The diagrams of the Sandy Hook sections, Nos. 4 and 5, show this sudden change very strikingly between positions 13 and 14, the probable minimum lying, however, inside of 13. So, also, the Cape Henry diagrams, Nos. 6 and 6 bis. The "cold wall" minimum and axis maximum are shown on diagram 10, on the same scale of miles at the top of the diagram, and temperature at the side. That the "cold wall" exists south of Hatteras is proved by the same diagram, where the Cape Fear, Charleston, and St. Simon's sections are compared with those for Cape Henry, Cape May, and Sandy Hook. The difference of temperature is less for the southern sections, but it is still strikingly marked.

In the cold water inshore from the Gulf Stream, Acting Master Jones, of Lieut. Maffitt's party, found a current setting southward, as also in the cold band outside of the axis. These results, if shown to be permanent, will be in the highest degree important. As it is, the existence of them at any time shows the cause of many anomalies noticed by navigators in relation to the currents of the Gulf Stream.

The investigations relating to currents remain to be made in detail, though some results have already been procured. It is important in work like this to confine the special attention of observers to a few problems at a time, that they receive close examination.

As the warm water of the Gulf Stream flows onward and outward from the axis at and near the surface, the stratum, as a general rule, becomes thinner. The current is then outward from the axis as well as onward.

5. The changes of the position of the remarkable points in the sections with the season, and other circumstances, are undergoing investigation, some results having been already collected. 6. Chart of the Gulf Stream.

The alternate bands of warm and cold water into which the ocean in and near the Gulf Stream is divided, are shown in the chart now presented, as deduced from the discussion already referred to. The higher temperatures are represented by the darker shades. The axis of the stream is marked by the darkest full shade, and the axis of the colder and warmer bands on each side of it by thinner shades, distinguished as stated on the chart. The axis, where it crosses the Sandy Hook section, is seen to take the general direction of the trend of the coast, which is even more closely followed by the "cold wall" axis. These lines are drawn with a free hand among the points by which they would be rigidly determined in the several sections, so as to give a general consistency to their form. The variations from the points rigidly determined are generally of the same order with the probable errors of those points. The probable outer limit of warm water is designated on the chart.

Within the "cold wall" minimum is a band of higher temperature crossing the Sandy Hook section, and generally well marked, followed by a minimum which appears pretty well determined on the northern sections.

The limits of the chart show the limits of the Gulf Stream explorations up to the summer of 1853, inclusive, the work being still in progress.

ART. VII.-Notice of Earthquake Waves on the Western Coast of the United States, on the 23d and 25th of December, 1854; by A. D. BACHE, Superintendent U. S. Coast Survey.

(Communicated under authority of the Treasury Department, to the American Association for the Advancement of Science.)

IN February 1855, I received from Lieut. W. P. Trowbridge, of the Corps of Engineers, Assistant in the Coast Survey, in charge of the tidal observations on the Pacific coast, a letter calling my attention to the singular curves traced by the self-registering tide-gauge at San Diego, on the 23d and 25th of December, and remarking that the irregularities of the curve could not be produced by disturbances from storms, as the meteorological records for the whole coast showed a continuance at that time of an ordinary state of weather, and the length of the wave was too great to be explained by such action.

"There is every reason to presume," he continues, "that the effect was caused by a submarine earthquake." No shock however has been felt at San Francisco.

When the record sheet of the self-registering gauge at San Francisco was received, similar irregularities in the curves for the same days were found upon it. The sheet for Astoria presented little or no special irregularity. These were the only self-registering gauges actually in operation at this time.

Waves of short period would of course escape detection by the ordinary hourly or half-hourly observations.

About the 20th of June, we received accounts from Japan of a violent earthquake on the 23d of December, the notice of which was more circumstantial than usual, from the damage to the Russian Frigate Diana, in the port of Simoda, in the island of Niphon, from the excessive and rapid rise and fall of the water.

A detailed account of the phenomena of this earthquake and of the rise and fall of the sea produced by it in different places on the coast of the Pacific, is much to be desired, and I have thought that by the publication of the results obtained by the Coast Survey, the publication of official reports of the phenomena

might be induced, perhaps even similar observations may have been made, and these registers of the self-acting tide-gange, will show what observations it is desirable to have for comparison.

Thus far we are left to the public prints for the information obtained, and the different accounts are quite discrepant where they give details, and are usually, as intended merely for general information, too vague in the statements to give satisfactory means of comparison.

A correspondent of the New York Herald writing from Shanghae gives the following notes, stated to be derived from an officer of the Frigate Diana.

"At 9 A. M. on the 23d of December, weather clear, thermometer 72°, barometer 30, a severe shock of an earthquake was felt on board the frigate, shaking the ship most severely. This shock lasted full five minutes and was followed at quick intervals by rapid and severe shocks for 30 minutes."

"At 9:30 A. M. the sea was observed washing into the bay, in one immense wave thirty feet high, with awful velocity. In an instant the town of Simoda was overwhelmed and swept from its foundations. * "This advance and recession of the water occurred five times. * By 2:30 P. M. all was quiet."

A communication in the same paper, purporting to give an extract from the log-book of the Diana, states that

"At a quarter past nine, without any previous indication, the shock of an earthquake, which lasted two or three minutes, causing the vessel to shake very much, was felt both on deck and in the cabin. At ten o'clock a large wave was observed entering the bay.

* * *

*

"The rising and falling of the water were very great; the depth varying from less than eight to more than forty feet, and these changes at intervals of about five minutes continued until noon. * * Scarcely had half an hour elapsed when the rising and falling of the water became more violent than before. Between this time and a quarter past two, (when the agitation again became much less,) the frigate was left four times on her side, and once while thus laid in only four feet of water.

