STUDIES ON HIGH AND LOW NON-DISJUNCTION 101 



X used is thus 33 — 5,r. units or 27,5 units and the maximum len^'th 

 is 44^4 — 1,5 or 42,0 units. For the case of simplicity we may approxi- 

 mate the percentage from table 14 as 10,3 % (which of course is 

 allowable since the mean error is 0,56 %). We have thus the following 

 two equations for the determination of the limits of A.- 



10,3 = 4,3 + 27,5 A- or A = 0,22 

 and 



10,3 = 4,3 + 42,9 A or A = (),u. 



From table 16 w^e likewise find that since the endpoints of A' 

 lie between scute and eosin and between cut and vermilion the mini- 

 mum length is 18,5 units and the maximum is 33 units. If also here 

 for the case of simplicity we approximate the percentage as 9,3 % 

 (wiiich is justifiable since the mean error is 0,9o) we have the following 

 equations for the determination of the limits of A 



9,3 = 4,3 + 18,5 A or A = 0,27 

 and 



9,3 = 4,3 + 33 A or A = 0,i5. 



In the first case the mean value for A is 0,i8 and in the second case 

 it is 0,21. The mean from both cases is thus A = 0,i95 or in order to 

 have a simpler value we may take A = 0,2. We have thus finally found 

 the relation between the percentage of exceptions and the length of 

 the part of the .Y with respect to wjiich the exceptional females are 

 homozygous to be of the form 



y = 4,3 + 0,2 z. 

 Let us now see how this equation fits the figures from the different 

 tables. That it fits the tables 14 and 16 is immediately clear since 

 we have derived the value k = 0,2 from these tables. 



In the tables 7, 8 and 9 we had the badly located right endpoint 

 of the part of the X of which we know^ only that it lies between tan 

 and forked. It is thus almost certain that our equation here must fit 

 the figures. In tables 7 and 8 the left hand end of the part of the X 

 lies between echinus and cut and the minimum and maximum length's 

 of the part of the X is thus in both these cases 7,5 and 51 units respec- 

 tively. In table 7 the percentage is 9,3 giving with the aid of our equa- 

 tion a length of 25 units for the piece of the A' and in table 8 the per- 

 centage is 6,9 giving a length of 13 units. In table 9 the left hand end 

 of the part of the X used lies between scute and eosin and hence the 

 minimum and maximum lengths of the X are here 26,0 and 56,5. As 

 the percentage in table 9 is 12,8 we find the length for the part of the X 

 used to be 42,5. 



