92 
HAROLD’S DISCUSSIONS. 
First, I made a barometer, sucli as was described in 
Book III, Harold’s Quests, and attached a graduated 
scale marked in inches and tenths of an inch, in order 
to read the height of the mercury in tlie tube at a 
glance. The greater the atmospheric pressure, the 
higher the mercury stands in the tube and the less there 
will be in the cup. Thus its level in the cup is con¬ 
stantly changing. The true height of the mercury 
column is the distance from the level of the mercury 
in the cu]) to the point at which it stands in the tube. 
Therefore, to get the correct height of the column, 
we must have a measuring wire so attached as to rest 
upon the surface of the mercury in the cup and move 
up and down with it. This I accomplished by attach¬ 
ing with little staples a fine wire about thirty-one 
inches long to the board beside the tube, so that it 
could move freely up and down. At the lower end 
of the wire I attached a piece of cork properly fitted 
into the cup so that the whole fioated on the mercury 
in the cup. At the upper end I attached the scale- 
card, by which the standing of the mercury in the 
tube could be easily read. When the air had little 
pressure, the mercury in the tube dropped and in¬ 
creased the amount in the cup. As this increased, it 
raised the scale-card so that the thirty-inch mark, for 
instance, was always that distance above the surface 
of the mercury in the cup. When the air-pressure 
increases, the scale-card settles with the mercury in 
the cup. 
A light board one foot square, suspended by wires, 
served to measure the force of the wind when blow- 
