94 



NATURE 



\May 29, 1879 



has once been obtained and the pupil is able to think in 

 another language than his own, the analysis and study of 

 the idiom should be carried on in the light of comparative 

 philology. He should be taught to see that the appa- 

 rently arbitrary phenomena of language are all subject to 

 strict law, and that the forms and words he uses all have 

 a history and a reason for being what they are. In this 

 way his intelligence as well as his memory will be excited 

 and quickened, his curiosity, that "fountain-head of 

 science," will be legitimately aroused and satisfied, and 

 above all the conception of law will be made familiar to 

 him from the first beginning of his education. When the 

 action of philological laws has been traced and illustrated 

 in modern languages, it will be easy to pass on to the 

 dead ones and show how they are but the earlier forms of 

 living speech, past links in the great chain of imbroken 

 development. 



Before parting from Miss Ottd, allusion must be made 

 to the reformed spelling of Dano-Norwegian and Swedish, 

 which she has adopted in her Manual, and of which she 

 has given an interesting account in her Introduction. 

 This reformed spelling is in the first instance due to 

 Rask, the great Danish philologist, but it owes its present 

 acceptance to the Stockholm Conference, called together 

 in 1869 by the exertions of Prof. Daa, and to the Dano- 

 Norwegian dictionary which was the result of it. The 

 vicious spelling of the past has now been superseded by a 

 consistent and fairly phonetic one, based upon a scientific 

 alphabet. In this respect the Danes have set us an 

 example which it would be well to follow. The " practical 

 men " of Scandinavia have at last condescended to listen 

 to the recommendations Of those who study language 

 scientifically, and the people consequently now possess 

 an orthography which forms no hindrance to learning to 

 read and write and throws no veil over the true nature of 

 speech. Let the Englishman who uses Miss Ottd's 

 Manual try to put himself in the place of a Dane who 

 wishes to learn English, and then consider whether he 

 does all in his power to facilitate the acquisition of at al] 

 events one language by the foreigner. A. H. Sayce 



LETTERS TO THE EDITOR 



[ The Editor does not hold himself responsible for opinions expressed 

 by his correspondents. Neither can he undertake to return, or 

 to correspond with the writers of, rejected manuscripts. No 

 notice is taken of anonymous communications. 



\The Editor urgently requests correspondents to keep their letters as 

 short as possible. The pressure on his space is so great that it 

 is impossible othenvise to ensure the appearance even of com- 

 munications containing interesting and novel facts. \ 



The Spectrum of Brorsen's Comet 



I AM much obliged to Mr. Christie for his answer to my 

 question. There can remain no doubt that Brori^en'.s comet does 

 not now give the same spectrum as, according to Huggins's obser- 

 vations, it did in 1868. The difference in position between the 

 brightest lines in the two spectra of carbon is, it is true, very 

 small, but if it were possible it would be a step gained to decide 

 which of the t«o lines the brightest comet band coincides with, 

 5198 4 or 5l65'S, since as far as experimental evidence goes at 

 present one uf these lines is due to carbon-vapom-and the otherto 

 an oxide of carbon. I fear that Prof. Pi;izzi Symth's theory that the 

 spectrum in question is caused by hydrocarbon must be rejected 

 for experimental reasons which I will presently recapitulate. I 

 have had no experience in cometspectroscopy, not liaving access 

 to any telescope of sufficient aperture, and I do not wish, there- 

 fore, to seem to make hght of the achievement of Mr. Christie 

 and Prof. Young ; but if it w«re possible to adjust the occulting 



bar so as to completely hide (but only fust hide) the least refran- 

 gible edge of the brightest comet- band, then I should imagine 

 that, on flashing in the spectrum of the alcohol-tube, its band 

 would be seen beyond the bar if the comet-spectrum be, as is 

 most probable, that which I have called " Carbon No. I." 



