that of the surrounding country, while partially true, is often exaggerated in the public mind. According to Professor J. Hann, unquestionably the leading authority on climate, city temperatures differ from those of the open country nearby in the following respects: The mean annual temperature of the air in places where there are many buildings from 1° to 2° F. too high, the differences being greatest in the morning and evening, and least at noon. The diurnal range is smaller in cities, especially in summer. The cooling by radiation, at night, is much greater in the open than in places which are built up. The cooling due to evaporation probably also plays a part. While it has been calculated that the burning of gas and coal in London develops sufficient heat to have an appreciable effect upon the temperature of the air in a stratum 100 feet thick over that city, no progressive increase in the mean temperatures of New York City and Boston can be found to form a parallel with the growth of those cities. The absolute winter minima are much less marked in the interior of cities than in the surrounding open country. A study of certain cold waves showed that the absolute minimum temperatures recorded in the cities of Toledo, Cleveland, Columbus, and Cincinnati, Ohio, were 20° to 25° F. higher than those noted in the country surrounding these cities. From this the investigator concluded that it would be well to put weather stations near, rather than in, large cities, and at a sufficient distance from them to be free from purely local conditions. It should be added that the temperature felt in the city, under the influence of the radiation from the heated walls of buildings and the reflection from the bare ground, is very different from that felt in the country. Other meteorological elements which show difference between city and country are sunshine, which, on account of smoke, is somewhat less in cities than in the surrounding country, and wind velocity. Every large city has one or more tall buildings about which the wind blows with frequent and violent gusts, even on comparatively calm days. As there is everywhere a rapid increase in wind velocity with height, the taller buildings tend to bring down the higher velocities from aloft. It is thus apparent that there is some degree of difference between the climate of city and country, but when due allowance is made for actual and sensible differences, the effect of the local control upon climate is seen to be small.
Concerning the course followed by a thunderstorm, there are many and varied misconceptions. It is often remarked that a thunderstorm, upon coming to a river valley or a mountain gorge, will divide into two parts, one moving up and the other down the valley, in other words, that thunderstorms tend to follow valleys. Another statement is to the effect that the center or most severe part of the storm passes not over the point of observation, but at some distance away. Instrumental observations fail to verify these and similar generalizations.