so that the observations accord very closely in fixing the height at a little exceeding 40 miles at disappearance, but a slight correction is necessary to allow for the Earth’s curvature. There are other methods of computing the heights, one of which is explained by Prof. A. S. Herschel in a paper entitled “Height of a Meteor” (‘Monthly Notices,’ vol. xxv. p. 251).

Meteoric Observations.—A large number of meteor-showers still await discovery, and there are features even in connection with the best known streams which remain to be elucidated. Such doubts as now exist are only to be cleared away by assiduous observation made with the utmost accuracy possible both of the directions and durations of meteors.

This attractive field of investigation has certainly been neglected in recent years, and the reason of this may perhaps be found in the complications inseparable from it, in the need of great patience and scrupulous care in observation, and the necessity of gaining experience before the observer can feel a reliance on his work, and draw safe conclusions. Meteors are so fugitive, so diverse and erratic in their apparitions, as to be quite beyond the scope of instrumental refinements. They must necessarily be observed under many disadvantages. Positions have to be fixed from very hurried and often imperfect impressions. But these drawbacks, formidable as they at first appear, may be severally overcome by practice, by careful regard for the conditions under which meteors are displayed, and the marked differences of aspect induced by these conditions. When the observer has acquired a practical knowledge he will proceed with confidence in his work, and avoid many of the difficulties surrounding it.

In recording meteor-tracks for the purpose of discovering the radiant-points, the chief feature in which precision is essential is the direction of flight. A perfectly straight wand, held in the hand for the purpose, should be projected upon the path of every meteor directly it is seen, and then when the eye has quickly noted the position and slope relatively to the fixed stars near, it should be reproduced on the chart or celestial globe. The time, mag., estimated duration, and details of appearance should be registered in a tabular form, with the R.A. and Dec. of the beginning-point and probable radiant. The end-point and length of path may be left until next day, in order to save valuable time. The wand is a great assistance to the eye in retaining the approximate directions and noting the places. If a meteor belongs to the slow, trained class, or if it belongs to the swift, streak-leaving order, the path may be very accurately noted, for the wand can be adjusted to its direction before the meteor or its visible offcome has died away. In the case of short, quick meteors, devoid of either streaks or trains, and generally shooting from radiants at high altitudes, they are more difficult to secure, as they vanish before one may turn, and the observer must rely upon the mere impression he received. But even these succumb to experience, and will be found to resolve themselves into a number of sharply defined radiants scarcely less certain than the positions derived from the streaked or trained meteors.

These positions are only to be fixed by the exercise of much cautious discrimination on the part of the observer, for the direction of the flight is not sufficient, alone, to indicate it. The visible aspect of the meteor has to be equally considered, for the place of its radiant imparts certain peculiarities to it which are rarely to be mistaken. First, the astronomical position of the radiant. If the radiant is at, or within 50° of, the Earth’s apex (a point 90° preceding the Sun along the ecliptic, and towards which the Earth’s motion is directed) the meteors generally leave streaks, especially the brighter ones, and move with great speed. They are usually white, exhibiting a high degree of incandescence. If the radiant is near the anti-apex or anywhere in the anti-apex half-sphere the meteors are streakless, they travel slowly or very slowly, and often leave trails of yellowish sparks. Bearing these facts in mind the region may be assigned in which any radiant is situated, if not the exact position of the radiant itself. If, say, on Aug. 10, at midnight a swift, streaked meteor is seen shooting from the Pleiades towards Aldebaran, just risen, the radiant is either in Musca, Triangulum, or Andromeda. But if the meteor is slow, with a train, then we must go further back in the direction of its flight, and seek the radiant in the S. or S.W. sky. If the motion is very slow, the radiant may be as far away as Aquila. Second, the sensible position of the radiant. A low radiant yields long-pathed meteors, characterized by slowness of speed and a flaky appearance either of the streaks or trains. A radiant near the zenith gives short, darting meteors, with rather dense streaks or trains. These nearly vertical meteors have a less extensive range of atmosphere to penetrate than the horizontal meteors, which are sometimes abnormally long. In the case of brilliant meteors, however, the paths occasionally extend over considerable arcs though the radiant may be high. Third, the position of the radiant relatively to the path of a meteor. If a meteor is close to its radiant its track is usually slow, and appears greatly foreshortened by the effects of perspective. It is travelling (approaching) nearly in the line of sight, and the streak or offcome of sparks is especially dense because it is seen through its entire depth; and the nucleus in such a case has a brushy diffused appearance. Such meteors often traverse sinuous, or curved paths of 2°, 3°, or 4°, and they are readily distinguishable from other meteors far from the radiants to which they belong.

A good method of tabulating meteor-tracks is that adopted by Lieut.-Col. Tupman in his catalogue published by the British Association in 1874. I have adopted the same form, and herewith append a copy of my register of a few isolated bright meteors observed in the autumn of 1890:—

Date
1890.		G.M.T.Mag.Observed PathLength
of
Path.		Dura-
tion.		Appearance.Probable
Radiant.
From
R.A. Dec.		To
R.A. Dec.
h mo oo oosec.o o
Oct.1710 37>1219+61255+65163·5V. slow, B. train.204+56.
1910 35161½+2644½+27½15½0·7Swift, streak.Orion.
1912 0½ ☽326 - 8319 - 1070·5Swift, streak.Orion.
2517 18>♃168+34180+2414½0·8Swift, streak.Lynx.
267 33♃329+69243+51424·0Slow.32+18.
Nov.17 1>1278+49244+11½466·0Very slow.50+15.
19 17>1345+11307+1½394·0Slow.50+15.
510 40>♀28½-2525½-29½5½0·7Swift, strk. 15 sec.Taurus.
1611 15♃274+77265½+67101·5Not very swift.Auriga.

The duration of flight is a most important element to estimate correctly, as it affords data wherewith the real velocity may be computed, and enables the nature of the orbit in which the meteor is moving to be definitely assigned. This feature is, however, one of the most difficult of all to derive with satisfactory precision. In the case of very slow meteors lasting several seconds, it is easy by means of a stop-watch, or by other methods, to get the times of flight within narrow limits of error, but the swifter class of meteors complete their visible trajectories in the fraction of a second, and are gone before any effort can be made to gauge their durations, so that a value has to be attributed which is little better than a mere guess.

Every adopted radiant-point should be based on at least five paths, unless the conditions are special, and these must show a very definite centre, and present family resemblances. It is often possible to detect a good centre from very few paths, when the radiant is low on the horizon, or when it occupies an isolated position.

In recording meteors the details of their appearances should also be appended to the paths. Foreshortened and crooked courses, fluctuations of brightness, halting motion, spark-trains, phosphorescent streaks, broken streaks, and other features must be invariably noted when observed, as likely to assist in fully comprehending these bodies. A streak will sometimes brighten up perceptibly after the head has died out.