or 200 leagues from the coast of America) and the meridian of the Marquesas, or Archipelago of Nouka-Hiva; and that the N.E. trade wind, which, commencing at about 100 leagues from the coast of America, extends to about the neighbourhood of the Marianne Islands.

Deviation in the Trade Wind.-The numerous groups of islands Westward of the islands of Nouka-Hiva, and near the Australian coaat, as well as those situated North of the equator in the same latitude, appear to effect a modification of the trade winds of the Western Pacific into periodical winds or monsoons, and also variable winds. Thus we find that in these groups, in the zone situated North of the equator, the trade wind only prevails from October to May; from March to October in that situated South of the equator, and that Westerly winds, storms, and rain occur during the rest of the year. We shall refer again to this important fact, and adduce some undoubted proofs of this from amongst numerous observations in the voyages of circumnavigators.

Zone of the Trade Winds.-The zone in which the trade winds of the Pacific prevail, where, as above observed, they blow with regularity, is comprised between the equator and the parallels of 30° North and 30° South latitude.

On consulting a large number of authorities, we have constructed the following table from voyages in the Pacific Ocean of ninety-two vessels which have crossed the equator between the meridians of 106° and of 147° West longitude. Incomplete as it is, it yet shows some important facts, although they should only be considered as approximations, having been deduced but from comparatively few observations.

The following table shows the limits of the trade winds every month, and the breadth of the zone of the variables of the equator.

Polar Limit of the Trade Winds.-From this table it is evident that the polar limits of the trade winds are variable, and that they are found most to the Northward and Southward according as the sun has a North or South declination.

Equatorial Limit of the Trade Winds.—It is also evident that the equatorial limits of the trade winds is equally variable, according to the season, and leads to the conclusion that the zone of the variable winds, which separates these limits, is not so broad in the winter as in the summer of the Northern hemisphere.

Zone of the Variable Winds of the Equator.-It shows also that the equatorial zone of the variables moves as much to the North of this circle as far as the results can do which we have collected. This zone also assimilates much with that of the Atlantic Ocean. It is larger on the meridians of 90° to 100° than on those further to the West, as 120° to 130° of West longitude. Hence the breadth of the zone decreases in proportion as it is Westward. There in fact calms and light airs, varying from N.W. to South by the West, are found, and frequently stormy weather.

The Northern limit of this zone is about the parallel of 8° North latitude. Its Southern limit is in 3° N. Nevertheless it sometimes happens that the N.E. and S.E. trade winds reach each other, and vessels pass from one to the other without experiencing calms.

The S.E. Trade.-Between the parallels of 30° S. and 4° S. lat., at 150 or 200 leagues off the coast of America, bad weather is not generally found. The trade wind, varying between East and S.E., blows constantly to the middle of this zone. In the summer the direction of the wind varies from E.S.E. to S.S.E., and never veers Northward. When the sun is in the Northern hemisphere, the trade winds of the Southern hemisphere blow directly from the S.S.E., and on the contrary they blow more from the East when the sun is in the Southern hemisphere.

In examining the foregoing table we see that the polar limit of the S.E. trade seldom attains the parallel of 30° South latitude. Its equatorial limit, on the contrary, often reaches to the North of the equator, as already observed. The result is, that the zone of the S.E. trade wind is generally larger than that of the N.E. The S.E. trade winds are more certain and less variable than those of the N.E. In the zone occupied by the former, calms and storms are much less frequent; lastly, we may observe, that in this zone the temperature has been found lower than in that of the N.E. trade, a fact which is confirmed by observations collected during recent voyages.

N.E. Trade. The foregoing table, compared with that drawn up for the Atlantic Ocean, proves a fact asserted long ago, that the polar limit of the N.E. trade winds of the Pacific Ocean is nearly the same as that of the same wind in the Atlantic Ocean; at least it differs from it very little.

In the northern hemisphere, when the sun has North declination, the N.E. trade comes more directly from the East, but varies to E.N.E.

and sometimes to the southward of it, but accompanied then by squally weather.

On the contrary, when the sun is in the southern hemisphere, the trade wind takes a more northerly direction, varying from E.N.E. even to N.N.E. In this case the equatorial limit of these winds approaches nearer to the equator than in any other season, but it never passes to the South of this circle.

The N.E. trade wind is always constant, and blows fresh towards the middle of the zone above-mentioned from October to May, and it extends to the Philippine islands. In which season a ship crossing the Pacific from the Eastward should keep between the parallels of 13° and 14° North latitude.

The N.E. trades are, however, generally less constant, and therefore less to be depended on than the S.E. trades. In their variations they sometimes come from the South, as already observed, and they blow as frequently from E.S.E. Calms and storms are also more frequent in the zone where they are found than in that of the S.E. trades.

Monsoons of the Pacific Ocean in the Zone of the Trade Winds.— It has been shown above that the S.E. trade winds are only regularly established in the space comprised between the eastern meridian of the Galapagos and that of the Nouka-Hiva group, and the N.E. trades in that which commencing at 100 leagues from the coast of America, extends to the Marianne islands. But in adducing this fact as a general law, it is not pretended that it is so in every part of it in the same manner, and precisely the same time.

We have said that to the North of the equator the N.E. trade wind blows over the whole of the Pacific Ocean from October to May, or nearly so; and South of the equator the S.E. trade wind blows from March to October, although we often find variations among the islands situated South of the equator.

We have added that these winds are replaced in the West part of the Pacific Ocean, during the rest of the year by variable winds or monsoons, contrary to the opinion which has been long established in this respect, and the following justifies our remark.

