April 1 6, 1885] 



NA TURE 



565 



The last Bulletin it la Sociiti de Geographie (l a Tnmestre, 



18S5), contains a paper by M. de Mailly-Chalon on a journey 

 in Manchuria. With two countrymen he left Peking for New- 

 chwang, and thence passing to the east of Moukden, through 

 Kirin to Ninguta, where the party turned to the south-east 

 along the Tiumen, towards the ocean, and reached Vladivostock. 

 The' journey the whole way was along the Corean frontier. 

 Leaving Vladivostock the travellers crossed Siberia to 1 omsk, 

 ft. mi which they went to Samarkand. From this point the story 

 of the journey is taken up by another member of the party, 

 Baron Benoist-Mechin, whose paper on the journey across 

 Turkestan succeeds M. Mailly-Chalon's. This journey led them 

 from Samarkand through Karshi, to Bokhara, thence to the 

 rya at Charjui. They followed the river then down 

 Uexandrovsk, whence they deviated to Khiva. From 

 the latter town they retraced their steps up the river, and from 

 Kurgan-Chin started across the Kara-Kum to Merv, and so to 

 Sarakhs and Persian territory at Meshed. The journey, here 

 barely indicated, lasted two years, i.e. from the departure from 

 Japan for Peking to the arrival in Teheran. M. Rabot writes 

 on Nordenskjold's expedition to Greenland, the paper being 

 compiled from the Professor's reports to Mr. Oscar Dickson, 

 published in the Journal ai the Swedish Society of Anthropo- 

 logy and Geography. M. Charles Iluber brings to an end his 

 long journeys in Central Arabia, between 1878 and 1882, to 

 which we have adverted in noticing previous numbers of the 

 Bulletin. 



At the meeting of the Paris Geographical Society on the 7th 

 inst., M. Giraud was received with great distinction, and detailed 

 his recent travels in Africa. The explorer has received the 

 gold medal ol the Society and the Cross of the Legion of 

 Honour. 



ASTRONOMICAL PHENOMENA FOR THE 



WEEK, 1SS5, APRIL 19-25 

 (For the reckoning of time the civil day, commencing at 

 Greenwich mean midnight, counting the hours on to 24, is here 

 employed.) 



At Greenwich on April 19 

 Sun rises, 4I1. 57m. ; souths, Ilh. 59m. 07s.; sets, 19I1. mi. ; 

 deck on meridian, 11° 20' N. : Sidereal Time at Sunset, 

 8h. 53m. 

 Moon (at First Quarter on April 21) rises, 8h. lom. ; souths, 

 l6h. 4m. ; sets, 23I1. 58m. ; decl. on meridian, 18 14' N. 



ponding 

 les from vet- 

 to right for 

 erted image 



The Occultations of Stars and Phenomena of Jupiter's Satellites are such 

 s are visible at Greenwich. 



April h. . , , 



19 ... 1 ... Saturn in conjunction with ant 4 I north 

 of the Moon. 



22 .. 3 •■ Jupiter stationary. o 



23 ... 19 ... Jupiter in conjunction with mid 4 37' 



north of the Moon. 



ON A REMARKABLE PHENOMENON OF 



CRYSTALLINE REFLECTION^ 



Introduction. 



TN a letter to me, dated March 29, 1854, the late Dr. W. 

 -*■ Bird Herepath enclosed forme some iridescent crystals of 

 chlorate of potash, which he thought were worth my examin- 

 ation. He noticed the intense brilliancy of the colour of the 

 reflected light, the change of tint with the angle of incidence, 

 and the apparent absence of polarisation in the colour seen by 

 reflection. 



The crystals were thin and fragile, and rather small. I did 

 not see how the colour was produced, but I took for granted 

 that it must be by some internal reflection, or possibly oblique 

 refraction, at the surfaces of the crystalline plates that the light 

 was polarised and analysed, being modified between polarisation 

 and analysation by passage across the crystalline plate, the 

 normal to which I supposed must be sufficiently near to one of 

 the optic axes to allow colours to be shown, which would 

 require no great proximity, as the plates were very thin. To 

 make out precisely how the colours were produced seemed to 

 promise a very troublesome investigation on account of the 

 thinness and smallness of the crystals : and, supposing that the 

 issue of the investigation would be merely to show in what 

 precise way the phenomenon was brought about by the oper- 

 ation of well-known causes, I did not feel disposed to engage in 

 it, and so the matter dropped. 



But more than a year ago Prof. E. J. Mills, F.R.S., was so 

 good as to send me a fine collection of splendidly coloured 

 crystals of the salt of considerable size, several of the plates 

 having an area of a square inch or more, and all of them being 

 thick enough to handle without difficulty. In the course of his 

 letter mentioning the despatch of the crystals, Prof. Mills 

 writes: "They [the coloured crystals] are, I am told, very 

 pure chemically, containing at most o't per cent, foreign matter. 

 They are rarely observed — one or two perhaps now and then in 

 a large crystallisation ... I have several times noticed that 

 small potassic chlorate crystals, when rapidly forming from a 

 str .ng solution, show what I suppose to be interference colours ; 

 but the fully formed crystals do not show them." 



Some time later I was put into communication with Mr. 

 Stanford, of the North British Chemical Works, Glasgow, 

 from which establishment the crystals sent me by Prof. Mills had 

 come. Mr. Stanford obligingly sent me a further supply of these 

 interesting crystals, and was so kind as to offer to try any 

 experiment that I might suggest as to their formation. 



On viewing through a direct-vision spectroscope the colours of 

 the crystals which I had just received from Prof. Mills, the first 

 glance at the spectrum showed me that there must be something 

 very strange and unusual about the phenomenon, and determined 

 me to endeavour to make out the cause of the production of 

 these colours. , The result of my examination is described in 

 the present paper. 



Section I. — Preliminary Physical Examination. — I. It will 

 be necessary to premise that chlorate of potash belongs to the 

 oblique system of crystallisation. The fundamental form may be 

 taken as an oblique prism on a rhombic base, the plane bisecting 

 the obtuse dihedral angle of the prism being the plane of sym- 

 metry. Rammelsberg denotes the sides of the prism by P, 

 and the base by C, and gives for the inclinations of the faces 

 PP = I04° 22' and CP = 105" 35'. The face C, which is perpen- 

 dicular to the plane of symmetry, is so placed as to bring three 

 obtuse plane angles together at two opposite corners of the 

 parallelepiped. The salt usually forms flat, rhombic or hexagonal 

 plates parallel to the C plane, the edges of the rhombus being 

 parallel to the intersections of the P faces by the C plane, and 

 the hexagons being formed from the rhombic plates by truncating 

 the acute angles by faces parallel to the intersection of the C 

 plane by the plane of symmetry. 



The plane angles of the rhombic plates, calculated from the 

 numbers given by RaLinrielsberg, are ioo" 56' and 79° 4', while 

 the hexagonal plates present end-angles of 100° 56' and (our 

 side-angles of 129 32'. These angles are sufficiently different 

 to allow in most cases the principal plane of a plate, or even of 

 a fragment of a plate, to be determined at once by inspection. 

 But in any case of doubt it may readily be found without break- 

 ing the crystal by examining it in polarised light. There are 



' Paper read at the Royal Society on March 19 by Prof. G. G. Stokes. 

 M.A., Sec. R.S., Lucasian Professor of Mathematics in the University of 



Cambridge. 



