Thus the N. W. wind largely prevails if the number of miles is taken into The following is the percentage of winds in California, Oregon, and

account.

The mean direction of the wind in the four seasons is as follows in the same western region of North America.

Thus in summer, westerly winds very largely prevail in this region, while in winter the ratio of resultant is much smaller in California and Oregon, and easterly winds prevail further north, as shown also by the map, Plate 8.

The geographical features of the North American continent are such as to exclude a great part of it from the influence of the Pacific Ocean. The mountainchains are higher in the west than in the east, and, what is more important still, there is a very extensive plateau occupying nearly all the western half of the continent, between 34° and 42° N. L. The eastern part of this plateau, in eastern Wyoming, Colorado, and New Mexico, and in northwestern Texas slopes gradually towards the east-the valley of the Mississippi-and is thus subjected to the influence of the Gulf of Mexico. This influence is especially felt in summer,

when the heated and rarefied air of the plains draws in that of the surrounding regions.

On the west these plateaus are walled in by ranges of mountains, and the indraught of air from the Pacific slope is thus prevented.

We know that there is a depression of the barometer in summer over the plateaus of the interior, but there are yet too few observations to decide as to the region where this depression is greatest. It is, however, most probable that it is in Utah.

There is also a low region, where pressure must be low in summer, that is the valley of the Gila and lower Colorado. The heat is extreme there, Fort Yuma and vicinity having the warmest summer in America, and the ascending current must be very powerful. Air is drawn in towards this hot region, and, owing to its geographical position, principally from the south, from the Gulf of California. (See also Map of Isobars, Pl. 14, and of Winds, Pl. 8 and 11.)

The following table gives the percentage of winds of the region cast of the

The predominance of southerly winds in summer, as shown by this table, is very great, and it must be remembered that the greatest part of this region is mountainous, and thence great local discrepancies should be expected. The period of observation was short in nearly all cases. Considering this, the agreement between the different regions is very satisfactory. (See Plates 8 and 11.) In Utah there are less southerly winds in summer, and still less in Montana. But this is easily explained. As Montana lies north of 44° N. latitude where there is no extensive plateau, and the mean height of the Rocky Mountains is less than to the south—the westerly winds from the Pacific can therefore readily reach Montana. We should also expect to see southwesterly winds in winter in Montana, as in California and Oregon. This is really the case. In Arizona and New Mexico, on the contrary, the winds are much more northerly in winter than in summer. I give below the mean direction of the wind in some of the regions here considered. (See also maps, Plates 5, 6, 8, and 11).

TEMPERATE ZONE OF NORTH AMERICA, EAST OF THE ROCKY MOUNTAINS.

This region has much in common with Arizona and New Mexico, as to the mean direction and percentage of its winds. In summer a strong current from the south sets in to supply the air which is rising on the interior plateaus. In the winter, on the contrary, the prevailing winds are N. W. and the mean direction generally between N. and W. In winter the winds are more variable than in summer, and even southerly winds are sometimes experienced. The boundaries of this region are the great axis of the continent on the W., the Rio Grande on the S.W., the Gulf of Mexico on the S. E., and the Mississippi on the E. The northern boundary is rather doubtful, but yet, as far as 45° N., winds from the S. E., S., and S. W. prevail in summer. (See also maps, Plates 5, 6, 8 and 11.)

In Texas the winds have nearly the same direction as in Arizona and New Mexico, but the percentage of southerly winds in summer and northerly in winter is much greater. The winds in Texas have very strong monsoon features. This is due in a great measure to the proximity of the Gulf of Mexico. The state, except its extreme western part, is wholly open to the winds from the Gulf, and they must be strongly drawn in towards the land in summer, as the continent is much warmer than the sea. We have seen that there is a monsoon drawn in from the small and narrow Gulf of California to supply the deficiency in the interior. We must expect a much more powerful monsoon from the Gulf of Mexico. Winds in Texas, other than S. and S. E., are all but excluded from April to September. In winter the winds are more northerly, but not N. E. or E. N. E. as in the trade-wind regions of the same latitudes, but N. and N. W., i. e. winds blow from the Staked Plain and other continental areas towards the Gulf of Mexico. Yet the prevalence of these winds, if we take the number of observations only, is not so great as that of the S. E. in summer. But the N. winds are extremely violent in Texas; they are the famous northers so well known and dreaded by seamen navigating the Gulf of Mexico, and also by travellers in Texas, especially because of the suddenness of their appearance. They are especially frequent in Central Southern Texas, about San Antonio, while the north winds east of the Guadalupe River are not so sudden and violent, resembling in fact rather the northwesters of the eastern States.

The cause of the violence of these winds must be sought to the southward in eastern Mexico. This country has not as regular a climate, with small barometrical variations, as other tropical regions of the same latitude. From December to March there are frequent storm-centres, with low barometer, passing there, as also on the eastern coast of Central America. A barometrical depression in Mexico or southward of it must draw in the air from the interior of Texas and New Mexico, where the pressure is high in the winter months. In April and May, when the barometrical variations are less in Mexico, the northers are less frequent, and cease altogether from June to September during the tropical rainy season, when barometrical variation is at minimum in Mexico. To illustrate this I give the mean and extremes of the pressure of the air at Vera Cruz.1 (See also Plate 14.)

1

From the observations by Dr. Berendt, manuscript collection of the Smithsonian Institution.

In the extreme south of Texas, at the mouth and in the valley of the Rio Grande, the S. E. winds are much more frequent, even in winter, than in the rest of the State. This is an intermediate region, partaking of some of the features of the Mexican climate, where easterly winds prevail the whole year. Yet the lower Rio Grande region is subject to violent northers. This seems to lead to the conclusion that in the other regions of Texas, where northerly winds prevail in winter, they are not all northers, there being also north winds of moderate force blowing towards the Gulf.

The mean direction of the winds in the different portions of Texas, is as follows:

The summer, as is shown by these tables, and the maps Plates 8 and 11, is the season in which the wind is most constant, the mean direction at all stations being between S. 7° E., and S. 46° E., and the ratio of the resultant very great, except in Western Texas. In the three last regions, nearest to the Gulf, the direction is more S. E., while in the more northern part of the State it is rather S. or S. S. E. The influence of the earth's rotation is here clearly seen. The wind begins as S. E., but soon is deflected to the south, and in its further course passes to the W. of S.

The agreement is not as exact in winter, probably because we have only the number of observations, and not the force of the wind. As the N. and N. W. winds are known to be the strongest, the mean direction would be much nearer each other in the different parts of the State, if we knew the force of the winds. Yet in all cases it would be seen to be more easterly on the lower Rio Grande near the Mexican frontier.

Spring and autumn are transition seasons, and in a country with monsoon winds, as Texas, there is very little to say about them. Generally spring is more analogous to summer, and autumn to winter. (See Plate 8.)

I must further remark as to the S. E. winds of the summer, that it would be an error to consider them merely as sea-winds blowing only during the day. They are stronger in the afternoon, while about sunset there is generally a calm. But about 9 P. M. the S. E. springs up again and blows till morning, when there is a second calm. I had occasion to observe this, in the summer of 1873, in the country between the Nueces and Guadalupe, and old residents of San Antonio informed me this was the regular course. (See the figures showing the number of observations and the force of the wind at 7 A. M., 2 P. M., and 9 P. M., at the last

direction.

Mean

Autumn.

Ratio of resultant.

Mean

direction.

Winter.

Ratio of

resultant.