PEDIGREE MOTHS. 61 



size or anything else), and whatever may be the kind of animal 

 or plant experimented on. It will depend on the amount of the 

 ancestral divergencies, measured with a special and relative unit 

 (' probable error,' as mathematicians call it), that I have often 

 written about, and cannot stop now to describe. This unit 

 enables us to treat on equal terms individuals of either sex, or 

 those in separate broods that have been affected by differences of 

 nourishment, &c. I have shown the rate of divergence to be the 

 same within the limits of statistical error, in the case of (1) weight 

 and size of sweet-peas ; (2) human stature ; (3) human eye-colour. 

 The course of investigation pursued is necessarily technical. It 

 will be found described in ' Law of Regression ' (Journ. Anthrop. 

 Inst., 1885); 'Family Likeness in Stature' (Proc. Royal Soc, 

 1886) ; ' Family Likeness in Eye-colour ' (Proc. Royal Soc, 

 1886)." 



Acting on the suggestion of Mr. Merrifield, Mr. Galton pro- 

 posed Selenia tetralHna7'ia {illustraria) as a suitable moth for his 

 experiment, but the majority of the Fellows, who took part in the 

 discussion following the reading of paper, did not concur in this, 

 and various other British and some exotic silk-producing species 

 were put forward. 



What is required is, as pointed out by Mr. Galton, a moth 

 that is " hardy, quickly breeding, of small size, easily measured 

 and preserved, and bearing broods of about 50 to 100 individuals." 

 Selenia tetralunaria possesses most of these requirements, but, 

 although it is double-brooded, it can hardly perhaps be called 

 quick breeding. Further, would not the results obtained from 

 the two broods of this insect be almost on a par with those 

 obtained from two distinct species ? — tetralunaria, the spring 

 brood, would produce cestiva {delimarla), the summer brood ; and 

 (estiva, tetralunaria certainly ; but the comparisons would be, 

 spring brood with spring brood, and summer brood with summer 

 brood. Mr. Galton, however, is of opinion that allowance can be 

 made for any difference in size known to exist between individuals 

 of the seasonal broods. 



Size is, no doubt, under the direct influence of nourishment, 

 and is perhaps the only character in the perfect insect that is 

 directly affected by food. As is well known, many breeders of 

 Lepidoptera pride themselves on rearing large specimens, and it 

 is not at all an uncommon thing to see bred specimens of a 



