BAROMETRIC PRESSURES ON THE GREAT LAKES 105 



next lake below in the chain, is shown for each lake by a line made up of suc- 

 cessive groups of two short dashes and a dot. This line is plotted from the 

 measured values of the rainfall on the lake surface, as determined by rain 

 gages around the lake, and from the measured inflow and outflow in the 

 streams connecting the chain of Great Lakes. Note the smoothness of this 

 line. In general, it rises or falls less than 0.02 foot in a day, usually not 

 more than 0.01 foot. In other words, the fluctuation of the elevation of the 

 mean surface of the lake due to rainfall, inflow, and outflow is normally at 

 the very slow rates stated. The extreme case anywhere on these graphs is 

 the rise of 0.19 foot in 3 days, October 4-7, 1910, of Lake Erie, due to ex- 

 tremely heavy rainfall over the whole lake surface on those days. 



ACCURACY AS TESTED BY GRAPHS. 



The mean surface of the whole of any one of the Great Lakes changes 

 only as the total content of the lake changes. That total content changes 

 from five causes only: (1) rainfall on the lake surface, (2) inflow from the 

 next lake above, (3) outflow to the next lake below, (4) run-off into the 

 lake from the surrounding land-drainage area, and (5) evaporation from the 

 lake surface. 



The dot-and-two-dashes graph, commented upon above and based upon 

 direct measurement, shows how slow and regular is the variation of elevation 

 of the mean lake surface due to the first three of the causes enumerated in 

 the preceding paragraph. It is believed that the variation in elevation due 

 to the fourth and fifth causes run-off and evaporation is even smaller and 

 more regular than that due to the first three. The run-off and the evapo- 

 ration in question have not been measured directly. From sources of in- 

 formation which are in part external to this investigation, it is estimated 

 that during the months June to October of each year the run-off into Lake 

 Erie from the surrounding land-drainage area is such as to produce a rise 

 from 0.004 to 0.040 foot per day in the mean lake surface, with only a small 

 percentage of days in which the rise is more than 0.020 foot. For Lake 

 Michigan-Huron the run-off expressed in the same terms is even more con- 

 stant. So, too, from external evidence, it is estimated that on either lake 

 during the season June to October the evaporation produces a fall in the 

 mean lake surface varying from but little more than 0.000 foot on some days 

 to 0.021 foot on days of extremely rapid evaporation. The considerations 

 indicated in this paragraph lead to the belief that the actual variation in the 

 elevation in the mean surface of either Lake Erie or Lake Michigan-Huron 

 is so slow and regular as to be properly represented by a graph but slightly 

 less smooth and regular than the dot-and-two-dashes graph shown in plates 

 7 to 13. It is believed that the actual variation of the mean elevation of the 

 whole surface of either lake is as a rule about 0.01 foot or less in a day, is 

 sometimes somewhat more that 0.02 foot in a day, and only on very rare 

 occasions exceeds 0.08 foot in any one day. 



