If any one of the four foregoing conditions be absent, tornadoes are not liable to occur. The reason why spring and early summer is the time when tornadoes are most frequent is because the earth and a thin stratum of air immediately next the earth are heated up rapidly with the gaining heat of the sun’s rays in the spring, while the air a short distance aloft still retains much of the cold of winter. At this time cyclonic action may bring together air masses of widely different temperatures, especially when the upper layers on the west side of the Low are drawn down and commingled with the hot and humid surface winds of the southeast quadrant.

Tornadoes Not Increasing. The writer does not indorse the theory that the number of these storms is increasing; that the breaking of the virgin soil of the prairie, the planting or the cutting away of the forests, the drainage of land surfaces by tiles, the stringing of thousands of miles of wire, or the laying of iron and steel rails have materially altered the climate or contributed to the frequency or the intensity of storms. To be sure, as population becomes more dense greater destruction will ensue with the same number of storms.

Difficult to Forecast Tornadoes. It is not possible for the forecaster to warn the exact cities and towns that will be struck by tornadoes without unduly alarming many places that will wholly escape injury. What we know is that tornadoes are almost wholly confined to the southeast quadrant of a cyclone, and that when the thermal, hygrometric, and time conditions are favorable, a region about one or two hundred miles square will be sacrificed by a number of these atmospheric twisters. One of the most destructive tornadoes of record devastated St. Louis in the afternoon of May 27, 1896. The abnormal heat and humidity of a rather small and weak cyclone centered in eastern Kansas on the morning weather map of that day, caused the Weather Bureau to distribute tornado forecasts at 10 A.M. throughout all of Missouri. The schools of St. Louis were dismissed and the children sent home on receipt of the warning, and although some eight or ten separate tornadoes touched various parts of the State and the people were prepared for their coming, so many people were terrorized by the warning in communities that were not harmed, that the writer, then Chief of the Weather Bureau, at once issued orders forbidding the specific forecasting of tornadoes in the future. Under tornadic conditions the forecast is for “conditions favorable for severe local storms.”

Freaks of the Tornado. The writer was in St. Louis the day after the storm and spent much time in examining the wreckage. He was impressed with the fact that some buildings were burst outward and that all four walls fell away from their bases, indicating that the tornado cloud must have lifted and dropped down over them in such a way that the partial vacuum that is created by the rotating cloud through centrifugal force so reduced the pressure of the air on the outside of the houses that the normal pressure of fifteen pounds per square inch exploded them. He saw bricks in a plastered wall that were neatly cleaned of all plaster by the expansion of the air inside the brick, as the air pressure from the outside was reduced. He saw a two by four pine scantling shot through five eighths of solid iron on the Eads Bridge, the pine stick protruding several feet through the iron side of the roadway, exemplifying the old principle of shooting a candle through a board. He saw a six by eight piece of timber driven four feet almost straight down into the hard compact soil, a gardener’s spade shot six inches into the tough body of a tree, a chip driven through the limb of a tree, and wheat straws forced into the body of a tree to the depth of over half an inch. Such was the fearful velocity of the wind as it gyrated about the small center of the tornado,—a velocity exceeding that of any rifle bullet. (See Figures 17, 18, 19, and 20.)

Some have advocated the planting of trees to the southwest of cities in the regions where tornadoes are frequent, so that the tornadoes may expend their energy in uprooting the trees before they come to the city, but this storm traveled through several miles of brick buildings, razing them to the ground and almost pulverizing them and still left the city apparently with greater force than it had on entering. The largest trees would offer no more resistance to a tornado cloud than would so many blades of grass.

When the official forecasts contain the statement that conditions are favorable for “severe local storms” it would be well to carefully observe the formation of portentous clouds in the west and southwest, between 3 and 6 o’clock in the afternoon, and if one with black, ragged fringes on its lower edge and accompanied with a noise like several railroad trains makes its appearance, seek safety in the cellar of a frame house.

General Rules for Forecastings. What has gone before in this chapter gives an idea of what guides the weather forecaster in making his deductions. In brief, he studies the developments and the movements of the Highs and the Lows during the past two or three days, as shown by preceding weather maps, and from the knowledge gained forecasts the future course and intensity of the fair and the foul weather areas for one, two, or three days in advance. By preserving the weather map each day and noting the movements of the Highs and the Lows, any intelligent person can make a fairly accurate forecast for himself, always remembering that the Lows, as they drift towards him, will bring warmer weather and sometimes rain or snow, and that as they pass his place of observation the Highs following in the tracks of the Lows will bring cooler and fair weather, except during periods of extreme summer heat, when the Lows bring showers that cool the parched earth; and except in the north Rocky Mountain plateau, where most of the precipitation occurs after the center of the Low has passed and northwest winds are blowing.

The amateur weather forecaster can closely anticipate the temperature of his region by remembering that the weather will be cool and the humidity low so long as the center of the predominating High (the High inclosing the greatest area within the thirty-inch isobar) is north of his latitude, either northeast or northwest, and that it will be warm so long as the High is south of the parallel of latitude that passes through his section of country.

