It will be noticed that the wall lines are only approximately rectangular. The outside dimensions of the village are as follows: Northeastern side, 203 feet; southwestern, 207 feet; southeastern, 182 feet; and northwestern, 194 feet. The northeastern and southwestern sides are nearly equal in length, but between the southeastern and the northwestern sides there is a difference of 12 feet, and this notwithstanding that the room at the western end of the southeastern row has been set out 3 feet beyond the wall line of the southwestern side. This difference is remarkable if, as the ground plan indicates, the village or the greater part of it was laid out and built up at one time, and was not the result of slow growth. As already stated, long occupancy of a village, even without increase of population, produces a certain effect on the ground plan. This effect, so strongly marked in all the ruins already described, is conspicuous in this ruin by its almost entire absence. The ground plan is just such as would be produced if a small band of pueblo builders, consisting of ten or twelve related families, should migrate en masse to a site like the one under discussion and, after occupying that site for a few years--less than five--should pass on to some other location. Such migration and abandonment of villages were by no means anomalous; on the contrary, they constitute one of the most marked and most persistent phenomena in the history of the pueblo builders. If the general principles, already laid down, affecting the development and growth of ground plans of villages are applied to this example, the hypothesis suggested above--an incoming of people en masse and a very short occupancy--must be accepted, for no other hypothesis will explain the regularity of wall lines, the uniformity in size of rooms, and the absence of attached rooms which do not follow the general plan of the village. The latter is perhaps the most remarkable feature in the ground plan of this village. The addition of rooms attached irregularly at various points of the main cluster, which is necessarily consequent on long occupancy of a site, even without increase of population, was in this example just commenced. The result of the same process, continued over a long period of time, can be seen in the ground plan of any of the inhabited villages of today and in most of the ruins, while a plan like that of the ruin under discussion, while not unknown, is rare. [Illustration: Plate XXIV. GENERAL VIEW OF NORTHERN END OF A GROUP OF CAVATE LODGES.]