and Pac. O. denotes Pacific Ocean; column 10th shows the highest pressure observed at any station at the date mentioned in column 2d; and column 11th indicates the station at which this pressure was observed. Each of these areas of low pressure appears to have moved eastward, and can be traced to the Atlantic coast. Column 12th shows the date at which the center of low pressure reached the Atlantic coast, and column 13th shows the latitude of the low center at that time.

A comparison of these cases shows that they all occurred during the six colder months of the year, and they were most numerous in January. In a majority of the cases the wind blew from the south, and in only three cases did the wind blow from any northern quarter. The greatest force of the wind in any case was thirty miles per hour; in six cases it rose as high as twenty miles per hour, and the average velocity was 8-2 miles per hour.

In forty-six per cent of the cases, the pressure at Portland was lower than at any other station at the same hour; in twentyseven per cent of the cases, the lowest pressure was at Virginia City or Fort Benton; in thirteen per cent of the cases, the lowest pressure was at Fort Sully; and in the remaining cases the lowest pressure was at some station still further east. A comparison of the observations at Virginia City with those at neighboring stations indicates that the readings of the barometer at Virginia City are too low, and accordingly they have all been increased by 0.26 inch.

With but two exceptions, all of these cases of low pressure appear to have originated north of Portland, and generally west of that station. In the table, this region is designated by the term British Columbia. It is probable that in some of these cases, and perhaps in all of them, the area of low pressure was first formed over the Pacific Ocean. No. 10 was apparently formed on the east side of the Rocky Mountains, but north of the United States, a region designated as Canada N. W. Nos. 8 and 17 were apparently formed over the Pacific Ocean, near the latitude of San Francisco.

In a majority of these cases there was an area of high barometer on the east side of Portland, at an average distance of about 1500 miles. In one case the barometer rose to 30 95 inches; in six cases the barometer rose as high as 30.75 inches; in one-third of the cases the pressure rose to 30.5 inches; and in more than two-thirds of the cases the pressure rose to 30-25 inches. In five cases (out of sixty-three) there was no station within the limits of the United States where the pressure rose as high as 30.15 inches at the dates mentioned. No. 17 is represented on Plate III accompanying the present paper, and No. 2 is represented on Plate I accompanying my fifth paper.

In each case the center of low pressure traveled eastward, and can be traced to the Atlantic coast. No. 11, being a depressed period of five days' continuance, should probably be regarded as consisting of two depressed areas, the second of which immediately succeeded the first, so that the two were united in Oregon, but traveled across the continent independently, one of them reaching the Atlantic four days later than the other. So also No. 12 apparently consisted of two depressed areas which were united in Oregon, but traveled across the continent independently, one of them two days later than the other. The average time of crossing the continent was five days, and the average latitude where the low center met the Atlantic was 45°.

The paths by which these areas of low pressure crossed the continent differed considerably from arcs of great circles. Starting from the Pacific Ocean, generally as far north as latitude 50°, the course was toward the southeast, until near the middle of the continent, and on the meridian of 100° from Greenwich the average latitude of the paths was 40°. Thence the course gradually veered northward, and upon reaching the Atlantic the average latitude of the paths was 45°.

Low barometer at San Francisco, California.

The observations made at San Francisco have been discussed in the same manner as those at Portland. The following table shows all the cases in which (during a period of twenty-six months) the barometer fell as low as 29.7 inches. The table is constructed in the same manner as that for Portland.

The number of these cases is twenty-nine, corresponding to nine different storms, and most of them occurred during the winter months. A single case is reported for the summer months, which apparently resulted from causes operating over the central portion of the North American continent. In threefourths of the cases the wind blew from some southern quarter, and in only three cases did it blow from a northern quarter. The average velocity of the wind was fifty per cent greater than at Portland, a result which may be ascribed to greater proximity to the ocean.

In more than one-third of the cases the pressure at San Francisco was the lowest reported at any station at the same hour. In eleven of the cases the greatest depression was on the east side of the Rocky Mountains, and in eight cases the point of greatest depression was situated about 1400 miles eastward. Five of these depressions appear to have originated over the Pacific Ocean; the remaining four appear to have originated north of the United States, two of them on the west side of the Rocky Mountains and two on the east side.

In nearly all of these cases there was an area of high barometer on the east side of San Francisco at an average distance of 1500 miles. In one case the barometer rose to 30.95 inches, and in two-thirds of the cases it was as high as 30.36 inches. In No. 4 a moderate depression of the barometer extended over the entire United States, with the exception of the southeast portion. This low area continued to cover a considerable part of the United States without much change during a period of fifteen days.

In each of these cases (with perhaps a single exception) the center of low pressure traveled eastward across the Rocky Mountains, and can generally be traced entirely across the continent, although in some cases the barometric wave experienced considerable modification in its progress. On account of the uncertainty attending the reduction of the mountain observations to the level of the sea, the progress of the barometric wave is best exhibited by the changes of pressure, without regard to

the absolute height of the barometer. Plate I exhibits the oscillations of the barometer for Nos. 2, 3 and 9 at San Francisco and several other stations, extending eastward to the Valley of the Mississippi, and a change of pressure of one-tenth of an inch is represented by one-tenth of an inch in the diagram. Plate II represents three other cases in which the minimum of pressure is pretty sharply defined, and the progress of the barometric wave is very distinctly indicated. These examples show conclusively that barometric waves sometimes travel from the Pacific coast across the Rocky Mountains into the Valley of the Mississippi, with so little change as to leave no doubt of their identity. It will be noticed, however, that the barometric oscillation gen. erally increases quite rapidly as soon as the wave reaches the Mississippi Valley; and in several of the diagrams palpable changes will be perceived in the form of the curves from one station to another. In several of the cases, not here represented, these changes are still more considerable.

The minimum at Salt Lake City usually occurs about sixteen hours later than at San Francisco, and at Cheyenne about one day later than at San Francisco. This indicates a velocity of forty miles per hour, which is greater than the velocity usually found for barometric waves; but it is probable that the motion of the center of low pressure was not parallel to the line joining San Francisco and Cheyenne, so that the velocity of the center of low pressure was less than forty miles per hour. It seems, then, to be clearly established that barometric waves frequently travel from the Pacific coast across the Rocky Mountains and reach the Mississippi Valley with but little modification. The Rocky Mountains form an uninterrupted barrier 6,000 feet in height from British America southward to latitude 32°, and the Sierra Nevadas present a barrier of the same height extending from British America southward to latitude 36°, with but three interruptions amounting in the aggregate to less than one hundred miles. The Rocky Mountains form a barrier of 10,000 feet in height, which extends nearly half the distance from latitude 49° to latitude 32°, and which is continuous for about 350 miles in the neighborhood of Colorado. The Sierra Nevadas also present short ranges of equal altitude, but the longest of them is less than 150 miles.

Thus we see that an unbroken mountain range of 6,000 feet in height cannot stop the progress of atmospheric waves; neither do ranges of more than 10,000 feet in height, broken as in North America, present any insuperable obstacle. A great barometric depression requires either a wind blowing with a hurricane velocity, or else a system of converging winds extending over a vast area. The mountain ranges between the Pacific Ocean and the Mississippi Valley present obstructions to the formation.