Page:The American Cyclopædia (1879) Volume IX.djvu/89

 HURRICANE 81 in a circuit abont the storm centre, contrary to the direction of the motion of the hands of a watch when the latter is laid on the ground with its face upward. Furthermore, Dove made the important remark that in the hurri- canes of the southern hemisphere the air re- volves in an opposite direction ; this general- ization, announced by him, apparently with some limitations, was by the labors of Reid (1838) converted into an accepted law. The law of the rotation of winds around the storm centre is considered to be of the highest im- portance in its practical bearings on the in- terests of navigation, and may be stated in other words as follows : If in the northern (or southern) hemisphere you stand with the cen- tre of the hurricane on your left (or right) hand, the wind will be on your back. The determining cause of this law of rotation, and of the distinction between the hurricanes of the northern and southern hemispheres, was imperfectly understood by early writers, as Taylor and Herschel, but was rigidly demon- strated in a remarkable mathematical memoir by Ferrel in 1858, who showed that the rota- tion of the earth on its axis affects the direc- tion not merely of north and south winds, but of every wind, in such a manner that in the northern hemisphere winds tend as they move forward to deflect to the right hand, but in the southern hemisphere to the left hand. This ten- dency, which is known either as Poisson's or as Ferrers law, is in large storms sufficient to determine the direction of rotation, while in storms of comparatively small dimensions acci- dental circumstances may conspire to annul or even reverse the direction of rotation. Thus we are provided with the means of harmoni- zing, at least in great part, the views of Hare, Espy, and others, with those of Redfleld and Reid. There are unfortunately but few actual measurements of the velocity of the stronger winds that occur within the limits of a hurri- cane. In general it appears that the velocity increases as we proceed from the outer limits toward the centre of the storm, but suddenly diminishes to feeble irregular winds and calms within the central space. From the observed destructive force of some gusts it has also been contended that a velocity of 10 m. per min- ute must have been momentarily attained, but such computations are not very satisfactory. The highest hurricane winds that have ever been actually observed have on the British coast attained a velocity of 130 m. per hour; in the comparatively small hurricane of August, 1871, the observers in Florida of the United States army signal corps recorded a velocity of 85 m. per hour ; all these winds of course were interspersed with gusts of great violence. The diameter of the region of calms varies from 30 m. to a much smaller size, and prob- ably even to nothing. It would seem that in some hurricanes, as frequently in the smaller tornadoes on land, the so-called axis of the storm rises temporarily above the surface of the earth. The central space in general, -ac- cording to Redfleld, increases in diameter as the storm moves away from the equator north- ward or southward. A heavy rainfall extend- ing far beyond the region of most violent winds attends all hurricanes. The quantity of water that falls during the prevalence of these storms forms a large percentage of the total annual rainfall over the hurricane regions, and in this respect they perform an important service to mankind. At Mauritius in the Indian ocean a single storm has been known to be attended by a rainfall of more than 10 inches. The area of cloud and rain is especially extended on the N. and E. quadrant of the storms of the North Atlantic ; it is sometimes much contracted, though rarely wanting, on the west side of the hurricanes of both the northern and southern hemispheres. The movements of the clouds have been carefully observed, especially by Redfleld (1832-'42) and Ley (1866-'70), and the result is well expressed by Reye (1872): " While on the earth's surface the storm wind in spiral curves gradually flows inward, it forces the flying storm clouds in spiral curves outward, and removes them away from the axis of the cyclone." This generalization was fully explained from a theoretical mechanical point of view by Ferrel, and was shown by him to be a consequence of the rising or np- ward movement of the masses of air that are drawn into the whirlwind. The clouds then must move in spirals opposed to the move- ments of the lower winds. Redfleld estimates the angle between the winds below and the clouds above to be about 22'5. The baro- metric disturbance is one of the most remarka- ble features of a hurricane. The nearer one approaches the centre, the lower is the baro- metric pressure, and at the centre the depres- sion is frequently two or three inches. The first notice of an approaching hurricane, when it is yet 100 to 400 m. distant, is usually given by the steady fall of the barometer; as we approach the centre the fall is more rapid. The law by which the pressure diminishes, as well as the variations from it, may be illus- trated by two examples, the first showing a very regular depression, the second giving a great and rapidly increasing rate of fall. The first example is Redfield's Cuba hurricane of Oct. 4-7, 1844, for which we have the follow- ing pressures : at the centre, 27'7 in. ; at 100 m. distance, 28-0 in. ; at 200 m. 29-0 in. ; at 300 m., 29-5 in. ; at 400 m., 29-8 in. The second example is from Buchan (1871), and re- lates to the Bahama hurricane of October, 1866. On the evening of the 1st of October we have the following pressures : at the cen- tre, 27'7 in. ; at 15 m. distance, or the radius of the central column, 27'8 in. ; at 300 m., 29-7 in. ; at 500 m., 29-8 in. ; and at 800 m., 30-0 in. The ratio at which at a fixed station the barometer falls on the approach of a hurri- cane differs from the preceding by reason of the progressive motion of the storm toward or