and point of recurvature will be determined by the position of the Bermuda Hyperbar, that is, the seasonal anticyclone of the Atlantic. This accounts for the swinging east and north of these tracks as the season progresses; for the hyperbar is slowly displaced east, the maximum displacement occurring in September.

Fig. 8. All Storms Lead to New England

Individual anticyclones also influence individual hurricanes. Thus a hurricane passing west over Havana, will go farther west if a vigorous "high" is spreading southeast over the Gulf States. And when this "high" passes seaward, the hurricane will work around the southwest quadrant of the "high," recurving and moving northeast.

STORM RENDEZVOUS

Altho storms originate or are first detected in nine different sections, it is a fact worth mentioning that they all leave the United States in the vicinity of New England or Nova Scotia. Some of the southern depressions starting near the coast, pass to sea south of New York, but in general an observer standing on Plymouth Rock can virtually encompass within a radius of 500 kilometres, 300 miles, the paths of ninety per cent of the storms that traverse the country.

Thus a storm that originates in Texas (7) will probably pass close to Cape Cod. Likewise, types (3) and (5); while the other types may pass a little to the north or south. See Chart, Paths of Storms.

STORM PATHS

Forecasting then would seem to be very easy; for one would only have to know the place of origin of the storm and the rate of travel, to foretell exactly the time of arrival. Unfortunately these are only the average paths; and as with most mean values, represent a value not often experienced in fact. These paths then are not paths which any given storm will follow. One must recall the story of the operating surgeon who gave the average age of his patients in the operating room as 35. There were but two patients, one 69 years old and the other 1 year old.

Fig. 9. Zephyros—The West Wind

As a matter of fact the path of any individual depression depends upon several factors, some of which are:—the prevailing eastward drift of the air; the extent and motion of some anticyclone advancing before the "LOW"; the duration and speed of relatively dry cold tongues of air from the north; and the supply of water vapor brought from southern waters by south winds. A depression can make little headway if to the north or east the normal path is blocked by what is known as a stagnant "HIGH." So therefore, if the anticyclone is a slow mover, a Texas storm, which would normally pass not far from southern New England, may be deflected farther north than when the HIGH moved rapidly east. So too, with the storms which originate in the western part of the country. A slow moving HIGH will prevent the LOW following it, from moving east at a normal rate along the usual path.

Anticyclones then, are the real weather controls. There are various types, but all drift from the north or west. Occasionally they enter the country from the Pacific, but the great majority come from Alberta and move leisurely southeast, often reaching the South Atlantic States; but more frequently recurving and passing to the north.

STAGNANT HIGHS

HIGHS are sometimes reinforced and this results in what is called a stagnant HIGH. A good illustration of such a slow moving HIGH and its consequences occurred during the last week of January, 1922.

A surge of cold air from Alberta or farther north reached the international boundary January 21st and spread slowly eastward, reaching the Great Lakes on the 24th and the St. Lawrence Valley two days later. Then seemingly it halted or moved slowly westward, retrograding. In three days, that is, on the 29th, the centre of the HIGH was apparently 500 miles west of where it had been on the 27th. After the 29th it followed a normal track, moving slowly southeast, reaching the Atlantic near Long Island.

Meanwhile a depression on the south coast of Texas on the 25th, moved across the Gulf of Mexico, passing over Southern Florida on the 27th and advanced steadily northeast, reaching Cape Hatteras in 24 hours. Owing to the presence of the anticyclone referred to above, the depression recurved off Hatteras. The result was a memorable snow storm in Northern Virginia and Maryland. At 8 p.m. January 27th, there had been a fall of 5 cms. (2 inches). Within the following twenty hours the average depth in the city of Washington was 66 cms. (26 inches). The weight of the snow caused the collapse of the roof of the Knickerbocker Theatre and the death of 97 persons.

The total snowfall in various coast cities was:

The table shows clearly how the snow was formed. On the east side of the LOW a stream of air, relatively warm, carried a load of water vapor, approximately 13 grams in each cubic metre.

Fig. 10. Paths of High and Low, Great Snow Storm of January 27-28, 1922]

This current was steered around the north side of the LOW and met the north-northeast wind. Under the new conditions the air saturated could hold only 2 or 3 grams; and so condensation and heavy precipitation resulted. The region of maximum snowfall was near Washington, and it will be seen that there is a proportional decrease north and south. The snowfall at Washington was the heaviest ever known at that city.

Unlike most storms, there was no strong cold northwest wind blowing into the depression. The temperature rose slowly. It was less a contrast of winds than a steady slow outward push of the anticyclone, and the consequent turning of the path of the cyclone eastward.

LAWS OF FORECASTING

Buys Ballot's Law.

"If you stand with your