OF THE GASES OF THE KING'S BATH SPRING. 3 



This consisted of four upright posts, fixed by means of cords to opposite angles of 

 the hand-rail which separates the central portion of the King's Bath from its sides ; 

 these posts being connected together by planks extending from one to the other, and 

 resting horizontally on the tops of each. 



This framework served as a point of attachment to the pulleys by means of which 

 the iron shield was raised and depressed ; and it at the same time enabled me to cover 

 over that portion of the bath in which my operations were conducted, with an awning, 

 to serve both as a protection from the weather, and as a screen from observation. 



In this manner I was enabled to carry on my experiments for the period of a 

 month, without being any hindrance to others, and with comparative ease to myself 

 and my assistants, choosing a time of day after the hours of bathing were concluded, 

 when the water was comparatively low, which, as the bath is emptied regularly once 

 a day, and afterwards replenished from the spring at the uniform rate of nearly six 

 inches per hour, could always be reckoned upon at a similar time in the afternoon. 



On one occasion, however, I caused the water to be kept low during the space of 

 twenty-four hours, in order to satisfy myself that there was no material deviation from 

 the mean quantity of gas evolved each day, at the particular period that had been 

 commonly selected ; and for this purpose I collected the gas at intervals of about six 

 hours, namely on Saturday at 1 p.m., again at 6 in the evening, again about mid- 

 night, again at 7 o'clock on Sunday morning, and lastly at noon of the same day. 

 Finding, however, a greater variation between the quantity collected at the same hour 

 on the two successive days, than between that obtained at different hours of the same 

 day, I saw no advantage in departing again from the usual routine. 



There might be more reason for suspecting, that the periods to which each obser- 

 vation extended were not sufficiently long to secure a fair average ; and fearing that 

 this might have been the case in the first trials, I gradually prolonged the duration 

 of them from five to fifteen minutes : so that the time occupied in the experiments 

 amounted altogether to more than four hours. 



If it be objected, that this period even might have been too short to insure exact 

 results, I may reply, that the correspondence which is frequently to be traced between 

 the amount obtained on two or even three successive days, affbrds in itself a pre- 

 sumption, that the numbers given furnish in general a pretty fair expression of the 

 mean quantity of gas emitted every minute from the spring. 



Assuming this to be the case, it would appear from the sum total of the observa- 

 tions recorded, ttiat the quantity of gas evolved averages 264 cubic inches, or, 

 rejecting observation 3rd, (which was noted at the time as of doubtful accuracy, on 

 account of some defect in the apparatus subsequently remedied,) 267 cubic inches 

 each minute. 



The largest quantity ever obtained in the course of one minute appears to have 

 been 530 cubic inches; the smallest in the same time, 80: but if we take the mean 

 of the observations, we shall find, that the quantity usually varied from 339 to 207, so 



b2 



