206 BOTANICAL GAZETTE [SEPTEMBER 
they looked like a very “bad catch” at lawn seeding, is one which few 
who have not been in the deserts in spring can imagine. Unlike the 
desert regions of our own country, with their sage brush, yuccas, and 
host of small tufts of grass and sedge, these deserts are for miles abso- 
lutely without a living plant. For days we steamed down the coast, 
but aside from an occasional garden made by irrigation in the neigh 
borhood of the towns, we saw no green plant of any description. 
From about Payta in Peru, to Carrizal in Chili, representing fourteen 
days of steamer travel, the coast presents one unbroken line of desert. 
_At Arica it is broken by a small fertile valley, and at Carrizal the 
desert ends in a scanty vegetation of cacti and low growing cushion- 
like perennials. 
At Mollendo we had an opportunity to see what this des 
have been had there been an abundance of rain. The two phot 
graphs are taken within a hundred meters of each other. One shows 
a private garden in Mollendo with apples, peaches, grapes, passion 
fruits, figs; in short a good collection of fruit and shade trees. The 
other is a representative view of the surrounding country 4 barren 4% 
a fresh lava bed. 
There are below this desert at Arica, and doubtless at other points, 
underground sources of water, for large pepper trees which a eh 
planted in the town square are growing as finely as they do in Gibral- 
tar, or southern California, and overshadowing the little clubhouse * 
ert might 
such a cool agreeable climate as this “zona sicca,” °F 
western South America. The contrast between the west and 
coasts of the continent at the same latitude is very T@ the same 
S. off Brazil, duck suits are necessary for comfort, while at : 
latitude off Peru thin flannels are quite comfortable. descrif 
For a systematic botanist, as may be judged from the abov® jologi- 
tion, there is not much of interest in this region, but from a phys! 
cal point of view it will yield some very interesting facts. ntion % 
Dr. Schimper, in his Pfanzengeographie p- 679, calls se studied 
the fact that this desert region of Peru has been very pun num 
from an ecological standpoint. It is probable that 4 ee s of squat 
ber of species will be found along this coast, and hundreés jp 
miles are absolutely without a living plant for year 
localities, however, favored by the fogs, are COV 
ered in the 
