METEOROLOGY. 553 



earth's radius, and the plane of the ring perpendicular to that ter- 

 restrial radius that passes through its center. A second ring in the 

 same plane, but of greater diameter, concentrically surrounds the first. 

 There are, therefore, five zones to be distinguished in the study of auro- 

 ras : (1) A circle with the radius of 8 degrees, whose center lies in or near 

 Smith's Sound, within which the aurora is seldom seen, and then only 

 as a light cloud m the horizon ; (2) A concentric ring 8 degrees out- 

 side of the preceding, where the single aurora arch is the ordinary 

 phenomenon ; (3) The ring between circles of 16 or 20 degrees radius^ 

 where the ordinary arch appears near the zenith, and as a faint light, 

 or else it appears as an arch to the north, while in the south another 

 aurora appears ; (4) The ring between circles of 20 and 28 degrees ra- 

 dius where auroral rays ordinarily attend the arch and perhaps even 

 extend to a second luminous ring ; (5) Between the circles of 2S and 

 33 degrees radius, the inner or lower arch is no longer seen, but the mag- 

 nificent drapery, with brilliant and moving rays, is the ordinary phe- 

 nomenon. (Z. 0. G. M., xvn, p. 232.) 



XI. — OPTICAL PHENOMENA. 



Kerber has studied certain optical phenomena by which to gain some 

 idea of the altitude of the earth's atmosphere. From the value of the 

 constant of refraction, i. e., 57".3, be deduces the limiting altitude 192.6 

 kilometers. But from a more critical study of the formula he deduces 

 189.0 kilometers. Al-Hazen's method of determining this height from 

 observations of the height of the twilight arc that rises in the east after 

 the sun sinks in the west, or vice versa in the morning, gives a much 

 lower altitude, but this is apparently due to the fact that the light of the 

 twilight arc comes from those lower strata that contain moisture or dust 

 enough to reflect a sensible amount of light to the eye. Nothing can be 

 determined from this as to how much higher up the attenuated air ex- 

 tends. (Z. 0. G. M., xvn, p. 72.) [It is to be hoped that photometric 

 methods may yet become applicable to this interesting problem. The 

 remarkable colors of the twilight and dawn during the latter half of the 

 year 1883 and over the whole world show that at very great heights our 

 atmosphere is subject to changes that only affect the vision, although 

 they may sensibly affect temperatures and weather at the surface and 

 are in some cases believed to have connection with terrestrial and cosmic 

 phenomena.] 



Exner gives some of the results of critical study upon the scintilla- 

 tion of stars, especially with the Arago scintillometer. After explaining 

 the various special phenomena observed by Arago, Marius, Nicholson, 

 Montigny, and Respighi, he proves that Arago's theory of interference 

 is not sustained, and we must abide by the views of Hooke, Newton, 

 Young, Montigny, namely, that the cause of scintillation is to be found 

 in the variable refractions experienced by the rays in passing through 

 the irregularities of the atmosphere, and- that the colored rays result 



