SECTION II.

ANNUAL VARIABILITY IN THE TIME OF ARRIVAL OF RAINS.

“ Each valley, every mountain range and local district, may be said to haveits own peculiar system of calms, winds, rains, and droughts.”— Maury.

“ All over the world local influences modify, but do not make, the generalcharacter of the weather. When the latter is weak, local weather may be thepredominant feature of the climate of any place; when it is strong, then localinfluences maybe entirely obliterated.”— Abercromby.

From the constant rains of the equatorial belt of calms tothe droughts which at all times hang over desert regions, theearth presents in its different parts every variety of weather.

In consequence of the annual excursion of the sun northand south, and the consequent shifting of belts of calms, andof constant precipitation, there results within the tropics analternation of dry and wet seasons of periodic recurrence.The periodicity is not, however, of the kind which distinguishesthe motion of its prime cause. Although the sun takes itswonted position of greatest northern and southern declination,or crosses the equator in a way that may be predicted to asecond of time, its effect—the rains—could not be predictedfor any one place to a day or a week, and sometimes not evento a month, so much do they vary from year to year.

The distinguishing feature of the rains, outside tropicalinfluence, is that, instead of the one or two seasons of rainand of dry weather which the shifting of the belt of constantprecipitation imposes, there occur fortuitously many dry andwet periods. Were it possible to mark any one of the periodswe should find the same variability in its time of arrival asdistinguishes the periodic rains. It is to the determination ofthis question that the present section is' devoted. As to thecause of the variability, we shall learn as we proceed.

The following is a brief description of my method of pro-cedure: First find for each year the dates of the lunar phasesproperly so called. Of the intervening six days apportion