In the study of the relation of clouds to rainfall are taken up: clouds preceding rain, clouds between intervals of rain and clouds following rain. The methods and cause of cloud formation were also carefully considered. The most important part of this special investigation is the use to which the cloud observations are put in the study of atmospheric dynamics, taking up in succession the questions: the relation of clouds to cyclones and anti-cyclones, having regard to the altitudes of the cloud levels; the annual and diurnal periods in the winds in general, at various levels as shown by direct observation near the ground and extended upwards to high altitudes by means of the observed cloud movements; the wind movements in cyclones and anti-cyclones from the ground up to the altitude of the highest clouds; the relation of the direction of the cirrus clouds to the existing temperature gradient; the relation between the velocity of storms, and the consequent variability of the weather, to the general movement of the atmosphere as shown by surface wind and cloud observations; the use of cloud observations in weather forecasts; and the frequency of winds from various directions at different heights above the ground, for different hours of the day, shown by wind and cloud observations.
The work of making observations of the atmospheric conditions in the free air by means of kites has been carried out with the success achieved only by the persistent endeavors of the observatory staff, not only in overcoming the difficulties in the mechanical construction of the apparatus employed, but also in the actual work of kite flying.
Experiments were undertaken as to the best forms of kites to use, the best materials for their construction, and the best lines to use for flying them. Special forms of self-recording meteorological instruments had to be so designed or so changed as to be adapted to the demands of kite work. Great care was exercised in so exposing the instruments that their possible errors would be reduced to a minimum. During the year 1897 there were thirty-eight successful kite flights, in 1898 thirty-five, in 1899 twenty-five, and in 1900 twenty-four; the average height above sea level at which records were obtained during the respective years being 7,350 feet, 7,400 feet and 8,450 feet, thus showing constant improvement in the methods employed.
The discussion of the Blue Hill observations has added very materially to our still meager knowledge of the distribution of the meteorological elements in the free air, and their variation with change in altitude. The average increase of wind velocities with increasing altitude was determined chiefly for those altitudes for which we have the fewest data because it is so difficult to make measurements there by means of the clouds. The change in direction of air currents at different levels was also clearly and accurately brought out by the changes in position of the kites as they ascended and descended. Such data as these are
