hown by
dotted contours. It will be noticed that while most of the mounds which
mark the sites of former structures rise but 10 feet or less above the
surrounding level, the profiles vary considerably, some being much more
smoothed off and rounded than others, the former being shown on the map
by even, "flowing" contours, while the latter are more irregular; and it
will be further noticed that the irregularity reaches its maximum in the
vicinity of the Casa Grande ruin proper, where the ground surface was
more recently formed, from the fall of walls that were standing within
the historical period.

External appearance is a very unsafe criterion of age, although in some
cases, like the present, it affords a fair basis for hypothesis as to
comparative age; but even in this case, where the various portions of
the group have presumably been affected alike by climatic and other
influences, such hypothesis, while perhaps interesting, must be used
with the greatest caution. Within a few miles of this place the writer
has seen the remains of a modern adobe house whose maximum age could not
exceed a decade or two, yet which presented an appearance of antiquity
quite as great as that of the wall remains east and southeast of the
Casa Grande ruin.

The application of the hypothesis to the map brings out some interesting
results. In the first place, it may be seen that in the lowest mounds,
such as those in the northwestern corner of the sheet, on the southern
margin, and southwest of the well-marked mound on the eastern margin,
the contours are more flowing and the slopes more gentle than in others.
This suggests that these smoothed mounds are older than the others, and,
further, that their present height is not so great as their former
height; and again, under this hypothesis, it suggests that the remains
do not belong to one period, but that the interval which elapsed between
the abandonment of the structures whose sites are marked by the low
mounds and the most recent abandonment was long. In other words, this
group, under the hypothesis, affords another illustration of a fact
constantly impressed on the student of southwestern village remains,
that each village site marks but an epoch in the history of the tribe
occupying it--a period during which there was constant, incessant
change, new bands or minor divisions of the tribe appearing on the
scene, other divisions leaving the parent village for other sites, and
the ebb and