NO. 9 BRIGHTNESS OF LUNAR ECLIPSES FISHER 39 



The mean grade of lO eclipses in the apogee group is 1.30; in the 

 middle group of 13, 1.46; in the perigee group of 14, 1.21. In so far 

 as the arithmetic mean averages out errors and non-periodic causes, 

 this would indicate that at middle distances the totally eclipsed moon 

 has been somewhat brighter than at greater or less, near apogee or 

 near perigee. This differs from the anticipation of L. Giinther,^ that 

 dark eclipses occur at perigee, bright at apogee. But the differences 

 are small, considering the coarseness of the grading, and it can only 

 be concluded that varying distance has not caused much difference in 

 brightness. 



Table 5 shows a comparison of the brightness of every total eclipse 

 in table i with the longitude of the center of the umbra at opposition. 

 The longitudes are in four groups, each in a quadrant of the ecliptic, 

 for the seasons as indicated. The quadrants are not separated by the 

 equinoctial and solstitial points, as it was found by plotting and trial 

 that a quadrant from 82° to 172° gave a mean grade greater than 

 any other ; a division based on this comes near, however, to a strict 

 division according to the seasonal points. 



The footings for the columns give as average grades. Autumn, 

 1.25, Winter, 1.89, Spring, 0.91, Summer, 1.33. So that, in the period 

 studied, winter total eclipses have been very bright, spring eclipses 

 only dim, autumn and summer eclipses intermediate and not much 

 different. Then the longitude effect suggested by Smith exists, but 

 its maximum phase is quite 45° away from that which he expected, 

 Jenkins (49) says :"".... of the seven recorded eclipses in which 

 the moon disappeared, none was later than June 15: May 15, mo, 

 June 15, 1601, April 14, 1642, May 18, 1761,-June 10, 1816, June i, 

 1863 .... middle of April and the middle of June .... it is only 

 in this period when the earth is approaching aphelion that the 

 phenomenon is possible." This fits in well with the dimness of the 

 spring group of table 5, but must be considered in connection with 

 volcanic dust. 



The curve of mean solar radiation published by Kimball ' shows 

 the mean relative intensity of received solar radiation, as measured 

 at a variable number of stations (all in north temperate latitudes) 

 for every individual month, 1882-1918. This mean is expressed as 

 a percentage of the mean for the like named months for the whole 

 period ; e. g., the mean for February, 1900, is divided by the mean for 

 all the Februaries of the period to get the relative intensity. The 



^L. Giinther, Weltall, i, pp. 101-103, 112-116, 127-131, 137, 1900-01. 

 ^ H. H. Kimball, Mon. Weath. Rev., 46, pp. 354-355, 1918. 



