392 



NATURE 



{Feb, 23, 1888 



fog-veiled horizon, perhaps some degrees deeper, and it may be 

 quite as possible that i have measured the diameter as a chord. 

 This supposition gives r = ^ 76° or 38°. The mean of these two 

 determinations is 38° 48', with a probable error surely not less 

 than ± 48', or half the difference. 



At 6h im. p.m., the circumstances were more favourable. I 

 measured Q.a = 76° 11', H = 8' ii', the breadth of the bow 2°. 

 The latitude was 67° 7'. From these data we have, by (2) 

 f — 38° 50'. Assuming the observed amplitude to have been the 

 diameter, which is very probable, we have r = 38" 5' -5. The 

 mean value is 38° 28' ± 22'. 



If the fog bow, like the rainbow, has always the same dia- 

 meter, we can join the three values thus found for the radius 

 into a mean result. We have thus, giving the single determina- 

 tions a weight inversely as the squares of their probable errors — 

 1875 August 7, 2h. 46m. r = 38° 38' ± 6' 

 1878 August 30, 5h. 20m. ;- = 38° 48' ± 48' 

 1878 August 30, 6h. im. r = 38° 28' ± 22' 



When I saw the fog bow, I had, I am sorry to say, no polari- 

 scope, so that I was unable to investigate the polarization of its 

 light, so characteristic for the rainbow. H. MoHN. 



"christiania, January 31. 



The Shadow of a Mist. 



The reticulated rippling shadow of the mist described in 

 Mr. Fawcett's letter (Nature, January 5, p. 224) reminds 

 me of a somewhat parallel phenomenon seen by me a few years 

 ago I saw a snow-storm some miles away crossing the valley be- 

 tween the Mendips and the Quantocks. It hung like a long, 

 heavy curtain partially obscuring the Mgki western sky. The 



Mean 



r = 38°38'±i'-4 



The breadth of the bow being 2°, with a probable error of ^V 

 or ± 6', we get — 



for the outermost red ring r = 39° 38' ± 6' "2 

 for the innermost blue ring r = 37° 38' ± 6''2, 



At 7h. p.m. the bow stood white against the blue sky, the 

 Sandhorn below it. At 6h. 40m. p.m., and sometimes before, 

 I remarked that my own shadow was visible on the fog wall. 

 In order to get a wider view of the phenomenon, I went up upon 

 the roof of the chart-house, where my eye was 27*5 feet (8'4m.) 

 above the surface of the sea. From here I saw how my shadow 

 distinctly imitated all my movements. The shadow of my head 

 appeared dark on a lighter white ground, and from a certain 

 distance surrounded by a concentric coloured glor}% in which the 

 colours were arranged in the order of the spectrum, so that the 

 outermost circumference was red, the middle yellow, and the 

 innermost blue. There was no white band in the glory. With 

 the sextant I measured the radius of the yellow ring, which was 

 the most intense, at 1° 31'. The intensity of the other coloured 

 rings was too feeble to allow their radius to be measured with the 

 sextant. From a comparison with the radius of the yellow ring 

 I judged that of the blue at 1° 15', and that of the red at 1° 45', 

 with a possible error on both sides of ± 5'. This phenomenon 

 is Ulloa's Ring. 



Taking all my results together, we have the following synop- 

 tical table : — 



Ulloa's Ring. Radius of inner blue 

 „ ,, ,, ,. yellow 



I 15 ±5- 



1° 31' ± 2'. 



,, ,, ,, „ outer red 1° 45' ± 5'. 



Fog Bow ,, ,, inner blue 37° 38' ± 6'. 



,, „ ,, „ middle 38° 38' db i'. 



„ „ „ „ outer red 39° 38' ± 6', 



The fog bow cannot be the rainbow with three or four inner 

 reflections, as these rainbows, if visible, would not be anthelic but 

 have, for the red rays, distances from the sun of 42° and 43°. 

 Moreover, the intensity of the fog bow is too considerable to be 

 the result of so many reflections in drops of rain. The ordinary 

 or first rainbow, with one inner reflection, has a radius or distance 

 from the anthelic point of 42° 30' for the red, and 40° 30' for the 

 violet rays, which gives, the sun's radius being 16', its innermost 

 radius like 40° 14'. The outermost red ring of the fog bow has a 

 radius of 39° 38'. Its distance from the ordinary rainbow is con- 

 sequently only 36'. This space we see sometimes covered by 

 the supernumerary rainbows, caused, as Sir G. B. Airy's inves- 

 tigations have shown, by the interference of the rays leaving the 

 raindrop. 



It seems probable that the smallness of the fog globules as 

 contrasted by the larger size of the raindrops, must enter as' an 

 essential part in the explication of the fog bow. In Giinther's 

 "Lehrbuch der Geophysik und physikalischen Geographic," ii. p. 

 151, he speaks of white rainbows, the description of which agrees 

 with the aspect and position of my fog bows, and for which 

 Bravais has given an explanation (" Sur le Phenomene de I'Arc- 

 en-ciel blanc," Ann. de Ckim et Phys. [3], vol. xxi. p. 348). 

 Not having Bravais' memoir at hand, I may only remark that 

 as far as I can see from Giinther, he assumes the fog drops to be 

 hollow, a supposition which is hardly in accordance with modern 

 investigations. 



'■'V'JW'JjJM-- 



light shining through the shower showed a fairly regular pattern. 

 On a reddish-brown ground the darker, because denser, parts of 

 the shower took the form indicated roughly by the accompanying 

 diagram. Was the snow falling in spiral streams, and would a 

 similar explanation apply to the shadow of the mist seen 

 by Mr. Fawcett ? Henry Bernard. 



The English Church, Moscow, January 31. 



Instability of Freshly Magnetized Needles. 



I MADE no attempt to investigate the fluctuations of the 

 dipping needle. They seemed to me to pass away after a few 

 minutes, and I therefore took that method to get rid of them, 

 supposing that the phenomenon was well known to other ob- 

 servers The variations that I observed amounted to three or 

 four minutes, I should think. It is true that the dip circle which 

 I used was of an ancient pattern ; as Prof. Riicker says, hardly 

 up to modern requirements. I did not send it back to the 

 maker for adjustment, as Mr. Whipple says he would have done, 

 because it was lent to me, and it was the best I was able 



° Recognizing the fact that we could not expect to get the best 

 results from our outfit, it was deemed best to make only one set 

 of observations at each station, and multiply the number of 

 stations as greatly as possible. This made it necessary to do the 

 work quickly at some stations in order to adapt our time to that 

 of trains, or in order to get the drudgery involved in camp-life 

 done within the twenty-four hours. It is probable that at some 

 stations we overdid the matter, and that tlie observations would 

 have been better if more time had been taken. The dip obser- 

 vations I always regarded as least satisfactory. But all of the 

 work has been published in such a way that its value can be es- 

 timated by anyone interested, and everyone is welcome to 

 place whatever value he pleases upon it. We did the best we 

 could under the circumstances, and the expense was met by 

 private means. , ^ , . , 



The dip circle was returned to Washington when we were 

 done with it, so that I am not n< w in a position to throw any 

 light upon the subject under discussion. For most of the stations 

 at which observations were made, I think the magnetic elements 

 were determined with as great precision as a single observation 

 would give them, and they seem to me to be as valuable as they 

 profess to be, and not very much more. The fact that so little 

 magnetic work had been done in the central part of the United 

 States seemed to me to justify the plan of making the number o 

 stations large, rather than of trying to attain the precision ot 

 observatory methods in field work at a f-^tauon. ^^^^^^_ 



