NATURE 



\_May lo, 1877 



to publish a text-book in accordance with the plan 

 adopted in this pamphlet. The objects are, " first to 

 present a new method of deriving the differentials of 

 functions by means of their algebraic characteristics with 

 the aid of a few elementary properties easily established, 

 and secondly lo show that the method of rates or lluxions 

 may be advantageously used for the purposes of instruc- 

 tion, and the use of infinitesimals, limits, and series en- 

 tirely avoided until the student is well grounded in the 

 elements of the calculus." 



The first seven articles under the head "the Newtonian 

 Method of Fluxions," treat of the methods in general use 

 at the present time, and contain extracts from Tod- 

 hunter, Lacroi.x, Carnot, and Cournot, especially directing 

 attention to the positive advantages of the Newtonian 

 method, as set forth by the last-named writer. The next 

 six articles are occupied with the " Proposed Method of 

 treating the Differential Calculus." 



The, remaining half of the pamphlet is given to alge- 

 braic and transcendental functions. It would be very 

 interesting to lay before our readers an account of the 

 ingenious methods adopted by our authors, but it would 

 take up too much space. Some idea of the original paper 

 (and there are no great differences, we fancy, between the 

 two publications) can be got from an account of it fur- 

 nished by Mr. J. W. L. Glaisher, F.R.S., in vol. iv. (pp. 

 58-64) of the Messenger of Mathematics (1S75). 



Altogether, on a review of the three books before us, 

 we anticipate that mathematical studies are destined to 

 occupy a more prominent position in the American 

 colleges and schools than they have in the past.' 



R. Tucker 



LETTERS TO THE EDITOR 

 {The Editor does not Iiold liimself responsible for opinions expressed 

 by his correspondents. Neither can lie undertal;e to return, 

 or to correspond with the writers of, rejected vtanuscripts. 

 No notice is taien of anonymous communications. 

 Tlie Eddor urgently requests correspondents to keep their litters as 

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

 is impossible otherwise to ensure the appearance rjen of com- 

 munieations containing interesting and novel facts. ] 



Visibility of the Ultra- Violet Rays of the Spectrum 



It is well known how sur[)risingly rich in rays of high refran- 

 gibility the spectrum of the electric arc formed between carbon 

 points is, above that of all other artificial flames ; and also how 

 far beyond the ordinarily discernible rays of the solar spectrum, 

 formed by a glass prism, light may be traced by eyes carefully 

 shielded, and raised to the highest susceptibility to perceive it. 

 The name of " lavender-grey " rays has been given to tliemfrom 

 a colour of that tint which they are considered to possess, Init the 

 term "ultra-violet," which is more commonly used, betrays 

 perhaps a lingering doubt as to the sensible existence of another 

 order of coloured rays in the spectrum distinct from the violet 

 and superior to it in refrangibility, which has yet been detected 

 by very close and careful observation. All doubts of this kind, 

 which from want of sufficient acquaintance with that part of the 

 spectrum I have mjself been hitherto rather too prone to enter- 

 tain, have lately been quite dispelled by frequent observations of 

 the spectrum of the electric arc between carbon points thrown 

 by a quartz prism on a white paper screen. The violet end of 

 the spectrum terminates rather abruptly, or at least beams with 

 bright colour that fades off very quickly ; and in the dark space 

 beyond it three more refrangible bright bands are visible with 

 more or less distinctness. Tlie middle one of the three is the 

 brightest, and from its perfect freedom from colour, in which it 

 contrasts most remarkably with the strongly-tinted light-belt 

 near it, and its distant separation from the violet termination 



» We are conhrmed in our views oa this subject by a perusal of Dr. Syl- 

 vester's characteristic address at the Johns Hopkins XJniversity on Com- 

 memoration Day, February 22, 1877. 



