38^ 



NATURE 



"^August 8, 1878 



These are : — 



1. That it is sometimes difficult to adjust the angle of the film 

 so as to get the best light on it. 



2. It is impossible to vary the distance of the film from the 

 mouth so as to use both loud and faint sounds. 



3. There is no means of adjusting the tension of the film. 



My phoneidoscope, which is free from these defects, and 

 which I have found to work exceedingly well, simply consists of 

 the hand and some soap-suds. The forefinger and thumb being 

 bent so as to form a circle, a soap film is drawn across them 

 with the other hand. By turning the wrist, the angle which the 

 film makes with the direction of the light can be accurately 

 adjusted, 



A motion of the elbow alters the distance from the film to the 

 mouth, and by slightly separating or bringing together the finger 

 and thumb, the tension of the film can be exactly regulated so as 

 to give any degree of sensitiveness that may be desired. 



The extra delicacy obtained by this adjustment much more 

 than counterbalances the absence of the tube and mouthpiece. 



Pixholme, July 30 J. E. H. Gordon 



Spectrum of the Electric Qablochkoff) Light 



I WOULD suggest that when your readers visit Paris they 

 should take their pocket spectroscopes. They will find a very 

 interesting spectrum in the electric lamps now being used for 

 lighting some of the principal public, places in the city. One 

 might have expected to have found from the brilliant spark 

 inclosed in a white opaline globe a continuous spectrum such as 

 is afforded by the voltaic arc. But the contrary is the case. The 

 Jablochkoff candle now in use in Paris, even when viewed by one 

 of Browning's small pocket instraments, presents a very compli- 

 cated and highly interesting spectrum. I had no chart with me 

 for comparison, nor did I, as I intended subsequently, make even 

 a rough record of the spectrum ; but speaking from memory, I 

 may say that several lines in the blue and green were very 

 marked and distinct, and, in fact, the whole spectrum was 

 traversed by bright and dark lines. I thought, probably, some 

 of these dark lines might be due to absorption by the white 

 opaline glass globe, but I have tested several specimens of this 

 white glass, and I find it does not alter in any way (except by 

 generally reducing its brilliancy) a continuous spectrum, nor 

 does it change the character of the solar spectrum. We must, 

 then, turn to the light itself and to the atmosphere surrounding 

 it for the cause of these phenomena. I believe that in a 

 chemical sense there is no difference between the ordinary electric 

 arc .between the carbon points and the arc of the Jablochkoff 

 candle, except that between the carbon points of the latter is a 

 rod of kaolin, which has, I think, a calcium base. This kaolin 

 is intensely heated by the current, and is volatised at the same 

 rate as the carbon reds by the alternative current which this 

 form of candle requires. The light, therefore, is a combination 

 of the electric and the lime light, the curi-ent taking the place of 

 the oxy-hydrogen elements. The surrounding atmosphere will 

 be the same in both case?, but the products of combustion 

 will obviously be different, and partly so from the composi- 

 tion of the kaolin. Still, I confess that I cannot suggest 

 the cause of this complicated spectrum, and I hope that 

 some observers who have moi^e accurate means and more ex- 

 perience will give us the rationale of the phenomena. 



I may say that there is not at present any Jablochkoff candles 

 to be seen in use in this country, but in the coiu-se of two or 

 three weeks they will be introduced into a large establishment, 

 where excellent means of observation will be afforded. 



Royston House, Tottenham, July 27 E. Walker 



P.S. — Since forwarding the above I have observed the Loutin 

 light now on view at the Gaiety Theatre. The spectrum is 

 somevv-hat similar to that of the Jablochkoff light, but much less 

 distinct. This is probably owing to the circumstance that at 

 the Gaiety the arc is inclosed in a small opaline globe, which is 

 itself encased in an ordinary ground glass lanthorn (the proper 

 lamps came to grief in transit), this diffusive ground glass 

 causing, by overlapping, the indistinctness. Still there are 

 absorption bands and some remarkable bright lines, ^liich, 

 with my small pocket instrument, I will not attempt to define. 