*

*

Since reading this paper I have received through the kindness of Commodore M. C. Perry, a copy of a letter from Captain H. A. Adams, U. S. N., who visited Japan in the Steamer Powhatan, to exchange ratifications of the treaty between Japan and the United States. Capt. Adams says: "Simoda has suffered dreadfully since your visit there. On the 23d of December there were several shocks of earthquakes; the sea rose in a wave five fathoms above its usual height, overflowing the town and carrying houses and temples before it in its retreat. When it fell it left but four feet of water in the harbor. It rose and sunk this way five or six times, covering the shores of the bay with the wrecks of boats, junks, and buildings. Only sixteen houses were left standing in the whole place.

"The entire coast of Japan seems to have suffered by this calamity. Yedo itself was injured, and the fine city of Osaka entirely destroyed."

Captain Adams then gives an account of the disaster to the Russian Frigate Diana, Admiral Pontiatine commanding, which was so injured in the harbor of Si. moda as to lead finally to her entire loss.

"Continuing to decrease in violence and frequency by 3 P. M. the agitation of the water and the motion of the vessel consequent thereon, were very slow."

*

"At this time a fresh west wind was blowing, the barometer stood at 29-87, and the thermometer was 10.50 degrees R. (about 55 63 degrees F.)"

The official report of the disaster to the frigate will probably contain further and more precise particulars of the phenomena.

Mr. P. W. Graves gives in the Polynesian a notice, for which I am indebted to Mr. Meriam, of an extraordinary rise and fall of the waters at Peel's Island, one of the Bonin Islands, on the 23d of December. The first rise noticed was fifteen feet above high water, followed by a fall which left the reefs entirely bare. The hour when this occurred is not stated. "The tide continued to rise and fall during the day, at intervals of fifteen minutes, gradually lessening" until the evening.

At Peel's Island the waters rose on the evening of the 25th of December to the height of twelve feet. I have not however seen any notice of an earthquake on that day.

I present to the Association a copy of the curves traced by the self-registering gauges at the Coast Survey tidal stations at San Diego, San Francisco and Astoria on the 23d and 25th of December, 1854 (see plate). The curves representing tides of short period being traced upon the falling or rising curve of the regular tide, their peculiarities are not so readily seen as when shown in the second diagram (see plate), where the regular tidal curve is represented as a horizontal line. The times of the San Diego curve are reduced to San Francisco time. The curve at San Diego presents many minor irregularities from the motion of the float not having been sufficiently checked to prevent the recording of the waves caused by the wind.

Upon a falling tide the crest of these waves will be met earlier and the hollows later than upon a horizontal surface and the intervals from crest to crest, or from hollow to hollow, will be affected by the change of rate of fall. Upon a rising tide the reverse will occur.

There can be no doubt that these extraordinary rises and falls of the water at short intervals, were produced by the same cause which determined the extraordinary rise and fall in the harbor of Simoda, in Japan, and at Peel's Island.

The San Francisco curve presents three sets of waves of short interval. The first begins at about 4h 12m, and ends at 8h 52m, the interval being 4h 40m. The second begins at about 9h 35, and ends at 13h 45m, the interval being 4h 10m. The beginning of the third is about 133, and its end is not distinctly traceable. The crest of the first large wave of the three sets occurred at the respective times of 4h 42m, 9h 54m, and 14h 17m, giving intervals of 5h 12m, and 4h 23.

The average time of oscillation of one of the first set of waves was 35, one of the second 31m, and one of the third about the same. The average height of the first set of waves was 45 of a foot on a tide which fell two feet; of the second 19 of a foot on a tide which rose three feet; of the third somewhat less than 10 of a foot, on a tide which fell some seven feet. The phenomena occurred on a day when the diurnal inequality of the tide was very considerable. The greatest fall of the tide during the occurrence of the first set of waves was 70 feet, and the corresponding rise 60 feet. In the second the corresponding quantities were 30 feet, and in the third 20 feet. These waves would not have attracted general attention.

There is a general analogy in the sequence of the waves of the three sets, which seems to mark them as belonging to a recurrence of the same series of phenomena. In the diagram No. 3 A (see plate), the heights of the successive waves of the first set at San Francisco are shown by the dots joined by full lines, and of the second by those joined by the fine dotted line. The full faint lines show the heights of the first series at San Diego and the broken faint lines the heights of the second. The heights in hundredths of a foot are marked at the side of the diagram, and those of the successive waves are placed at regular intervals, the waves being numbered from 0 to 7 at the top of the diagram. The height is the mean of the fall from a crest to a hollow and of the succeeding rise from the same hollow to the next crest. The times of oscillation from one crest to the next succeeding, are placed on the same diagram, the times being written at the right hand, and the wave being designated at the lower part of the diagram No. 3, B. The full line represents the times of the first series at San Franciso and the broken line the times of the second. The full and broken faint lines represent the times of the first and second series at San Diego.

The intervals between the times of occurrence of the crests of the successive waves in the first and second series diminish from 5h 10m to 4h 48m by irregular differences.

The effect of the rising or falling tide upon which these waves occur is of course greater in disturbing the heights than the times. The series itself looks like the result of several impulses, not of a single one, the heights rapidly increasing to the third wave, then diminishing, as if the impulse had ceased, then renewed, then ceasing, leaving the oscillation to extinguish itself.

If we had a good scientific report of the facts as they occurred at Simoda, the subject would lose the conjectural character which must otherwise belong to it. Although we have no account of the place where the earthquake had its origin, the violence of its effects in Japan and the diminished effects at Peel's Island, show that Japan was certainly not far from the seat of action.