if L is taken as its thickness, 450 centimeters, and X is the distance measured 

 upward from the till underlying the peat. F is determined to be 2. 14 x 10 A 

 by equation (2). 



It is not unlikely that the peat acts as a wick, conducting water upward 

 at about the rate at which it evaporates from the surface. Assuming this to 

 be the rate at which sea water evaporates at 40° N. , a - 3 x 10 cm. /sec. , 

 which is only one tenth the velocity of the thermal wave, and consequently 

 would not greatly influence its determined value. 



Substituting this value in equation (2) the values of the requisite associ- 

 ated eddy diffusivity. A, may be estimated as it varies with depth, -t- as shown 

 in Table 2, Except close to the surface eddy diffusivity becomes sufficiently 

 small, relative to the thermal diffusivity to have little effect on the observed 

 values for the latter. 



The preceding considerations do not lead to unique solutions for the 

 water motions. They indicate however that on reasonable assumptions the 

 distribution of chloride may be reconciled with the observed thermal cycle on 

 the assumptions that: 



1. The interstitial water is very nearly stagnant. 



2. There is a movement of ground water upward at a rate not greater 

 than that of evaporation. 



3. The interstitial water is being mixed by weak eddy motions which 



decrease in intensity with depth below the surface. 



References 



Joos, G. -Theoretical Physics, New York, 1934. 



Arons, A. B. and H. Stommel-A mixing length theory of tidal flushing. 

 Trans. A. G. U. 32, 419, 1951. 



DISCUSSION 



Chapman: I am most interested in the contribution that Dr. Redfield has 



given. It raises a number of important points I think. The 

 first question I would like to ask him is the distance that these 

 observations were made from a creek, because one of the im- 

 portant things that we have found out, not only on the New Eng- 

 land marshes but also on the New Zealand and British marshes, 

 is that the up and down movement of water varies with the dis- 

 tance from the creek, and results that you have obtained here 

 might hold well for a certain distance from a creek. The ques- 

 tion is, would they then hold for a greater proximity to a creek? 

 The second point I would like to ask Dr. Redfield is whether he 

 has any views on this: there are two effects on the water table 



81 



