8o 
K. Aichi and T. Tanukadate, post-graduate students in 
physics, with the following results :— 
Considered as Power Series of 9: 
A,=1°3704 (Tanukadate) 
A,=1°3704 (Aichi) 
For x=6: 
A, =2'6512 
For 2=7: : 
A) =2°6512 
Considered as Power Series of No: 
A,=1°3750 (A. and T.) 
For z=4: A,=2°6509 
— A,=1'3704 (A. and T.) 
Hote erAW— 2 (OG TL 
It thus appears that the number of points for n=4 is in- 
sufficient, but for n=6 or 7 the approximation becomes 
very close, so that the values of these two integrals are :— 
A, =2-6512 A, =1-3704. 
Maxwell’s value of A, is about 1/300 too large, and that 
of A, is about 1/600 too small. Such small differences will 
not materially affect the theoretical results in which these 
two integrals enter, but it will be worth while to notice 
that the actual values are slightly different from those 
usually given in works on the kinetic theory of gases. 
; H. Nacaoxa, 
Physical Laboratory, Imperial University, Tokyo, 
October 8. 
Leonid Meteor Shower, 1903 
Tne return of the Leonids was clearly ,observed here on 
the night of November 15. The night of November I4 was 
also clear, though at times a dark belt of cloud, which con- 
cealed the lower part of Orion, extended itself along the 
eastern horizon to the foot of Leo, and occasionally also 
small patches of cloud dimmed or caused a momentary dis- 
appearance of stars in or around Gemini. These slight 
impediments to observation continued also on the next night 
until between two and three o’clock on the following morn- 
ing. It had been intended to commence observations: here 
as early as the night of November 11, as the writer had 
anticipated that the Leonids would put in an early appear- 
ance in the present year; cloudy skies, however, prevented 
the possibility of knowing if these anticipations were 
realised. During a watch from 1oh. 20m. to 12h. 30m. on 
November 14 (local time) eleven meteors were observed 
almost half of which were Leonids as bright as stars of the 
first or second magnitude. The radiation from Leo was re- 
garded as surprisingly good, considering the hazy or clouded 
appearance of the horizon, which rendered that constellation 
invisible until after midnight. In a forecast made by the 
writer for the present year he found that, so far as the 
Leonid epoch of November 14-16 was concerned, the maxi- 
mum would fall between the hours 13 and 14 on November 
153 but it was considered that the display on the latter 
night would be weak, owing partly to the reported in- 
significance of the Leonid shower on the night of November 
15, 1902, the preparations for the due observation of which 
were generally frustrated by unfavourable atmospheric con- 
ditions. When, therefore, the radiation from Leo was 
found to be so pronounced during the early hours of the 
night of November 14, the writer found that he had entirely 
underestimated the probable strength of the shower for 
1903. The watch, however, was not prolonged beyond 
12h. 30m. on the night of November 14, as there was no 
possibility of a star shower taking place on this night 
theugh, no doubt, there were manifest signs that the 
Leonids might be unusually numerous. The first watch on 
the following night lasted from gh. 15m. to roh., during 
which time four meteors were seen, one of them being a 
Leonid as bright as a first magnitude star. Its appearance 
was revealed by the rich streak it left in its wake as it slowly 
rose from the invisible radiant. The watch was resumed 
at eleven.o’clock, and whatever misgivings might have been 
felt for abandoning the lookout on the previous night were 
quickly dissipated by the appearance of as many Leonids 
in the first quarter of an hour’s observations as had been 
seen during a period about five times as long on the pre- 
ceding evening. Between rth. and rth. gom. the meteoric 
rate was twenty-two per hour for one observer facing due 
east, but by midnight it had fallen to sixteen per ‘hour, 
NO. 1778, VOL. 69] 
NAT ORE 
[ NovEMBER 26, 1903 
though ‘sporadic meteors were included in the count. 
