4H 



NA TURE 



[Augttst 30, 1888 



Dense glass shows the same defect, besides its great liability to 

 become tarnished. 



Another advantage possessed by phenyl -thiocarbimide is its 

 very high boiling-point, viz. 222 C. It is practically non- 

 volatile at ordinary temperatures, and a lighted match may be 

 put into it without setting it on fire. It has, of course, the 

 pungent smell characteristic of all the mustard-oils (to which 

 class it belongs) ; but this, from the slight volatility of the 

 substance, gives no practical inconvenience. 



It dissolves iodine freely, but, from the complexity of its 

 molecule, we cannot expect it to be so diathermic as carbon 

 disulphide. It also readily dissolves metacinnamene (one of the 

 most highly-refractive resinous substances that I have been able 

 to meet with ; fx n = 1 "6, nearly), and the viscous solution is use- 

 ful in determining refractive indices by Wollaston's total reflexion 

 method, or Bertrand's modification of it. 



It can be safely used in the ordinary hollow prisms, as it has 

 no action on the mixture of isinglass and sugar with which these 

 prisms are cemented. H. G. Madan. 



Eton College, August 28. 



Michell's Problem. 



I HAVE read with considerable interest the short letter in 

 Nature of August 9 (p. 342), in which Mr. Joseph Kleiber 

 refers to a paper of his own on the controversy between Michell 

 and Forbes, and notices what lie believes to be a mistake in my 

 paper of July 19 (p. 272). Mr Kleiber shows that the experiments 

 of Forbes on random distribution by scattering rice over a chess- 

 board, and also some additional experiments of his own on 

 numbers taken successively from a table of logarithms, are in 

 accordance with the ordinary formula ior finding the probable 

 number of squares containing r grains, where m is the number 

 of squares and n the total number of grains — 



tnp r — m \n (i/m) r (i - i/m)"~ r /\r \n-r. 



He concludes that "the theory of probabilities does not affirm 

 that ' a perfectly uniform and symmetrical disposition of stars 

 over the sky would (if possible) be that which could alone afford 

 no evidence of causation or any interference with the laws of 

 random.' " 



Forbes, throughout his paper, is not attempting to controvert 

 either the theory of probabilities, of which he himself makes 

 frequent use, or the result arrived at by Michell and Struve, but 

 only Michell's method of applying the theory of probabilities 

 to prove his point. Hence it seems to me that Mr. Kleiber's 

 paper is not so convincing as he takes it to be ; one may agree 

 with his experiments, mathematics, and conclusion, without 

 admitting the truth of Michell's argument. 



Mr. Kleiber's second objection is, I think, founded on a mis- 

 conception due possibly to too great regard on my part for the 

 exigencies of your space. As an example of distribution, I 

 suggest a number of stars shot at random from the centre upon 

 the interior surface of a sphere. The idea may be roughly repre- 

 sented by the explosion of a small uniform shell of shot in the 

 centre of a globe lined with clay. I then attempt to prove that 

 the chance of exactly uniform distribution is nil, and proceed : 

 " Michell, however, seems to assume this probability to be 1, 

 or certainty." Mr. Kleiber strongly repudiates any such assump- 

 tion on the part of Michell. It is always, I admit, a little 

 doubtful to attribute to anyone an opinion which is not dis- 

 tinctly stated, but to Forbes, as well as to myself, the assump- 

 tion seems to be clear. For if the distribution is not uniform, 

 and any groups of stars are formed, Michell's argument applies 

 in its entirety, and he would prove, of course with greater or 

 less probability, a posteriori that the arrangement is due to a 

 cause, while a priori from the datum of shooting out at random, 

 the distribution is due to chance. Sydney Lupton. 



Remarkable Rainbows. 



On Saturday, at 3.15 p.m., there was a very brilliant primary 

 rainbow, and a faint secondary bow above. Inside the primary, 

 at first to the right of the centre, afterwards over the entire 

 centre, were two other very faint bows, their colours in the 

 same order as those of the primary, but with no distinguishable 

 red, the violet of the upper bow seeming to touch the orange of 

 the bow below in each case. Green was the most striking colour 

 in the two inner bows, whose breadth appeared equal to each 

 other, but considerably less than that of the primary ; part of 



this effect being due to the loss of the red, probably all the 

 remainder to irradiation. The perfect primary arch lasted fifteen 

 minutes ; an arc of the eastern side half an hour. The sun 

 being comparatively high, the centre of the arch was low, and 

 the bow looked flat. There was no wind, and many of the 

 rain-drops were large, others mere dots. 



