Jtme 2 1, 1877] 



NATURE 



143 



should give the strongest reflection, owing to a difterence in density 

 produced by stamping," and to the brassworker's solution, not 

 only are the mirrors not stamped, but cast ; but it is the thicker 

 parts, those which stand out on the back in highest relief, which 

 reflect the most light. I have one on the back of which there 

 are two large characters as it were laid upon a background of 

 trees. These letters have been carefully ground flat and polished 

 like the front. Their figures nut only appear more distinctly 

 than those of other less highly raised adornments, but actually, 

 in the sun, throw off a brilliant white light, which contr.asts very 

 markedly with the comparatively subdued spectrum from the rest 

 of the plate. 



Is it possible that there may be some difference in molecular 

 arrangement during the consolidation of the melted metal in the 

 thicker (relieved) and other parts of the plate ? And yet, the 

 thick rim of the mirror does not reflect a rim of light. 



One of my specimens has suffered a little oxidisation, and I 

 observe that this appears on the face to follow certain of the 

 larger masses of relief on the back. This mirror does not 

 " show the pattern through," but there is a curious bright rim 

 reflected from the edge of each scar of injured surface. 



Manchester R. D. Dariiishire 



Colour-Sense in Birds 



As the fact of the preference of sparrows for ydlmv crocuses 

 still excites interest and requires explanation, perhaps you will 

 allow me to call attention to the following remarks of Gilbert 

 White in his " Observations on Nature " : — 



" Birds are much influenced in their choice of food by colour, 

 for though white currants are a much sweeter fruit than red, yet 

 they seldom touch the former till they have devoured every bunch 

 of the latter." 



The obvious criticism that the craving for sweets which dis- 

 tinguishes the human biped is not equally predominant among 

 his feathered friends, and consequently, that their selection of 

 the less sweet but more highly coloured fruit may be due to some 

 t.iste other than the aesthetic, does not detract from the import- 

 ance of White's generalisation that birds are much influenced in 

 their choice of food by colour — a generalisation which, there is 

 no reason to doubt, was based upon his own keen and repeated 

 observation. Paul Henry Stokoe 



Beddington Park 



OUR ASTRONOMICAL COLUMN 

 Variable Siars. — The following are Greenwich mean 

 times of visible geocentric minima of Algol, for July, 

 August, and September, according to the elements 

 adopted by Prof. Schonfeld : — 



h. m. h. .n h. m. 



July II 1151 ... Aug. 20 15 II ... Sept. 12 1339 



.. 31 1331 •■• •' 23 II 59 ... ,, 15 10 2ii 



Aug. 3 1020 ... Sept. 9 16 51 ... ,, 18 716 



Minima of S Cancri occur on September 8 at ish. im., 



and September 27 at Hh. 14m: 



A minimum of Mira Ceti will fall on July 23, according 

 to Argelander's formula of sines, the same perturbations 

 being applied as in the case of the maximum of the year, 

 which is computed to occur November 97. 



Mr. John Tebbutt, writing from Windsor, N.S. Wales, 

 on April 13, states that in consequence of remarks on the 

 probable variability of /j Doradiis, in Nature, vol. xv. 

 pp. 14 and 28 1, he examined the star on February 26, and 

 March 14, and found it of the 8th magnitude. There is 

 a star, estimated gth magnitude, about thirty seconds of 

 time west, and twelve seconds north of it. With such an 

 instrument as was employed by Lacailie at the Cape 

 of Good Hope in 1751, /i Doradiis, with its present 

 brightness, would hardly have been visible. Lacailie 

 calls it a fifth magnitude. 



