388 Roi/al Society. 



as to allow of the water in the cistern being withilravvn, without dis- 

 turbing that contained in the tube. 



The boiler was set with brickwork, in a proper position, over a 

 small fire-place. It was nearly filled with distilled water, which was 

 made to boil thoroughly so as to free it from air ; and the cock being 

 then closed, the water was raised in the tube by the pressure of the 

 steam collected in the upper part of the cistern. The tube, when 

 filled, was hermetically closed at the top : a proper scale, constructed 

 by Newman, was applied to it, great care being taken to determine 

 its height and to ensure the accuracy of its adjustments, and the 

 precision of its measurements, by an exact mode of reading; and 

 also to provide proper corrections for temperature. The water in 

 the cistern was protected from contact with the air by being covered 

 with pure castor oil to the depth of half an inch. The mercurial ba- 

 rometer employed as a standard of comparison, was of a portable 

 construction, and was provided with a platina guard. 



An account is then given of some of the results of the observations 

 made with this water barometer, arranged in several sets of tables. 

 The great object was to obtain good and uninterrupted series of ob- 

 servations, taken, at least once a day, at a fixed hour. The regis- 

 ters given by the author, contain such observations, continued for 

 nearly a year and a half, namely, from October 1830 to Marcii 

 1832. Some curious results are afforded by these observations. In 

 windy weather the column of water is found to be in perpetual mo- 

 tion, not unlike that from the breathing of an animal. Many consi- 

 derable fluctuations in the pressure of the atmosphere are rendered 

 sensible bv the motions of an aqueous column, which would totally 

 escape detection by the ordinary mercurial barometer. Mr. Hudson 

 remarked in the course of his observations, that the rise and fall of 

 the water-barometer precedes by one hour the similar motions of the 

 mercurial one. The most striking result of the comparison between 

 the two, is the very near coincidence of the elasticity of the aqueous 

 vapour, as deduced from the experiments, with its amount, as deter- 

 mined from calculation, in a range of temperature from [iS° to 74'^. 

 But a gradually increasing difference was at lengch perceptible, show- 

 ing that gaseous matter had by some means insinuated itself into 

 the tube. When this became no longer doubtful, the boiler was 

 opened, and it was found that a portion of the liquid oil had escaped .• 

 and that the remainder had become covered with large flakes of a 

 mucilaginous substance, by means of which it is probable that a 

 communication had been established between the air and the water. 

 The water had, however, retained its purity, and no indication was 

 afforded of the metal having been anywhere acted upon. The au- 

 thor recommends that if these researches are prosecuted, the water 

 .should be covered with a stratum of oil of four or five inches in depth, 

 which he has reason to think will form an effectual barrier to all 

 atmospheric influence. 



11." Hourly Observations on the Barometer, with experimental 

 investigations into the phsenomena of its periodical oscillation," by 



