NOTES ON TREE PLANTING AT SAN JORGE, URUGUAY. 223 
Long droughts prevail sometimes, and sometimes long-continued 
rains; and neither can be forecasted. In 63 days, between 
lst February and 4th April 1881, only 0°86 inch of rain fell ; 
in such hot weather this amount would have no perceptible 
influence on growth; it might suffice to wash dust off leaves and 
grass, and on the parched ground would run into the land-cracks. 
As there is a decided growth of grass in autumn, when rains 
assist, that year there was but poor winter feed for cattle. In 
65 days from 28th June to 2nd September 1886, only 1:01 
inch of rain fell. This drought caused little or no harm, as 
little or no growth can be expected in these months. A more 
serious drought succeeded from 21st September 1886 to 13th 
January 1887, 114 days, when only 6°40 inches of rain fell ; this 
term following a dry winter, and embracing all spring and half 
summer ; and this was not all, as the remaining half of the same 
summer, with the following autumn and the beginning of winter, 
167 days from 14th January to 30th June 1887, only obtained 
12°76 inches of rain. 
There was plenty of rain in spring 1880 and early summer 
following ; but had the season, July 1880 to June 1881, when I 
first began tree-planting, proved as dry as the seasons 1886-87 
and 1887-88, I should probably not have repeated the experi- 
ment; the season, however, on the whole, was fairly moist, and 
the rains were distributed with some degree of equality. Thus, 
though we had had no experience as to the best mode of planting 
trees, our success encouraged us to try again; and when in after 
years newly-planted trees, on two or three occasions, suffered 
from long-continued frosts, or from prolonged summer droughts, 
I have looked on these failures as on incidents that may occur in 
all businesses. As examples of frosts, I extract from my records 
that for five nights running in July 1886 the average minimum 
thermometer was 32°, the highest minimum record being 34° in 
a sheltering instrument box, 4 feet above grass. In this same 
July, for another period of seven nights running, the average 
minimum was 29°:3, the highest minimum record being 33°, In 
the following August, one period of five nights averaged 31°, with 
a highest minimum of 34°; and another period of six nights 
averaged 30°°3, with a highest point of 32°. The lowest tem- 
peratures to which the trees were exposed on different nights of 
these periods were 29° four times, 28° twice, 27°, and 24°. 
Amount of planting done, chiefly since July 1880, for though 
