( 57 ) 



outer part ot' a cyclone and also — as follows from (2) — for 

 constant winds and straight isobars. 



In Table III the values of /■ and m, calculated in this wav are 

 giveji. 



TABLE III. 



\'alues of the friction /• and the angle of deviation in. 



d e Bi 1 t. 



The friction coefficient / = itk = k X '^•'^ X 10— •"' then becomes : 



de Bilt 6.09 3.75 3.32 7.57 4.58 



Terschellino-erbank 12.03 4.80 4.44 8.05 6.86 



Xio- 



'&^ 



Oberbeck assumes arbitrarily / = 12 X J0~' > Wien') / =: 8 X JO—"» . 



It is, of course, doubtful whether a friction coefiticient, as 

 calculated for the case of feeble periodic winds may be used also 

 for non periodic winds of any force ; but the fact that Van Ev^erdingen") 

 on determining the angle of deviation directly from weather charts 

 and for strong gradients (one mm, and morel for de Bilt and taken 

 over the whole year, comes to almost exactly the same value (69°) 

 as that given in Table III (68°), can hardly be considered as a merely 

 accidental coincidence. 



6. With the aid of form. (5) the quantities // and /, /; being 

 known, can easily be calculated; this calculation has been made only 

 for the mono-diurnal variation. Putting 



dp dp 



— ~ = G„ (gradient), —~ = (t, 

 Oy O.v 



further 



1) WiEN. Lelirbuch der Hydrodynainik, 1000, p. 294. 

 ^) Phys. Z. S. 9, J 908. Veiii. d. D. Pliys. Ges., 190S. 



