namely, that the water in it may be cooled below the freezing point and yet not frozen. A snowflake or ice crystal falling into it may suffice to start a sudden congelation, just as we may see ice needles dart in all directions when the chilled surface of a still pond is disturbed. We liken this monstrous cloud to a huge gun loaded and quiet, but with a trigger so delicately set that a falling snowflake would discharge it. The sudden puffs, gusts, and elongations of the thunder cloud may have their origin in this way. Again, there is every reason for believing that electricity plays an important part in the enlargement and subsequent history
Cumolo-Nimbus.
of this cloud. We have ourselves measured with sensitive quadrant electrometers the pull in volts experienced by the air between one of these clouds and the ground. The approach of the cloud can be foretold without seeing it and the sky mapped out roughly by the changes in the electrical potential caused by the passage of the cloud.
From what precedes it will be readily understood that cloud motion is not always a true exponent of air motion. Meteorologists know that it is not safe to obtain the motion of the air currents from the motion of the clouds, for the latter may move faster or more slowly, or even apparently stand still in the wind, as in the "table-cloth" cloud on Table Mountain at the Cape of
