|3;2 G'REytTEST ©EXSITY OF WATEK. 



second is that in which the liquid still descends, but so as 1» 

 produce re^ilar strata from the bottom, increa*ing in tempera- 

 ture upwards to the middle, where it at last becomes of that 

 temperatuvc~which is of the same specific gravity as the upper 

 liquid at 32" ; the third is that in which all the heat acquired 

 nsccnds into the upper half of the vessel. Let us now see 

 whether the 5th experiment of Dr. Hope will more favour the. 

 notion of greatest density at 36 or at 40°. In transcribing hi* 

 results, I will add a column denoting the temperature at the 

 middle of the vessel, such as I apprehend it would hine been 

 found by a thermometer. It must therefore be noticed, that 

 the middle column is an imaginary one, an<!fencft obtained from, 

 Ur. Hope's observations. 



The observations were : 



Bottom. Middle. Top. 

 At corFAmencement, — 32 — 32 — 32 

 In iO minutes, — 35. — o-t — 32 

 ]5 36 — — 35 — 52 



20 — 36-}- — V>6 — 32 



05 37 __ 37 _ 33 



30 38 _ 38 — 3;> 



38— 38 -j- — 40 — 33 



45 39— — 42 — 33 



60 • 39+ — 44 — 44 



55 39+ — 46 — 45 



Co 39+ — 48 — 4S 



Induction that Now if we take 30" as the point of greatest density, the first 

 denfity^^atseoF^^^^ °^ *^"^^' ^'^^ ^^ 20 minutes, after which, the botton*. 

 and not 40". thermometer was found at 56°+, the middle probably 35 or 

 36", and the top one unaffected at 32"; the second period will 

 terminate about 38 or 40 minutes, when the bottom thermo- 

 meter was 38-1-, tiie middle 40" (indicating the same specifie 

 gravity as 32"), and the top 33°, having only gained 1"; the 

 third terminates with the cx}jerimcnt ; during this period the 

 bottom thermometer has gained 1", which may easily be acr 

 counted for by the natural tendency to equalization of tem- 

 perature in the lower half of the jar; the middle of the jar^ 

 being the focus of heat, may be supposed about equal to the 



top 



