August 25, 1887] 



NA TURE 



407 



* 



I 



which constantly depresses the average size of any useless struc- 

 ture ; and that in a comparatively few cases, where changed con- 

 ditions of life have rendered a previously useful organ actively 

 injurious, the influence of selection may not only be withdrawn, 

 but reversed. And if in justification of these views I were re- 

 quired to adduce any single tests as crucial, I should point on 

 the one hand to the neuter ants, and on the other hand to the 

 bower-birds. For the neuter ants prove to demonstration the 

 fact of developing such imp")rtant structures as enlarged and 

 strengthened jaws through the agency of selection, and of totally 

 losing such i'liportant structures as wings through the cessation 

 of selection— in both cases under circumstances which effectually 

 preclude the possibility of any inherited effect? of use and disuse. 

 On the other hand, the b^wer-birds no less conclusively prove 

 the fact of developing highly elaborate and most remarkable in- 

 stincts, which are entirely without reference to any life-preserving 

 function, and therefore can be ascribed only to the inherited 

 effects of functionally acquired peculiarities. 



If this paper has been at all successful in its objects, 

 it must have brought into prominence one point which I am 

 particularly anxious to make clear — namely, that it is a pre- 

 carious thing to differ, in any point of biological doctrine, from 

 the matured judgment of Charles Darwin. The more deeply 

 his work is studied, the more profoundly is the conviction im- 

 pressed, that even though he did not always give it, he always 

 ,liad a reason for the faith that was in him. Therefore, before 

 his followers venture to question a doctrine which was sanctioned 

 by him, common prudence should dictate a careful pondering of 

 the matter. Some of the readers of Nature may have been led 

 to suppose that as to this I am myself living in a glass house. 

 For my recent suggestion of an additional "factor of organic 

 evolution " has had the effect of bringing many stones about 

 my head with regard to this very point. But these have mostly 

 been thrown by men who have not taken the trouble to acquaint 

 themselves with the exact nature of Mr. Darwin's final judg- 

 ment upon the points in question. As a matter of fact, there is 

 only one point upon which I have deviated at all from the latest 

 editions of Mr. Darwin's works — namely, as to the degree in 

 which free intercrossing is inimical to natural selection — and, 

 curiously enough, this is just the point which my critics for the 

 most part disregard. I am blamed for my arrogance in dis- 

 puting the universally adaptive character of specific distinctions, 

 in affirming the generality of some degree of sterility between 

 species, and so forth ; but all these criticisms only serve to 

 exemplify the truth of what I am now saying — namely, that 

 before anyone ventures to write about Darwinism he should 

 take the trouble to ascertain exactly what it was that Darwin 

 thought. George J. Romanes. 



THE AUGUST METEORS OF 1887. 



HTHE circumstances "attending the recurrence of this celebrated 

 •^ meteoric display were by no means favourable in the pre- 

 sent year. On August 10 and 11 the moon rose before 11 p.m., 

 so that during later hours the smaller and more numerous class 

 of meteors, many of which would have been visible on a dark 

 sky-grouni, were obliterated. Apirt from this, the night of 

 the nth was much overcast, and comparatively few observations 

 could be secured. But, making every allowance for hindrances 

 of this character, the recent shower has proved itself decidedly 

 inferior to many of the conspicuous returns recorded in previous 

 years. 



But if this notable stream has been deficient in numerical 

 strength, it has exhibited some features which, though previously 

 observed, have never been capable of being so definitely and 

 satisfactorily traced in their development as during the present 

 year. I refer to the displacement of the apparent radiant point 

 amongst the stars, and to the visible duration of the shower, 

 both of which form important elements in determining the 

 physical nature of the system and in theoretical investigations 

 as to the perturbations which our earth may have exercised 

 upon it during the frequent ren<-ontrcs with its ma'.erials in past 

 ages. 



The very clear weather recently experienced enabled the pro- 

 gress of the display to be watched on fourteen nights between 

 July 19 and August 14, and the radiant point on each one was 

 <letermined separately, as by combining the results of several 

 nights the changes in its position would have been rendered 

 more difficult of detection. I first pointed out this change in 

 the radiant in Nature, vol. xvi. p. 362, and subsequently 



further details were 'published in the Monthly Notices for De- 

 cember 18S4, pp. 97-98. In Nature, vol. xxxiv. p. 373, will 

 also be found the observations of this peculiarity made here in 

 1886, but they were not so complete as during the present year, 

 when the radiant centres were successively derived as under. 



