LIGHTNING. 262 discharges observed by Kood as constituting one Hash were simply a grand example of Henry's oscillatory dicharge. The truth of this liypoth- esis is coiilirmed by the iiumcruus photographs of lightning that liave been taken williin the past ten years. Une of the most remarkable of these was taken by W. H. Jennings, of Philadel- phia, August 12, 18'J2, from the rear platform of a railroad train in full motion crossing the prairie of North Dakota ; this was simply an example of what is now called 'ribbuu lightning.' A tine illustration of a rililMjn Hash is shown in the accompanying Ugure. This pliotograph was taken by J. " il. Justice, of Franklord. i'a., between 9:00 and 9:30 P.M., August 10. 1899. The plate shows a small single flash to the left of the centre and a vivid broad multiple Hash a lit- tle to the right of the centre. The latter Hash was between 2000 and .'{000 feel distant from the camera; it descended from the cloud to the earth at an inclination of from 20 to 40 degrees, ap- parently striking the ground within a mile to the north, Tlie angular breadth of the flash, as photographed, is about 32 minutes of arc near its summit and al)out IS minutes near the horizon, representing a linear breadth of between 14 and 24 feet. Two distinct explanations have been ofTcred of multiple flashes as seen in photographs. Ac- cording to one, the first flash makes, as it were, a hole in the air, and succeeding discharges quickly pass uj) and down through this tube, while the general wind of the atmosphere carries it along for a few feet. According to the other explanation the hole or tube of rarefied air lasts so short a time that the motion of the wind is inappreciable, but it is of such a great diameter that the successive discharges passing up and down within it side by side, or one within the other, appear as separate streaks of light. Neither of these explanations can be said to be satisfactory. Somewhat similar flashes can be produced artificially by blowing across a dis- charge between two electrodes. Photographs often show a dark flash among the bright ones, but this is known to be a photo-chemical effect on the sensitive plate and not atmospheric. The researches of Trowbridge seem to show that the resistance of atmospheric air to elec- trical discharges becomes less as the voltage increases beyond some high number such as a mil- lion Aolts; when the first or pilot discharge has taken place the air offers less resistance to the succeeding discharges. Under powerful electric stress a resisting medium seems completely to lose its power of resistance and is immediately ruptured. Considt: Lodge. Lifihtirinfj Conduct- ors and Liffhtning flnnrdn (London. 1802), and the bulletin of the United States Weather Bureau on Trotfction from Lightning, by Alexander Mc- Adie. The frequency of occurrence of lightning and thunder storms is published regularly in the United States Monthly Wrathrr licrirw. and the annual summary shows that the number of days on which thunder-storms were reported varies from 100 at any station in Louisiana and Florida to 5 or even 1 per annum at special stations in Arizona, California. Maine. Nevada. North Da- kota, Oregon, and Washington. It is suspected that there is some relation between the aurora and the lightning, since on many occasions numer- ous thunder-storms have been recorded around a LIGHTNING. where clear skies central region where clear skies and atnoras were observed. LIGHTNING, Accidents from. Statistics have been kept for many years, both in Kurope and America, showing the frequency of liglitning strokes and the destruction done by them. Thus, according to the report of the Registrar-General for England, iu 1871, 28 persons were killed; all but 5 were men and were chielly laborers iu the open air. In 1875 17 were killed, and in 1877 only 10. Out of 103 deaths in live years (lSo2-uU) there were 38 in July and 22 in Au- gust. According to Helluiann, in Scbleswig- Holstein. during the decade 1874-83, the annual average of destructive lightning strokes per million of buildings was 163 for slate or metal, roofs, 38G for wooden or thatched roofs, (3277 for chimneys, 8524 for windmills, and 30li for fac- tories and steam chimnej's. The danger to build- ings in the open country is five times greater than in the cities. The number of deaths of human beings on the average of fifteen years per mil- lion is 4 in Prussia and Baden and 3 in France and Sweden. The danger to buildings erected on chalk or marl is very small, but on clay or sand larger. Among forest trees the oak is most frequently struck and the beech least frequently. The records from 1833 to 1882, according to Von Bezold, show that there is a steady increase in the recorded number of strokes ; possibly this may be explained by the increased care in collect- ing data. According to Prof. A, J, Henry, the records of the United States Weather Bureau show that during 1898 throughout the United States, excluding Alaska, the number of death.s by lightning was 307. and the number of in- juries 491. The proportion of deaths to the popu- lation was greatest in the Upper ]VIissouri Valley and portions of the Kocky Mountain region. The proportion of deaths by lightning to the total population was about 5 per million, which is higher than the average of most countries, owing to the large proportion of our agricultural population. Nine hundred and sixty-six barns or sheds, 735 dwellings, stores, or oHices, 95 churches and schools, and 70 other buildings were struck and damaged, the approximate loss being about $1,500,000. Of the buildinf;s struck 40 were provided with lightning-rods, 855 were not ; and in 952 cases this item was not recorded. Nine hundred and sixty-four head of cattle. 300 horses, 30 mules, 426 sheep. 110 hogs were killed, the total value of the stock being .$48,257. The continuous barbed wire fences that are used over the Western prairies are said to increase greatly the number of cattle killed by lightning, as these seek shelter in .some corner of the field and the lightning runs to them along the wire from very considerable distances. LIGHTNING, Pkotectiox from. When a flash of lightning strikes any object on the groimd it generally does some destruction, apparently due to the fact that when the electric discharge passes through a poor conductor it heats that substance intensely. Thus when a discharge passes through a living tree all the sap may be converted into vapor and all the passages filled with air or filled with sap burst open. When the flash rung along a metallic water-pipe or gas-pipe it heats and melts the solder at the junctions. When the flash strikes a shingle roof it sets fire to the wood. or. striking the chimney of a house, it ex- pands the air within the brickwork or evaporates
