of ascertaining the situation of a ship in a whirlwind, from the strength and changes of the wind, with the view, no doubt, of enabling the vessel to resist its fury, and escape from its vortex.

These observations, valuable though they be, seem to have excited no interest either in this or in other countries; and the next philosopher who directed his attention to the subject was led to it by independent observations, and in the course of more extensive meteorological inquiries. Mr. W. C. Redfield, of New-York, whose position on the Atlantic coast gave him the finest opportunities not only of observing the phenomena, but of collecting the details of individual storms, was led to the same conclusion as Colonel Capper, that the hurricanes of the West Indies, like those of the East, were great whirlwinds. He found also, what had been merely hinted at by Colonel Capper, that the whole of the revolving mass of atmosphere advanced with a progressive motion from south-west to north-east; and hence he draws the conclusion, that the direction of the wind at a particular place forms no part of the essential character of the storm, and is in all cases compounded of both the rotative and progressive velocities of the storm in the mean ratio of these velocities. Mr. Redfield was conducted to these generalizations by the study of the hurricane of September, 1821; but, in order to corroborate his views, he has taken the more recent hurricane of the 17th August, 1830, and, by the aid of a chart, he has exhibited its character, and traced its path along the Atlantic coast, as deduced from a diligent collation of accounts from more than seventy different localities.

Interesting as these details are, our limits will only permit us to give a few of the leading facts, along with the results at which Mr. Redfield has arrived. The hurricane of 1830 seems to have commenced at St. Thomas on the 12th of August at midnight; and, continuing its course along the Bahama Islands and the coast of Florida, it passed along the American shores and terminated its devastations to the south of the island of St. Pierre, in long. 57 deg. west, and lat. 43 deg. north. It performed this long journey in about six days, at the average rate of about seventeen geographical miles per hour. The general width of the tract, which was more or less influenced by the hurricane, was from 500 to 600 miles; but the width of the tract where the hurricane was severe was only from 150 to 250 miles. The duration of the most violent portion of the storm at the several points over which it passed, was from seven to twelve hours, and the rate of its progress from the island of St. Thomas to its termination beyond the coast of Nova Scotia, varied from fifteen to twenty miles per hour.

The rotative character of this storm, which always moves from right to left, is amply proved by the varying directions of the wind at the different points of its path; but a striking evidence of this was exhibited in its action on two outward bound European ships, the Illinois and the Britannia. On the 15th August the Illinois experienced the swell which preceded the hurricane advancing from the south; but as the ship had a fair wind and was impelled by the Gulf Stream, while the storm lost time by making a detour toward Charleston and the coast of Georgia, the ship outran the swell; but on the 17th she was overtaken by the hurricane blowing furiously

from the south, while at the same moment it was blowing hard at New-York from the north-east. The Britannia, which left NewYork in fine weather on the 16th, met the hurricane on the same night, having the wind first at north-east, then ENE., and after midnight from the south-east.

After describing other hurricanes which led him to the same conclusions, Mr. Redfield remarks that their axis of revolution, or gyral aris, as he calls it, is probably inclined in the direction of its progress. This inclination he ascribes to the retardation of the lower part of the revolving mass by the resistance of the surface; in consequence of which the more elevated parts will be inclined forward, and overrun to a very considerable extent the more quiet atmosphere which lies near the surface. Hence we see the reason why vessels at sea sometimes encounter the sudden violence of these winds upon their lofty sails and spars, when all upon the deck is quiet

One of the most important deductions which Mr. Redfield has made from the facts and illustrations to which we have referred, is an explanation of the causes which produce a fall in the barometer at places to which a hurricane is approaching, or more immediately under its influence. This effect he ascribes to the centrifugal tendency of the immense revolving mass of atmosphere which constitutes a storm. This centrifugal action must expand and spread out the stratum of atmosphere subject to its influence; and toward the vortex or centre of rotation must flatten and depress the stratum so as to diminish the weight of the superincumbent column which presses on the mercury in the barometer.* Mr. Redfield also conceives that whatever be the upward limit of the revolving mass, the effect of its depression must be to lower the cold stratum of the upper atmosphere, particularly toward the more central portions of the storm; and by thus bringing it in contact with the humid stratum of the surface, to produce a permanent and continuous stratum of clouds, with an abundant precipitation of rain, or a deposition of "congelated" vapors, according to the state of temperature in the lower region.

