April 23, 1903] 



NA TURE 



535 



During the past century the Lyrids have been subjected to 

 pretty close observation. The star shower seen in America 

 on the morning of April 20, 1803— just 100 years ago— seems 

 to have far excelled in brilliancy its Lyrid successors, though 

 a display witnessed, it is supposed, in 1S00 in the equatorial 

 regions of Africa is described as having rivalled in splendour 

 the November meteor-shower of 1866. Shooting stars were 

 seen in unusual numbers in America on April 20, 1838, and 

 Prof. Forshey observed a Lyrid display in Louisiana on the 

 night of April 18, 184 1, when he counted sixty meteors in 2k 

 hours, which gives a mean rate of twenty-four per hour for 

 one observer. On the morning of April 21, 1863, these meteors 

 were reckoned by an English observer as appearing at the 

 rate of forty per hour. On the night of April 18, 1876, a party 

 of American students casually noticed that shooting stars 

 were unusually numerous during the hours 10 to 12. Lyrid 

 meteors were also conspicuous on the night of April 20, 

 1874. Mr. Denning has recorded important appearances of 

 Lyrid meteors in 1882 and 1884, especially in the latter year 

 on the night of April 19. The same observer has also stated 

 that the Lyrid radiant was unusually active in 1893 and 

 1901, in the former on the nights of April 20 and 21, and in 

 the latter on that of April 21. The foregoing are the most 

 important displays on record since April 20, 1803. Periods 

 of somewhat different lengths have been proposed with 

 respect to the Lyrid showers, but the true period seems to 

 be one which overlaps, and consists of nineteen years. 

 Thus, from 1803 to i860, we have exactly three periods of 

 nineteen years, and from 1803 to 1841, two periods of the 

 same length. Again, thirty-eight years, or twice nineteen 

 years, separate the showers of 1838 and 1876. The nineteen- 

 year period also connects the displays of 1863 and 1882, of 

 1874 and 1893, and of 1882 and 1901. This nineteen-year 

 cycle is specially interesting, as it is completed at the Lyrid 

 epoch of the present year, reckoning from the somewhat 

 important display of April 19, 1884. A calculation made 

 by the writer indicates that the maximum in 1903 is on 

 April 19, ioh. 30m. G.M.T. The Lyrid radiant ought there- 

 fore to be found active in the early part of the night of April 

 19, probably from the hours 9 to 12. There is no prospect 

 of Lyrids being numerous on the nights of April 20 and 21. 



John R. Henry. 



Unlike the August Perseids, the Lyrid meteor-stream, 

 like those of the Ouadrantids, Orionids and Geminids in 

 January, October and December, seldom exhibits an 

 abundant shooting-star display, more nearly resembling in 

 that respect the Leonid and Bielid meteor-systems than the 

 stream of August Perseids, its materials appearing to be 

 still collected in one or more dense clusters in its orbit. Its 

 brightest as well as its ordinary apparitions are also, like 

 those of the Leonids, of remarkably short duration, so as to 

 be very liable to escape observation unless splendid enough 

 to arrest attention at some observing station on the globe. 

 The great shower seen in America on the morning of April 

 20, 1803, only lasted in full splendour for two hours, from 

 ih. to 3(1. a.m. ; and a rather sensational abundance of the 

 Lyrids on the morning of April 21, 1863, was entirely con- 

 fined to the night of April 20, when 11 meteors, chiefly 

 Lyrids, were seen at Hawkhurst in 45m., and 7 bright and 

 several smaller ones were observed in 30m. at Weston-super- 

 mare, between nh. and I2h., and in a quarter of an hour 

 after ish., at Hawkhurst, 11 shooting-star tracks were 

 noted, the meteors falling too rapidly then in all directions 

 to be all recorded ; the radiant point obtained from that 

 night's tracks, and from a few Lyrids mapped on April 19 

 (23 Lyrid paths together, some of which may perhaps really 



have diverged from other centres), was at 27 



-342°. c!ose 



to the position which was first obtained of it " near a Lyrae, 

 by Prof. E. C. Herrick, in America, 24 years earlier, on 

 the morning of April 19, 1839. On the preceding night, of 

 April 19, the hourly rate of meteors from ioh. to nh. was 

 only ordinary, and on the night of April 22, not a single 

 meteor was seen in an hour by either of two observers who 

 watched the clear sky simultaneously from nh. 15m. to 

 i2h. 15m. in London and at Hawkhurst for hoped-for 

 accordances. 



