22 



been evened up with a cement resembling that in the walls. The back 

 rooms were half-filled with rocky debris from roof and cliff. 



While busied with my negatives, the others had prospected the ledge 

 in opposite directions, coming upon ample evidence of its having been 

 quite thickly peopled. Euins of half a dozen lesser houses were found 

 near by, but all in such exposed sitnations as to be quite dilapidated. 

 Some had been crushed by the overhanging wall falling upon them, and 

 others had lost their foot-hold and tumbled down the precipice. One lit- 

 tle house in particular, at the extremity of this ledge, about fifty rods 

 below the one described above, was especially unique in the daring 

 of its site, filling the mind with amazement at the temerity of the 

 builders and the extremity to which they must have been pushed. 

 Careful views of this having been secured so as to show as well as pos- 

 sible its almost complete inaccessibility, we felt impelled to hurry on to 

 new developments. Apparatus was carefully lowered to the patiently- 

 waiting mule, and adjusted to the pack-saddle, then, mounting our own 

 animals, pushed on down the caGon. Below it opened out into quite a 

 valley, side-caiions opening in from either hand, adding much to the 

 space. Every quarter-mile, at the most, we came upon evidences of 

 former habitations, similar to those already described; the greater major- 

 ity occurring in the level bottoms and on the low spurs of the escarp- 

 ment. 



Two or three miles below the house in Fig. 12, we discovered a wall 

 standing in the thick brush upon the opposite side of the river. Consid- 

 erable difficulty was experienced in crossing; in some places having to 

 cut our way through the entangling vines with our belt-knives, and then, 

 when the stream was reached, had to follow it some distance before an 

 opportunity occurred to emerge. 



The wall, or walls rather, before us were a portion of an old tower, 

 (see rig. 1, Plate I,) in the midst of a group of more dimly marked ruins 

 or foundations, extending some twenty rods in each direction from it. 

 As seen in the figure referred to, the tower consisted of two lines of 

 walls, the space between them divided into apartments, and a single 

 circular room in the center. The outside diameter of all was 25 feet, 

 that of the inner circle 12 feet, and as the walls were respectively 18 

 and 12 inches in thickness, left a space of 4 feet for the small rooms. 

 This outer circle was evidently divided into six equal apartments, but 

 only the divisions marked in the diagram could be distinguished. In 

 the i)lace where they should have occurred, the walls were so broken 

 down and covered Avith debris as to render all details indistinguisha- 

 ble. Where the walls were standing, they showed small window-like 

 doors opening into the inner circle. The highest portion of the inner 

 wall was not more than 8 feet, and of the outer about 15, From 

 the amount of debris, I should not judge it to have been much 

 higher — not more than 20 feet at the most. The space between 

 the walls was filled with debris. Outside there was very little, ex- 

 cept where the wall was totally ruined, and inside the inner circle was 

 more; but, as that had probably been an underground apartment, or 

 excavation, and was still a little below the ordinary level, it was not 

 easy to judge how much had gone into it. 



The stones of which this tower was constructed were irregular in size 

 and shape, but with the outer face dressed to a uniform surface, and of 

 the same average size as those already described. The mortar and 

 " chinking " had been worn out entirely from the more exposed portions, 

 giving the wall the appearance of having been^dry-laid ; but uj^on pull- 



