BAROMETRIC PRESSURES ON THE GREAT LAKES 121 



fourth (continuous) graph shows the oscillations of the water surface, under 

 the influence of inertia, about the instantaneous positions of equilibrium of 

 the water surface fixed by the barometric gradients and the winds. 



Consider the information shown on plate 14, pertaining to Buffalo : 



Note that the barometric effect, as shown by the dotted graph, was 

 +0.22 foot at 1 a.m. on August 4, 1910, that it increased steadily to +0.46 

 foot at 8 a.m., then began to decrease at once, and decreased steadily to 

 +0.27 foot at 1 p.m. Thereafter the barometric effect fluctuated but little 

 and slowly until 1 p.m. on August 6. From 1 p.m. on August 6 to 11 a.m. 

 on August 7 the barometric effect decreased slowly and steadily from +0.28 

 foot to 0.11 foot. Thereafter the barometric effect remained nearly 

 constant. 



Note that during these 4 days the wind effect, as shown by the dashed 

 graph, was very small until 5 a.m. on August 4, increased rapidly after that 

 time to +0.32 foot at noon on August 4, then decreased rapidly to +0.01 

 foot at 9 p.m. on August 4, and thereafter increased slowly to +0.08 at 9 

 a.m. on August 5. From 9 a.m. on August 5 the wind effect increased 

 rapidly to +0.65 foot at 5 p.m., and then decreased still more rapidly to 

 +0.03 foot at 11 p.m. on August 5. Thereafter, during August 6 and 

 August 7, the wind effect fluctuated but little. It was not more than 0.01 

 foot at any time between 6 p.m. on August 6 and the end of August 7. 



From the graph of observed elevations, note that the water responded to 

 the barometric and wind impluses in the forenoon of August 4 and to the 

 reversed wind impulse of that afternoon. Note that the continuous graph 

 of corrected elevations is very irregular after these impulses, and that during 

 the period 9 p.m. on August 4 to 9 a.m. on August 5, when the water was 

 comparatively free from new impulses, there is some indication of a seiche 

 with a period between 3 and 4 hours and a range of more than 0.2 foot. 



Note that the water surface responded to the sudden large upward wind 

 impulse which occurred between 9 a.m. and 5 p.m. of August 5 and to the 

 still larger and more sudden downward wind impulse between 5 p.m. and 

 11 p.m. on August 5. Note that thereafter, during August 6 and 7, with 

 no large new impluses, the continuous graph (corrected elevations) shows 

 clearly a long-period seiche in progress. The period of this seiche is ap- 

 parently slightly in excess of 13 hours, as identified from the record of these 

 2 days. It is believed to be the 13.1 hour seiche for which the evidence is 

 summarized on pages 114-116. Note that the total range of this seiche on 

 August 6 is 0.7 or 0.8 foot and on August 7 is about 0.5 foot. Note its 

 comparatively regular form on August 7. 



Examine plate 15, showing the four graphs for Buffalo on October 21-22, 

 1909. Note that the water surface, as shown by the observed elevations 

 (the dot-and-dash graph), responded to the large and gradual barometric 

 impulse. The most striking thing shown on this plate is the short-period 

 seiche which was evidently in progress throughout the 2 days, continually 

 being modified and distorted but persisting. On the continuous graph 



