150 



NATURE 



\yune 2 1, 1877 



normal direction is changed by whatever causes may change the 

 direction of the winds. 



Prof. William Ferrel, of Washington, delivered a communica- 

 tion on the progressive movements of storms ; the object being 

 to show that the movement of great storms is (determined by the 

 currents — especially the upper currents — of the atmosphere. 



Prof. Pickering presented the results of an investigation made 

 in connection with Prof. W. A. Rogers on systematic errors in 

 star declination. A comparison with the mean both of the 

 earlier and later catalogues rendered probable the existence of 

 systematic errors in the Gesellschaft catalogue. 



Prof. Simon Newcomb presented a communication on the 

 secular acceleration of the moon and its increasing deviation 

 from uniformity through many years. He revie\^'ed the existing 

 theory on the subject ; the calculation of Laplace according with 

 Halley's estimate of the acceleration as about io| seconds of 

 time, to be multiplied by the square of the centuries for a given 

 period ; also the Adams theory, which reduces the explanation 

 of Laplace to about 6 seconds, leaving more than 4 seconds 

 to be otherwise accounted for. In ascribing the surplus accelera- 

 tion to diminished rotation of the earth, we are dealing with a 

 subject where the evidence should be carefully weighed. Much 

 dependence seemed to be placed on the records of ancient 

 eclipses. Prof. Newcomb considered these echpses separately. 

 The most promising of the Gieek scalar eclipses was that of 

 Agathocles, tyrant of Syracuse, occurring at the commencement 

 of his voyage to attack Carthage. l!ut we do not know on 

 which side of Sicily he sailed ; according to whether he was on 

 one or the other side of the coast, the difference of time for that 

 eclipse may be calculated as justifying the 10 seconds or the 6 

 seconds acceleration of the moon. The eclipse known as that of 

 Thales has a record open still more to criticism, because it came 

 to its historian by hearsay, and probably through tv.o or three 

 generations after the lapse of a hundred years. It seems curious 

 that if Thales predicted the year (by an estimate of lunar periods) 

 he did liOt also predict the day. Each of the ancient solar 

 eclipses yielded similar elements of doubt on careful examination. 

 From the lecords of lunar eclipses if all uncertain features be 

 weeded out, the old estimate of acceleration will be reduced one 

 half. The Arabian records of lunar eclipses were published at 

 Leyden in the early part of this century. The work is very rare. 

 Altitudes ol sun and moon are constantly given in it. Calcula- 

 tions from these eclipses give the smaller estimate of acceleration. 

 From all the data he has been able to study Prof. Newcomb 

 concludes that the whole amount of acceleration is about 8'4 

 seconds. He hopes to make further estimates from modern 

 records, having had the good fortune to pick up in Paris care- 

 fully compiled data of occultations going back to 16S0. 



In introducing his communication on " a Proposed New 

 Method in Spectrum Analysis," Prof. S. P. Langley, of Alle- 

 gheny Observatory, said that in giving this title to his description 

 of his method, he believed, and, so far as he could ascertain, v/as 

 justified in believing, that the method in this special application 

 of it, was quite new. The process consists not only in placing in 

 juxtaposition, simultaneously, the spectra of light from two 

 opposite edges of the sun's disc (which had been done before), 

 but also in determining that when these spectra are taken re- 

 spectively from east and west edges, the atmospheric spectrum 

 lines still coincide, while the solar lines of the two spectra do not 

 coincide. Prof. Langley was anxious to disclaim any intention 

 to abate one jot of the praise due to Prof. C. A. Young for con- 

 clusively demonstrating that the difference of wave- lengths from 

 the east and west edges of the sun can be measured and its rota- 

 tion thereby be proved. The history of this line of discovery 

 was briefly given. Zollner, Secchi, and Hastings thought they 

 had perceived a change in the refrangibility of the light, and 

 Vogel, using Zollner's reversion spectroscope, obtained a dis- 

 placement of from "oS to o'i5 of one of Angstrom's units. 

 Finally, Prof. Young, using a Rutherfurd grating, showed a 

 velocity of the sun's equator of ini. 42s. ; and also that inde- 

 pendent measurements of solar and atmospheric lines gave 

 different results for these two classes. Prof. Langley's new 

 method has the advantage of great security against instrumental 

 errors, since the two classes of lines under like instrumental 

 conditions, betray their diverse origin. In 1S75, while studying 

 the selective absorption of the solar atmosphere. Prof. Langley 

 constructed an apparatus for comparing honiofjeneous light Irom 

 different parts of the solar disc ; use being made of two pairs of 

 prisms of total reflection, connected with a spectroscope so as to 

 give spectra from different parts of disc side by side. A photo- 

 metric apparatus was attached to compare the relative intensities 



of light in diflerent parts of these spectra. The whole apparatus 

 was not intended at first for the comparison of individuil lines 

 of the spectrum, a purpose for which somewhat similar arrange- 

 ments had been used by Lockyer, Hastings, and perhaps others ; 

