and at an elevation of 6,400 feet, we found numerous rude ground plans,
some of which showed rubble walls fifteen inches thick. They formed
groups of four or five apartments, each ten by twelve feet. But on
the north side of that summit there was a larger plan, nearly eighteen
feet square; however, the outlines of the entire settlement were not
distinct enough to enable us to trace its correct outlines.

Many fragments of pottery lay about, but neither in number nor in
interest could they be compared with those found near the ruins in the
southwest of the United States, for instance, near the Gila River. Some
of the potsherds were one-third of an inch thick, and large enough to
show that they had been parts of a large jar. They were made of coarse
paste, either gray or brown in colour. Some had a kind of rude finish,
the marks of a coarse fibre cloth being clearly discernible on the
outside. Others were primitively decorated with incisions. One sherd
of really fine thin red ware was picked up, but there was no trace
of ornamentation on it. We found, besides, a few cores of felsite
and some shapeless flakes and several fragments of large metates.

In the valley formed between the mountains on the upper Bavispe River
we met with very many such houses. The clusters which we came across
seemed to have been composed of a larger number of houses. Parapets,
also built of undressed stones and surrounding these villages,
now became a constant feature. Even within sight of our camp was
such a parapet, six feet high, and house ruins were near by. We also
discovered an ancient pueblo consisting of thirty houses, all of the
usual small dimensions, but not all alike in shape. Some were round,
others triangular, but most of them were rectangular, measuring eight
by ten feet. Along two sides of this village ran a double wall, while
the other two sides were bound by a single wall constructed on the