168 



♦ KNOWLEDGE 



[June 1, 1889. 



necessary, as if the same flash were photogi'aphed from two 

 or more points, we should be able to fix the position of the 

 flash in the air, and determine its size. The time interval 

 between the lightning flash and the thunder should always 

 be noted, as this will aflbrd the means of ascertaining how 

 far ofi" the lightning is. For all practical purposes we may 

 allow 5 seconds to a mile. It is desirable, if possible, not 

 to have more than one flash upon a plate. Particulars 

 should also be recorded of the direction of the storm, 

 the movement of clouds, the direction and force of the 

 wind, &c. 



More is likely to be learnt with respect to the structure 

 of these ribbon and dark flashes if they can be photographed 

 on a large scale. There is a limit to the size of the camera 

 which can be conveniently used, but it is evident that the 

 nearer the flash the larger will be the angle that it sub- 

 tends upon the plate. For this reason it is not photographs 

 of flashes upon the horizon that ai-e so much needed as 

 photographs taken in the midst of a storm. With a storm 

 immediately overhead, the points of reference necessary for 

 determining the parallax and height of the flash by com- 

 parison with another photogi-aph might be obtained by 

 noting the place of the camera and including chimneys in 

 the picture, or telegraph-wires or strings stretching across 

 the sky. 



The subject of lightning flashes is only part of the inves- 

 tigation which the R. Meteorological Society has on hand 

 in connection with thunderstorms. The Society is collecting 

 observations on British hail and thunderstorms with the 

 view of gaining some knowledge of the nature and causes 

 of the diiferent kinds of thunderstorms, and discovering the 

 localities where hail and thunder are most frequent and 

 destructive. Much valuable information has been already 

 received which is now being collated and discussed. Mr. 

 Marriott would be glad if the readers of Knowledge would 

 favour him with observations made during thunderstorms ; 

 even simple records of the times when thunder, lightning, 

 rain and hail liegin and end, and when they are loudest, 

 brightest, or heaviest, would be accepfeible. Notices of 

 distant or sheet lightning, with the time of its occurrence 

 and the direction in which it is seen, are also of service in 

 tracing the path of a thundei-storm. This was well illus- 

 trated in the case of the storm which occurred on the night 

 of May 18, 1888. This thunderstorm came up from the 

 English Channel about 8 p.m., passed across the country in 

 a north by we.st direction, and reached the Firth of Forth 

 by 4 A.M. on the 19th. The dist.ance traversed was over 

 400 miles, the rate of progression being 50 miles an hour. 

 Many observers in various parts of the country only saw 

 lightning ; but where they gave the time and direction in 

 which the lightning was seen, it was found that these 

 agreed precisely with the positions of the thunderstorm at 

 those times, although the observers, in two cases at least, 

 were more than 100 miles from the storm. 



The INIeteoi-ological Society is anxious to secure as large 

 a collection as possible of lightning photographs. Eeaders 

 of this paper who wish to assist in the research should 

 forward the original negatives if possible, or, if not, prints 

 from them, to Mr. W. Marriott, secretary of the R. Meteoro- 

 logical Society, 30 Gi-eat George Street, M^estminster, S.W. 

 Any other information bearing on the subject of thunder- 

 storms will also be welcomed by him. It would be well to 

 mark the top of the sensitive plates before they ai-e put 

 into the dark slide for exposure, so that there may bo no 

 doubt as to the orientation of the picture if no buildings or 

 trees are included. But pictures with reference marks, 

 from which the parallax and the actual size and position of 

 the flash c;xn be determined, cannot be too much insisted 

 upon. The time of the flash, and the interval between it 



and the thunder, should, in every instance, be noted, and 

 the observer's watch should be compared as soon after the 

 storm as possible with post-office time. 



TOTAL SOLAR ECLIPSE OF 1889. 



JANUARY 1. 

 Br A. C. Ranyard. 

 IIOTOGRAPH No. 14 on Plate III. is copied 

 from a transparency on a flexible film, which 

 has been kindly sent me by Professor W. H. 

 Pickering. The tran.sparency was made 

 direct from the original negative, and shows 

 a great deal more coronal structure than is 

 reproduced in our illustration. The difference 

 of brightness between the coronal details and the back- 

 ground on which they are seen is so slight that there is 

 always difliculty in reproducing them, and copies of coronal 

 photographs are always much inferior to originals in show- 

 ing such details. 



The original negative was taken at Willow, California, 

 with a lens of thirteen inches aperture, and an exposure of 

 ten seconds on a Carbutt A plate. It is the largest photo- 

 graph of the corona which has yet been taken in the primary 

 focus of an instrument, though in the early days of eclipse 

 photography more than one photograph on a still larger scale 

 was taken in the secondary focus of instruments, but the 

 extent of corona shown in such early photographs was very 

 small as compared with that shown in Professor Pickering's 

 photograph. The sun's axis is as nearly as possible vertical 

 on the page, the north pole being uppermost. It will be 

 noticed that this corona is not quite symmetrical mth 

 respect to the sun's axis, the chief equatorial extension 

 being a little northward of the solar equator on the west 

 side, and towards the south on the east or left-hand side of 

 the picture. There has been a similar want of exact 

 symmetry with respect to the sun's axis in other 

 coronas. The po.sition of the sun's axis is derived 

 from the observation of the motion of sun-spots in the 

 equatorial regions ; but the corona, both in its equatorial 

 regions and polar rifts, has in several instances shown a 

 tendency to symmetry with respect to a line which does not 

 conespond by some 10° or 15° with the axis of rotation as 

 derived from the observation of sun spots. 



In the transparency sent by Professor Pickering, four 

 synclinal groups of structure are distinctly shown, sym- 

 metrically placed with respect to the general axis of symmetry 

 of the corona. The northern edges of the two northern 

 synclinal groups and the southern edges of the two 

 southern synclinal groups show marked contrary flex- 

 ure. The rays bounding the polar rifts are greatly in- 

 clined to the radial at the point where they spring from 

 the sun's limb, as in the case of the 1871 corona, and are 

 concave towards the axis of the synclinal groups in their lower 

 regions and convex above ; a similar contrary flexure is 

 noticeable in the rays bounding the polar rifts of the 1871 

 corona. The corona of January L'^81i corresponds to a period 

 of solar quiescence, as measured by sun-spot disturbances. 

 It is similar in many respects to other coronas, which have 

 been observed at other periods of sun-spot minima. Theie 

 is great equatorial extension of the corona, and there are 

 broad polar rifts. On the transparency sent over by Pro- 

 fessor Pickering, though a great many curved rays are 

 visible, no great tree-formed structures, with spreading 

 heads and narrow stems, can be recognised. Such struc- 

 tures seem to be associated with periods of greater sun-spot 

 development. 



