96 T. H. MORGAN 



between yellow and lethal occurs four times and the other cross- 

 over, viz., that between lethal and white (to give gray white 

 males) is not realized at all. The sons of the non-lethal sisters 

 involve only two pairs of characters, hence there are two contrary 

 single cross-over classes, viz., yellow eosin males and gray white 

 males. These are expected in equal numbers as is practically 

 reahzed (7YR and 8GW). 



The data for non-lethals permits us to calculate again the dis- 

 tance of yellow from white (or eosin). The sum of all the male 

 classes (812 + 760 + 7 -F 8 = 1587) into the sum of the two 

 cross-over classes (7 + 8 = 15) gives a percentage of .95 which 

 is very close to the value (1.1) given by other experiments. 



The question naturally suggests itself as to whether the re- 

 appearance of lethal I was due to mutation, or to descent from 

 a common original stock. The stock used by Miss Rawls ('12) 

 was wild stock that had been in the laboratory about a year 

 when she found her high ratios running in it. Now the yellow 

 stock, the white eyed stock, and the stock with abnormal abdo- 

 men had arisen before 1912, and had been maintained as pure 

 stock. Hence the new lethal appears to have arisen as an inde- 

 pendent mutation. Contamination is excluded on the grounds 

 that it is well nigh inconceivable that a lethal could be inserted 

 between yellow and white by a double cross-over, after con- 

 tamination in the yellow white stock. 



Summary of linkage data of Part I 



From the preceding pages the data bearing on the linkage 

 values have been collected and may here be summarized. In the 

 following table for each linkage value the total number and the 

 number of cross-overs is given; from which the percentage of 

 crossing-over is calculated and is given in the last column : 



total 



Yellow lethal 1015 



Lethal white 2559 



Yellow white 5564 



Yellow miniature 131 



Lethal miniature S14 



White miniature 994 



