Second, the black clouds and lightning which indicate the approaching hurricane are seen to the S., S. E., and E. S. E., according to the season of the year, as we see them at the westward. During the rainy season, and when the storm, as is usual at that period, is small, and the S. E. trade blows more eastwardly, the wind at the Windward Islands, possibly, may set in at the north, and back round by the east as it progresses. So Colonel Reid thinks it sometimes does, at Barbadoes. But when the belt of rains is south, and the hurricane comes from the south-east, and is larger and more violent in its action, and the north-east winds prevail, the first effect is an increase of these trades. Soon, however, the wind hauls to the north and north-west, in opposition to its course, bearing the same relation to it that our east and north-east winds bear to storms in the United States; and the wind hauls around during the passage of the storm to the west, south-west, and south-east, and at the latter point it clears off. Mr. Edwards in his History of Jamaica says--and as a resident, his authority should be decisive as to this Island--"_that all hurricanes begin from the north_, veer back to the W. N. W., W., and S. S. W., and when they get around to the S. E. the foul weather breaks up." Doubtless the same is true of the class of northers of which we are speaking on the Gulf of Mexico. _But with this class the barometer does not rise during the gale, and in proportion to its length and violence._ With the other class of N. W. winds--the northers of winter--it does. The following description of two winter northers, copied from Colonel Reid's valuable work, will illustrate what has been said. _Precisely such changes from S. E. rains to N. W. winds, with blue sky and detached dark clouds--fair-weather N. W. scud--occur every autumn in October and November_, and the falling of the thermometer and rising of the barometer, after rain, and a change of the wind, are perfectly characteristic.