2 per cent, from their mem value, nor 

 by 4 per cent, from the general mean. I prefer the opinion 

 of Sir Edmund Beckett (now Lord Grimthorpe), that the 

 result of the Harton Colliery experiments " cannot be 

 accepted," and is " not to be compared in value " with those 

 obtained by the Cavendish experiment. 



The ingenious Michell, to whom science owes the first 

 satisfectorv' reasoning about the architecture of the sidereal 

 heavens, devised the method of weighing the earth which is 

 commonly named after the eminent chemist Cavendish, who 

 first successfully applied it. 



( To he concluded.) 



COAL. 



By W. Mattieu Williams. 



-MINERS' LAMPS AND COLLIERY EXPLOSIONS. 



APPILY for themselves, few or none of my 

 r&aders have any practical acquaintance 

 with absolute darkness. "With the excep- 

 tion of a skilfully devised horror — the dark 

 punishment cell in some of our prisons — 

 scarcely any place can be found above 

 ground where such darkness prevails. But 

 in a coal-pit, without lamp, it exists in perfection. The 

 lighting of a coal-mine is a serious problem, not because 

 illuminating gas is dear, but for the opposite reason. Lead, 

 copper, and other mines in which metals and their ores are 

 worked, are lighted by the primitive device of wearing a 

 candle in the front of one's hat while travelling down the 

 shaft or along the working.*, and sticking it in a miner's 

 candlestick, a lump of clay, which is superior to ordinary 

 candlesticks, inasmuch as it may rest on the ground or con- 

 stitute its own bracket by being dabbed against a wall, or 

 take any other position required. 



Primitive oil-lamps of pattern closely resembling those 

 found so abundantly in Pompeii, or metal lamps with a hook 

 at one side for attachment to the hat or hanging to ledges, 

 are used, and others with a spike below for simdar purpose; 

 but the candle and lump of clay is the general favourite. 

 Paraffin lamps have been lately introduced. 



Candles and such lamps may be used in some coal-mines, 

 but these are exceptional, the majority of coal-mines beinw 

 " fiery." This means that hydrocarbon gas, which has been 