Records of bright Lyrid showers are therefore of peculiar 

 interest, as they may not improbably represent clusters of 

 meteor-dust along the Lyrid stream, like some which appear 



NO. 1747, VOL. 67] 



to have been noted in the stream of Leonids' on the morn- 

 ings of November 15 and 14, in 1871 and 1872, on November 

 13, 1879, and on the morning of November 14, 1S88, when 

 in 'a watch of 2|h. until daybreak, at Bristol, Mr. Denning 

 noted the appearance of 17 Leonids, although such strong 

 recurrences of the shower are only rarely seen in the interval 

 of some thirty years between the maximum Leonid displays. 

 But the comet 1861, I., of which the Lyrid shooting-stars 

 are supposed to be the streaming wake of pulverised 

 materials, is one of those which it was pointed out by Prof. 

 G. Forbes in his important paper in the Observatory, 1888, 

 on the probable existence of an ultra-Neptunian planet, may 

 presumably have been captured by such a planet, and would 

 thus be moving now with long periodic time in a very long 

 elliptic orbit ; and this would seem to be a rather serious 

 objection to the short period of 19 years assigned in Mr. 

 Henry's letter to the meteor, unless it should be really true, 

 which seems hardly probable, that the meteors and the wake 

 of dust-materials of the comet are only accidentally in ex- 

 tremely near agreement in their radiant points, and may 

 yet not be actually associated together with each other in 

 a common orbit. 



In its two last returns in 1901 and 1902, the Lyrid shower 

 was very distinctly observed to attain its greatest bright- 

 ness on the night of April 21, and as this retardation of a 

 day from its usual date of April 20 accords like the present 

 similar retardation of the January, August, October and 

 December showers with the postponement of all annual 

 astronomical events by one day, since February, 1900, from 

 the omission at the end of that month of the usual four- 

 yearly leap-year day, attention should certainly, in the 

 reasonable expectation of its fixity, be directed again to the 

 night of April 21, in the approaching Lyrid period, as well 

 as to that of April 19, which the very interestingly detailed 

 evidence presented in Mr. John R. Henry's letter shows 

 also to be one on which an unusually bright display of the 

 April Lyrids may perhaps be expected. 



A. S. Herschel. 



Observatory House, Slough, April 13. 



Mendel's Principles of Heredity in Mice. 



I appreciate Prof. Weldon's reluctance to defend his 

 position in a short letter, and I look forward with peculiar 

 interest to the number of Biometrika where I gather this 

 task will be undertaken. 



Though deferring a reply on the simple matter of the 

 eye-colour in the Oxford mice, Prof. Weldon finds space 

 to ask an " explanation " of two over-lying complexities. 

 To debate these finer points with one who doubts the Men- 

 delian nature of the phenomena taken as a whole is like 

 discussing the perturbations of Uranus with a philosopher 

 who denies that the planets have orbits. Still, at the risk 

 of diverting attention from the main issue, I will suggest 

 how these complications may be regarded — scarcely " ex- 

 plained." 



(1) The " lilac " mice illustrate that resolution, and 

 partial disintegration, of characters commonly witnessed 

 when a compound colour is crossed with an albino. The 

 statistical value of the "lilacs" and their place in the colour- 

 system can only be determined by further breeding. The 

 appearance of " lilacs " or analogous types is what we 

 expect, though their absence in the offspring of hybrids x 

 albinos constitutes a certain problem. This and • other 

 genuine difficulties call for careful statement and analysis. 



(2) The diversity of coats in the first crosses points to 

 heterogeneity among the gametes of one or both " pure " 

 races. The nature of that heterogeneity is the question. 

 Each race may breed true to colour, but the cross-bred off- 

 spring of the two is not necessarily uniform. The pig- 

 ment excreted by heterozygotes may, as I could easily 

 demonstrate, depend on factors (probably determinable) 

 other than the visible colours of the parents, and having an 

 independent distribution amongst their gametes. Also, 

 while we are comprehensively assured that the coloured race 

 was Dure, the precise, if as yet uncontrolled, testimony of 

 the records that certain individuals were not, seems to have 



1 From a table of principal observations of the Leonids from 1870 to 

 1896, in a p' rtion of Mr \V. F. Denning's admirable review of the whole 

 history of " The Great Meteoric Shower of November" ; the Observatory, 

 vol. -\.\. p. 201, May, 1897. 



