largely if we take into account the number of miles. And this may be done safely, as the expedition of Capt. Hall had an anemograph of Robinson's plan. The east winds then seem to be a weak local land-wind, caused by the difference of temperature of land and sea. The N. E. winds, on the contrary, are the true polar currents, flowing towards the barometric depression about Iceland. In summer the S. W. wind prevails as to time, but the excess is on the side of the N. E., if the number of miles is considered, but of much less amount than in winter and spring. In the second winter station of Capt. Hall's party, Lifeboat Cove or Polaris House, as also in Hayes's Station, Port Foulke, in the vicinity, the N. E. prevail even more than in Polaris Bay in winter and spring. The W. and N. W. are entirely wanting. In the tables of Professor Coffin, the winds at Rensselaer Harbor, Kane's winter station, were recorded with reference to the magnetic direction. As the magnetic declination is known to be 108° 12′ W., I give below the true mean direction of the wind in this locality, and also that recently calculated by Dr. Bessels for Polaris Bay. In the Map, Pl. 2, the true direction is given. The observations of Rensselaer Bay are thus shown to agree, to a considerable extent, with those of the surrounding stations. The winds are more easterly than at Polaris Bay at all seasons, and do not vary as much as at that station, the difference between winter and summer being only 91° instead of 161°. See Map, Pl. 2. The Danish settlements of Northern and Southern Greenland (all on the west coast of the island), Upernavik, Jacobshavn, and Godthaab, have largely prevailing east winds (from the land) in winter, and west winds (from the sea) in summer. As the force of the winds has not been accurately ascertained, we cannot say whether the N. E. are much stronger than the East, as in Polaris Bay. In the summer the rocky surface of the interior (as Greenland is not all covered with ice) is highly heated by the sun, it draws in the air from the colder sea, which is cooled by the large number of icebergs floating southward. We know much less about Eastern Greenland, the country being entirely uninhabited. Yet the 2d German polar expedition having passed a year near Sabine Island, 75° L. N., near the coast, we are able to say that the prevailing winds are N., especially in spring, autumn, and winter, while S. winds are nearly as frequent as N. in summer. The N. prevail here to a less degree than the N. E. at Lifeboat Cove and Port Foulke; but it would be rash to decide from so short a period and so few observations that the polar winds are really less prevailing in the east than in the west of Northern Greenland. The eastern coast of the island being nearer to Iceland, where pressure is low, we might infer that the contrary should be the case, if all local influences were eliminated. Nearly all the storms near Sabine Island come from the N., and the mean force of this wind is very much greater than that of any other wind.1 The constancy of the polar current in Northern Greenland is indirectly proved by the small precipitation of rain and snow. The quantity of snow falling at Polaris Bay and Lifeboat Cove was scarcely measurable, according to Dr. Bessels. He thinks the glaciers of Northern Greenland are the remnant of a former age, when the climate was different. The snow and ice that melt in every summer are not now replaced by new snow, so that the glaciers must be decreasing. The German expedition did not encounter a heavy snow-fall, and the parties who, in sledges, explored the interior, were quite astonished at the constant brilliancy of the sunshine of the Greenland summer, In Arctic countries the sea is warmer than the land in the mean of the year; during a very short time only, in summer, are the conditions reversed. The pressure is generally higher on land, so that we must expect to see a prevalence of landwinds in the mean of the year. In looking at the map of the polar regions (Plate 2) an easterly mean direction is seen to prevail in all stations in Greenland, that have the open sea to the westward; and a westerly in the stations of the Arctic Archipelago, which have the sea to the eastward. By sea, is meant here the more or less open waters of Baffin's Bay and Davis Strait, and not the more ice-bound straits and inlets of the archipelago. Ikogmut and St. Michael in northern Alaska have easterly winds, directed towards Behring Strait. In Ustyansk, in the extreme north of eastern Siberia, the mean yearly direction is nearly due south-as we might infer from the fact that the Arctic Ocean lies to the north of this place. Hammerfest, Vardo, and Bossekop, in extreme northern Norway, have also prevailing southerly winds for a similar reason. The extreme prevalence of land-bound (Mediterranean) seas, north of the North American continent, greatly affect the character of the region considered in a climatic point of view. As land-bound seas in these latitudes will be also ice-bound, the air over them would cool as over a continent, so that places situated on the shores of such seas will have a cold continental climate in winter, spring, and autumn. This cold will not, however, be followed by a comparatively warm summer, as is the case on polar continents far from the influence of the sea. The melting ice over the sea absorbs the heat of the sun's rays. Thus we have a continental climate during three-quarters of the year, and an oceanic during the remaining summer quarter. This is the case in the Arctic Archipelago. It has one of the coldest climates of the world, the winter being even colder than in northern Greenland, and only a little warmer than in Iakutsk in eastern Siberia, and the summer also extremely cold. The percentage of winds is as follows: 1 See "Die Zweite Deutsche Nordpolarfahrt," Leipzig, 1874. 