torical, and general iniormaiion concerning uic isiauu, cumjjucu 

 from official and other trustworthy records by Mr. S. P. Musson 

 and Mr. T. Lawrence Roxburgh. 



We have received the first number of a new journal published 

 at Oporto, and entitled Aitnaes de Sciencias Naturaes. The 

 articles are mostly written in Portuguese, and among them we 

 notice one on the flora of Oporto, and another on the birds of 

 Portugal, as well as numerous notes on natural science 

 matters. 



Messrs. Macmillax and Co. have issued the thirty-first 

 volume of the "Statesman's Year-Book," edited by Mr. J. 

 Scott Keltie. The statistics have been well revised, and re- 

 newed in cases where recent information rendered such a course 

 desirable. These changes, and the many additions that have been 

 made, bring the volume in touch with current topics and maintain 

 its character as an indispensable work of reference on all statisti- 

 cal and historical matters relating to the States of the world. 



Several new crystallised compounds of hydroxylamine with 

 the chlorides and sulphates of cobalt and manganese have been 

 isolated by Dr. Feldt in the laboratory of the University of 

 Berlin. The chlorides are analogous to the salts containing 

 zinc, cadmium, and barium described some few years ago by 

 Crismer, being constituted upon the type RCl... 3NH0OH. The 

 sulphates, however, only contain one molecular equivalent of 

 hydroxylamine, but contain two molecules of water of crystal- 

 lisation. The compound CoClo.aNHjOH is obtained by digest- 

 ing in a flask through which a current of hydrogen is passing, 

 and which is heated by a water bath, an alcoholic solution of 

 cobaltous chloride with four molecular equivalents of hydroxy- 

 lamine hydrochloride and a few cubic centimetres of an alcoholic 

 solution of free hydroxylamine. Air requires to be excluded, 

 IS brown subsidiary products are otherwise produced. The 

 liquid after a short time deposits the new compound in beautiful 

 rose-coloured acicular crystals, which are fairly stable, and may 

 be preserved for months out of contact with the air. They 

 detonate somewhat violently, however, when heated, owing to 

 sudden decomposition. The manganese salt MnCU-aNH.iOH 

 may be similarly obtained, and is more stable than the cobaltous 

 compound. It explodes at 160°. The sulphates cannot be pre- 

 pared in alcoholic solution, owing to the sparing solubility of 

 the constituent sulphates in alcohol. By employing aqueous 

 solutions salts of a similar nature are obtained, but with the 

 difference of composition above mentioned. Both the salts 

 CoSO^.NH20H.2H20 and MnSO^.NH.OH.aHoO are similar 

 in appearance to the chlorides, and are considerably more stable 

 in their nature. The most interesting of Dr. Feldt's pre- 

 parations, however, is a salt CoCl3.6NH^OH, analogous to the 

 well-known luteo-cobalt-ammonium chloride. When aqueous 

 or alcoholic solutions of cobaltous chloride and hydro.xylamine 

 are mixed in contact with air, the rose-coloured precipitate 

 rapidly darkens, taking up oxygen in all probability to form the 

 compound C0OCI.2NH0OH. If this substance is suspended in 

 strongly cooled alcohol, and a^similarly cooled alcoholic solution 

 of hydrochloric acid is allowed to fall slowly in, a dark green 

 liquid is produced, which^eventually deposits'a yellow crystalline 

 powder. This precipitate dissolves readily in dilute aqueous 

 hydrochloric acid, and the solution yields on evaporation the 

 luteo-salt in large, well-formed, bronze-coloured crystals belong- 

 ing to the monociinic system. This somewhat remarkable com- 

 pound is a particularly stable substance, which yields a crystalline 

 precipitate of the corresponding oxalate, €0.1(0004)3. laNH^.OH, 

 upon the addition of ammonium oxalate solution. Full details 

 of the work are contributed to the Berichle. 



The publisher of " Der Botanische Garten zu Buitenzorg auf 

 Java," and "Fine Botanische Tropenreise, Indo-malayische 



NO. 1273, VOL. 49] 



XXC13C3tVI££.Cll, IJUll^CLl 111 lllC^C L.W1UI1I1 



last week, is W. Engelmann, of Leipzig. 



