12 EFFECTS OF WINDS AND OF 



has been adopted for these contours coinciding with the location of the isobar 

 marked 29.90. Note, also, that the interval between contours is 0.0113 

 foot, which corresponds, in accordance with equation (4), to the interval of 

 0.01 inch between the isobars shown on the upper part of the plate. Note, 

 that the surface gradient of the water of Lake Erie is shown as upward to the 

 northeastward in accordance with the barometric gradient shown on the 

 upper part of the plate as downward to the northeastward. 



Plate 1 is a concrete illustration of the relation between isobars and surface 

 contours at any water surface which must exist if the water is in equilibrium 

 under the influence of gravity and barometric pressure. 



BAROMETRIC EFFECTS IN TERMS OF BAROMETRIC PRESSURES. 

 In setting up observation equations to express the relation between ob- 

 served fluctuations in water elevation and barometric pressures as shown on 

 the forecast maps it would be desirable to utilize accurately the relations 

 shown in equation (4) and illustrated in plate 1, if limits of time, expense, 

 and accuracy did not prevent it from being feasible. But a reconnaissance 

 of the problem (including an attempt, on a small scale, to utilize the exact 

 relations) indicated that in order to keep within limits imposed on this in- 

 vestigation it was necessary to introduce an assumption, called assumption 

 No. 1. 



ASSUMPTION No. 1. 



It is assumed that, with sufficient accuracy for the purposes of this in- 

 vestigation, the isobars at the instant represented by any forecast map 

 (8 a.m. or 8 p.m., 75th meridian time) are straight and uniformly spaced 

 within the limits of the lake under consideration. 



Assumption No. 1 is nearly true for the actual case illustrated by plate 1. 

 In general, for each of the Great Lakes assumption No. 1 is nearly true. 

 The size of any one barometric low-pressure area or of any high-pressure 

 area is usually many times as great as that of any one of the lakes. Within 

 the area occupied by the lake the isobars curve but little and there is but 

 moderate departure from uniform spacing. The most serious departures 

 from assumption No. 1 in these two respects ordinarily occur when the center 

 of a well-developed low-pressure area is over a lake. One may verify these 

 statements by studying the forecast maps. 



Assumption No. 1 was introduced to save time and expense. The ulti- 

 mate effect of its introduction in reducing the accuracy of the final computed 

 barometric effects is believed to be moderate only. 



Assumption No. 1 combined with the relations between isobars and con- 

 tours on the water surface which have been commented upon and which are 

 fixed by equation (4) makes the contours on the water surface straight and 

 uniformly spaced. In other words, under assumption No. 1, whenever the 

 water is in equilibrium under the influence of gravity and barometric pres- 

 sures its surface is plane. 



In the case illustrated in plate 1, for the actual isobars there shown one 