From these views Mr. Redfield is led to speculate on the cause of the hurricanes which prevail on the Atlantic coast. He conceives that they "originate in detached and gyrating portions of the northern margin of the trade winds, occasioned by the oblique obstruction which is opposed by the islands to the direct progress of this part of the trades, or to the falling in of the northerly or eddy wind from the American coast upon the trades, or to both these causes combined."

Such is a brief analysis of the first and most important memoir of Mr. Redfield. The second paper contains a very short notice of the hurricane which, after raging with great violence at Barbadoes on the night of the 10th August, 1831, passed over St. Lucia, St. Domingo, and Cuba, and reached the northern shores of the Gulf of Mexico, in about 30 deg. of north lat., where it raged simul

* Hence we see the reason why the mercury in the barometer always rises again during the passage of the last portion of the gale, and reaches its greatest elevation after the storm has passed.

taneously at Pensacola, Mobile, and New-Orleans. Here it entered upon the territories of the adjoining states, where it must have encountered the mountain region of the Alleghanies, and was perhaps disorganized by the resistance which these elevations offered to its progress. It seems, however, to have caused heavy rains over a large extent of country to the north of the Gulf of Mexico; and if its peculiar action was continued beyond New-Orleans, it must have been confined to the higher atmosphere, as no violent effects were produced at the surface nearer than the southern states. This hurricane, which revolved from right to left, passed over a distance of 2000 nautical miles in about 150 hours, which gives an average velocity of more than 13 1-2 miles an hour. The rotative character of this storm was finely exemplified in the effects which it produced at Barbadoes. The trees which it uprooted near the northern coast lay from NNW. to SSE., having been thrown down by a northerly wind in the earlier part of the storm, while in the interior and some other parts of the island they were found to lie from south to north, having been prostrated in the latter part of the gale.

In his third memoir, Mr. Redfield directs our attention to the different points which he considers as established in reference to the principal movements of the atmosphere which constitute a hurricane. The following is a condensed summary of his observations:

1. The severest hurricanes originate in tropical latitudes to the north or east of the West India Islands.

2. They cover simultaneously an extent of surface from 100 to 500 miles in diameter, acting with diminished violence toward the exterior, and increased energy toward the interior of that space.

3. South of the parallel of 30 deg. these storms pursue toward the west a track inclined gradually to the north till it approaches 30 deg., where their course changes abruptly to the north and eastward, their track continuing to incline gradually to the east, toward which point they advance with an accelerated velocity.

4. The duration of a storm depends on its extent and velocity, and storms of smaller extent advance with greater rapidity than larger ones.

5. The direction and strength of the wind in a hurricane [for the most part] are found not to be in the direction of its progress.

6. In their westward course, the direction of the wind at the commencement is from a northern quarter, and during the latter part of the gale from a southern quarter of the horizon.

7. In their northward and eastward course, the hurricane begins with the wind from an eastern or southern quarter, and terminates with the wind from a western quarter.

8. North of 30 deg., and on the portion of the track farthest from the American coast, the hurricane begins with a southerly wind, which, as the storm comes over, veers gradually to the westward, where it terminates.

9. Along the central portion of the track in the same latitude the wind commences from a point near to south-east, but after a certain period changes suddenly to a point almost directly opposite to that from which it had been blowing; from which opposite quarter it blows with equal violence till the storm has passed. Under this

central portion the greatest fall of the barometer takes place, the mercury rising a short time previous to the change of wind.

10. On the portion of the track nearest the American coast, or farthest inland, if the storm reaches the land, the wind begins from a more eastern or north-eastern point, and afterward veers more or less gradually by north to a north-western or westerly quarter, where it terminates.

11. From these facts, it follows that the great body of the storm whirls in a horizontal circuit around a vertical or somewhat inclined axis of rotation, which is carried onward with the storm, and that the direction of this rotation is from right to left.

12. The barometer in all latitudes sinks under the first half of the storm in every part of its track, except, perhaps, its northern margin, and thus affords the earliest and surest indication of the approaching tempest. The barometer again rises during the passage of the last portion of the gale.

Our readers will naturally inquire, What are the phenomena which take place within the vortex, or in the axis of the revolving storm? It is well known that in the heart of a storm or hurricane in the open sea, violent flaws or gusts of wind alternate with lulls and remissions of its violence; and here Mr. Redfield conceives that the vortex or rotative axis of a violent gale or hurricane oscillates in its course with considerable rapidity in a moving circuit of moderate extent near the centre of the hurricane; and he conjectures that such an eccentric movement of the vortex may be essential to the continued activity or force of the hurricane.

