THE SHIFTING OF CLIMATIC ZONES. 



209 



In table 10 these cases are marked as neutral, but in reality it might be fair to mark 

 them as showing agreement. The next event in the history of the trees is an increase 

 in the rate of growth lasting for thirty years. During this period a tunnel carried off the 

 surplus waters of the lakes, and therefore no inundations are recorded, even though it is 

 possible that some would otherwise have occurred. In 1675, at the time when the rate of 

 growth of the trees was increasing most rapidly, a flood is said to have so injured the tunnel 



B. C. A. D. 100 



300 



15m 



400 500 600 700 800 900 1000 1100 1200 1300 1400 



Fig. 72— Changes of Climate in California for 2,000 Years. 



This figure 13 the same as the part of figure 50 after 100 u. c. but is plotted with a threefold greater vertical scale. 



that it caved in. Nevertheless, Mexico City was not flooded, probably because even after 

 the collapse of the tunnel it still functioned. In 1697 and again in 1707 two inundations 

 occurred, which are markedly out of harmony with the apparent rainfall in California. About 

 1730 the trees again rapidly increased their rate of growth, and Mexico suffered from 

 two inundations. It is probable that a somewhat general increase of rainfall stimulated 

 the Mexicans once more to attempt to get rid of the water, for in 175.5 a long series of 

 schemes for draining the lakes culminated in a really successful tunnel. After this time 

 the increasing number of artificial constructions makes it impossible to judge of the chmate 

 from the fluctuations of the lakes. Of the three inundations recorded between 1750 and 

 1800 the first occurred in a decade when the trees showed a marked decrease in growth, 

 whereas the other two occurred at a time when the trees increased their rate of growth. 



To sum up the matter we find records of twenty inundations, great and small, between 

 the times of Montezuma and Humboldt. All the great inundations took place at times 

 when the trees were growing rapidly. Including great and small, we find four cases where 

 there is a disagreement between the trees and the lakes, seven in which the matter is open 

 to question (since the data for individual years are not available for comparison), and ten 

 showing agreement. The suggestive point about the whole matter is that, in spite of minor 

 disagreements, the main eras of high water in the lakes seem to correspond with periods of 

 rapid or increasing growth in the trees. So far as this single line of evidence is concerned, 

 it seems to suggest that when the winter rainfall increases in the zone of prevailing westerlies, 

 the summer rainfall increases on the borders of the equatorial zone. It must be remem- 

 bered, however, that in some cases, such as 1623, the floods were due to winter rainfall, 

 suggesting that winter storms came farther south than usual. If the conclusion here 

 reached is really true, it appears that at certain past periods not only Mexico City but such 

 places as Yucatan probably had more summer rain than at present, while during the winter 

 not only did all parts of the country enjoy greater variations of temperature than is now the 

 case, but the parts now receiving rains from the trade-winds were drier than to-day, and 

 hence were covered with jungle and bush instead of forest and jungle. 



How great a direct effect upon the inhabitants may have been produced by these in- 

 ferred climatic changes it is hard to tell. In a later volume I hope to discuss this question 

 fully and to present the results of a large series of observations upon the state of human 

 activity under widely different climatic conditions. The results there to be given will 

 lie based upon absolute mathematical measurements of human efficiency. At present, 

 however, we must be content with a single example which illustrates one of the important 

 15 