The Pomoutou and Nouka-Hiva Groups.In the month of July, 1838, the corvette Astrolabe, in the latitude of the Pomoutou group, met with the S.E. trade wind in 19° 50' South latitude. From the 15th of July, the time when she was in this latitude, and in 114° W. longitude, to the 28th of July, she experienced less regular winds, varying from N.N.E. to S.S.E., most often light, and intersected by winds from N.W. to S. W.b. W., blowing fresh.

From the 28th of July to the 5th of August, being in sight of the islands of Manga Reva, she experienced nine days of N.W. wind, varying to S.W. by the West, the N.W. wind bringing strong breezes with fine weather in general, although cloudy. To the 15th of August the prevailing winds were those from N.W. to S.S.W., the weather cloudy and sometimes rainy. The winds from W.S.W. were the

strongest. These winds were interrupted by light breezes from N.N.E. to S.E.b.E.

From the 15th of August to the 20th of the same month, the period when the Astrolabe was in sight of the island of Clermont Tonnerre, she had the same variable winds that she experienced in the archipelago of Nouka-Hiva. At the end of August and the beginning of September winds varying from W.N.W. to S.S.W. by the West.

M. Duperrey, in the same latitudes, experienced in April South winds varying from West to W.N.W. by the West. It thus appears by these observations that near these two archipelagos the S.E. trade wind is not the most frequent. They also confirm this remark of Cook, to the effect that, in the neighbourhood of islands the trade is not a regular wind, and that it does not extend beyond the parallel of 20° S.; that further South in these latitudes the winds blow from the West, and the trade wind is hemmed into so narrow a space that it blows with greater violence. Near these limits the trade wind is

much lighter.

In the archipelago of Nouka Hiva, Krusenstern also mentions S. W. winds, which are tolerably constant.

Tahiti Group.-In the Tahiti Group the winds during part of the year vary from E.S.E., E.N.E., and sometimes (especially in July) East to S.S.E. The trade wind blows strongly in this archipelago; the weather is then cloudy and we sometimes have rain. In October, November, December, and part of January the winds blow from West to S.W. They are interrupted by calms and light breezes from the East and sometimes by storms and rain. In December and January, when the sun is in these latitudes, the wind and weather are very changeable. It is then that moderate winds from W.N.W. to N.W. last five or six days. With these winds the sky is cloudy, often with rain. If they pass to the North they are much slackened and replaced by winds from W.S.W. to S.W. If they take a more southerly direction they become strong. The S.W. and W.S.W. winds are more frequent than those from the North.

A second but all important observation presents itself in the circumstance that this archipelago is very close to that part of the Pacific Ocean where the trade winds are regularly established throughout the year.

We might adduce a similar observation in respect of the Tonga or Friendly Islands. In September we find S.E. winds there, followed in October by S.W. to S.S.W. winds, and even by those from N.N.W. The same observation has been made in regard to New Caledonia. In December, 1838, the Astrolabe experienced, near the Solomon Islands, winds from W.S.W. to S.S.W., sometimes blowing very strong, with cloudy and rainy weather.

Solomon Islands.-Among the Solomon Islands the N.W. monsoon commences in December or January. Sometimes these two months are tolerably fine. In February and March fresh breezes, with heavy squalls and abundance of rain, may be expected. April is generally a

fine month.

The winds are then variable until May; during which

the weather is often fine.

In June the S.E. monsoon is very strong, with frequent rain squalls; and bad weather, interrupted by a series of fine days, lasts until the end of August. In September the monsoon begins to slacken, and from this time to the return of the N.W. monsoon the breezes are moderate.

Eastward of these islands the N.W. monsoon is not so constant. It commences later, towards the end of January, and blows fresh during sixteen or eighteen days, when it is followed by the wind from East, which blows during the same number of days. The West wind then sets in with a little bad weather and continues, with storms and rain, constantly blowing hard. These winds from East and West last alternately until the end of March, the time when the regular S.E. winds commence. The further to the East the more irregular is the N.W. monsoon, and it is not found beyond the Nouka Hiva Group.

New Britain and New Ireland.-At New Britain and at New Ireland the S.E. trade wind is found to prevail only from March to October. When the sun has South declination the West winds, varying from N.W. to S. W., replace it during the rest of the year.

New Guinea.-At New Guinea two monsoons are found, which seem to be the continuation of those from the China Sea. One is the N.W., the other the S.E. The S.E. monsoon prevails from the middle of May or June until October; that from N.W. from November until May. In December we generally find the wind varying from N.N.W. to N.E.

Eastward of New Guinea the wind is from the S.E. during the time that the S.W. monsoon is blowing North of the line.

Torres Strait.-In Torres Strait the N.W. monsoon commences towards the end of October and lasts till the middle of March. The S.E. trade blows in April and strengthens till June, being sometimes very strong in this latter month and also in July; they then slacken and gradually moderate.

Marianne Isles.-In the Marianne Islands we find a monsoon from West, beginning from the middle of June and continuing to the middle of October. The rainy season lasts during the months of July and August, but often only known by a succession of heavy showers. During the westerly monsoon the wind varies very much to the time of the syzygies, and then blows hard from all points of the compass. The typhoons and hurricanes met with in the China Sea occur iu the latitude of these islands at the time of the changes of the monsoons. The East limit of these storms is near their meridian. From the 15th of October to the 15th of June the weather is fine and the Easterly monsoon prevails throughout this interval.

From these facts, which if we were to quote more of them would oblige us to exceed our general observations, it appears that with regard to the S.E. trade winds the zone where they in a manner permanently prevail is bounded on the West by the meridian of the