I 254 ] 
remark, that the quantity of ice depends materially on the 
Weather; and confequently, it has fometimes happened, 
that no congelation took place. At others, perhaps, half the 
quantity will be frozen; and I have often fecn the whole 
contents formed into a perfe6l cake of ice : the lighter 
the atmofphere, and the more clear and ferene the wea- 
ther, the more favourable for congelation, as a frequent 
change of winds and clouds are certain preventives. P"or 
I nave frequently remarked, that after a very fliarp cold 
night, to the feel of the human body, fcarce any ice has 
been formed ; when at other times the night has been 
■calm and ferene, and fenfibly warmer, the contents of 
the pans will be frozen through. The flrongeft proof 
of the influence of the weather appears by the water in 
one pit being more congealed than the fame preparation 
for freezing will be in other fltuations, a mile or more 
diftant. 
To reafon phyflcally upon this procefs of making ice, 
it may be faid, that had the thermometer been fufpended 
in the air, free' from every other body capable of com- 
municating heat, in fome parts of the night during the 
cold months of December, January, and February, the 
quickfllver might have defeended to the freezing point, 
hnd that water, being artfully placed in a flmilar fltua- 
'tion, contained in thin porous pans, and fupported by a 
fubftance little capable of communicating heat from the 
earth, might alfo freeze, and continue in a- flate of con- 
■gelation till the heat of the morning came on. I fay this 
may be poflible ; but at the fame time I mull beg leave 
to 
