1822.} the Quantity of Rain falling in adjaeent Places, 19 
one inch deep of rain is equal to 28-274 cubic inches (nearly one 
int wine measure), and in weight at the temperature of 60° to 
7139-1850 grains ; consequently } inch = 1784-7962 or 17843 
grains nearly. Let, therefore, the glass jar be exactly balanced 
m good scales, and then filled with raz or distilled water to 
the perfect equipoise of 17843 grains, marking the height on 
the side of the measure, which will be the indication of :25 or 4 
inch of rain in the guage. Were the cylinder of equal diameter 
throughout, it would only remain to divide the space so marked 
off into 25 equal parts; but as these glasses are seldom quite 
uniform, it is necessary to check the measure by weight to every 
05 of an inch, which is easily done by weighing with 14273 ers. 
for -20, with 10703 for -15, with 714 for -10, and with 357 grs, 
for -05 of an inch, marking these divisions severally and accu- 
rately on the side of the measure, and then dividing each into 
five equal parts, observing to allow in the lowest for the bulb 
usually found at the bottom. Thus the measure will be gra- 
duated to the hundredths of an inch, and if it be about 14 inch 
in diameter, the spaces will be large enough to halve, so that the 
register may be conveniently kept in three places of decimals; 
thus instead of -101, set down -105, being so many thousandths 
of an inch. If there is any difficulty to cut the glass, the gradua- 
tion may be marked with a pen on a slip of paper pasted on the 
side of the measure, observing that it should be quite dry before 
the operation takes place. 
It is thus easy to provide an accurate pluviameter ; but to find 
a suitable situation for fixing it ‘‘at a sufficient distance 
from trees, buildings, or any object that might obstruct ” the 
free current of the wind, is a matter of great difficulty, and of the 
greatest importance. For this reason, the tops of the highest 
buildings have been heretofore selected. There is, however, 
cause to suspect that they are the most ineligible. Inthe reports 
from Kinfaun’s Castle, the upper is made to indicate more rain 
than the lower guage. This is contrary to all similar observa- 
tions, and to the nature of things. The error must, therefore, be 
in the different capacity or situation of the instruments used. 
To this phenomenon, which has excited so much speculation, 
our observations have been particularly directed. Three guages 
and measures, exactly alike in form and size, have been used. 
The first was already fixed onthe top of our Museum (from this 
our annual reports are drawn), higher than the level of the adja- 
cent chimney stacks, and consequently free from lateral obstruc® 
tions. The house is open in front, joined on each side to others, 
and its back towards the continuous buildings of the town. The 
guage is 45 feet.above the surface of the ground, and 143 above 
the level of the Sea. The two other guages are fixed at the level 
of ihe ground, each in a garden, and at a distance of about 60 
yards from any building or high trees, being in respect of all 
circumstances, that seem to affect the fall of rain, similar each 
c 2 
