2o8 Mr . Walker’s- Experiments on the 
I now embraced the lower part of the tube with my hand a 
few feconds, refting it upon the upper part of the bulb ; and 
upon taking it away, I found that. the whole of the mercury 
had iubfided. into the bulb, which it did not now quite fill,, a 
fmall fpace at the top of the bulb remaining empty. I them 
took out the thermometer glafs, and applied my hand to the 
tube ; but the mercury remained ftationary until I funk my 
hand fo as to communicate heat to that part of the bulb which 
is immediately connected with the tube, when the thread of 
mercury dropped entirely into the bulb. It was now im- 
merfed again for a Ihort time, then taken out, and the (hell of 
glafs beaten off, which expofed a globe, of folid mercury, nearly 
an inch in diameter. This bore feveral very fmart ftrokes with, 
a hammer before it began to liquify, but. was not perfectly 
malleable. 
In the courfe of thefe experiments, feveral fragments of the 
folid mercury were thrown into mercury in its ordinary liquid 
ftate, and were found, to fink with confiderable celerity. 
In continuing my refearches refpe£ting the means of pro*- 
ducing artificial cold, I have found that phofphorated natron * 
produces rather more cold by folution in the diluted nitrous 
acid than the vitriolated natron i 
At the temperature of + 50% four parts of the diluted ni- 
trous acid (prepared by mixing ftrong nitrous acid with half its 
weight of water)* required eight parts of that neutral fait in* 
* 
fine powder to be added, in order tocaufe the thermometer to. 
fink, to 6° ; and again* by the addition of five parts of ni- 
trated ammonia in fine powder, the thermometer funk fb low 
as —1 6°, in the whole fixty-fix degrees* 
A mixture of this kind made the thermometer fink from 80® 
((he, temperature of the materials before mixing) to o°. 
I,- was- 
