492 



METEOROLOGY. 



bright ones has attracted considerable attention. 

 The phenomenon has been referred to photo- 

 graphic reversal, due to extreme brilliancy; to 

 a predominance of infra-red radiations: to the 

 existence of flashes deficient in actinic rays; to 

 changes in the density of the air. occasioned by 

 the spark when a dark line with a bright line 

 within it is shown if the air is compressed and 

 a dark line inclosing a light one if it is rarefied; 

 and to some qualities of the photographic plate. 

 The first real light was cast upon the subject 

 by experiments described by Mr. A. W. Clayden, 

 who, having photographed some electric sparks 

 of different intensities, before developing the 

 plates, exposed them to the diffused light of a 

 gas flame. The brilliant sparks then yielded 

 images which might either be called normal with 

 a reversed margin or reversed with a normal 

 core, while the fainter images were completely 

 reversed or, in other words, came out darker 

 than the background. The "fogging" of the 

 pictures to produce the reversal must be done after 

 the image of the flash is impressed, for if it is 

 done before the image appears lighter than the 

 background. This effect, called " the Clayden ef- 

 fect." is accepted as a satisfactory explanation 

 of the phenomenon by tw T o of the authors who 

 have most studied it Dr. W. J. S. Lockyer and 

 Prof. K. W. Wood, of the. University of Wiscon- 

 sin. Prof. Wood, on repeating Mr. Clayden's 

 experiments, obtained dark flashes without any 

 difficulty, but as they failed to appear when the 

 light of an incandescent lamp was substituted 

 for the electric spark he concludes that there is 

 something in the spark essential to the reversal. 

 Dr. Lockyer summarizes his conclusions by say- 

 ing that dark lightning flashes " do not exist in 

 Nature, but their appearances on photographs are 

 due to some chemical action which takes place in 

 the gelatin film." 



A peculiar form of lightning sometimes revealed 

 by the photograph is the ribbon flash, of which 

 a photograph was obtained by Mr. W. H. Os- 

 bome, of Chardon, Ohio, in the summer of 1898. 

 The picture showed three flashes, of which a dis- 

 tant and faint one at the right and a bright one 

 at the left were simultaneous, while the middle 

 one occurred a few seconds earlier. The thunder 

 was light, yet the bright flash had come to the 

 earth not more than 15 rods from where the 

 observer stood. No marks of the discharge were 

 found upon the surface of the earth. The ribbon 

 of light (the bright flash at the left) was about 

 8 feet wide. It was seen to consist of 6 lines, 

 approximately parallel, of unequal brightness, a 

 pair of lines being at each edge and a pair near 

 the middle. The space between these pairs was 

 occupied by many nearly horizontal lines and a 

 few oblique ones, while that between the right- 

 hand pair was crossed by oblique lines only. The 

 horizontal lines at the right of the center were 

 curved downward; and this, with the increased 

 brightness of the whole toward that side sug- 

 gested that the ribbon of light did not lie in a 

 plane, but was concave toward a point at the 

 observer s left. It was also evident that the rib- 

 bon did not stand at right angles with the line 

 of sight, but was nearer the observer at the right- 

 hand edge. A similar lightning flash was de- 

 scribed and pictured in the Electrical World for 

 Oct. 28, 1898, by Mr. A. F. Kennelly, who sug- 

 gested a natural explanation for it based upon the 

 condition of the air produced by the wind and 

 the electrical discharges passing through it 

 Trowbridge, of Cambridge, suggests that 

 many apparent phenomena of lightning may be 

 >f purely optical or physiological origin 



Weather Forecasts. The possibility of a sci- 

 entific w-eather prediction for several days in ad- 

 vance has been discussed by Dr. Van Bebber, with 

 special reference to the requirements of agricul- 

 ture. Instead of dealing with the more mobile 

 areas of barometric pressure, as is the usual way, 

 the author keeps in view the behavior of the 

 simpler and more persistent areas of high pres- 

 sure, among which he distinguishes five principal 

 types in Europe, according to the position of the 

 anticyclonic lines and the direction of the de- 

 pressions from them. Tables are given showing 

 the frequency in days of the weather types dur- 

 ing twenty years and their mean duration. The 

 cases in which the high pressure is situated in 

 the western half of the horizon are more fre- 

 quent than those in which it lies to the eastward, 

 especially in summer. The duration of the differ- 

 ent types varies considerably in the twenty years' 

 period, but the average time is about three days. 

 Although the study refers particularly to the 

 weather of Continental Europe, the method might 

 be employed with advantage in other countries. 



In a paper relating to polydiurnal weather 

 types and forecasts Mr. Douglas Archibald has 

 pointed out the existence of types of weather 

 which persist for several days before breaking 

 up, and for which the meteorological conditions 

 can be definitely foreseen. These types have been 

 discussed by Profs. Koppen and Van Bebber and 

 the late Hon. Ralph Abercromby, who divided 

 them into 20 specific sorts. The author shows, 

 that they may all be included as subtypes of 4 

 primary kinds of weather. Their average dura- 

 tion in England is about four days. Some of 

 them seem to succeed other types or to recur. 

 A science of w r eather types is therefore growing 

 up by which even now the weather may, with 

 due regard to a sudden change of type, be pro- 

 visionally forecasted in general terms, and par- 

 ticularly for agricultural purposes, for half a 

 week or more. 



One of the great needs of the weather services, 

 of the United States and Europe, particularly 

 with reference to forecasts, is that of stations 

 in the oceans. The islands of the Atlantic afford 

 sites where a few such stations might be pro- 

 vided, though if they were all occupied they would 

 be long distances apart, and still have vacancies 

 which would have to be filled before a complete 

 service could be constituted. The Central Weather 

 Bureau at Paris has been for several years re- 

 ceiving weather telegrams from the Azores, where 

 two stations have been established by the Portu- 

 guese Government at the suggestion of the Prince 

 of Monaco. One of these stations is on the island 

 of San Miguel, in the eastern part of the archi- 

 pelago, and the other is on the island of Flores, 

 about 300 miles west of this. The station of San 

 Miguel has been telegraphically connected since 

 1893 with several of the meteorological stations, 

 of Continental Europe, and has a supply of instru- 

 ments, which are added to every year. The sta- 

 tion at Flores, established in 1897, is about 1,200 

 miles from the European Continent, afld, being 

 farther advanced in the ocean, is regarded as 

 more important than that of San Miguel. By the 

 aid of these observations the weather stations 

 of Europe will be able to receive notifications of 

 important storms fifty hours in advance. 



The investigations of the meteorology of the 

 ocean by the British Meteorological Council in 

 1898 related to the meteorology of the southern 

 ocean between the Cape of Good Hope and New 

 Zealand and the meteorology of the South Atlan- 

 tic and the west coast of South America. The 

 council's report represents that the results of the 



