occur about the time of the autumnal equinox, September 21, when the sun crosses the celestial equator to the southern hemisphere. East of the Rocky Mountains rain occurs on an average about once in three or four days, while in the North Pacific states it occurs once in every two or three days, taking the year as a whole. Throughout these large areas the latter part of September is a transition period, with autumn conditions replacing those of summer, and occasionally with the first occurrence of a storm of the winter type. The latter are usually characterized by relatively high winds, rain on two or three successive days, and followed by a considerable fall in temperature. Bearing in mind the average frequency of rainy days and of winter storms, it is apparent that it would be abnormal should no rain occur during the week preceding or the week following September 21. The so-called equinoctial storm is a fiction.
Indian Summer is another popular superstition. Characterized by high temperatures, light winds and calms, and a hazy or smoky atmosphere, it is generally supposed to be a particularly pleasant period of indefinite length occurring in October or November. That there is frequently a return of summer-like conditions during the late autumn can not be denied. But to affirm that Indian Summer is a period of several weeks in duration, recurring each autumn, and easily recognized by the occurrence of heat, calms and haze, can not be proved by climatological records. It is a peculiar fact that while the recurrence of summerlike conditions in autumn has given rise to this tradition, and even the name as a season, the similarly frequent recurrence of winterlike conditions in spring has not been popularly recognized. Summerlike periods in autumn and winterlike periods in spring can in every individual case be explained by the weather map in terms of barometric distribution, paths of storms, resulting winds and calms, the height of the sun, the length of days, and the unequal distribution of heat over the continent and the bordering oceans.
Another false notion, particularly common in rural districts, is the belief that various animals, through some particular dispensation of Nature, have a previous knowledge of coming weather changes. As a result, many proverbs have arisen, based upon observations of the behavior of animals. For example, it is sometimes stated that a cold winter is portended when the musk-rat or the beaver builds the walls of his home thicker than usual, or when the squirrels or the non-migratory birds hide large quantities of food during the autumn. Again, the remark is often made that a storm is imminent when the chickens go to roost early or when the house-cat seeks a warm place beside the fire. Even the human feeling of comfort occasionally gives rise to presentiment. Persons afflicted with recurrent rheumatism claim to feel the approach of a storm long before it appears, and people of