At present the observations at the Royal Observatory seem to 

 point to a coincidence with the second spectrum, but it was the 

 frst which Prof. Young employed and which Prof. Iluggins 

 also employed. A reference to Prof. Huggins's account of his 

 experiments shows that the comparison-spectrum was obtained 

 by taking the electric spark in olefiant gas at tlie ordinary pres- 

 sure ; and he further observes (Quart, your. Science, April, 

 1869) that "the same spectrum is given by the spark in 

 cyanogen." 



The difference between the spark in olefiant gas and in olive 

 oil, shown in Huggins's diagram, is simply one of detail — the 

 separate lines being distinctly seen in the spectrum of the oil and 

 not in that of the gas. 



Prof. Piazzi Smyth's alcohol-tube seems to differ from Mr. 

 Christie's in containing besides the lines of spectrum No. II. (if 

 he will allow me to call it so) the green band seen in the blue 

 base of a candle flame — that is the band beginning with 5l65'5. 



This, I believe, is always the case if the vapour be at a some- 

 what high pressure. A reference to tlie Phil. Trans, for 1865, 

 or the Phil. Mag. for October, 1869, will show that the tubes 

 \\ ith which PlUcker worked contained lines of both spectra — and 

 that he did not succeed in completely separating the two. But a 

 tube containing pure carbonic oxide at a small pressure (one or ^^ 

 two millimetres) shows no trace of this green band even ^Kk 

 "end-on." iKt 



I cannot accept Prof. Piazzi Smyth's theory that this green 

 band and the remainder of the lines in spectrum No. I. are due 

 to hydrocarbon, for the simple reason that they are obtained ^Hji 

 brilliantly from substances which do not contain hydrogen, viz., T^Kk 

 cyanogen, carbonic oxide, and sulphide of carbon. ^^' 



There is no more magnificent spectrum than the "carbon 

 spectrum No. I.," obtained by burning cyanogen and oxygen 

 together at the nozzle of an oxyhydrogen blow-pipe. 



I should like to refer Prof. Smyth for other arguments than 

 my own and for experimental evidence to a paper to be found 

 in the Ann. Ch. Phys. for 1865, t. 4, p. 305. 



Giggleswick, May 27 William Marshall Watts 



A Universal Catalogue 



The last April number of Nature contains an article on a 

 Universal Catalogue, which seems to be still under discussion. 



So great a work, when undertaken, should to a certain extent 

 be complete, so as not to necessitate the same thing having to be 

 done fifty times. With a really universal catalogue of books and 

 memoirs existing, it would be quite easy for each library to form 

 its own catalogue in a much abbreviated form. For instance : — 

 Brewster, Optics. 1831. P. 2350, or 



Hauy, Crist, et Propr. phys. Enclase. 1819. Min. 6430, 

 would be quite sufficient to stand for — 



Brewster, Treatise on Optics. London, 1831. Catalogue of 

 Physical Science Papers, 2350. and 



Hauy, Memoire sur la Cristallisation et sur les Proprietes 

 physiques de I'Enclase, Paris, Mus. Hist. Nat. Mem. v. 1819, pp. 

 278-293. Catalogue of Mineralogical Papers, 6430. 



So in the library catalogues no cross references and main titles 

 would be necessary, and no double and treble lines for titles of 

 books or memoirs, five or six words and two numbers being 

 sufficient to characterise each publication, while now, there being 

 no general catalogue, each library desiring to give its catalogue— 

 an undertaking which is highly desirable— is obliged to spend 

 disproportionate cost, time, and space for such a purpose. 



But scientific workmen would also be much better served in 

 this way, as may be shown by the following facts : — 



Putting the whole number of titles in the British Museum 

 catalogue at 3,750,000 (1,250,000 real ones, 2,500,000 cross 

 references, &c.), they may be classed into old and modern works, 

 the former 750,000, the latter 3,000,000. Now, putting the 

 number of special branches which deserve and imperatively 

 demand special catalogues of subjects — as mathematics, botany, 

 statistics, &c., &c. — at 50, and supposing that old books extend 

 even 15 to 20 branches each, every special branch is represented 

 by 15-»-°-«i>-20 = 300,000 titles of old publications and 5iL»_Ll»_» = 

 60,000 of new ones. 



So, one seeking for books of a single branch does not find 



