NA TURE 



[July 9, 1896 



July, with a fine Venus-like shooting-star just similar to this 

 present meteor (its streak remaining visible 5 sees.), observed at 

 Bristol on June 14, 1887, among them, that Mr. Denning con- 

 siders it most probable that the real course of this bright meteor 

 was directed. Its path and that of the just similar one observed 

 at Bristol, prolonged backwards, intersect each other at 

 354°. + 36°, close to this well-marked shower-posilion ; and if 

 additional notes of the meteor have been obtained at other 

 stations, it will be an interesting question to examine if they bear 

 out this conclusion. 



I take this opportunity of the returning abundance of meteors 

 of the August period to beg to correct some errors of calculation 

 and projection in my letter on " Heights of August Meteors" in 

 Naturk of September 5, last year (vol. lii. p. 437), which gave 

 confused descriptions, without, however, tripping (except slightly 

 in the case of the second meteor) in the real values of the heights 

 there found, of two meteors doubly observed at Tring and at 

 Slough on August 1 1, 1S95. The places over which the meteors 

 were vertical when first seen, and at disappearance, were laid 

 down in proper relation to Tring and Slough on a map of 

 England, but from left to right instead of from right to left of 

 the Tring and Slough meridian ; so that although both their 

 radiant-points were really easterly, their paths were described 

 as from Oxfordshire to Middlesex, and from west to east across 

 the northern part of Buckinghamshire, in my letter. The 

 corrections applied to the observed paths of the second meteor 

 were somewhat incorrectly chosen, and gave thereby resulting 

 heights and a length of path which were somewhat faultily 

 arrived at : while the radi.ant-point of this meteor (a I'ersei), 

 given by the observations directly was a little departed from, 

 and its altitude and azimuth were also given as 45", 34° north of 

 east, instead of 35°, 44° north of east. As uncertainties from 

 errors inherent in meteor-observations (rarely very small ones) 

 are unavoidable in comparing them together, a recalculation of 

 both paths will, it seems probable, serve better to correct this 

 mare of errors and mistakes, than to try to rectify individually 

 the above principal ones, and a few lesser faults which the paths 

 deduced last year contained. 



Two other accordances, of a Perseid .shooting-star between 

 Tring and Slough, and of a small fireball between Tring and 

 Sydenham, in .August last, and one of a rather notable firijball 

 in November last, have aflbrded, since the path-descriptions in 

 my former letter, .some fairly good materials for height-deter- 

 minations, although the data of the two fireballs' paths are of a 

 little looser kind than those of the two foregoing and the one 

 now newly-added shooting-stars. In the accompanying table I 

 have grouped together what appear to be the real courses 

 (including the two shooting stars' recalculated paths as Nos. 2, 

 and 3, in the table), of all the meteors found here to have 

 furnished accordant observations in la,st August and November ; 

 and I may add, in gladly expressed acknowledgment of the aid 

 supplied abundantly by accuracy of the original observations to 

 the general certainty of these 'determinations, that nearly .all the 

 present li.st of meteor-heights (only excepting those of the fireball 

 of November 9) is due to some weeks' actively continued watch, 

 with careful records of descriptions and well-mapped courses by 

 the stars, .successfully maintained at Writtle and at Tring liy Mr. 

 J. A. Hardcastle in order to obtain them, on the Perseuls and 

 occasional meteor-members of other showers more or less con- 

 spicuous during the period of the display from Perseus, in 

 August last. 



