June 1, 1889.] 



♦ KNOWLEDGE 



167 



apparently parallel to a fixed line in space through the 

 whole of the loop and the convolutions of the knot or con- 

 torted region where the two flashes join and branch off. 



A similar appearance of parallelism of the stri;e to a 

 fixed line may be recognised in the other photographs of 

 ribbon flashes ; possibly the strise always make the same 

 angle with the horizon, for there is some doubt as to 

 whether pliotographs Nos. 1 and 2 are the right way up 

 upon the page. Photographs of such contorted ribbon 

 flashes from two points would enable us to determine 

 whether there is any connection between the direction of 

 the strite or folds and the earth's magnetic axis, or whether 

 the direction is determined by the prevailing wind or other 

 meteorological conditions. Similar folds are shown in 

 many drawings of the Aurora. Some good woodcuts of 

 Auroral curtains have recently appeared in the German 

 periodical, " Himmel und Erde," according to which the 

 Auroral strise are nearly perpendicular to the horizon. But 

 the drawings were made in the neighbourhood of the earth's 

 magnetic pole. 



If No. 4 be examined carefully it will be seen that in addi- 

 tion to several ordinary white fla.shes, there is a dark flash 

 of precisely the same character as the bright flashes. A 

 suggestion has been made that this may be the result of a 

 very bright flash so over-exposing the plate as to produce the 

 well-known revei'sal efiect by over-exposure, as when the 

 disc of the sun appeals black on a positive print instead of 

 white. Professor Stokes, who has examined the photograph, 

 says : " The dark flash was perhaps produced by a previous 

 flash before the cap over the lens was taken ofT, wliich had 

 produced nitrous oxide along its track. The actual flash 

 photographed would then illuminate the background of cloud, 

 but this light could not pass through the nitrous oxide, and 

 thus would give a dark track due to absorption." If the dark 

 flash were due tn over-exposure, its edges and fainter parts 

 would be bright, but there is no such evidence of photographic 

 reversal having taken place. In the case of the curtain form 

 of flash, No. 2, witli a dark line running along it parallel to 

 the bright edge, this may possibly be due to a previous cur- 

 tain flash in front which produced the nitrous oxide gas 

 which cut out the light of the second flash. 



It is to be hoped that some one will be able to make experi- 

 ments to test this theory. Professor Stokes has suggested that 

 during a thunderstorm in the daytime a camera should be 

 charged and the cap taken ofi" the lens just after a flash has 

 been seen ; then, according to his theory, a dead flash should 

 show itself on the plate. 



More than one photograph of such a flash would be 

 of great interest in showing how the dead flash di.sappears. 

 The tracks left by meteors in the upper air occasionally 

 remain visible for half an hour or moie, and are seen to be 

 bent and drifted by the wind. The behaviour of the dark 

 track might, perhaps, give some clue as to whether it 

 is due to the smoke of bvirnt dust particles floating in the 

 air, or whether it behaves as a column of gas might be 

 expected to behave while being lost by diflusion into the 

 surrounding air. If the dark column is due to smoke a 

 photograph taken with a large double image prism, placed 

 in front of the camera lens, would, if the column of smoke 

 were illuminated by a side light, probably show a diflerence 

 in brightness of the two images, due to polarisiition of the 

 light dispersed by the smoke particles, and we should 

 expect to find such a column darker in the lower air and 

 in cities whore inflammable dust exists than in the case of 

 lightning photographed from a mountain side ; but, during 

 the tliun<lersti)rm season, the air in the open country is 

 frequently clKii'gc-d with polloa to a, considerable altitude. 

 It might lie well to (est the character of the dust by burning 

 the sediment from rain collected durinir the storm. 



Nos. 3, 5, and 6 are further examples of the meandering 

 form of lightning. No. 6, which was taken at Clapham by 

 Mr. J. Guardia, is not only a very good specimen of the 

 ribbon-type, but is a remarkable example of the strange 

 course sometimes taken by lightning. The fla.sh seems to 

 have twisted and turned about in a very curious manner, 

 and to have apparently taken an oval course, and come back 

 upon itself. It would have been very interesting if we 

 could have had another photograph of this same flash taken 

 froin a diflfercnt point of view. We could then have fixed the 

 position of the flash, and ascertained its length. The length 

 of a lightning flash may be some miles in extent. Mr. 

 L. P. Muirhead states that during the thunderstorm 

 which occurred at Glasgow on May 19, 1888, he observed a 

 very long and bright flash which by the Ordnance map he 

 judged to have been six miles long. 



There is a form of lightning of which at present no photo- 

 graphs have been obtained, viz., ball or globular lightning. 

 ■This appears as a ball or globe of fire, varying in apparent 

 size from a cricket ball to a football ; it moves very slowly 

 — in fact, some people have stated that they have been able 

 to get out of its way— and, as a rule, finally explodes with 

 ^eat violence. Arago, in his " Meteorological Essays," 

 gives sevei-al accounts of this phenomenon. Thus at Milan 

 in 1841, one of these globes moved along a street so slowly 

 that the spectators walked after it to watch it, and the 

 narrator, who first observed it from a window, ran down- 

 stairs, and saw it for three minutes before it struck the 

 cross on a church steeple and disappeared. Such state- 

 ments are very frequent, and there is no reason why 

 they should be doubted. Dr. Tripe, of Hackney, says 

 that on July II, 1874, while watching a thunderstorm, 

 he saw a large ball of fire rise apparently about a mile 

 distant from behind some low houses. It first rose slowly 

 then accelerated its pace as it ascended, and gradually 

 acquired a very rapid motion. When it had risen about 

 45°, it started ofT at an acute angle towards the west, with 

 such rapidity as to produce the appearance of a flash of 

 forked lightning. 



A very remarkable account of globular lightning in the 

 Glendowan Mountains, in county Donegal, was reported by 

 Mr. M. Fitzgerald some ye;irs ago to the Royal :Meteoro- 

 logical Society. He s;iys :— " From its fii-st appearance till 

 it buried itself could not have been more than twenty 

 minutes, during which time it travelled leisurely, as if 

 floating with an undulatory motion through the air-, over 

 one mile. It appeared at first to be a bright red globular bail 

 of fire, about two feet in diameter, but its bulk became 

 rapidly less, particularly after each dip in the soil, so that it 

 appeared not moi-e than three inches in diameter when it 

 finally vanished." 



From what has been already said, it will be seen that the 

 photographs have brought out several facts about lightning 

 which liad hitherto been entirely unsuspected. Additional 

 photographs would probably still further increase our know- 

 ledge, and also confirm or disprove some of the conclusions 

 already arrived at. As the thunderstorm season is now ap- 

 proaching, it is to be hoped that many photographers will take 

 up this interasting subject and secur-e a goodly number of 

 photogi'aphs of flashes of lightning. As the plate can only 

 be exposed at night, the lens should be focussed for some 

 distant object dirring the daytime, and a mark made on the 

 camera, so that it can be easily adjusted at night. 



A rapid rectilinear lens with full apertuiv should be used, 

 and if possible some buildirjg or landmark should bo included 

 on the plate to give the position of the horizon. The exact 

 position and aspect of the camera should always be car'efully 

 recorded, as well as the precise time of any flash that may 

 come within the field of view. These pjuticulars are very 



