region. Investigation, however, proved that it was artificial, as
well as the fields. Its presence may be due to a temporary connection
between the upper and lower levels of ancient irrigation canals.

Hour after hour our pack train painfully climbed the narrow, rocky
zigzag trail. The climate is favorable for agriculture. Wherever the
sides of the canyon were not absolutely precipitous, stone-faced
terraces and irrigation had transformed them long ago into arable
fields. Four thousand feet above the valley floor we came to a very
fine series of beautiful terraces. On a shelf near the top of the
canyon we pitched our tent near some rough stone corrals used by
shepherds whose flocks grazed on the lofty plateau beyond, and near
a tiny brook, which was partly frozen over the next morning. Our
camp was at an elevation of 14,500 feet above the sea. Near by were
turreted rocks, curious results of wind-and-sand erosion.

The next day we entered a region of mountain pastures. We passed
occasional swamps and little pools of snow water. From one of these
we turned and looked back across the great Cotahuasi Canyon, to the
glaciers of Solimana and snow-clad Coropuna, now growing fainter
and fainter as we went toward Parinacochas. At an altitude of 16,500
feet we struck across a great barren plateau covered with rocks and
sand--hardly a living thing in sight. In the midst of it we came to
a beautiful lake, but it was not Parinacochas. On the plateau it was
intensely cold. Occasionally I dismounted and jogged along beside my
mule in order to keep warm. Again I noticed that as the result of my
experiences on Coropuna I suffered no discomfort, nor any symptoms
of mountain-sickness, even after trotting steadily for four or five
hundred yards. In the afternoon we began to descend from the plateau
toward Lampa and found ourselves in the pasture lands of Ajochiucha,