1823.] the Maximum Density of Water. 405 



small globule of mercury was introduced, which perfectly 

 answers that purpose. 



As the momentum of an ascending ball is very apt to cause its 

 upper extremity to rise above the water, and however free of 

 any thing unctuous, it will there remain too long, a slight tap or 

 blow by a small hammer, on the under side of the table, will 

 obviate this incident. 



In cooling water for such experiments, it ought to be kept as 

 still as possible ; agitation to procure uniformity of temperature 

 has a bad effect by charging it with air ; bubbles may settle on 

 the ball during the experiment, and must be closely watched for, 

 as their effect may be apprehended, if detected occasionally rising 

 through the water. Knowing the degrees at which a ball might 

 be expected to rise or fall, I have frequently lifted it to the sur- 

 face of the water a short time before, in order to free it of any 

 thing which though imperceptible might have affected its gra- 

 vity. The thermometers were sometimes placed one at the top, 

 and the other at the bottom of the vessel, in order to ascertain 

 beyond doubt the temperatures at the initial points, or the 

 extremes of the above-mentioned intervals. 



When it was considered how uncertain the indications might 

 have been, had I succeeded in adjusting a ball to seeming equi- 

 librium at the maximum gravity, owing to the minute variations 

 near that point, there was little cause for regretting my failure, 

 especially when the method by varied extents of intervals seems 

 so satisfactory ; still, since writing the above, another effort was 

 made, when the following appearances took place. 



Water in the jar being near 42, and the ball as seen by the 

 naked eye in apparent equilibrio, it was observed with the read- 

 ing glass as seen over a slight scratch on the side of the jar; it 

 was then very slowly descending ; having two or three times 

 breathed on the part of the jar nearest the ball, the consequent 

 dimness was removed by a camel hair brush, but before this 

 could be done and the eye-glass applied, the ball had decidedly 

 begun to ascend, which it continued to do for a few seconds, 

 and after a momentary pause again began to fall. This was 

 repeated several times, the thermometer meanwhile ranging 

 from 42 to 42-5' ; from this and other circumstances, I with due 

 deference incline to think, that 42*3 is very near the true point 

 in the scale of temperature, where the maximum density of water 

 takes place. In this last trial it may not be improper to men- 

 tion, that the increment of weight producing the approximating 

 effect, was a mere speck of leaf gold, attached to the side of the 

 ball by means of spirit-varnish, and fixed by applying a mode- 

 rate heat. 



Having thus given an explicit account of these experiments, 

 apology on my part for having too minutely done so, will be 

 deemed quite unnecessary by any one who repeats them. I 

 shall only add, that the thermometers having been made pur- 