 Nor, as it is a matter for careful observation, will I speculate 

 further than to suggest — seeing that the Loutin light is from the 

 carbon points only — that the ^hite opaline glass may exercise a 

 selective power over the spectrum given by this high state Tof 

 incandescence which it does not in ordinaiy cases, and may give 



us also, to an extent, the actuahwave due to a particular element 

 rather than its obscuration. If so the Loutin light should differ 

 somewhat from the Jablochkoff light, being deficient of the 

 kaolin. E. W. 



The Meteor Showers of July 



The prominent shower of Aquarids mentioned in my letter 

 in Nature, vol. xviii. p. 356, had become extremely feeble on 

 July 31 and August i, for of 136 shooting stars seen on those 

 nights only three or four were conformable to that radiant point 

 which, from a careful re-examination of all the paths recorded 

 from it, is situated exactly at 341° — 13°, near S Aquarii (from 

 fifty-four meteors). 



Between July 26 and August 2 403 shooting stars were re- 

 corded here, of which no less than sixty-three (including one 

 perfectly stationary) belonged to a very sharply-defined radiant 

 near x Persei, at 32° + 53°. Fourty-four of these were noted 

 on the three nights, July 30-31 and August i, when the shower 

 appeared to attain its full intensity. The meteors were very 

 swift with short paths (of about 7°), and almost invariably left 

 streaks of 3° or 4°. They were shorter and less bright than the 

 August Perscids at 43° -f 58°, and in strong contrast to the long, 

 slow meteors of Aquarids seen on the few preceding nights. 

 This very rich stream at 32° + 53° has escaped previous detec- 

 tion, for, being near the date and position of the August 

 Ferseids, there can be no doubt that its meteors have in past 

 years been attributed to that well-known shower, and given it 

 an undue extension of period. The two radiants are, however, 

 quite distinct, and it is now easy to explain the statements of 

 some observers that there are many Perseids visible during the 

 latter part of July. I believe that but few of the old Perseids 

 can be seen before August 6 or 7. My own observations this 

 year show that only seven or eight were seen before August 2, 

 though I watched that region in which the radiant lies very 

 carefully, and noted 400 meteors amongst the constellations 

 there ! 



Thus at the end of July we may expect two special meteor 

 showers : one of Aquarids, at 341° — 13°, the other of Perseids, 

 32° + 53°. The former comes to a maximum two or three 

 nights earlier than the latter, which may be called the ^^ Perseids 

 II.," in order to distinguish them from the old Perseids of Heis. 



Ashleydown, Bristol, August 3 VV. F, Denning 



The Rainfall of Brazil and the Sun- Spots 

 An examination of the scanty records of rainfall obtainable 

 in Brazil proves that the relation between rainfall and sun-spots, 

 which has been pointed out in India by Dr. Hunter and others, 

 holds good for the inter-tropical portion of Brazil. 



The only stations from which I have been able to obtain 

 records for a series of years are the city of Fortaleza (better 

 known in Europe as Ceara), in latitude 3°42'S., and Rio de 

 Janeiro, in latitude 23' S. The rainfall of these two stations is 

 shown in series in the following table : — 



The northern provinces of Brazil outside of the Amazon 

 valley, and notably that of Ceara, are subject to severe and pro- 

 longed droughts, of which that of 1877 is one of the most 

 tet-rible on record. The annals of Ceara make mention of 

 thirty years of drought since 1 71 1, many of which, however, 

 were only partial or slight, and many of which occurred in 

 groups of consecutive years, there being one group of five dry 

 years, another of four, and four groups of two years each. 

 Twelve notable floods are also recorded smce 1776. Ihe 

 droughts and floods are distributed as follows, among the groups 

 of the years of the sun-spot cycle, proposed by Dr. Hunter :— 