Great Perseid Radiant Point 1887. 



Night. Raiian'. Meteors. Night. Radiant. MetCDrs. 



a I 



July ij 

 22 

 23 

 27 



23 



29 

 31 



19 + 51 

 25 + 52 

 25 + 52 



29 -\r 54 



30 + 55 



31 + 5li 

 35 + 54 



August I 



6 

 7 



3'' +56 



42 -f 55 



43 -t- 56 

 43 + 56 

 42* -I- S7i 

 45 -f 57i 

 53 + 57 



It will be noticed that these figures d) not show a perfectly 

 regular progression of the radiant in the direction of east-north- 

 east. This is, however, en.irely owing to observational errors 

 which cannot be wholly eliminated from such determinations. 

 Thus the radiant given above for August 6 is no doubt slightly east, 

 and the one for August 10 slightly west, of the true positions. 

 But these trivial discordances in individual positions do not affect 

 the general result, which shows in the clearest manner possible 

 that there is a rapid advance of the radiant from night to night. 

 From all my observations since 1867, which include several 

 thousands of Perseids, I believe this shower extends over a 

 duration of at least forty days, from July 13 to August 22. The 

 earliest visible meteors of the stream emanate from a point 

 between Cassiopeia and Andromeda, while the latest ones 

 diverge from the space separating Auriga and Camelopardus. 



From its first oncoming to the epoch of culmination on the 

 night of August 10 it does not gradually intensify but reaches a 

 somewhat sudden maximum. I have sometimes found these 

 meteors rather scarce on August 6, 7, and 8, and not much ex- 

 ceeding their observed frequency at the end of July. But on 

 August 9 there is a marked increase, and on the following night 

 it is apparent the shower attains its most brilliant effect. As to 

 the displacement of the radiant this seems to be accelerated 

 during the declining stages of the display. In July I find the 

 degrees of right ascension of the shower nearly correspond with 

 the days of the month, the diurnal advance being equivalent to 

 about l°of R. A., whereas on nights succeeding the maximum 

 the change amounts to 2° of R.A. or even more. This difference 

 in place is so striking that any observer may determine it for 

 himself by watching the region of Perseus at the right epoch and 

 charting, with the utmost accuracy, the directions of such meteors 

 as presumably originate from the Perseid stream. These meteors 

 generally leave streaks which furnish a ready means of fixing 

 the paths with a degree of precision that could not be otherwise 

 attained. 



In Nature for August 4, p. 318, I described my observations 

 up to July 29 last. On July 31 I recorded 42 meteors in a watch 

 of 3i hours, but the moonlight interfered considerably with the 

 work, as it also did on following nights. The Perseids forme<I 

 one-fourth ef the visible meteors on July 31. I saw 25 meteors 

 on August I in 3^ hours, but the Perseid display was only just 

 recognizable. At I2h. i8m. I observed a splendid fireball 

 passing somewhat slowly from 338° -f 43° to 164° -I- 70". It 

 left a bright streak or thick train in the latter part of its course, 

 and it was evidently a member of the July Aquariads. At first 

 it was scarcely brighter than a third magnitude star, but when 

 near Polaris it became very brilliant, and afterwards lit up the 

 northern sky with a flash much stronger than the moonlight. I 

 saw 7 other Aquariads on the same night. 



On August 6 observations were continued, and 28 meteors 

 were seen in 4^ hours. Besides the usual shower of Perseids I 

 was much interested in finding a companion radiant at 31 " + 49°, 

 which was very sharply defined. I observed a shower on 

 August II-13, 1880, from 30° -}- 46° which may be the same ; 

 and there is a great probability that this system is connected 

 with Comet I. 1870, which passed near the earth's orbit and 

 would give a radiant near that of the meteor shower and at the 

 same epoch. 



On August 7, 23 meteors were seen in 2f hours. O.ily 5 

 Perseids were recorded. On August 8, 14 meteors were seen in 

 2I hours during moonlight, and of these one appearing at loh. 

 34m. was as bright as Jupiter. Its course was from 6° 4- 67^" to 

 302° -f 60^°, and it left a bright streak. At iih. 28:11. a fireball 

 was seen moving rather swiftly from 349° -|- 15° to 9° -1- 141", so 

 that its path was one of 20' just above 7 Pegasi. At its end 



