SOUTHERN MEXICO AS A TEST CASE. 97 



side toward Chapultepec, where the water was too shallow. The small boats engaged in 

 ordinary traffic sailed everywhere, however, not only on Tezcuco, but on the other lakes 

 and on the connecting rivers. It is not evident whether this was a permanent condition, 

 but in 1.553, as appears below, we have evidence of a special inundation. 



Two or thi-ee generations after the Spanish conquest the condition of the City of Mexico 

 had changed. It had ceased to be an island, the canals had become dry, and wheeled 

 vehicles had taken the place of canoes. This result was due in part to the construction 

 of additional dikes, but nature apparently had been the main agent in the matter. Such 

 seems to have been the opinion of Torquemada. He is quoted by Prescott (page 33) : 



"As God permitted the waters which had once covered the whole earth to subside, after man- 

 kind had been nearly exterminated for their iniquities, so He allowed the waters of the Mexican 

 lake to subside in token of good will and reconciliation after the idolatrous races of the land had 

 been destroyed by the Spaniards." 



The waters rose again, however, for, to quote Torquemada,* "in this same year 1604, 

 it rained so much in the month of August that the lake of Mexico was filled with all its 

 plains, so that the waters covered nearly all the city and reached such a point in some 

 streets that people passed in canoes, and I myself passed San Juan in this manner. The 

 inhabitants Uved carelessly, and forgetful of the previous danger of the same kind in the 

 year 1553 when Don Luis de Velasco, the Fu-st, was governor. ..." This wet period 

 continued, for we are told that in 1607 the town of Tultitlan was inundated for the tliird 

 time with great loss of houses and fields. To prevent such occurrences in the future a tunnel 

 was built to carry off the surplus water of the Cuautitlan River. It might be supposed 

 that, after the construction of the tunnel, the lake would never return to its natural condi- 

 tion. In 1629, however, during a season of uncommonly heavy floods, the tuimel was 

 stopped up completely. The City of Mexico was flooded for a time and was in great straits 

 during a period of rainy years lasting till 1634. Thereafter it became dry once more, 

 although neither the tunnel nor the old dikes were in a condition to prevent the rise of the 

 water. Again, from 1675 to about 1755, the tunnel was closed, being filled with earth for 

 an unknown distance. At the same time also the dikes were in poor repair, breaking 

 whenever the water rose higher than usual; yet the city continued to stand on dry land, 

 though sometimes a year of exceptional rains caused the water to rise sufficiently to flow 

 into some of the streets, but not enough to do any serious damage. Taken as a whole 

 the history of the lake appears to have been characterized by fluctuations of considerable? 

 magnitude. How far these fluctuations agree with those in regions farther north will 

 appear in a later chapter after we have considered the data derived from trees. 



The evidence just presented is in itself too slight to justify any conclusion derived 

 from it alone. Only by bringing together many diverse lines of evidence can we ascertain 

 the truth even approximately. Fortunately, Mr. Manuel Gamio, under the direction of 

 Professor Franz Boas, has recently been engaged in archeological excavations on behalf 

 of the International School of American Archeology and Ethnology, and has done some 

 work which is significant for our present purpose. The hamlet of San Miguel Amantala, 

 near the village of Azcapotzalco, lies on the edge of the lacustrine plain of the City of 

 Mexico, not far from the base of the hills on the west. This portion of the plain is dotted 

 with little mounds which mark the sites of villages or small groups of houses built by the 

 Aztecs and fuU of the typical pottery, images, and other relics of that people. Elsewhere 

 the plain is strewn with the scattered fragments of another and older type of civihzation, 

 which is known as that of San Juan Teotihuacan, from the great pyramids of that name 

 on the eastern border of the basin of Mexico. The San Juan reUcs never occur in mounds 

 of the Aztec type except for a few stray bits which have been carried in by accident. This 

 indicates that the two are of distinctly different dates, as indeed we know from other 



8 * Loc. cit. Book v, Cli.ap. uc, p. 728 b; and Chap, lxx, p. 756. 



