ON THE MAXIMUM DENSITY OF WATER; 29 



Water expands by heat from fome point (whatever it may. Count Rum- 

 prove to be) by a law which is nearly that of the fquare of ^J "^ a I i '„ edf 

 the temperature from the faid point, as is evident from Sir on Mr. Dalton'* 

 Charles Blagden's table. Confequently the force of afc^^ftttery 

 which water acquires by temperature, is at firft very fmall, \j tt \ t change of 

 but increafes to a very confiderable amount before ebullition, dimenfion* by 



J n c heat or c0 '® near 



The cohefion of the particles of water is a conltant force; its max< denfity. 

 there will therefore be a point of equilibrium between thefe 

 two forces; that is, a point at which the increafed temperature 

 will be but jufl fufficient to counteract the tenacity, in which 

 cafe no internal motion can enfue. Whether a diminution in 

 denfity in water to the amount of one hundredth or one thou- 

 Jhndth, or one ten thoufandth or more, is the point alluded to, 

 is to be determined only by experiment. I apprehend that This change at 

 water of 40° is about one ten thoufandth part lighter than water ^^0™*^,." 

 of 32°; but that this force of afcent is but juft fufficient a current j 

 to counteract the tenacity, and confequently no motion takes 

 place ; in fuch cafe the diffufion of heat through water is the 

 fame as through a folid body. Whenever the difference in 

 denfity exceeds thut jufl: mentioned, internal motion is thecon- 

 fequence, and that greater in proportion to the difference of 

 denfity, which we know may amount to J ? of the whole. 



Count R.'s experiments therefore will be explained by ob- — and therefore 

 ferving that the thermometer acquired heat by the proper con- ^d"^^* 

 dueling power of water, as if it had been metal, or any other wards as if thro* 

 folid body; the temperature acquired was greater in the 2d jfht't*** m 

 experiment than in the lft, becaufe the heat of the ball was atures there wat 

 greater; but in the 3d experiment the heat of the ball was a current, which 

 fuch as to produce a current upwards that almoft precluded defcent. 

 the defcent of heat, by carrying away the heated particles as 

 foon as formed. 



The circum (lances of the two thermometers by the fide of p lfficu ^y re- 



., , „ , jL ,, c a • , » • • 'pectingthe two 



the ball and the cup, in the two nrft experiments not acquiring thermometers. 



any temperature, is certainly remarkable, and not eafy to be 



explained, even upon Count R.'s principle; for, the fuppo- 



fed defcending current of warm water fliould have filled the 



cup and overflowed, fo as to affecf the collateral thermometer. 



One moft important experiment Count R. has omitted, and important expe- 



which it is particularly defirable that he, or fome one in pof- L et the wTieVbe. 



feffion of a fimilar apparatus, would perform, efpecially as it at 40 and the 



would go further than any other to eftablifh th# doctrine of cur- °' at 3 Z • 



rents 



