above given, with that of portions of Texas lying within the same parallels of latitude. At Fort Brown, near the mouth of the Rio Grande, and where the summer heats are much modified by the strong coast-breezes, the mean annual range is 67°, or from 95° in summer to 28° in winter. In the interior of Texas, the range is very great, the summer heats rising to a mean of 112° in the valley of the Rio Grande at Laredo (Fort McIntosh), and to 103° at the posts on the upper plains. It falls to a mean of 20° in winter on the Rio Grande valley, and to 10° on the plains, making the range near 92° in each case. At Fort Smith, Arkansas, the range is 108° having the summer temperature of 103°, and the winter of -5°."

The above extract was written in October, 1855, and may be taken as a fair comparison of the temperature of Florida with that of other sections of the United States.

The table of daily mean range of temperature for each month, for the year begining October, 1873, and ending September, 1874, calculated from the daily bulletins of the Signal Office, show it to be comparatively small. For the seasons, the greatest range is in the winter and spring. The daily range diminishes as the peninsula is descended, and at Key West there is but little difference for all seasons, while the daily range is small.

Another element in a climatological consideration is humidity of the atmosphere, and the amount of rain-fall. As to the rain-fall, its mode of precipitation—whether rapidly or slowly-is also important as materially affecting the state of the weather. Regarding the humidity of the atmosphere, the amount of data I possess-only for one year (see table)-is too insignificant, and the positions of the few signal stations too unlike and peculiar individually, to allow of any attempt at generalization. Yet their general tendency is to correct an erroneous impression entertained even by some physicians in the northern parts of the State, and by the public generally, to wit, that there is greater humidity for the winter the lower the peninsula is descended. The table of mean monthly percentage of relative humidity gives pretty nearly the same for all four of the signal stations. But it must be remembered, as previously intimated, that that for Punta Rassa-from its littoral position-cannot be considered as fairly representing the interior of the peninsula, from the greater humidity of the atmosphere always present on the While there is no great variation from the annual mean for any season, yet spring shows the least percentage. The rela

coast.

tive percentage of humidity is, however, not in excess of that of the Atlantic States, nor of California during winter as observed at San Francisco and San Diego, so far as a cursory examination enables me to judge. The saline impregnation of this moisture, arising from the Atlantic and Gulf, doubtless imparts to it also antiseptic and salubrious properties.

The rain-fall in Florida is not characterized by uniformity as to amount for different years and same seasons, nor as regards sections and localities. This is particularly noted on examining the tables for separate years of the meteorological register embodied in the Surgeon General's reports, published in 1856, 1860. This variation from the mean is greater in excess than deficiency. For instance, the mean at Tampa is 55 inches. In 1840 the rain-fall at this place was 89 inches, and in 1854, 69 inches, yet the minimum rain-fall has never been below 40 inches during the same period of observation. Again, at Pensacola in 1855 it amounted to 77 inches. The exposed position of St. Augustine immediately on the Atlantic, accounts to some extent for the smaller average rain-fall at that place as compared with other points in the State. The same applies to Punta Rassa as a signal station, though hardly with equal force, as the latter is not quite so exposed to the open The well-known fact that there is less rain on the coast than in the interior is a sufficient explanation. The rain is not equally distributed through the year, but is so excessive in the summer as to specially denominate that the rainy season. According to Loomis' meteorology, Alabama, Louisiana, and Oregon exceed Florida in the annual amount of rain-fall; while Virginia, the Carolinas, Tennessee, and Kentucky equal her. By this author, the annual mean rain-fall for the whole State is put at 48 inches, which is in conformity with observations so far made. As to the distribution of rain throughout the year the same author says:

sea.

"Throughout most of the United States east of the Rocky Moun tains, the rain is pretty equally distributed through the different months of the year, but the rain of summer is everywhere some. what greater than the rain of winter, including the melted snow. In New England, the difference between the rain for these two seasons is less than 10 per cent.; in the State of New York it is nearly 50 per cent.; in Virginia and the Carolinas it is 100 per cent.; in Florida it is 200 per cent.; in Texas it is 75 per cent.; in Ohio it is 25 per cent.; in Michigan and Wisconsin it is 140 per cent.; while in Iowa and Kansas it is 300 per cent.; that is, the fall of

rain in summer is four times as great as it is in winter. On the Pacific coast, this law is reversed. In California, the rain of winter is more than twenty times as great as that of summer, and in Oregon, it is seven times as great."