The additions to the Zoological Society's Gardens during the 

 past week include a Diana '^lonksy {Cercopithec us ciiaiia 9 ) from 

 West Africa, presented by Miss L. D. Summerbell ; a Wild 

 Cat {Felis cattis 9 ) from Inverness-shire, presented by Mrs. 

 Ellice ; two Collared Peccaries {Dicotylcs tajacii 9 9 ) ; a 

 Globose Curassow (Cra.v ^/o/'/cd'ra (5 ) from British Honduras, 

 presented by H.C. Sir Alfred Molony, K.C.M.G. ; two Cape 

 Bucephalus {Bucephalus capetisis) ; a Cape Viper {Causus 

 rkoDibeatus) from South Africa, presented by Mr. J. E. 

 Matcham ; two Crossed Snakes [Psammophis crucifer); a 

 Smooth-bellied Snake {Ho>n(iloso»ta lutrix) ; a Rhomb-marked 

 Snake {Psavimophylax rhombeatus) from South Africa, pre- 

 sented by Messrs. H. M. and C. Beddington ; a Crossed Snake 

 {Psammophis criicifir); a Hoary Snake {Coronella cana) ; a. 

 Puff Adder ( Vipera arietans) from South Africa, deposited. 



OUR ASTRONOMICAL COLUMN. 



Comet-Spectra as affected by Width of Slit. — 

 The unaccountable difterences between the spectrum of burning 

 or electrically glowing carbon and the carbon bands observed 

 in comets are successfully explained by Prof. H. Kayser in the 

 Astronomische Nachi-ichteu. The chief diff'erences observed 

 between the cornetary and terrestrial spectra are the following : 

 — The carbon flutings in the laboratory have a bright edge on 

 the red side, which in the comet spectrum is displaced 

 towards the red. But the maximum of luminosity in the 

 latter is more refrangible than the bright edge in the former. 

 Whereas in the true carbon spectrum the first fluting is 

 the brightest, in cornetary spectra the second has often 

 appeared brighter than the first. It is suggested that all 

 these anomalies are due to the fact that in. astronomical spectro- 

 scopy the slit cannot be closed so far as in the laboratory, when 

 the objects observed are as faint as comets usually are. If we 

 suppose the true spectrum to be that produced by a very narrow 

 slit, we may reproduce the impure cometary spectrum by slid- 

 ing a wide slit along the true spectrum, and adding up for every 

 position the strips of the true spectrum covered by the slit. We 

 shall thus obtain the portion of the impure spectrum corres- 

 ponding to the centre of the slit. When the wide slit 

 encounters a band with a bright edge towards the red, it will at 

 once begin to indicate a brightness, which will gradually in- 

 crease until the slit is completely filled with light. The 

 maximum will then have been obtained, and we see that it does 

 not correspond to the bright edge, but to a line within it. Thus 

 the first two anomalies are accounted for. Finally, if the slit is 

 so wide that it comprehends two carbon bands at the same time, 

 the maximum will not be obtained when the first or the second 

 band occupies its centre, but when the first is just leaving and 

 the third just entering. This accounts for the third anomaly. 

 The experiment may be easily performed in the laboratory, by 

 observing the arc spectra of calcium or iron. On widening the 

 slit the line spectra of these elements show the same positions 

 for the widened lines, but the carbon bands are diffused towards 

 the red, and their maxima are displaced towards the violet. 



The Astigmatism of Rowland's Concave Gratings. 

 — The astigmatism of the Rowland concave grating gives to 

 this form of spectroscope the advantage of showing no dust lines 

 along the spectrum, and of broadening out the spectrum of a 

 star or a small electric spark into a band ; but the same pro- 

 perty makes it unsuitable for the simultaneous observation of 

 two spectra by the usual method of illuminating one part of the 

 slit with one source of light, and the other part with another 

 source. By a special device. Prof. Rowland has no difficulty in 

 obtaining photographic comparison spectra, but his method 

 only holds good lor photography. In a recent pamphlet by Dr. 

 J. L. Sirks (Amsterdam : Johannes Miiller), however, it is 

 shown that a slight modification of the ordinary method will 

 enable the desired comparison to be made, at least in the first 

 and second order spectra. The comparison prism, or equiva- 

 lent arrangement for introducing a second source of light, needs 

 only to be placed some distance from the slit, at a point de- 



