moisture of the atmosphere; increasing amount of cloud and an ominous heaving of the sea, are all conditions forerunning the occurrence of the typhoon.

The average tracks of the various classes of typhoons, together with the frequency and the season of appearance of each class are to be found on Pilot Charts of the Pacific Ocean. For a more complete account of typhoons consult the North Pacific Pilot Chart for July, 1898.

55. Cautionary Remarks.-It must be borne in mind that although the region and season of the year would render the navigator very cautious, yet every strong wind or gale met with, particularly in the tropical regions, must not be treated as a cyclone. When there is reason to suspect the advance of a cyclonic storm the safest proceeding is to lie-to and carefully watch the barometer, weather indications, and shiftings of the wind. A decided drop of the atmospheric pressure of at least inch, together with marked shiftings of the wind, should be experienced before the storm can be concluded as cyclonic.

NOTES ON WEATHER INDICATIONS

56. Indications by a Mercurial Barometer.-The use of the barometer as a weather glass is common both on sea and on land. But only those that have long watched and carefully compared its indications with the prevailing weather conditions are able to foretell more than that a rising barometer indicates less wind or rain; a falling barometer, more wind or rain, or both; a high barometer, fine weather; and a low one, the reverse. But useful as are these general conclusions, in most cases, they are sometimes

erroneous.

By attending to the following brief observations, any one not accustomed to the use of a barometer may do so with less hesitation and with immediate advantage:

The column of mercury in a good barometer usually stands, on an average, some tenths of an inch higher with

or before polar and easterly winds than it does with or before equatorial and westerly winds (of equal strength and dryness or moisture) in all parts of the oceans. The terms polar and equatorial are here used with reference to winds blowing from the nearest polar direction, or from the equatorial parts of the earth.

This peculiarity of the barometer causes many mistakes to be made. The barometer is high, perhaps, but falling. Wind or rain, or both, are expected in consequence, yet neither follows to any decided extent. A change of wind, only, from one quarter to another takes place. Reversely, the barometer is low, but rising. Fine weather is expected; yet, instead of that, a strong wind, accompanied perhaps by rain, hail, or snow, rises from the polar direction. By such changes as these, seamen are often misled, and calamity, caused by unpreparedness, may sometimes occur as a consequence.

There may be heavy rains or violent winds beyond the horizon, and even within the view of an observer, by which his instruments may be affected considerably, though no particular change of weather occurs in his immediate locality. Sometimes, severe weather from an equatorial (southerly in north latitude, northerly in the southern hemisphere) direction, not lasting long, may cause no great fall of the barometer, because followed by a duration of wind from polar regions; and at times the mercurial column may fall considerably with polar winds and fine weather, apparently against the rule, because a continuance of equatorial winds is about to follow.

57. As a general rule the barometer rises for northerly winds (included between the northwest and northeast); for dry or less wet weather; for less wind; or for more than one of these changes, except on a few occasions, when rain, hail, or snow comes from the northward with strong wind.

The barometer falls for southerly winds (included between the southeast and southwest); for wet weather; for stronger wind; or for more than one of these changes, except on a

few occasions, when moderate wind with rain or snow comes from the northward.

There is little variation of the barometer between the tropics because the wind blows generally in the same direction and with equal force, and no contending currents of air cause any considerable change in the temperature or density of the atmosphere. For violent storms or hurricanes, however, within the tropics, the barometer falls very low, but soon returns to its usual state after the storm center has passed.

It has been observed on some coasts that the barometer is differently affected by the wind, according as it blows from the sea or from the land, the mercury rising on the approach of the sea breeze and falling previously to the setting in of the land wind.

58. Indications by Appearance of Sky.- Some young seamen hardly appreciate sufficiently common rules about weather, which are as true as they are trite; namely, that a red sky at sunset presages fine weather; a red sky in the morning bad weather, or much wind, if not rain; a gray sky in the morning, fine weather; that soft-looking or delicate clouds foretell fine weather, with moderate or light breezes; hard-edged, oily looking clouds, wind; that a dark, gloomy blue sky is windy, but a light, bright blue sky indicates fine weather; that, generally, the softer the clouds look the less wind, although rain may be expected; and the harder, more "greasy," rolled, tufted, or ragged, the stronger the wind will prove. Also that a bright yellow sky at sunset presages wind; a pale yellow, wet; and that, by the preponderance of red, yellow, or gray tints the coming weather may be foretold very nearly-indeed, if aided by instruments, almost accurately.

These indications of weather, afforded by the colors of the sky, seem to deserve more critical study than has yet been given to the subject.

59.

Indications by an Aneroid Barometer.-A rapid rise indicates unsettled weather.

A gradual rise indicates settled weather.

A rise, with dry air and cold increasing, in summer, indicates wind from the northward in north latitudes, but from the southward in south latitudes; and if rain has fallen, better weather may be expected.

A rise, with moist air and a low temperature, indicates wind and rain from the northward in north latitudes, but from the southward in south latitudes.

A rise, with southerly winds, indicates fine weather in north latitudes, the conditions being reversed in south latitudes.

60. A steady barometer, with dry and seasonable temperature, indicates a continuance of very fine weather.

A rapid fall indicates stormy weather.

A rapid fall, with westerly winds, indicates stormy weather from the northward.

A fall, with a northerly wind, indicates stormy weather, with rain in summer and snow in winter.

A fall, with increased moisture in the air and the temperature rising, indicates wind and rain from the southward. A fall, with dry air and cold increasing, in winter, indicates

snow.

A fall, after very calm and warm weather, indicates rain with squally weather.

61. All indications appertaining to the fall of the aneroid apply to the northern latitude. In southern latitudes the wind directions are reversed.

The

62. Meteorological Observations at Sea. United States Government is conducting an extensive system of ocean, as well as of land, meteorological observations through its Weather Bureau. It seeks the cooperation of all navigators, requesting them to take one observation every day at a prescribed moment, which is simultaneous for every part of the globe. These simultaneous observations are charted and published by the Weather Bureau at Washington in its international weather charts and bulletins, copies of

which, with other publications of the Weather Bureau, are sent regularly by mail, without charge, to every seaman making such observations and forwarding them to the Chief of Weather Bureau, Washington, D. C. By entering into this arrangement and taking part in the observational work, every seaman may contribute materially to this scientific enterprise and further the elucidation of the "law of storms," as well as secure for his own use a large supply of valuable nautical information. Blanks for recording the observations (and instruments also, if necessary) are furnished to applicants, and it is hoped that all captains and officers interested in meteorology will enlist in the service.

OCEANIC CURRENTS

63. Classifications of Currents.-Oceanic currents may be conveniently divided into two classes, viz., drift, or surface, currents and stream currents.

64. Drift, or surface, currents are produced directly by the wind and move along the surface of the water, only, in a direction more or less parallel to the wind. Drift currents, therefore, are shallow and slow and can run in no other direction than before the wind that produces them. The following extract from a publication by the United States Hydrographic Office, relating to drift currents, will no doubt prove very instructive:

"For our knowledge of the surface currents of the ocean as they actually exist we are dependent largely upon ships' observations, and concerning these it may be said that, strictly taken, they represent not the actual current experienced during the preceding 24 hours, but only the difference between the position of the vessel as determined by astronomical observation and as determined by dead reckoning. This difference is made up primarily of the current actually encountered, secondly by the errors in the dead reckoning, arising from incorrectly estimated course and distance-here the deviation of the compass, often poorly determined, plays