 but Prof. Young's success suggested to Prof. Langley another 

 and cognate method of using the principle of Doppler, to which 

 this apparatus is well adapted. For six months Prof. Langley 

 has been engaged in overcoming the instrumental difficulties of 

 this conception. Only within a few days has he been able to 

 produce complete results. When the apparatus is pointed so as 

 to receive the light from the north and south poles of the sun, the 

 lines are continuous in the two spectra ; but when the instru- 

 ment is rotated so as to take light from the east and west sides, 

 all the solar lines are found discontinuous at the junction of the 

 spectra, while the atmospheric lines remain continuous. If the 

 instrument is rotated iSo" the solar lines again appear discon- 

 tinuous, but the spectiTim whose solar lines were before shifted 

 to the right as compared with the other, will after such rotation 

 show them shifted to the left. In order to keep clear of any 

 bias of judgment. Prof. Langley was careful not to know before- 

 hand which way the instraraent was pointed ; but tire displace- 

 ment in every observation tallied with the theory. Essential aid 

 was given in the construction of the instrument by the use of the 

 choicest of glass gratings, ruled 8,600 to 17,200 lines to the 

 inch, which Mr. Rutherfurd, of New York, sent for the purpose 

 of this investigation. In the higher spectra of these admirable 

 gratings thirty-one lines are discriminated in the E group where 

 Angstrom and Kirchhoff have a dozen. On actual comparison 

 for the fine lines of that group more have been counted with the 

 grating than with the most powerful spectroscopes consisting of 

 trains of twelve or more prisms. The method of analysis by Prof. 

 Langley's instrument seems less adapted to quantitative work than 

 Prof. Young's, but in this respect it is hoped also to make it 

 useful by employing the micrometer upon the double displace- 

 ment obtainable in right and left hand spectra of the same order 

 presented simultaneously and in combination. By the observed 

 displacement or fixity of any line we can now discriminate cer- 

 tainly ;as to its solar or telluric origin. It is hoped that a 

 ready means of mapping atmospheric lines will thus be afforded, 

 since indeed they are already mapped by this process before the 

 eye of the observer. 



Gen. J. G. Barnard, U.S.A., contributed a mathemaijcal 

 essay, also in part historical, on the internal structure of the 

 earth as affecting the phenomena of precession and nutation, 

 supplementary to an article under this head in vol. xix. of the 

 '* Smithsonian Contributions to Science," being the third of the 

 " Problems of Rotary Motion." The paper shows that Gen. 

 Barnard has coincided in Sir William Thomson's change of 

 view. The formation of a diurnal tide in the lluid earth is called 

 in doubt by this paper. In general it presents work of the kind 

 that Sir William Thomson was longing for in his Glasijow 

 address — a solution, coherently worked out, of the problem 

 above indicated. 



Prof. O. N. Rood, of Columbia College, New York, con- 

 tributed two papers giving details of his researches concerning 

 colours. Prof. Rood used a set of brilliantly coloured circular 

 discs representing the chief spectrum colours, and aUo purple. 

 By combining in successive proportions with tr.ese colours, a 

 white disc, and giving the combined discs rapid rotation, the 

 foU'^wing results were attained : the lighter shades of vermilion 

 became purplish ; of orange, more red ; of yellow, more orange j 

 of greenish yellow, unchanged ; of yellowish green, more green; 

 of green, blue ; of cyanogen blue, less greenish and more bluish ; 

 of cobalt blue, a more violet blue ; of ultramarine, violet ; of 

 violet, unchanged ; of purple, less red and more violet. Exactly 

 similar effects were produced when violet instead of white was 

 used to reduce the colours. Hence the mixture with white is the 

 same as if the colours were moved towards the violet end of the 

 spectrum. Prof. Rood thinks his results tend to indicate violet 

 as one of the primary colours, which cannot be said of Maxwell's 

 third fundamental colour, an artificial ultramarine, or Bezold's, a 

 blue violet, careful tests of those colours having been fully 

 carried out. The foregoing results were laid before Mr. Charles 

 S. Peirce. He has reported at considerable length on the 

 mathematical principles involved. He regards the results as in 

 accordance with Fechner's law, that the sensation is proportional 

 to the logarithm of the excitation. When the objective brilliancy 

 of any light is varied, the specific subjective brightness is not 

 changed in the least ; but the only effect on the sensation is to 

 add to, or subtract from it a variable amount of a certain con- 

 stant sensation, which Mr. Peirce designates as the "colour of 



