BAROMETRIC PRESSURES ON THE GREAT LAKES 123 



and that the corrected elevation for this day was automatically rejected by 

 the criterion. 



Compare the graphs for Cleveland on October 27, 1910, as shown on plate 

 16, with those for Buffalo on that plate. Note that the barometric effects 

 and wind effects were much smaller at Cleveland on that day than at 

 Buffalo. Note that the minimum observed elevation occurred at Cleveland 

 at 8 a.m., within 2 hours of the maximum observed elevation at Buffalo, and 

 that the maximum observed elevation occurred at Cleveland at 8 p.m., 

 within 1 hour of being simultaneous with the minimum observed at Buffalo 

 at 7 p.m. The rise of 1.57 feet at Cleveland between 8 a.m. and 8 p.m. 

 corresponded to the fall of 4.60 feet at Buffalo between 10 a.m. and 7 p.m. 

 This is in accordance with the idea that this was largely an inertia effect, 

 the early part of a new seiche affecting the whole of Lake Erie and of the 

 character indicated on page 118. Note that in such a seiche Cleveland and 

 Buffalo water surfaces should normally be changing in opposite directions, 

 and that Cleveland minima should be simultaneous with Buffalo maxima, 

 and vice versa. 



GENERALIZATIONS. 



In the course of this investigation, during the consideration of the many 

 details of the evidence which it is not feasible to set forth here, and in the 

 efforts to develop an adequate theory, the writer has made certain general- 

 izations in regard to barometric effects, wind effects, and seiches. These 

 generalizations have not been fully established in some parts. Even in 

 such parts as are fully established it is not feasible to set forth the con- 

 siderations on which they are based, except in part. Yet even under these 

 circumstances it is desirable to give the generalizations as a guide to others, 

 to be used with such degree of caution as may seem best to them. The 

 generalizations are accordingly given in what follows. 



GENERALIZATIONS AS TO BAROMETRIC EFFECTS. 



Barometric effects upon the elevation of the water surface at a given 

 point are primarily proportional to the distance of said point from the center 

 of gravity of the area of the lake surface. Note the formulae for R w and #, 

 (17), on page 16. 



Barometric effects at a given point are also dependent to a considerable 

 extent upon the shape of the bottom and the configuration of the shore in 

 all parts of the lake. This influence shows in tending to make the P w and 

 P n proportionality factors in (19), page 17, depart widely from unity. 

 Note the value of the factors found in this investigation in table No. 16, 

 page 70. It is probable that these proportionality factors tend to be 

 much greater than unity, and the barometric effects correspondingly large, 

 on all lakes having long natural periods of oscillation, say more than 6 hours. 

 If the natural period of oscillation is much less than 6 hours, the water will 

 tend to respond to the rather slowly changing barometric gradients in such 