Between 12h. and 13h. drifting patches of cloud probably 
prevented several meteors from being observed, and the rate 
did not rise above thirty per hour, but at r5h., when the 
sky had become quite clear, Leonids were appearing at the 
rate of one per minute. During the next half hour forty- 
one shooting stars were counted, and this high rate was 
more than maintained for the next two hours; indeed, it 
was estiniated at one time that the meteoric rate was easily 
200 per hour for one observer. The brilliancy of the dis- 
play was as remarkable as its numerical strength. When 
the shower was at its maximum, few of the shooting stars 
seen were less bright than the second or third stellar 
magnitude; indeed, most of them, if observed apart, would 
have merited individual description, and almost every third 
or fourth meteor might be called a bolide. To an observer 
looking eastwards the radiation from Leo was very marked, 
but a few instances were characterised by a centripetal 
rather than the usual centrifugal motion as regards the 
well-known radiant. The most noted of these exceptional 
cases occurred betwcen one and two o’clock, when a bolide 
of surpassing splendour passed slowly downwards, leaving 
a rich trail across the stars « and ¢ Leonis. When close 
to the ‘* Sickle ’’ it exploded with a vividly white flash that 
imparted to it an almost startling brilliancy, and an instant 
afterwards a meteor as bright as Sirius made its appearance 
about twenty degrees further on, shooting down towards 
the horizon in a path that seemed to be a production of that 
of its more brilliant predecessor. Another meteor brighter 
than Jupiter shortly afterwards moved slowly downwards 
from within the “‘ Sickle,’’ passing between y and 7 Leonis 
one-third nearer the latter than the former star, and pur- 
suing a course parallel to the line joining # and Regulus. 
This west-to-east motion of the most brilliant members of 
a meteor display (for it has been noticed on other occasions 
by the writer) appears very significant. Several shooting 
stars shot from e Leonis to 8 Canis Minoris, or slightly 
below the latter star. There seemed to be a second centre 
of emanation much lower down in the ‘* Sickle’? than that 
indicated by the foregoing meteors, and there was certainly 
another radiant altogether far away from Leo, and situ- 
ated probably in or near Perseus. Several bolides passed 
cut of sight overhead, arresting the attention only by their 
exploding flashes that momentarily illuminated the whole 
heavens. The pear-shaped appearance of meteors in the 
morning hours was very remarkable. These left rich trains 
which, lilke the meteors themselves, appeared of a yellowish 
tinge in the light of the waning moon or in the increasing 
twilight. Other members of the star shower dissolved in 
bright streaks, or made their appearance as vivid flashes 
of light, in the latter case generally at a great distance 
from Leo, bursting forth at one moment near the Great 
Bear, and in the next in the neighbourhood of Sirius. The 
largest number of meteors visible at one time was four. 
At six o’clock the activity of the shower, though consider- 
ably diminished, was still, even in the morning twilight, 
very noticeable. 
The somewhat prolonged duration of the meteor shower 
affords some measure of its intensity, and it is probable that 
it has been widely observed, notwithstanding the fact that 
the notion of the supposed connection of the Leonids with 
the comet of 1866 precluded the possibility of such a striking 
meteoric occurrence in 1903, for the present is unquestion- 
ably the brightest Leonid display that has been subjected to 
European, and very probably also American, observation 
since the brilliant star showers of 1866-8. It is noteworthy 
that the present shower resembles very much in intensity 
and also in other particulars a bright display in 1865, in 
which year it was estimated that one thousand Leonids 
might have been counted by observers in England on 
November 13. A lapse of thirty-eight years separates the 
two events, and this interval suggests the nineteen year- 
period which has already been noticed (Nature, April 23) 
in the case of all the April meteor-displays of the past 
century, and has also been shown (English Mechanic, 
April 3) to connect several important Leonid star-showers ex- 
tending over the same time. If this be so, it is possible 
that the years 1904 and 1905 may be marked by even richer 
meteoric occurrences than that which has taken place on 
the present occasion. Joun R. Henry. 
Dublin, November 18. 
