378
THE POPULAR SCIENCE MONTHLY.
Here the horizontal lines represent the number of feet below the plane of 1838.
It requires but a glance at the diagram to show that some relationship exists between the lake and rainfall periods. The first impression conveyed is that the curves are in opposition—that the high in one corresponds with the low in the other, and the reverse. But I think
the true relation will be made to appear when we notice the important fact (which I endeavor to render more apparent by dotted lines), viz., that the water extremes lag behind the rainfall extremes—following them at intervals of from two to four years. Thus the seeming want