of the continuous spectrum, I at first hastily ascribed it to a 

 "ghost," or faint image of the slit, indirectly refracted and 

 reflected through the prism, and thrown with the spectrum on 

 the screen. That it is not so, however, is shown by the action 

 of these three lines on fluorescent substances, of all of which that 

 I have tried they excite the fluorescence most strongly, especially 

 that of fluorescin, eosin, rose of Magdala, and other solutions, all 

 of which alike show thes e rays to be clearly defined and well- 

 insulated spectral bands. In particular, the solutions of a;sculin, 

 paviin and amido-terephthalic acid are only excited by these 

 "ultra-violet" lines, and not by any rays in the spectrum of 

 lower refrangibllities, clearly showing that the vigorous fluor- 

 escence that they produce is not the effect of any ordinary light- 

 beam of common refrangibility, irregularly transmitted by the 

 prism, but that they are well-marked rays, probably of carbon, 

 in the spectrum of the voltaic arc. The light of the middle band 

 is bright enough to be easily reflected and examined separately 

 fiom the rest of the spectrum on a white screen, where it is so 

 nearly grey or colourless that it scarcely admits of being ranged 

 in any colour scale, although the name " lavender-grey " per- 

 haps expresses better than any other term the faintest possible 

 tone of colour which, if it exhibits any at all, this almost purely 

 neutral, or steel-grey band of rays may possibly be suspected to 

 possess. It is a little more strongly absorbed by ordinary plate- 

 glass than the neighbouring violet b.ands ; but it remains visible 

 in the spectrum of the arc formed by an ordinary flint glass 

 prism, though much spread out and enfeebled by the dis- 

 persion, which gready exceeds that of a quartz prism of the 

 same refr.acting angle. It is perhaps for this reason that it is not 

 perceptible in the spectrum of the arc as usually projected on a 

 screen with a fluid-prism of bisulphide of carbon, but if the 

 latter is replaced by benzine, which disperses the light less than 

 flint glass, it forms a pretty conspicuous grey band in the spec- 

 trum. The other two lines or bands are so much fainter than 

 the principal one, that in general they can only be found with 

 the help of a fluorescent substance, and where so faintly visible 

 it is not possible to speak positively as to their colour. The 

 less refrangible is very near the violet termination of the spec- 

 trum, and when well seen it shares its violet tinge ; the more 

 refrangible one is nearly as far beyond the principal grey band 

 as this band is beyond the margin of the violet, and as far as its 

 weak light allows one to distinguish, it is of the same colour as 

 the brightest band. In order to determine their positions, some 

 measurements were made of metallic lines, and of the spectra of 

 sodium, lithium, thallium, and strontium in the arc, with the 

 result that the violet part of the continuous spectrum extends to 

 the closely-neighbouring positions of the hydrogen line H5 (/j), 

 the potassium flame-spectrum line K0, and the last violet 

 line in the arc spectrum of a salt of strontium. The first 

 faint outlying ray occupies nearly the position of Hj in the solar 

 spectrum, and it is therefore in the true violet region of the 

 spectrum, as its colour faintly indicates. The prominent grey 

 line begins with its brightest edge about as much further bey ond 

 this, from the end of the violet field ; and becoming weaker from 

 there, it is about twice as broad as the distance between the two 

 Fraunhofer lines H, its mean position in the spectrum being 

 nearly as far from H as II is from //, reckoning the distances as 

 they would be seen with the quartz ptism and with solar light. 

 The third faint line occurs about as far again from the violet as 

 this band ; and it lies at least as far beyond H as the distance 

 between G and 11 in the solar spectrum. Yet it is visible there 

 by glimpses, like the first faint member of the group, which it 

 does not yet by any means surpass in the strength with which it 

 produces fluorescence. 



If any fresh proof was needed of the characteristic grey appear- 

 ance of visible rays in this portion of the spectrum rt was soon 

 presented in one of the metallic spectra used to determine their 

 positions. The spectrum of mercury exhibited a bright line 

 (beautifully distinct when a fluid prism of benzine was used with 

 a refracting angle of between 50" and 60 ), much brighter than 

 the principal grey carbon band, considerably more refrangible, 

 and of the same tintless, and perfectly neutral grey appearance. 

 Though not so distant from the violet as the most refrangible 

 faint carbon line, it is yet according to the best mea~urements and 

 identifications that were made, about as far beyond II j in a 

 prismatic spectrum as Hi is from G ; and radiation of thrs high 

 degree of refrangibility is evidently strongly luminous, when 

 sufficiently intense, with homogeneous grey light characteristic 

 of this regiou, and contrasting conspicuously m its appearance 

 with the zone of violet colour, which often borders closely upon 

 it in electric spectra. 



