Sixth maximum. Interval 24. 1870, Sept. 14, 16 spots, 23 for the whole interval, from 3 observations.

When we take the mean numbers of spots seen at each interval, the curve is comparatively flat, rising but little above the mean line of 7.9 spots visible at any interval, and this is about the mean number visible on any evening. The flatness of the curve is not accordant with an increase of spots dependent on an increase of solar altitude or greater angle of illumination, otherwise the apex would be much more decided. We may, however, trace from the number of spots actually seen and contributing to the maxima of the spot-curve, as well as from the observations adduced above, that the change of illuminating angle does exercise an influence on the appearance of spots, inasmuch as on a few occasions the largest number of spots have been seen with higher illuminations. The actual curve, however, derived from two years' observations is not sufficiently decided to refer the appearances of the spots to this agency.

By dividing the whole of the twelve hourly intervals into six series or groups, and taking spot No. 1 as the standard of comparison, we have the data for computing Tables III. and IV. containing the visibilities of each spot for each group of intervals: sunrise, or 0 to 60 hours, altitudes 0° to 17°; 60 to 120 hours, altitudes 15° to 34°; 120 hours to meridian passage, altitudes 31° to 42°; meridian passage to 120 hours, altitudes 42° to 31°; 120 to 60 hours, altitudes 34° to 15°; and 60 to 0 hours, or sunset, altitudes 17° to 0°. From the results in these Tables, Table V. has been formed, in which we have a bird's-eye view of the visibilities during the luni-solar day. Generally the visibilities are low during the first 60 hours, i. e. compared with spot No. 1, the smaller spots are but seldom seen; and this is so far indicative of solar light in some way developing or bringing the spots into visibility. During the next 60 hours some spots have risen considerably in visibility, while others have been seen more frequently during the afternoon hours of the luni-solar day. The numbers are, however, too irregular to allow us to conclude that the smaller and less frequently seen spots are influenced in their visibility by further changes of illuminating angle beyond their first development; and this is very strikingly manifested by the curves which these numbers furnish; for example, the diurnal curves of spots Nos. 5, 14, and 16 in fig. 2 generally agree in exhibiting greater visibilities from 60 hours to meridian passage, while spots Nos. 9 and 11 are more frequently seen from 120 to 60 hours before sunset. These, as well as the peculiarities of the other curves, strongly suggest that the variations of visibility of certain spots are not to any great extent dependent upon an increase of intensity of solar light, but rather upon some agency more particularly connected with the spots themselves. It is important to remark that another series of observations may furnish totally different diurnal curves, should the variations in visibility depend upon local lunar action.

In nearly every case the spots seen during the first 60 hours of the lunisolar day have increased during the day in visibility, i. e. they were seen less

Fig. 3.

frequently during this group of intervals than during the succeeding sixty hours. This increase, however, has not been regular, which it would have been from changes of illuminating angle alone, some spots having been seen, as before stated, more frequently during the second group of intervals, while others have declined in visibility and not attained their maxima until the period 120 to 60 hours before sunset. The diurnal curves of spots Nos. 14, 5, and 16 in the first category, and those of Nos. 9 and 11 in the second, have already been referred to; that of spot No. 22 (fig. 3) differs from the others by its showing an increase of visibility from sunrise to 120 hours before sunset. The visibilities of many spots are lower during the last 60 hours of the luni-solar day.

The curves of visibility during the luni-solar day are essentially different from the curves of visibility as deduced from the observations of twenty-four lunations, although both lead to the same result; and from both a very important conclusion may be drawn, viz. that upon assuming other agencies to be in operation than changes of illuminating angle, such as present activity, the epochs at which such activity was manifested varied to such an extent, and were so far separated from each other in time, as to coincide, in the case of spots Nos. 14, 5, and 16, with the period in the luni-solar day of 60 to 120 hours after sunrise, while the activity manifested by spots Nos. 9, 11, and 22 occurred at a later period of the luni-solar day, 120 to 60 hours before sunset. So far as the variations of visibility of spots Nos. 14, 5, 16, 9, 11, and 22 are concerned, they do not appear to depend exclusively on changes of illuminating angle, even if a certain intensity of solar light contributes generally to render the spots visible.

While the four craterlets Nos. 1, 3, 30, and 17 are visible during the whole of the luni-solar day, the spots on their sites are seldom seen until the sun attains an altitude of about 30°, and then they appear as "bright round disks;" and this characteristic attaches as well to the craterlets as to other spots when the sun attains this altitude. With altitudes between 30° and 40° a different class of phenomena is manifested; the sharp and distinct cha

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

2: : : : : : : ON: 2:

.04

1.00

⚫04

-07

.04

16

19

⚫06

25

.66

63

.91

⚫06

•22

·09

⚫03

•28

•12

•12

.37

03

12

'06

'06

'06

.31

⚫09

•12

25

.14

racter of the craterlets is no longer observed. Some put on a hazy appearance, and they all assume the same aspect as those spots which have not been observed as craterlets. This state of things continues until the declining latitudes approach the limit at which the crater form was lost in the advancing day, then it once more appears accompanied by a disappearance of most of those spots which came into visibility as the sun rose higher. We have an analogous phenomenon to this in the well-known crater Aristarchus. Shortly after sunrise its outline is sharp and distinct, while its interior is partly covered with a well-marked shadow and partly glowing in strong sunlight. As the sun rises above its horizon these characteristics are lost; the ridge extending from it to Herodotus becomes brighter, and to some eyes, and with some instruments, it is confounded with the interior, the whole appearing as a very vivid brush of light. The exact solar altitude at which the 1871.

F