The fourth and last memoir of Mr. Redfield has for its object the illustration of his preceding labors, by delineating on a chart the route of several storms and hurricanes, as derived from numerous accounts of them in his possession, by which their progress is specifically identified from day to day during that part of their route which appears on the chart. The following is a list of the storms thus projected :

1. The hurricane which visited Trinidad, Tobago, and Grenada, on the 23d of June, 1831.

2. The hurricane of the 10th August, 1831, already referred to. 3. The hurricane which passed over the Westward Islands on the 17th August, 1827, and terminated about Sable Island and Porpoise Bank on the 27th; having traveled over 3000 nautical miles in about eleven days, at the average rate of about eleven miles an hour.

4.* The hurricane which swept over the Windward Islands on the 3d September, 1804, the Virgin Islands on the 4th, Turk's Island on the 5th, the Bahamas on the 6th, the coast of Georgia and the Carolinas on the 7th, Chesapeake and Delaware Bays, and Virginia, Maryland, and New-Jersey on the 8th, and the states of Massachusetts, New-Hampshire, and Maine on the 9th; becoming a violent snow storm in the high lands of New-Hampshire. It performed a journey of 2200 miles in about six days, at the average rate of about fifteen and a half miles per hour.

* This is, by an oversight of Mr. Redfield, described as No. V. in the text, while No. V. is described as No. IV. We follow the chart.

5. The hurricane which ravaged Antigua, Nevis, and St. Kitts on the night of August 12th, 1835, and reached Metamora on the coast of Mexico on the 18th, after passing over St. Thomas, St. Domingo, and Cuba. Its velocity was fifteen and a half miles per hour, having moved through 2200 miles in six days.

6. This is the memorable gale of the 12th August, 1830, already referred to. It visited St. Thomas on the 12th, and reached the Porpoise and Newfoundland banks on the 19th, having traveled through more than 3000 nautical miles, with an average rate of eighteen miles an hour.

7. This hurricane, which swept over the Atlantic in 1830, was encountered to the north of the West India Islands. It passed along a more eastern route than any of the rest, and reached the Grand Bank of Newfoundland on the 2d of October, after having caused great damage and destruction to the many vessels which occupied its widely extended track. The length of its route is about 1800 miles, and its average velocity twenty-five miles per hour.

8. Is the path of a much smaller, but more violent hurricane, which was encountered off Turk's Island on the 1st September, 1821, and reached the state of Maine, having passed over 1800 miles in sixty hours, with a velocity of thirty miles an hour.

9. A violent and extensive hurricane, which was encountered north of Turk's Island, on the 22d August 1830, passed north of the Bahamas on the 23d, and was off the coast of the United States on the 24th, 25th, and 26th. A great deal of damage was done on the ocean by this storm, but it scarcely reached the American shores. It appears to have moved more slowly than other storms. 10. Is the course of a violent hurricane and snow-storm on the 5th and 6th December, which swept along the American coast from the latitude of 30 deg.

11. Is a portion of the general route of the violent inland storm which swept over the lakes Erie and Ontario on the 11th of November, 1835.

After some general remarks on these hurricanes, which our limits will not allow us to notice, Mr. Redfield makes the following observations:

"It will hardly escape notice that the track of most hurricanes, as presented on the chart, appears to form part of an elliptical or parabolic circuit, and this will be more obvious if we make correction in each case for the slight distortion of the apparent course in the higher latitudes which is produced by the plane projection. We are also struck with the fact that the vortex of the curve is uniformly found near the 30th degree of latitude. In connection with this fact, it may also be noted that the latitude of 30 deg. marks the external limit of the trade winds on both sides of the equator: and perhaps it may not prove irrelevant to notice even further, that, by the parallel of 30 deg., the surface area, as well as the atmosphere, of each hemisphere, is equally divided, the area between this latitude and the equator being about equal to that of the entire surface between the same latitude and the pole."

Independent of the scientific interest which is attached to inquiries such as these we have been considering, they deeply involve the still higher interests of humanity. Mr. Redfield has, therefore, labored to deduce some practical rules by which the unfortunate mariner may extricate himself, with the least hazard, from the im