He will find that the centers of the Lows will follow closely the direction indicated by the isotherms that lead eastward out of their centers, and that they move across the country from the west in quite regular succession, and that the frequent changes from sunshine to clouds and from warm to cold are the result of the mixing of the air by these atmospheric eddies.

Experience will teach him that Lows from the southwest are reasonably sure of causing precipitation, and that if his temperature be sufficiently low—anywhere from zero to 20°—the fall will be in the shape of snow; that Lows that only skirt our northern border will be deficient in precipitation, even if they cause any at all; that the slow settling of a High over the South Atlantic States means heat for all the rest of the country east of the Rocky Mountains in degree that will be dependent upon the magnitude and the intensity of the southern High; that the heat will continue, even if temporarily interrupted by showers, so long as this High retains its location in the southeast; that tornadoes occur in the spring of the year when Lows have excessive heat and humidity in their southeast quadrants; that V-shaped Lows cause violent local storms, if not tornadoes, and often deluges of rain; and that frosts may be expected in the country when a minimum temperature of 40° is forecast for the city; and that the severity of cold waves modifies as they come eastward, and that they will only flow as far south as the area covered by the Low that preceded them,—that is to say, by that part of the Low included in the thirty-inch isobar, or by a close approximation to such area.

National Forecaster E. H. Bowie, known to the writer as one of the ablest forecasters ever developed by the Weather Bureau, in a recent most valuable publication by the Bureau, entitled “Weather Forecasting in the United States”, formulates rules for forecasting as follows:

1. When there is an area of high pressure over the southeast and a cold wave in the northwest threatens, there will be a storm development in the southwest and precipitation will be general.

2. If a storm form in the southwest and be forced to the left of a normal track (Charts 10 and 11), another storm will immediately begin to develop in the southwest and it becomes a sure rain producer. Storms that develop in the southwest and move normally are quickly followed by clearing weather.

3. Troughs of low pressure moving from the west are of two types—the narrow and the wide. The former moves eastward slowly and storm centers develop in the extreme northern and the extreme southern ends. When the trough is wide, the development of an extensive storm area is not uncommon, especially if the wide intervening area between the Highs shows relatively high temperatures.

4. When the northern end of a trough moves eastward faster than the southern end, the weather conditions in the south and southwest remain unsettled and the chances are that a storm will form southwest of the High that follows. When the southern end moves faster than the northern end, settled weather follows.

5. Storms that start in the northwest and move southeastward do not gather great intensity until they begin to recurve to the northward. At the time of recurving they move slowly, as a rule, and care must be exercised in predicting clearing weather.

6. Marked changes in temperature in the southeast and northwest quadrants imply an increase in the storm’s intensity. Small temperature changes do not indicate a further development of the storm.

7. Abnormally high temperatures northwest of a storm indicate that it will either retrograde or remain stationary.

8. East of the Rocky Mountains, a storm which moves to the left of its normal track increases in intensity.

9. Storms with isobars closely crowded on the west and northwest generally move slowly and to the east or southeast, and the precipitation and high winds are maintained unusually long in the northern and western quadrants.

10. Storms with the isobars closely crowded in the south and southeast quadrants move rapidly northeastward and the weather quickly clears after the passage of the storm center.

Rules for Making Local Forecasts. As an illustration of what may be done by the local observer or the layman in formulating rules of weather forecasting for his immediate vicinities, the following rules, which were evolved by the writer in 1892, while serving as the Weather Bureau local forecaster for Milwaukee, Wisconsin, are subjoined:

1. In summer warmer weather occurs after the center of the Low has passed a little to the east, and southwest winds are blowing, because the easterly winds, which otherwise would be the warmest winds, are cooled by passing over the lake.

2. A Low from the northwest that reaches western Minnesota and western Iowa without precipitation or clouds will pass over Wisconsin as a dry Low, unless the isobars are closer than five eighths of an inch.

3. Light frosts will occur on clear, quiet nights in the cranberry marshes when minimum temperatures at Duluth and La Crosse fall to 40° and 45° respectively. When these stations record five degrees lower the frost will be killing in the cranberry marshes and light in the tobacco fields of the southern counties of the State.

4. No frost will occur in the counties bordering on Lake Michigan until the temperatures at the Weather Bureau stations fall close to the freezing point, such is the influence of the lake in storing up heat and slowly radiating it during the night; and on the eastern side of the lake its protecting influence is much greater.

5. When the wind sets in from points between south and southeast and the barometer falls steadily, a storm is approaching from the west or northwest, and its center will pass near or north of the observer within twelve to twenty-four hours, with wind shifting to northwest by way of south and southwest. When the wind sets in from points between east and northeast and the barometer falls steadily, a storm is approaching from the south or southwest, and its center will pass near or to the south of the observer within twelve to twenty-four hours, with wind shifting to northwest by way of north. The rapidity of the storm’s approach and its intensity will be indicated by the rate and the amount of the fall in the barometer.

