754 J. A. BROUN ON THE LUNAR DIURNAL VARIATION OF 



before and three days after the day of the moon's age which it represents ; and 

 we have seen in the results of this paper, the danger of trusting in mean results 

 where different laws may be in question, and one movement overlies another. 

 It was sought then to determine with more distinctness the change of form 

 from one day of the moon's age to another, and for this end, the lunar diurnal 

 variation for the twelve single days of new moon was obtained, and that for 

 each of the days for which the moon was, one, two, to seven days old/"" 



52. Though the number of days is scarcely sufficient to give means free from 

 irregular disturbances, the results appear little affected by them; they are pro- 

 jected in the second line of curves, Plate XXVIII. Here we find that the 

 minimum takes place at sunrise on the day of new moon, and next day; but 

 when the moon is two days old, the minimum is half an hour after sunrise; on 

 the third day it is an hour and a half after sunrise, and there appears an inflexion 

 in the movement westward ; on the fourth day the movement has already turned 

 westward at sunrise, which is the case also on the fifth and sixth days ; on the 

 seventh clay the movement westward begins only one and a half hour after sunrise. 



53. These results appear then to show, that the turning point is, as it were, 

 attracted by sunrise, and that the change from an easterly movement to a 

 westerly movement at sunrise occurs within an interval of about two days. 



54. So curious and unexpected a fact required a still more careful arrange- 

 ment of the observations. On the days of new moon the observation near 

 noon was always marked as h. for the moon on the upper meridian; the next 

 observation as the moon on the meridian of 1 h., and so on. As, however, the 

 moon is not generally on the meridian at the time of the observation nearest 

 noon at new moon, being sometimes half an hour before or after, an error of an 

 hour occurred easily in the lunar hour corresponding to the following observa- 

 tions, so that in the combinations considered previously, the observation nearest 

 sunrise was not always entered under the true lunar hour angle for a given day 

 of the moon's age. Another method then was now employed, in order to avoid 

 any error from this cause, and to insure that we have the exact effect produced 

 at sunrise. 



The observations nearest sunrise were made at 5 h. 58 m. a.m. in 1854 and 

 1855, and at 6 h. 28 m. a.m. in the following years.t In order to avoid any com- 

 plication due to this difference of hour, the observations in 1854 and 1855 were 

 not employed in the following combination. All the periods for which the new 

 moon occurred, between the 10th December and the 20th January (the time 



* As no observations were made on Sunday, there were twelve days' observations for new moon, 

 and for the moon four and seven days old, eleven days for the moon six days old, ten days for the 

 moon three days old, and only nine days for the moon one, two, and five days old. 



t The time of sunrise varies at Trevandrum from 6 h. 8 m. m.t., on the 15th December, to 6 h. 

 20m. on the 15th January, a change due nearly altogether to" the equation of time, the interval betwixt 

 sunrise and true noon, being (refraction effect included) nearly 5 h. 48 m. during the whole month. 



