110 



KNOWLEDGE 



[June 1, 1891. 



many millions to one) that the stars are really collected 

 to_G;etber in clusters in some places, where they form a 

 kind of system, whilst in others there are either few or 

 none of them, to whatever cause this may be owing, 

 whether to their mutual gravitation, or to some other law 

 or appointment of the Creator ; and the natural conclusion 

 fioni hence is, that it is highly probable in particular, and 

 next to a certainty in general, that such double stars as 

 appear to consist of two or moi-c stars placed very near 

 together do really consist of stars placed near together, 

 and under the influence of some general law." 



Herschel subsequently greatly enlarged the list of 

 telescopic double stars which he sought for and measured 

 in the hope of finding some in which the distance and 

 position angle might vary in the course of the year, 

 indicating a parallax as the earth moves round its orbit. 

 Dut instead of finding a yearly oscillation in the distance 

 and position angle as he expected, he found to his 

 surprise, in many cases, a regular progressive change 

 which indicated that one of the stars was slowly describ- 

 ing a regular orbit round the other. This discovery was 

 annoimced in Herschel's paper in the L'liil. Trans, for 

 lKO;-i and 1804. In speaking of this discovery Herschel 

 said that ho " went out like Sanl to seek his father's 

 asses, and found a kingdom," the dominion of gravitation 

 extending to the stars. .\ dominion which, it should be 

 noted, the Eev. -Tohn llichcll had six-and-thirty years 

 before prophesied would be found to exist, and which in 

 his paper of 17S1 he bad still more coufideutlj' asserted 

 must exist. — A. C. Ixanyaed.] 



STAlfdNAKV RAinATIiiX OF MRTKORS. 

 To thf EiUtar of Knowledge. 



Sir., — I note some remarks in your current number, 

 bearing on the Stationary Radiation of Meteors. The 

 difficulty of reconciling this feature with theory is well 

 known, and the question has been debated whether the 

 fixed radiants cannot be explained by successive showers 

 accidentally placed in nearly the same apparent points of the 

 firmament. After uivestigating the observational part of 

 the subject as fully and carefully as circumstances allowed, 

 I found that, allowing for tiivial errors in determining 

 positions, the showers occurred from identical centres, and 

 coi-tainly could not be ascribed to chance grouping. 



A large proportion of the observed radiant points are of 

 this fixed and long-enduring character. As a rule their 

 individual meteors move with great velocity when the 

 radiants are near the earth's apex, but, with increasing 

 distance from it, their speed sensibly moderates, and they 

 finally become slow, and very slow. Thus my stationary 

 radiant at 47° -f 4-1° yields rcni snift streak-leaving meteors 

 in .July and August, while in November and December they 

 are venj shnr. But this peculiarity is not exemplified in all 

 cases, for there is a shower at 01°+ 41)'' which discharges 

 very swift meteors at the end of November and early in 

 .January, whereas in September I have recorded them as 

 Hifift. This is, however, an exception, and the rule appears 

 to be that the showers meeting the earth give swift meteors, 

 whUe those overtaking it give slow ones. It is clear from 

 the varied phenomena observed both in the major and 

 minor systems, that one and the same explanation will not 

 suffice to explain them all, for every possible diversity of 

 meteor shower may exist and display radiation in the 

 firmament. We may find sporadic meteors of great 

 velocity, and coming from distant space. There may be 

 hyperbolic, parabolic, and elliptical streams, also meteors 

 comparatively isolated, and forming the remnants of past 



groups dismembered by planetary perturbations, for the 

 vicissitudes which these small bodies encounter must be 

 very considerable. Some of the elliptical streams are very 

 wide, and of this the August Perseids, with their shifting 

 radiant, afford a prominent instance. The apparently 

 stationary shower which I have seen at various times, 

 coinciding in position with that of the Perseids on August 

 10th, is, of course, entirely different in character to the 

 cometary shower, and needs a different explanation. The 

 latter presents a vastly richer display as well as a moving 

 radiant, and these features readily distinguish it from 

 contemporary showers. 



We require a vast amount of additional observation in 

 this field. If an energetic observer, living in a finer 

 climate than England, took up the subject, and watched 

 the sky assiduously for several years, he would undoubtedly 

 obtain sufficient data to clear up some of the features which 

 now present such difficulties. I do not tlsink the results 

 of past observation would be controverted, but that the 

 new evidence might enable satisfactory theories to bo 

 formed. 



As to the accuracy with which radiant points may be 

 determined, I believe that observers of long experience are 

 likely to be the best judges of this. The probable error is 

 different in difl'erent cases, for scarcely any two observers 

 exhibit the same degree of skill. Training would never 

 make some individuals accurate in this difficult branch of 

 work. It takes fully two years of habitual observation 

 before anyone can acquire desirable proficiency in record- 

 ing meteors, and confidence in assigning their radiants. 

 Speaking for myself, I believe my positions are within 2°, 

 and frequently within 1", of the real centres. I should 

 regard 2 as a lanjc error in ascribing the radiant of a 

 well-defined meteor shower. The Andromcdes of Novem- 

 ber 27th, with their widely-difi'used radiant, form a very 

 exceptional shower, from which it would be unsafe to 

 judge of the character of others. Many of the minor 

 systems exhibit contracted and sharply-defined radiants 

 which may be accurately determined by the careful 

 observer. 



Yom-s faithfully, 



Bristol, Mav 20th, 1S!J1. W. F. Dexnixg. 



i[KTF.0R-EADIAXT8. 

 To the Editor of Knowledge. 



Sir, — I am afraid that I cannot throw any light on the 

 limits of error as regards the determination of meteor- 

 radiants. It would be interesting if three or four practised 

 observers would make their observations on the same 

 night, at the same place, and then compare their results. 

 But whatever the hmits of error may be, I do not see that 

 errors would, on the whole, produce a greater clustering 

 of meteor-radiants than if the positions were accurately 

 known. I should add that Mr. Dennmg's catalogue 

 contains only positions determined by observations on a 

 single night. But he has a column entitled " Other 

 nights of observation," i.e., other nights on which meteors 

 were observed as coming fi-om the radiant thus deter- 

 mined, which I used in my previous letter. The number 

 of meteors used in determining the radiant is in each case 

 mentioned in bis catalogue. 



I did not intend to maintain that these stationary or 

 long-enduring radiants are in all cases active throughout 

 the year. The evidence, at present, at least, does not go 

 that far. But the reasons for not obsernng meteors from 

 particular radiants at certam seasons of the year are pretty 

 evident. We cannot expect to trace such meteors when 