Vast Extent of the Area Brought Under Observation. It is a wonderful panoramic picture of atmospheric conditions which, by the aid of the electro-magnetic telegraph and two hundred simultaneously reporting stations, is presented to the eye of the forecaster. Each day the kaleidoscope changes and a new graphic picture comes into view. Nowhere else in the world can the student of the weather find such opportunities.

Early meteorologists studied only the storm of low levels and humid airs, where convection only needed to carry the moist air currents to but a slightly higher elevation before cooling by expansion would produce condensation and an immediate acceleration of the cyclone by the liberation of latent heat within the region of the upward-moving air in its central area. They never had seen the cyclones of the arid northern Rocky Mountain plateau move down to our Great Lakes with rapidly increasing energy, notwithstanding the fact that there had been little condensation, and hence no addition of the latent heat that Espy supposed was essential to a continuation of storms.

The widely differing elevation, topography, temperature, and moisture of the broad region under observation by the United States Weather Bureau present conditions unequaled for the study of every phase of storm development and translation, or at least such as may be comprehended from data taken on the bottom of the atmospheric ocean; and it is but a matter of a short time when the data for extremely high levels will be added.

Here we see summer cyclones formed under the intense solar radiation that beats down through a nearly diathermanous atmosphere upon the wastes of the Rocky Mountain plateaus; cyclones that, if they form in the northern part of the plateau region, move eastward to our Lakes and thence eastward to the St. Lawrence with scant rainfall; cyclones that, if they have their origin farther south in the region of Colorado, move into the Ohio Valley and thence to New England with considerably more precipitation; and cyclones that, if they have their origin anywhere in our southwest States or Texas, or enter our region of observation from the South Pacific Ocean, can always be expected to cause general rainfall when they reach the Lower Mississippi Valley and later as they pass up through the central portions of the continent.

Here also one may view the great winter cyclones that originate in the Pacific between Hawaii and the Aleutian Islands and come under our vision as they successfully surmount the formidable barriers of the Rocky Mountains with but little diminution of energy, sweep across our continent with increasing force and heavy precipitation, and within three days pass beyond our meteorological horizon at the Atlantic seaboard only to be heard from several days later as boreal ravagers of Northern Europe.

The great anti-cyclones that constitute the American cold waves drift into our territory from Canadian Northwest provinces, and are studied under rapidly changing conditions during three thousand miles of their course.

West Indian hurricanes, at sea level and in humid air, which are the most violent of all storms except the American tornado, intrude themselves into the domain covered by the weather map at Florida or the East Gulf coast and usually pass off to the northeast with high winds skirting our southern coast stations.

Permanent Highs and Lows in the Pacific Are Great Centers of Action. Near the end of Chapter XII reference is made to the fact that there is a barrier in the Pacific Ocean that interferes with the movement of storms from the Orient, but which does not entirely stop their progress. Extensive Highs and Lows, sometimes called “Centers of Action” because they do not migrate like the traveling Highs and Lows that cause the alternations of weather that we experience from day to day, are also called Sub-permanent Highs and Lows. They are the parent systems out of which come many of the Highs and Lows that cross the North American continent, and they act as a bar to the free passage of storms from the Far East. As these Sub-permanent areas shift their centers a little to the north or to the south they change the character and the line of movement of the storms and cool waves that come to us, and they alter the general character of the weather for thousands of miles to the east of them. In the region of Iceland is the center of an extensive Sub-permanent Low that has much to do in controlling the weather of Europe, and there is a Sub-permanent High central at or near Bermuda in the southern part of the North Atlantic Ocean. Whenever the latter is built up by having a migrating High from the North American continent join with it, the whole United States experiences what is called a “hot wave”, and the heat continues as long as this Sub-permanent High remains unusually high and extends its western limits to include our South Atlantic States.

The matter in the foregoing paragraph is so important that it will be restated in slightly different form: Whenever either the High or the Low Center of Action (Sub-permanent High and Low), out of which comes nearly all of the migrating Highs and Lows, shifts its normal seasonal position, then storms are erratic and unusual weather occurs over the North American continent and farther eastward. The reason why much the greater number of the storms that cross the United States, the Atlantic Ocean, and Europe originate either in our Rockies, the Canadian Northwest, or just off the Alaskan coast is due to the fact (Chart 1, page 99) that the Low center of action is normally over the middle and northern Rocky Mountain plateau in summer, and over the Aleutian Islands (Chart 2, page 100) in winter. The High that follows the migrating Low in winter either separates from the center of action central over the Canadian Rockies (Chart 2), or from the one central at Honolulu; if from the latter, the weather will be simply cooler after the passage of the Low, but if the High separates from the center of action in the Canadian Rockies it will constitute a cold wave as it follows a Low southeastward into the interior of the United States and then eastward to the coast.