Pedigree Moth-breeding . 21 



afford a standard whence the divergencies of the A and Z 

 lines in each generation can most suitably be measured. 

 M will be affected, together with A and Z, by all the 

 influences that affect the entire stock, and will therefore 

 exhibit in an unmixed degree the A and Z peculiarities. 

 It will easily be understood how important it is to attend 

 to the requirement already laid down that the three 

 lines of descent should be carried on under idea lira! 

 conditions. 



The broods in the M and Z lines will be mounted in 

 the same way as those in the A line. Therefore for 

 every generation there will be three compartments in 

 one or more trays, each containing (say roughly) 100 

 moths, and, as there are two generations in a year, the 

 result of each year's breeding will be to fill six com- 

 partments. 



After the sixth generation or thereabouts has been 

 reached in each of the three lines of descent, it is 

 further desired to proceed conversely, by breeding 

 from medium specimens in each of the three lines, and 

 again from medium specimens of their several broods, 

 and so on until all trace of the A and Z peculiarities 

 shall have disappeared from their respective descendants. 



I have spoken of one or more pairs, because the moths 

 do not emerge simultaneously, and yet they must be 

 paired soon after they emerge. The ideal pair in the A 

 line would be the very longest-winged male and the very 

 longest-winged female of the entire brood, and in the 

 Z line the very shortest-winged. In practice we must 

 be content with an approximation to this. Two or three 

 separate matings will have to be made between the 

 most suitable of the brood at the time when they 

 require to be paired, but the eggs that are to be pre- 

 served and reared will be those of only one or two of 

 the most suitable of the trial pairs. The produce of 

 different pairs ought not to be mixed. 



There will be little difficulty about the M pairings, as 

 mediocrities are numerous. 



The wing-length is the characteristic with which it is 

 at present proposed to deal, as being more definite and 

 easily measured than others. A similar treatment 

 might be adopted in respect to other characteristics, 

 such as the area of the wing, or the area of the patch 

 of colour in the wings of such moths as may possess it. 