Minor Planets and Comets of Short Period.— 

 Dr. von Asten, in the course of his recent researches on 

 the motion of Encke's comet, found that, although in the 

 interval 1819-68 the comet had experienced in each period 

 of revolution an almost exactly equal amount of accelera- 

 tion, and that this might be attributed to the existence of 

 a resisting medium, yet in order to connect the last two 

 appearances in 1871 and 1S75 with the previous ones, it 



is necessary to have recourse to the hypothesis of an ex- 

 traordinary perturbation which, in the period 1868-71, 

 counteracted the influence of a resisting medium. For 

 certain reasons Dr. von Asten is led to conjecture that 

 about the middle of the year 1869, when the comet was 

 in the region occupied by the numerous group of small 

 planets (the radius-vector being about 3'2), it made so 

 close an approach to one of these bodies, as yet undis- 

 covered, that a sensible effect on the comet's mean mo- 

 tion was the result. 



In connection with this hypothesis it may be interesting 

 to note that the late Prof. Hubbard, whose masterly in- 

 vestigations on the motion of Biela's comet appeared in 

 Gould's Asti-oiio}ii!Cal Journal, came to the conclusion 

 that the separation of the comet into two distinct bodies, 

 by whatever cause effected, took place in all probability 

 in a heliocentric position corresponding to about longi- 

 tude 3i8"6, latitude -f 1 2°'o, with radius-vector 4'36, 

 which position the comet occupied in November, 1844 

 {Ast. yoiirii.. No. 140). It is stated in some works that 

 the comet in 1846 separated under the very eyes of 

 astronomers ; nevertheless it is upon record that the 

 companion was first recognised on December 29 by 

 Herrick and Bradley at New Haven, but was not -again 

 seen until Maury rtfound it on January 13 ; and its not 

 having been remarked when tlie comet was first glimpsed 

 in the Northumberland and other powerful telescopes 

 may well have been owing to distance and faintness. 



A radius vector of 4'36 would, until quite recently, have 

 been considered as placing the comet rather outside the 

 probable superior limit of distance of the minor-planet 

 group, but the discovery of Hilda by M. Palisa in No- 

 vember, 1875, considerably extended the limit, this body 

 in aphelion being distant from the sun 4'6. Although the 

 separation of Biela's comet, if it really took place at the 

 epoch assigned by Prof. Hubbard, could not have been 

 owing to an encounter with this particular planet, yet the 

 position indicated for the occurrence is clearly a possible 

 one fora meeting with an unknown member of the giou i. 

 In saying thi? much we are of course aware that the 

 separation may have been owing to a very different 

 cause, indeed it might be supposed that such a rencontre 

 would have left a more sensible effect upon the mean 

 motion of the comet. 



Meteoric Fire-Balls in America.— Prof. Daniel 

 Kirkwood in a communication to the American Philo- 

 sophical Society, on March 16, gives some particulars of 

 meteoric fire-balls which appeared in unusual number in 

 the United States in the latter part of 1876 and beginning 

 of the present year. The circumstances attending the 

 appearance of eight conspicuous meteors are inclu led : 

 the dates were 1876, July 8 (two fire-balls), December 16 

 and 21, January 3, 20, and 23, and February 8. The train 

 of the larger meteor of July 8 was visible at least forty 

 minutes, the mass having been apparently dissolved or 

 dissipated in the latter part of its track ; the motion 

 about the sun was retrograde, but sufficient materials 

 were not forthcoming for determining the orbital velocity 

 or the nature of the orbit. The fire-ball of December 16 

 had been visible but a few seconds near San Francisco 

 when it apparently plunged into ihe Pacific at no great 

 distance trom the shore, the lall being f )'lowcd by a loud 

 detonation. The mettor ut December 21 wa^ rt-m.irkable 

 for the length of iis track, between 1,000 and 1,100 miles, 

 one of the longest upon record, and, n.oreover, the track 

 would appear to have been somewhat curved. When 

 crossing Indiana the principal fireball was followed 

 by a train of smaller meteors, many of which ex- 

 ceeded Venus and Jupiter in apparent magnitude ; the 

 breadth of the cluster, as seen from Bloomington, was 3^, 

 and the length at least 20°, from which Prof. Kirkwood 

 concludes that the true diameter was five rniles ; and the 

 length about forty miles ; several explosions occurred 

 during the passage ol the meteorite over Indiana and 



