appears to subsist between it and the Formation of Clouds. 45 



clouds after their formation moved gradually to the south, where 

 they soon disappeared and were succeeded in the same manner 

 by others. These phenomena took place in a part of the sky 

 otherwise quite clear. 



"April 21, h 30 m a.m. Luminous cirri scattered over the 

 sky. Faint auroral pulsations flashing up to the coronal point, 

 which was occasionally distinctly marked. During the day the 

 sky was well covered by cirri and cirrocumuli. A dark haze 

 extended to about 30° above the horizon; wind gusty, E.S.E. 

 8 h p.m. Cirrus and cirrostratus streaks extended across the sky 

 from W.N.W. to E.S.E. After dusk the cirri became decidedly 

 luminous. 10 h to ll h p.m. Luminous sheets flashing wildly up 

 to the zenith, chiefly from W.N.W., along the line of the cirrus 

 and cirrostratus streaks, the latter showing very dark, and the 

 former acquiring an additional fitful brightness. Pulsations also 

 proceeded from the N.N.W. point. No appearance of a distinct 

 corona. The flashes were fainter than those of the aurora gene- 

 rally are, and no well-defined ' streamers ' were seen. The prin- 

 cipal seat of the phenomenon was also more westerly than usual. 

 Stars shone faintly through the luminous cirri. ll h 30 m . The 

 cirrus and cirrostratus bands approaching nearer the magnetic 

 meridian than before. 



"April 24. During the day wind E., gusty; sky clear. 

 911 4Q m p.m. Clear, with the exception of an auroral band about 

 10° broad, extending from a point a short way to the west of the 

 magnetic meridian (which it crossed between 45° and 55° above 

 the N.N.W. point) to about midway to the E.N.E. point. No 

 other auroral appearance visible. It continued nearly in the 

 same position, though frequently disappearing and reappearing, 

 till 10 h p.m. It was affected by rapid pulsations. At 10 h p.m., 

 immediately after its final disappearance, a slender luminous 

 streak suddenly appeared, extending from the place where the 

 former phenomenon intersected the magnetic meridian to near 

 the ordinary coronal point. This soon assumed a cirrocumulus 

 or rather compact cirrous appearance, grew broader, and moved 

 slowly from nearly E.S.E. to W.N.W. Its length lay nearly 

 N.N.W. to S.S.E. Some other luminous patches then appeared 

 scattered over the sky. 10 h 30 m . The cloud formed as above 

 described now much broader, and resembling a compact cirro- 

 cumulus. ll h p.m. Broad luminous bands of thin cirri extended 

 from N.N.W. to S.S.E. ; wind N.E., moderate. 25th. During 

 the forenoon bands of cirri passing into cirrocumuli (same as 

 last night) extending from about N.W. by N. to S.E. by S., and 

 moving from about E.S.E. ; wind E. 



"Nov. 11, 9 h p.m. Wild and irregular gusts from N.E. 

 Aurora up to at least 30°, but clouded underneath. Cirrostrati 