At 5.5, during a thunder-shower, there was a fairly bright 

 primary with perfect arch, and two faint arcs of a secondary 

 bow, but I saw no trace of inner bows. This primary differed in 

 height and brightness from the other, and the rain was a 

 downpour of heavy drops. 



Judging by relative brightn -ss, the inner bows should seldom 

 be seen. L. J. H. 



Rock Ferry, August 27. 



I have been unable to send sooner the following, which you 

 may perhaps think worth inserting in Nature. 



On the 18th of July, at 7.30 p.m., I saw a most remark- 

 able rainbow. A sudden light fell on the book I was reading, 

 so powerful that I thought it must be some neighbouring house 

 on fire. It was a rainbow coming across the mountains opposite 

 (Savoie), and ending in the lake just at the " Bee de Peilz," 

 which some of your readers will know. It was only a section 

 of the rainbow, and was not continued in any other part of the 

 sky, and it was so small a section that it scarcely appeared bent, 

 but looked like a fiery column coloured as a rainbow, but having 

 the peculiarity of not sJioiving the mountain through it : it cut it 

 sheer off, and yet the mountain was looking unusually dark, 

 and indeed the brilliancy of the sunset was such that all came 

 out in strong relief. The sky was covered with stormy c'ouds 

 with breaks of brightness, and above this column they hung in 

 golden radiance such as only painting could faintly convey an 

 idea of. Certainly I never saw so beautiful or curious a sight. 

 It lasted about six minutes. M. C. C. 



La Tour de Peilz, August 17. 



Sun Columns. 



I have nevtr before seen the phenomenon of sun columns in 

 such splendour as on the nth inst. The day wa-; very hot, the 

 wind a pretty stiff westerly one, and the sky perfectly cloudless. 

 After sunset, which (according to the calendar) took place at 

 7h. 27m. p.m., several sun columns became visible. They were 

 seen to grow in length, and at 7I1. 40m. they extended over the 

 whole sky. The columns were five in number, and pretty 

 regularly distributed, so that one passed through the zenith, two 

 on the north, and two on the south of it at equal distances. A 

 very small cloud was visible at that time in the west-north-west. 

 The colour of the shades was dark blue, and their width in 

 the zenith from 2° to 4 . The lighted parts of the sky had a 

 pale violet colour. The rays extended over the whole sky like 

 meridians on a globe, and all five columns were seen to meet in 

 one point in the east-south-east, about 5° above the horizon. 

 The phenomenon could be seen well in all its extension, as I 

 watched it from a hill 688 metres above the sea-level. The 

 intensity of the colour of the colu ims was at its highest at 

 7h. 45m. (Prague local time), and it disappeared at 7I1. 50m. 



I hope, Sir, that you will be able to mention this not very 

 common phenomenon in the columns of Nature. 



St. Benigna, Bohemia, August 14. B. Brauner. 



Meteor. 



The most brilliant meteor I have ever seen flashed across the 

 sky h re from east to west, about 7. IO p.m. on the 30th ult. I 

 was riding along a dark road, looking downwards, when 

 suddenly the road was so brightly lit up that I thought the 

 lamp-lighter had lit another lamp. Seeing neither lamp nor 

 lamp-lighter, I looked up, just in time to catch a glimpse of the 

 meteor. It was of an intense white colour, with a train or track 

 of white behind it. When about 45 above the horizon, it 

 appeared to burst like a sky-rocket, but not so violently. It 

 lasted about two seconds. H. W. L. HlME. 



Coonoor, Madras, August 1. 



P. S. — August 2. Since writing my letter of yesterday I am 

 informed by Lieut. M. de Montmorency, Hampshire Regiment, 

 that the meteor I mentioned burst with a loud noise. I can only 

 suppose that the noise of my horse's hoofs prevented me from 

 hearina it.— H. W. L. H. 



