12 
HARRINGTON. 
especially and remarkably true for October, when what rain 
falls is very heavy. 
The succession of the seasons on the more southern coasts 
in this region are thus described by Findlay in his Direc¬ 
tory of the North Pacific: 
On the coast during the fine season, which commences in November 
and ends in May, the land and sea breezes blow alternately, with a clear 
sky and but little rain. Strong winds rarely occur during this period. 
* * * Occasionally a strong breeze from the northward may be ex¬ 
perienced. 
In the rainy reason, May to November, heavy rains, calms, light vari¬ 
able breezes, with a close, sultry atmosphere, heavy squalls, with thunder 
and lightning, and not unfrequently strong gales from the southwest, are 
prevalent. 
During the fine season the land and sea breezes set in regularly ; the 
former are called el Terral and the latter la Virazon. The only winds to 
be guarded against at this season are the northers [more properly the 
Papagayos]. These violent gusts give no warning but the noise created 
by them a few.moments before they burst. Sometimes a thick fog sets 
in beforehand, which is dissipated at the first gust. These gusts are 
more frequent near the Gulf of Tehuantepec or abreast of the Gulf of 
Papagayos. 
In the rainy season calms are frequent and the sea and land breezes 
which are felt on fine days have no regularity. The prevalent winds 
then are from southeast to southwest, blowing strongly and in squalls, 
bringing bad weather and torrents of rain for twelve or fourteen days in 
succession. During this season, nearly every afternoon about 3 or 4 
o’clock a violent gust sets in from the northeast and lasts until daylight. 
These gales are called chubascas, and resemble the tornadoes of the African 
coast. 
On the Mexican and Guatemalan coasts the invierno rains 
often come on strong, squally winds from the southwest, 
called Cordonazo de San Francisco (Castigation of St. Francis), 
and suspected of being occasional extensions of the southern 
trade winds. They occur from July to October. 
It may be noted that throughout the region of Type I an 
idea is prevalent that the rainfall is decreasing and that this 
decrease is due to the rapid deforesting of the country. This 
idea is not justified by Table II, which gives the annual 
rainfall for the years, and its basis is probably to be found 
in the marked periodicity already pointed out. 
