January 8, 1886.] 



SCIENCE. 



41 



ter fifteen or sixteen degrees colder than its neigh- 

 boring sea-level stations. Here it is as much too 

 warm as it was too cold in the first type. Al- 

 though it is near the storm-centre, where the 

 winds are supposed to ascend obliquely, the air on 

 the mountain is evidently not derived from the 

 low-level stations near by ; for, independently of 

 the evidence furnished by the wind's direction and 

 velocity against such a conclusion, the tempera- 

 tures disprove it. If a current of air ascend from 

 sea-level to the top of Mount Washington, its tem- 

 perature must fall at least eleven degrees, even if 

 the cooling from expansion were retarded by con- 

 densation of vapor through the whole ascent. The 

 surface source of the wind, if it come from the 

 surface at all, must therefore be sought many 

 miles south of New England, in the southern 

 states or on the Gulf Stream, where the tempera- 

 ture is fifteen or twenty degrees higher than in 

 Mount Washington latitudes. Then, as in the first 

 case, the temperature on the mountain is largely a 

 matter of importation ; but now the cooling by 

 ascent abruptly up the mountain sides, or gradu- 

 ally in the cyclonic whirl, acts to destroy the im- 

 ported characteristics of the wind, instead of to 

 confirm them, as before. In the pronounced ex- 

 amples of this type, when it is warmer on Mount 

 Washington than at Boston, we find illustration 

 of the inversion of temperature, that is generally 

 held to be peculiar to anticy clonic weather, as will 

 be explained below ; and although such cases are 

 not, so far as I know, characteristic of other 

 mountain stations, they are not rare on Mount 

 Washington. Examples may be found on the old 

 maps for Dec. 3, 27, 1873 ; Jan. 7, 8, 27, 28, 1874. 

 The warm waters of the Gulf Stream, and the 

 rapid decrease of temperature with latitude along 

 our eastern coast, must be chiefly responsible for 

 this. Another factor of hardly less importance is 

 the fivefold greater velocity of the winds at the 

 height of Mount Washington over those at the 

 earth's surface. As a storm-centre draws near, 

 the winds on the mountain may be derived from 

 a source four or five times as distant to the south 

 as that which supplies the low-level stations. 

 Thus the ordinary decrease of temperature with 

 height is overcome. Montreal is decidedly colder 

 than the three other stations at such times ; for it 

 is well to the north of the storm-centre, and draws 

 its winds from northerly sources. 



The third type is one that has attracted much 

 attention in Europe of late years, on account of 

 the very abnormal temperatures that accompany 

 it. It appears when a centre of high pressure — 

 an anticyclone — passes over the mountain, and, 

 when fully developed, it causes a remarkable in- 

 version of weather elements. We are accustomed 



to see mountain-tops cold and cloudy, while the 

 valleys about them are warmer and clear ; but 

 anticyclonic weather places the cold and the clouds 

 in the valleys, while the peaks rise into brilliantly 

 clear, warm, dry air. Dr. Hann was the first to 

 give a full explanation of the facts, in 1876, and I 

 follow him in this statement. In an anticyclone, 

 the few lofty clouds that are observed generally 

 move towards its centre ; the surface winds move 

 outwards to all sides ; with converging currents 

 above, and diverging below, there must be a 

 descending current about the centre ; the descent 

 is probably slotv, but it undoubtedly exists. This 

 type, therefore, involves the consideration of the 

 temperature of air derived from regions of the 

 atmosphere far above the mountain-tops. The 

 first opinion that one would have of such tempera- 

 ture would probably be to place it well below 

 freezing, for we are all familiar with the excessive 

 cold experienced in very lofty mountain ascents 

 and balloon voyages. But this is wrong. Although 

 undoubtedly cold while up aloft, air that de- 

 scends from the upper regions is compressed as 

 it comes under greater atmospheric weight near 

 sea-level, and it is thereby warmed. A current 

 coming down from a moderate altitude in sum- 

 mer might be cooler than the surface air ; but in 

 winter it would be in practically all cases de- 

 cidedly warmer. The statement of this fact is 

 not particularly new, but its recognition and 

 general application are a recent progress in 

 meteorology. More than forty years ago, Arago, 

 Pouillet, and Babinet reported to the French 

 academy that "it is proved by the investigations 

 of Mr. Espy that one should not hereafter attempt 

 to adduce, in the mean state of the atmosphere, a 

 descending current of air as a cause of cold." 



It is, then, to the descent of air from aloft that 

 we are to look for the abnormal warmth and dry- 

 ness of mountain-tops in anticyclones. It remains 

 to account for the extreme cold that prevails at 

 the same time in the neighboring valleys. An 

 illustration of the contrast is given by Professor 

 Upton in the second Bulletin of the New England 

 meteorological society. On the morning of Dec. 

 27, 1884, when the winds were everywhere light, 

 and the pressure higher than on the days before 

 or after, the temperature on Mount Washington 

 was +16°; at the low-level stations north of 

 Massachusetts, it was —10°, or colder. On con- 

 sulting the records, I find Grafton and Littleton, 

 N.H., — 18° ; Hanover, N.H., and Newport, Vt., 

 —20° ; Woodstock, Vt., —27° ; Portland, Me., 

 + 7°. The lower cold must therefore be in spite 

 of, not on account of, the down-cast current ; and 

 we are forced to believe that it is caused by rapid 

 cooling of the ground, and of the air close to it, by 



