PROGRESS OF METEORIC ASTRONOMY 455 



or even more. As these enormous appendages are wheeled 

 rapidly round during a comet's circum-solar flight, the material 

 capable of producing shooting stars may be distributed over 

 a large area. In fact, though the nucleus of a comet may 

 have a perihelion distance of only 50 millions of miles, the 

 extremities of the tail may spread out sufficiently far to envelop 

 the earth, and possibly may give rise to shooting stars. We 

 cannot therefore limit the operations of a comet in this respect, 

 because the material of certain of these bodies extends out 

 indefinitely to almost inconceivable distances, and may influence 

 regions of the solar system apparently far removed from the 

 position of a comet's head. And observations and calculations of 

 cometary positions are always made with special reference to the 

 head or nucleus. Having been engaged for some years in col- 

 lecting and collating records of meteors, in comparing radiants, 

 and in making observations, I may perhaps venture to express 

 an opinion as to the quality of existing materials. I regret 

 that it does not appear to me to be satisfactory. The majority 

 of the radiants at present determined are of very doubtful 

 accuracy. In catalogues of radiants the results of bad, good, 

 and indifferent observers are mixed up together ; and thus it 

 happens that really reliable positions are spoiled by combination 

 with erroneous ones. The observers who have made mistakes 

 can well plead the difficulties of the work in extenuation of their 

 wild efforts. But I think that in some individual instances a 

 fair degree of accuracy has been attained ; in fact, when the 

 conditions are allowed for, it is surprising that the results are so 

 correct. It would be well if photographic data could be obtained 

 to displace the faulty naked-eye records of flight ; but failing a 

 better method, observers have certainly done their best. In 

 preparing my general catalogue of metoric radiants (published 

 in 1898) I included all the positions I could find; and there are 

 many very doubtful among such a miscellaneous collection. I 

 lelt that it was not for me to sit in judgment upon my colleagues 

 in this field, and altogether expunge some of their contributions. 

 But there is no doubt that while some of the observers could 

 not derive radiants to within f or 8° of probable error, others 

 could and did do so within 2 or 3 of error! 



With regard to the observation of meteors, it is a most 

 inviting branch of research and one holding out the prospect 

 of new and interesting discoveries. It undoubtedly requires 



