iQ2o] SHULL— SEEDS 385 



that plotting logs of velocities against temperatures will yield some 

 kind of a curve, but there are not enough data at hand to determine 

 anything as to the character of the curve. The general conclusion 

 to be drawn from this part of the work is that the evidence, as far 

 as it goes, is rather against the assumption that the velocity of 



i 



intake is an exponential function of the temperature. 



Another point that deserves notice is the nature of the curves of 

 water intake. Brown and Worley called their curves paraboloid 

 and described them as running out toward a common asymptote. 

 The language, of course, must have been intended in a very loose 

 sense, for parabolic curves passing through a common point of 

 origin, as theirs do, could never have a common asymptote. It was 

 found impossible to fit a parabolic formula to the intake data pre- 

 sented, but from the figures given in tabular form (tables VII-X and 

 XIV) it is evident that the logarithmic curve y = a log IO (bx+i)+c 

 may be made to fit the data very closely. Furthermore I have 

 taken the 3. 8° barley data and attempted to fit to it both the 

 logarithmic and a hyperbolic equation made to pass through the 

 origin and the second and fourth values of their data. I have 

 found that the logarithmic equation fits much closer to their data 

 than the hyperbolic equation. The two sets of values and the 

 original data are given for comparison. The time and data 

 columns are from Brown and Worley. 



™. 1-, , Computed Computed 



Ume Data (logarithmic) (hyperbolic) 



5.58 hours 4 



24.75 hours n 



48.83 hours 18 



72.25 hours 23 



96.50 hours 27 



144.25 hours 34 



42 4 



82 11 



52 18 



42 23 



42 27 



02 33 



41 5.21 



82 11.82 



49 17-99 



43 23.42 



56 28 . 78 



89 38.99 



The logarithmic equation used in this comparison is ^ = 48.6 

 logic (0.0252+1), and the hyperbolic equation, y = 0.2024 



l 7 ^4-112. 9882. 



Considering the closeness of agreement which is obtainable with 

 the logarithmic formula, it seems more reasonable to consider the 

 curves of water intake, even in the case of barley seeds, as logarith- 

 mic rather than hyperbolic. 