From this estimate it is seen that the summer rain in Florida is three times more than that of winter. This, taken in connection with the mean annual rain-fall of the Atlantic Gulf States and the winter rain-fall of the Pacific States, demonstrates the winter climate of Florida to be a comparatively dry one, especially that of the peninsula. As nearly all the posts where meteorological ob servations were made are on the peninsula, it is presumable that this estimate more fully represents that section than it does the northern portions of the State. It is only the peninsular portion of the State that has a climate of its own markedly different and distinct from that of the northern portions of the State, which possess no distinctive features from similar sections, contiguous to the coast, of the other Gulf States. The winter climate of the peninsula is dry, and this dryness becomes more marked as the peninsula is descended. Blodget, in his Climatology of the United States, says:

"By reference to the general tables it will be seen that there is a marked tendency toward the development of a winter dry season even in the States next to Florida, and that in southern Florida. this is quite decided and almost as fully developed as in the recog nized tropical climates. As before remarked, however, there are great irregularities, and often quite contradictory results for single years. . . . The most prominent feature of rain distribution in Florida, in distinction from other parts of the United States, is this great irregularity, which prevents us from getting clear general views from periods embracing but two or three, or a few years. . . . . The first approach of this dry season is made in October, and from the minimum, which appears to occur in November, there is a partial resumption of the rains apparent in midwinter, to be followed by other months of less rain. But as a whole, the winter from October forward is a dry season on the peninsula of Florida. It appears to be a climate ordinarily of a division into two principal seasons in regard to the rains, the wet summer and the dry winter, yet either may be interrupted by the extremes of an opposite character much greater than those occurring in any other known district. .

"Beyond the plains the characteristic features of the districts of VOL. XXVI.-27

periodical rains are strikingly exhibited, and Southern Florida gives strong evidence that it should be identified with the districts of periodical rains rather than with the area of constant precipitation of the Eastern United States. . . . . From the Gulf coast northward there is also a decided increase of annual precipitation at some points, and for a moderate distance; as also from the coast of St. Augustine, Florida, northwestward towards the interior of Georgia. . . . The south of Florida alone gives as soft a climate for the winter as that of the south coasts of the Mediterranean, and at a point far enough south to do so, the tropical features of a dry winter and rainy summer become instituted. At Tampa the average temperatures are nearly those of Cairo, Egypt, and the difference of latitude two degrees."

The winter dry season in the Florida peninsula is generally extended into the latter part of May or first of June. Another distinctive feature of the peninsula climate in connection with the rains of winter—and this I wish particularly to emphasize as being the case on the Gulf side-is that they generally assume the character of showers, and are not diffused over several consecutive days, in a light but constant rain, as is the case in the northern sections of the State, and in the Southern Atlantic and Gulf States generally. As a natural consequence the number of wholly cloudy days in the peninsula is much less also. Fogs are also comparatively rare, especially on the Gulf side of the peninsula, as compared with other sections of the State. The absence of fogs becomes more marked during all seasons as the peninsula is descended.

The frequency of dense and persistent fogs on the Atlantic and along the St. John's is doubtless dependent on the reduction of the atmospheric temperature on that side by the shoaling of the Arctic under current. The U. S. coast surveys, as well as the deep sea soundings of a late date of H. B. M. ship Challenger, have conclusively determined the returning deep Arctic current, the counter-movement of the Gulf Stream, obtaining in its course towards the equator a westward tendency from the rotation with increased diameter of the earth, shoals up on the eastern coast of Florida. The result must necessarily be to lower the temperature of the superjacent atmosphere, and the latter coming in contact with the excessive humidity naturally afforded by a low level section of country, with its abundance of still and sluggish waters, could hardly fail in producing fogs and rain. This also accounts,

no doubt, for the number of rainy days on the eastern side of the peninsula being much larger than on the Gulf side, as is clearly shown by the registry of the weather embodied in the Surgeon General's Report for 1860. Besides, the more elevated and rather hilly nature of the middle and western slope of the peninsula, rendering the soil naturally drier, may also have some effect in rendering the climate of these sections less subject to excessive humidity.

The table of winds shows how many days the wind was from one of the cardinal points of the compass, and its monthly mean rate of velocity for one year. The winds are generally very changeable, except very low down on the peninsula, which is slightly within the belt of the trade-winds. The upper two thirds of the peninsula constitute the calm belt, or that which has no regular movement with either the westerly or trade-winds.

By reference to the table it will be seen that the range of barometrical pressure, at the four signal stations, for two years, is very inconsiderable; and that there is but little difference in the yearly mean for the several stations. September has a low pressure for Punta Rassa and Key West, while May has the lowest pressure at all the different stations. Storms and such like are not more common in Florida than in the other Southern States. Most of the storms originating in the Gulf follow the Gulf stream, and consequently their outer circles only reach the peninsula, seldom amounting to much.

Lightning with thunder is quite common in the summer season. The variety known as zigzag is the most common kind; very little ball or sheet lightning; while heat lightning in the early part of the night may be observed for a short time preceding the commencement of the summer rains. As to the electrical tension of the atmosphere I have no observations to offer.

So far as my observations extend, tests for ozone give positive results for all seasons.

When the climate of Florida is compared with that of any of the United States its superiority becomes apparent. As to dryness of winter climate the peninsula compares favorably with California, the rain fall for this season being about the same. It is only the winter isothermal of 55° which traverses Northern Florida and Lower California. The winter isothermal of 65° crossing the peninsula does not even appear in Texas on the lower Rio Grande. If compared to the drier ones of Arizona, New Mexico, and the lower