XVIII XX] Introduction 



Interpolating for a <r r first, 



xxxix 



log %- = 2-0737, 

 ,- = 6 O ov=-0514 log X " = 2-2537. 

 Hence for log ^ 2 = 2'124'J : 



We conclude tliat the equiprobable correlation is '53. 



(4) Imbecility and Deaf-mutism. 



log x* = 3-9039 <r,. = '0175, 



r = 0-95, O ov = -01, log x 2 = 4'3673 ; O ov = '02, log %> = 3'7660. 

 Hence: r = Q-<J5, ,<r r = '0175, Iog x 3 = 39163. 



Again : 



r = 90, O a r = -01 , log x 2 = 4-2207 ; <7 r = "02, log x 2 = 3'6197. 

 Hence: r = O'iM), <r r = '0175, log X 3 = 37699. 



Interpolating log x - = 3'9039 between 3 9163 and 3 7699, we find 



r p = 0-946. 



(5) Developmental Defects and Dullness. 



log x * = 3-5 1 28, u <r r = -0201 . 



r = 0-8, 0< 7, = -02, log x s = 3'4097 ; <r r = '03, log X 3 = 3'0598. 

 Hence : O o-,. = -0201, log x 2 = 3'4062. 



r = 0-!, <7 r =-02, log x == 3-6197; O o- r = -03, log X 5 = 3'2690. 

 Hence : log x -' = 3'6162, for O o- r = '0201. 

 Thus, by interpolating log x" = 3-5128 between 3'4062 and 3'6162, we find 



r f = -851. 

 We have accordingly the following results : 



It will be Been that equiprobable r p confirms generally the results from P, i.e. 

 the tables for ' Datura ' and ' Houses ' give no sensible association. r t also confirms 

 this view and shows that ' Houses ' is even lower in the scale than ' Datura.' The 

 order of r p is the same as that of Yule's coefficient of association Q, but neither 

 r p , r t , C s , P or Q support the conclusions stated to flow from the percentages on 



