As regards longitudinal distribution, an unusually large proportion of storms prevailed between the 50th meridian and 105th meridian, west; 37 per cent. or one-third of all the storms of the Northern Hemisphere occurring within this region. A second belt of comparative storm frequency obtains from the meridian of Greenwich eastward to the 30th meridian; over which region 15 per cent. of the entire number of storms occurred.
Only four hundred, or less than 9 per cent. of the entire number of storms, entered the American continent from the Pacific ocean, while about thirteen hundred storms, excluding the West India hurricanes, passed eastward off of the American continent. Over nine hundred storms entered Europe from the Atlantic ocean, of which probably four hundred and fifty, or ten per cent. of the whole number recorded, were developed over the Atlantic ocean. Probably not thirty storms, or less than three per cent. of those which entered Europe from the Atlantic, crossed over the continents of Europe and Asia to the Pacific ocean. Fully two-thirds of the storms which enter Europe from the Atlantic are dissipated as active storm-centres before they reach the Asiatic frontier.
The tendency of great bodies of water, when surrounded wholly or largely by land, to generate storms or facilitate their development, is evident from the unusual prevalence of storms over the great lakes, the St. Lawrence bay and the Gulf of Mexico in North America; over the North and Baltic seas, Bay of Biscay and the Mediterranean in Europe; the Bay of Bengal, and over the China and Okhotsk seas.
Undoubtedly a considerable proportion of these storms are drawn towards these regions owing to the effect of evaporation upon the humidity and temperature of the superincumbent atmosphere, so that a very considerable proportion of the storms credited to these squares have not originated therein, but have been drawn up from neighboring quarters. This tendency is marked in North America, as storms pass over the lake region and St. Lawrence valley, whether they have originated in the Gulf of Mexico, along the central slope of the Rocky mountains in the United States, or further north in the Saskatchewan country. In like manner storms pass southeastward to the Mediterranean from the Bay of Biscay, and northeastward from the Atlantic ocean to the same sea, and then later show a very marked tendency to pass over the Black and Caspian seas.
This tendency of storms originating in diverse sections to move toward the lake regions in the United States, is very evident from the normal storm-track charts for April, May, June, August, November and December.
The opinion that gales rarely, if ever, occur upon the equator is confirmed by these storm-tracks. The most southern storm in the North Pacific ocean, developed in July, 1880, between the Island of Borneo and Mindanao, an excellent account of which is given by Père Mark Dechevrens, S. J., in the Bulletin Mensuelle of Zi-Ka-Wei Observatory. The most southern storm over the North Atlantic ocean, in November, 1878, was remarkable for its origin, duration, length of its path, and its enormous destruction of life and property. It was central on the 1st, as a violent tropical hurricane near Trinidad, the barometer being 29.05, the lowest ever recorded there, and, from its intensity and velocity, it is more than probable that it originated considerably to the eastward, and possibly somewhat to the southward of that island. The storm was described in the U. S. Monthly Weather Review for September, 1878.
The writer looks with considerable interest to the results which may follow from a discussion of the annual fluctuation of the atmospheric pressure as shown by the mean monthly pressures deduced from the ten years' International observations. As far as these means have been examined they show that the periodicity of atmospheric pressure is largely in accord with the results set forth in 1885 in The Report of the Lady Franklin Bay Expedition. The conviction expressed in that year is still adhered to—that, at no distant day, the general laws of atmospheric changes will be formulated, and that later, from abnormal barometric departures in remote regions may be predicted the general character of seasons in countries favorably located.
The success of long-time predictions of this class for India, has been set forth in a previous part of this report. It is believed that a further discussion of meteorological phenomena on a broad basis, by means of International Weather Charts, both in daily and monthly form, must eventually result in important and fundamental discoveries. It is gratifying to American pride to know that in this international task of outlining the geography of the air, the United States has liberally provided the labor and means for presenting these ten years' meteorological data in such tabular and geographical forms as to render them available for study by all.
Acknowledgment is due to Professor Thomas Russell, for valuable translations, especially from the German; which translations have been of material value in preparing this report.
December, 1888.