128 
HAROLD’S DISCUSSIONS. 
pliere, which may be compared to water flowing 
out of a tub or basin through a hole in the bot¬ 
tom. I filled a large stationary wash-bowl with 
water, and after the liquid had become perfectly 
quiet I pulled the cork out very carefully. Soon 
the water began to move in a spiral direction just 
above the hole, and quickly developed a funnel- 
shaped opening. This experiment was performed 
thirty-eight times. Thirty-three times the water 
moved in a spiral direction contrary to the hands 
of a watch, and but five times in the opposite direc¬ 
tion. I could easily see why it should move spirally, 
but why it should choose one direction rather than 
the other I was not able to tell. 
As the plug is pulled out, the water begins to 
drop through the hole, and the liquid rushes in from 
all sides to take its place. As it does so, the opposing 
currents collide with and pass each other, producing a 
rotary motion. 
A cyclone is very much like this, except that 
where the water flows downward through the open¬ 
ing, the atmosphere flows upward and outward in the 
center of the cyclone, as the rotary motion tends to 
produce a vacuum. 
Why should the rotary motion be contrary to the 
hands of a watch ? As the air particles flow toward 
the storm center, often from a great distance, in obe¬ 
dience to the suction, they are turned to the right of 
a direct course (Fig. 64) by the rotating influence of 
the earth. Its path to the center would therefore be 
a portion of a narrowing spiral (Fig. 64). Air par- 
