METEOROLOGY. (WINDS.) 



537 



desists on the summit of Bohul mountain, in 

 the Caucasus range, 12,000 feet above the sea. 

 During a storm of hail and lightning following 

 rain, a bright violet ball, surrounded by rays 

 which seemed about two yards long, struck the 

 top of the peak. A second and a third stroke 

 followed, and the whole summit of the peak was 

 soon covered with an electric light which lasted 

 not less than four hours. One of the party was 

 prostrated by the first stroke of lightning, which 

 left marks upon his body, but he recovered. At 

 midnight, a camp to which the party had re- 

 moved was struck by similar globular lightning. 



Globular lightning was imitated by M. Plante 

 with his secondary batteries, and more recently 

 by Herr von Lepel, with static electricity ob- 

 tained from an influence machine. When the 

 pointed conductor ends were held at certain dis- 

 tances from the opposite sides of an insulated 

 plate of mica, ebonite, or glass, small luminous 

 balls appeared moving about, now quickly, now 

 slowly, or sometimes still. Even better effects 

 were had with a glass or paper disk which had 

 been sprayed with parafnne. Small particles of 

 liquid or dust seemed to be the carriers of the 

 light. A slight air current made the spherules 

 disappear with a hissing noise. 



Measurements of different auroras by A. Paul- 

 sen at Godthaab give heights varying from 0'6 

 to 67*8 kilometres. A series of observations made 

 by Garde and Eberlin at Nanortalik, near Cape 

 Farewell, gives from 1-6 to 15-5 kilometres. The 

 results obtained by the Swedish International 

 Expedition at Spitzbergen vary from 0-6 to 29-2 

 kilometres. These observations, therefore, lead 

 to the conclusion that auroras are not confined 

 to the highest parts of our atmosphere, but occur 

 at all altitudes. Paulsen gives accounts of several 

 appearances of auroras beneath the clouds and 

 the summits of mountains. Flogel gave as limits 

 from 150 to 500 kilometres ; Reismann from 800 

 to 900 kilometres ; Nordenskiold, a mean height 

 of 200 kilometres; while Lemstrom has seen 

 auroras as low as 300 metres, and Hildebrandsson 

 has seen them in a clouded sky. Paulsen thinks 

 that in the temperate zone auroras appear only 

 in higher strata, while in the auroral zone they 

 appear in lower strata of the atmosphere. 



Winds. The sea breeze is discussed as one of 

 the minor climatic features of New England by 

 W. M. Davis, L. G. Schultz, and R. DeC. Ward 

 in the Observations of the New England Meteor- 

 ological Society. Its occurrence depends on the 

 general weather of the region ; it appears most 

 distinctly on warm, clear, quiet days, and is ab- 

 sent on cool, cloudy, and rainy days, and on days 

 with strong winds of any direction. It comes 

 into the shore from the sea, working its way 

 against a belt of calm air, as is the case with the 

 tropical sea breeze, and it exhibits the veering 

 with the sun as the day passes that is noticed 

 with winds of its kind elsewhere. It reaches the 

 shore commonly between eight and eleven o'clock 

 in the forenoon with a velocity of ten or fifteen 

 miles an hour. Its velocity rapidly diminishes 

 inland. Its inland advance from the shore line 

 is made at first at a rate of from three to eight 

 miles an hour, but slower afterward when ap- 

 proaching its greatest penetration of ten or 

 twenty miles in the late afternoon. It produces 

 a distinct and agreeable depression of tempera- 



ture on the coast, but this effect is not carried 

 inland as far as its wind extends ; neither is the 

 effect as great as that produced by the "sea 

 turn," or easterly cyclonic wind of our coast. 

 The district of most persistent occurrence and 

 penetration of our sea breeze is from Boston to 

 Cape Ann, along what is known as the " North 

 Shore," where the northeast trend of the coast 

 line favors its development in combination with 

 the prevalent southwesterly wind of summertime. 

 South of Boston or Cape Ann, the southwesterly 

 wind often reverses it or drives it away in the 

 afternoon. The origin of the breeze is to be 

 looked for in the diurnal excess of the tempera- 

 ture of the air over the land above that over the 

 sea. The breeze is part of a littoral convectional 

 circulation ; but in the morning, while the tem- 

 perature over the land is rising rapidly and the 

 convectional circulation is in process of estab- 

 lishment, the outward expansion of the land air 

 holds the incoming breeze off-shore for a time, 

 thus causing its first appearance to be not close 

 on the coast line, but in the offing like " a fine, 

 small, black curl upon the waters, as when all 

 the sea between it and the shore not reached by 

 it is as smooth and even as glass in comparison," 

 as Dampier long ago observed. 



The hot southwesterly day winds of Kansas 

 are injurious to many crops. Mr. G. H. Allen re- 

 gards the quality of intense heat in them as of 

 purely local origin, and supposes that they are 

 caused by a lack of moisture in the earth and the 

 air above it. They may therefore be prevented by 

 any means which will supply the earth and air 

 with abundant and constant moisture. This view 

 is strengthened by the fact they are never felt 

 during the night; but when they are most se- 

 vere the nights are coolest and most pleasant. 



The first prize of the American Meteorological 

 Society for an essay on tornadoes has been 

 awarded to Lieut. J. P. Finlay, who has reached 

 the following general results : Tornadoes gener- 

 ally accompany an area of low barometer. Their 

 progressive motion to the northeast arises from 

 the fact that as they always form in the south- 

 east quadrant of an area of low barometer, they 

 must come within the influence of the general 

 drift of the atmosphere on that side of the low 

 barometer, which is always to the northeast. A 

 hail storm is an incipient tornado in the cloud 

 region of an area of low barometer. As the area 

 of low barometer progresses eastward, the region 

 lying on an average about 350 miles to the south 

 and east of the general storm is the region with- 

 in which tornadoes may be expected. Tornadoes, 

 with hardly an exception, occur in the afternoon, 

 just after the hottest part of the day ; the de- 

 structive power of the wind increases rapidly from 

 the circumference of the storm to its center. The 

 months of greatest frequency, as determined from 

 a period of over two hundred years, are April to 

 July ; the average frequency of the storms does 

 not appear to have changed within that time. 

 The shortest time occupied by the tornado cloud 

 in passing a given point varies from an instant to 

 about twenty minutes, the average time being 

 seventy- four seconds. 



Two incidents illustrating the effect of local in- 

 fluences in tornadoes are related in the "American 

 Meteorological Journal " by Mr. A. Sharpless, of 

 West Chester, Pa. The first is on the authority 