In cases of loose, and of partially inconsistent descriptions, 

 much latitude of choice is left to adjust the observed tracks to 

 e.ach other: but the limits of this choice become, in general, 

 narrowly restricted if the ordinary descriptions and characters of 

 meteor-flights are kept in view, and if any wholly improbable 

 results, as of rising upwards, or of paths abnormally at variance 

 with the usual heights, and not reconcilable with the acknow- 

 ledged astronomical velocities of meteors, are rejected. The 

 course of the fireball No. 5, of November 9, descending almost 

 vertically in E. N.E. at each of the .stations, Peckham, Slough, 

 and Reading, where it was ob.served, along nearly the same 

 apparent path (but beginning at those staliims at nearly equal, 

 and ending at very unequal altitudes), was situated so unfavour- 

 ably for determining its radiant-point by the observed paths' 

 mutual intersections, that only the ordinary conditions of 

 meteor flights (attained to here by trial and error combinations 

 of more and more appropriate paths to satisfy them and 

 the tolerably accordant observations also) could be usefully 

 resorted to as the sufficient additional criterion needed to define 



NO. 1393, VOL. 54] 



the radiant-point, or the line of real directi(m of the meteor's 

 course. 



The observed paths at Writtle and at Slough of the Pcrseid 

 shooting-star. No. i, are also in the unfavourable po.sition of 

 nearly " fore-and-aft " conjunction, but deviate sufficiently from 

 it to yield the radiant-point noted in the list, by their mutual 

 intersection, and from the heights, and meteor-speed given in 

 the list, which are computed from it, this position of the point 

 appears to stand in no further need of very material amendment 

 by the proof test of the atmospheric and astronomical criterion. 



In the case of the small fireball No. 4 (if the same meteor, 

 as assumed from their resemblance, was really referred to in 

 both the observations), besides a pretty large divergence from 

 parallax direction between the two path-iiositions, making a 

 rather considerable reconstruction of them both essential, a 

 nearly lengthwise conjunction (as in Nos. I, and 5) also once 

 more occurs, introducing a large uncertainty, scarcely less than 

 in the case of the fireball No. 5, into the po^ition of the radiant- 

 point, and into the consequent heights and length of path 

 computed from it ; which could only be overcome, as is done, it 

 is to be hoped successfully, by the same tentative method of 

 proceeding as that which was used to fix the real path and 

 radiant of that latter fireball. With heights not far removed 

 from sixty miles, and speeds not far fnim parabolic ones 

 admitted, the construction's freedom becomes so greatly 

 narrowed in these two fireballs' otherwise most equivocal- 

 looking cases, that nearly as great, though (by unadjustable 

 difierences in the observ.alions) still questionable dependence 



may be placed on these two fireballs' real ]iaths and radiant- 

 jjoints as on those a.scribed to the Perseid and Piscid shooting- 

 stars in the present list. 



The position thus found in Gemini, near Castor, of the 

 radiant point of the fireball of November 9 last, perhaps affords 

 a clue to the striking dark green light, not unlike the "signal 

 green " colour of ships' lights and railway lamps, with which 

 the fireball was seen to shine at Slough in the end-half of its 

 course; since large-sired meteors of the "deminid" shower 

 diverging between the end of November and the middle of 

 December from the immediate neighbourhood of the fireball's 

 radiant-point position so determined, are notable for frequent 

 occurrence of green hue in their nuclei. The low height of 

 eight miles assigned to the meteor at its disappearance would 

 nearly break record' of a fireball's deep penetration towards the 

 earth before extinction, if the extraordinarily bright detonating 

 fireball of December 14, 1890, had not been shown undoubtedly 

 to have ended its shining course at not more than seven or eight 

 miles above the earth, and that, too, by a curious coincidence, 

 like the present meteor's end-point, over BiUericay or some |ilace 

 close to BiUericay, in Essex I Should observers in that latter 

 county, or in any of the adjacent East Coast districts, have noted 

 particulars of its appearance on November 9, which happeneil 

 on a clear night during a prolonged display of a widely-observed 

 and rather fine aurora (and, though the memory is a sad one to 

 recall, especially on a date which was an annual one of rejoicing 



> The fivelmll which accompanied the fall of aerolites .-u Weston, Con- 

 necticut, in the United States, on Deceinber 14, 1807. w.is found by Dr. 

 Bowditch to have remained luminous in its descent until only about three 

 or four miles above the places on the earth's surface where the aerolites were 

 scattered. 



