_Heights of Meteors._—Usually the height of meteors at their first appearance is less than 90 miles, and at disappearance more than 40 miles. From a comparison of a large number of computations I derived the following average values:— Beginning height 76·4 miles (683 meteors) End height 50·8 ” (756 ” ) But if fireballs and the smaller shooting-stars are separated I find the usual heights at disappearance are:—fireballs, 30 miles; shooting-stars, 54 miles. Fireballs therefore approach much nearer to the Earth’s surface before disruption than the ordinary falling stars. [Illustration: Fig. 59. Large Meteor, and successive appearances of its streak, seen at Cape Jask, in the Persian Gulf, on June 8, 1883, 7^h 51^m to 8^h 33^m.] A very slight acquaintance with trigonometry will enable anyone to compute the real path of a meteor if two or more observations, made at distant stations, are available for the purpose. The observed courses of the meteor should be marked upon a celestial globe, and extended backwards to the point where they mutually intersect; this will be the _radiant-point_. The globe having been set for the time and latitude, the apparent tracks should also be prolonged in a forward direction until they meet the horizon, this will indicate the _Earth-points_, or azimuths of the place where the meteor would have been precipitated on the Earth had it been enabled to continue its flight so far. The azimuths and altitudes of the beginning and end of the path, and the azimuths of the Earth-point should then be read off, and by means of a reliable map and a protractor their points of intersection over the Earth’s surface may be readily found by lines drawn from the two places of observation. From the spot where the Earth-points intersect a straight line should also be drawn in the direction of the radiant, and it is along this line the meteor’s motion was directed. The coordinates of the observed points of appearance and disappearance of the meteor, at the two stations, would intersect this line at identical points were the observations perfectly accurate, but this is rarely the case. The distance between the observer’s station and the places over which the meteor began and ended is easily derived from the map, and the height of the object may be found by adding the logarithm of the distance to the log. of the tangent of the altitude. Thus, if the end of a meteor is witnessed from London in azimuth 130° W. of S. (alt. 25°), and from Bristol in azimuth 216° W. of S. (alt. 30°) the place of intersection on the map will be at Warwick, so that the meteor must have disappeared when vertically over this city. London is distant from Warwick about 86 miles, and from Bristol 70 miles, and the resulting height of the meteor is:—