36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 76 



tion seems to be due to Smith/ 1825. Finally, the transparency of 

 the air being affected in some way by emanations from the sun, there 

 should be a relation between the brightness of the eclipsed moon 

 and the solar cycle.' 



The effect of volcanic eruptions being supposed non-periodic and 

 desultory, in a long series of well-reported eclipses it should average 

 out. So the proposed astronomical causes of variation in brightness 

 will be examined first. 



The results of table i are arranged in table 2. The eclipses are 

 grouped in three columns, headed North, including eclipses in which 

 the moon passed wholly north of the geometrical center of the 

 shadow, Central, in which the moon passed over the center. South, 

 in which the moon cleared the center on the south side. Each group 

 has columns headed Magnitude and Grade. At the foot the magni- 

 tudes and grades are averaged. This is a rough arrangement accord- 

 ing to the moon's latitude, the center of the shadow lying on the 

 ecliptic. 



The footings show that south eclipses, mean grade 1.62, have been 

 during 1860-1922 decidedly brighter than central eclipses, mean grade 

 1.32, and these again than north eclipses, mean grade 0.64; the 

 mean magnitudes, south 0.77, north, 0.71, show that the difference 

 between north and south eclipses can hardly be laid to differences 

 in the moon's immersion. In this connection 13 pairs of consecutive 

 eclipses are collected in table 3. The footings of this table show that 

 while the mean magnitudes, 0.68 and 0.72, are not far from equal, the 

 mean grades are wide apart, north 0.54, south 1.54. 



From these two tables the conclusion lies near that during the 

 period 1860-1922 in general the southern zone of the earth's shadow 

 has been brighter than the central, and this again than the northern. 



Credit for first noticing this inequality must be yielded to the 

 French amateur Rudaux (127), whose observations and sketches of 

 1895 III 10 showed the center of darkness to be displaced northward 

 from the geometrical center of the umbra. This inequality shows 

 itself in a very marked way in the photograph of 1909 XI 26, taken 

 by Metcalf, (258), figures 2 and 3 at end of this paper. For Rudaux's 

 conclusions, see table i. 



Table 4 shows a comparison of the brightness of every total eclipse 

 in table i with the moon's equatorial horizontal parallax at opposition. 



^M. Smith, Phil. Mag., (i) 66, p. 168, 1825. 



'This idea appears, but without suggestion of novelty, in an unsigned note, 

 Nature, 46, pp. 64-65, 1892. 



