faces of the houses, and just on the brink of the narrow ridge.  On the south side the ridge was precipitous, but toward the north it ran out in a gentle shallow slope toward the next higher hill.  The building material here is a close-grained felsite, and huge fragments of it have been used in the construction of the parapets.  These boulders were, on an average, thirty-five inches long, twenty-five inches thick and fifteen inches wide; while the stones used in the house walls measured, on the average, fourteen by nine by seven inches.

On the western end of the ridge is a small house group, which, for convenience sake, I will designate as “Mason’s Ruins.”  They showed a decidedly higher method of construction, and the walls were better preserved, than in any we had seen so far.  The ground plans could be readily made out, except in a small part of the southwest corner.  These walls stood three to five feet high, and the stones here too were dressed only by fracture.  They were laid in gypsiferous clay, a mass of which lay close to the southwest corner.  This clay is very similar to the material used by the Moquis in whitening their houses.  The stones themselves were felsite, which abounds in the locality.  The blocks have an average size of twelve inches square by six inches thick.  It should be noted that no regard was paid to the tying of the corners and the partition walls; but considerable care had been taken in making the walls vertical, and the angles were fairly true.  The walls were almost twelve inches thick, and on the inner side they had evidently never been plastered.

Being coated with some white plaster, these ruins look white at a distance, and the Mexicans therefore called them casas blancas.  I heard of an extensive group of such buildings near Sahuaripa, and there are also some ruins of this category near Granados, and in the hills east of Opoto.  Undoubtedly they belong to a more recent period than the rude stone structures described before.  Most of the ancient remains of the Sierra are remnants of tribes that expanded here from the lowlands, and only in comparatively recent times have disappeared.  I also perceived that they were built by a tribe of Indians different from those which erected the houses in the caves of the eastern and northern Sierra Madre, and in the country east of it, and may safely be ascribed to Opatas.

In spite of the rest here, the animals did not seem to improve on the grama and buffalo grass.  It was rather perplexing to note that they grew weaker and weaker.  The grass of the sierra, which was now gray, did not seem to contain much nourishment, and it became evident that the sooner we proceeded on our journey, the better.  To save them as much as possible, we loaded only half the regular weight on the mules and donkeys, and sent them back the next day to fetch the balance of the baggage.  In this way, and by strengthening the poor beasts with a judicious use of corn, I managed to pull through and overcome this most serious of all difficulties, which, at one time, threatened to paralyse the entire expedition.