8 11 16 15 15 10 28 18 6 9 6 15 5 5 4 30 ་ 7 6 7 5 21 33 S. E. The prevalence of the N. and N. W. winds is here strongly marked, especially in winter. At two of the stations more than half of all the winds come from the N.W. The exception presented by Port Bowen, where E. winds largely prevail in winter, is explained by the large land-mass to the E. The winds of the inland and western stations of Arctic America, as well as the Arctic Ocean in their vicinity, show more irregularities. From the foregoing table it appears that in Northern British America (Forts Norman, Simpson, Enterprise, Reliance, Franklin, and Anderson) there is no accordance in the direction of the winds. They seem to vary much according to locality. This is a very cold region, and being continental, calms are much more prevalent in winter than in the Archipelago. We must expect to find here higher pressure in winter than further to the east, because the depression about Iceland is not so near. The great distance of the Atlantic depression and the mountains which lie between this region and the Pacific depression, also explain the undecided character of the winds in winter. We have fewer observations in the summer. Among these, Fort Franklin has prevailing E. winds, coming from Great Bear Lake, where the ice does not melt till the end of the summer. In Alaska monsoon winds are seen to prevail from the N. E. (the land) in winter, S. W. Fort Tongass S. W. in summer. In Behring Strait southerly winds are also more numerous in summer, while the Arctic Ocean northward of it has northerly winds at the same season. In the last two stations lying near Hudson's Bay, a monsoon influence is exhibited in the S. winds of winter. Hudson's Bay does not freeze entirely, and thus the wind will blow towards it from the land. (See Maps, Plates 5, 6, and 14.) TEMPERATE ZONE OF AMERICA WEST OF THE ROCKY MOUNTAINS. On the coast of Alaska and further south in Washington Territory, the winds have a monsoon character. The cause of this is the difference of temperature and consequently of pressure on land and sea, producing a current of air from the land in winter, and from the sea in summer. It is necessary to remember that the warm current of the Kuro-Sivo, the Gulf Stream of the Pacific, passes, in its return to the south, near to this coast, and there must be a diminished pressure over the region, at least in the colder part of the year. The interior of the continent is very cold at that time, and therefore the pressure of the air must be high there. In the summer there is a narrow cold current passing between the coast and the Kuro-Sivo, while at the same time the interior of the continent has a great excess of temperature over the coast, and, as in other dry and warm continental areas, the pressure must be low. There is no country of the world where the temperature of the summer increases so much as we go from the coast to the interior as on the Pacific slope of America, from Alaska to Lower California. The summer isotherm of 59° passes near San Francisco on the coast of California, and is supposed to reach the polar circle on the Yukon River, in the interior of Alaska, a difference of 28° in latitude. Fort Miller, in the interior of California, has a summer temperature of 85°.5, and Monterey, on the coast, and in the same latitude, but 59.0; difference 26.5 F. The percentage of winds in Alaska and Washington is given below, and, with the help of the maps, Pl. 5 and 6, will serve to illustrate the winds of this region. Plate 14 gives the atmospheric pressure. Isl. of St. Paul, Alaska, Beh- Iluluk, Aleutian Islands' Fort Wrangel Sitka . N. W. Washington S. W. Washington If, as was said before, the winds of this coast have monsoon features, these monsoon winds do not overpower others, especially in winter. At that season of the year the pressure is high in the latitude from 25° to 35° N. on the coast of California, and in the same latitudes on the Pacific Ocean. Winds from this region are quite frequent, and passing over the warm waters of the Japanese current, give a very warm climate to the whole coast. The winter temperature of Sitka is equal to that of New York, and above that of St. Louis. It seems to me that the S. E. winds which are so frequent on this coast, are, partly at least, the deflected S. W. winds of the Pacific. The mountain-chains give them a direction from the S. S. E. The Aleutian islands are very near to the centre of lowest pressure on the Pacific, at least in winter. They occupy a position similar to that of Iceland in the Atlantic; the same may be said of the island of St. Paul in Behring Sea. The storms are frequent and severe, and the winds polar and equatorial in turn, without a marked predominance of either. In summer the centre of depression moves to the northward and inland, and accordingly the winds are principally from the south. In Washington Territory the winds of the coast-region are very similar to those of Sitka. In the interior of Washington and Oregon the winds have no strongly marked monsoon character. (See also Maps, Pl. 5, 6, 8, and 11.) The S. W. is here the prevailing wind, winter and summer, as in the same latitudes on the oceans and in Europe. We must see in these winds a continuation of the equatorial current of the Pacific, which crosses the coast-ranges and descends into the valleys, while part of it is deflected by these mountains and appears as a S. E. wind at Sitka. The winds of California differ in some respects from those of the northern Pacific coast. They are westerly at all seasons of the year, more S. W. in winter and N. W. in summer. The winds of the summer are very strong and steady, giving to the California coast a peculiar climate—a summer colder than anywhere in the same latitude even in the southern hemisphere. In some places the prevailing winds in summer are S. W., and the mean direction also south of W. This is probably due to the position of the coast, so that the S. W. seems to be a local sea-wind. At San Diego the number of miles was also observed, and I have calculated separately the percentages for the number of observations and for the number of miles, in the three summer months. |