ODEZIA ATRATA AND ITS VARIATION. 223 



From these eggs I bred some five or six dozen moths. The early 

 stages presented some points of interest. As a matter of fact O. 

 atrata being one of our common moths, no one seems to have thought 

 it worth while to note its early stages, and nothing much is, I think, 

 recorded about it in our language. There is Buckler's account of the 

 larva, Avhich stands out almost alone, unless we refer to Mr. Tutt's 

 account of the egg {Knt. Bee, vol. xv., p. 838), which shows 

 meagreness, due to the use of a hand-lens in the examination. There 

 is, hoAvever, also a very curious oversight, viz., the remarkable sulci, 

 of which the egg possesses one on each side, are described as " a deep 

 longitudinal depression running up the whole length of one side of 

 the egg, making it exactly like a grain of wheat in shape." That 

 there was a sulcus on each side appears to have escaped observation. 

 The egg is, therefore, only half like a grain of wheat, whichever side 

 you look at is like the grooved face of the wheat, neither side is like 

 the rounded back of the grain, though the larger side of the egg is not 

 unlike it, the resemblance in appearance to a grain of wheat may 

 therefore stand, but there is no complete resemblance in forni {see 

 pi. xix., fig. 1). 



The imagines reared at Reigate present several females as well 

 marked as their parent, and several males not, at first glance, very 

 different from English specimens of 0. atrata, the majority approach 

 jn/renaica more nearly than typical atrata. The largest of the bred males 

 is about 84mm. in expanse, and the group averages altogether larger 

 than English atrata, which Meyrick notes as 24ram. to 26mm. (an 

 inch). Swiss specimens I have are even larger than the Pyrenean 

 (bred) examples. The var. pyrenaica is characterised by an abundant 

 sprinkling of brownish-yellow scales, giving it a paler and speckled 

 appearance — they are so prominent a feature that they at once attract 

 attention. None of my bred specimens were more densely clothed 

 with yellow scales than their mother, and some were very much less 

 so. Whether this is the normal state of matters in the wild state, or 

 whether the loss of yellow scales was due to breeding in England in 

 captivity I cannot z'^y. No specimens were without yellow scales. 

 One feature that I have never noticed in English specimens, and is 

 most marked in the Pyrenean ones with a medium supply of yellow 

 dotting, less so in the others, is that the upper wings are more plenti- 

 fully sprinkled than the hind ones, and the effect is to deprive the 

 insect of the appearance of all the wings being of the same colour and 

 texture (as usual in butterflies, and some other day-flying insects), and 

 to give the definite appearance of upper- and underwings, so usual in 

 Noctuids and many Geometrids. 



I have already called attention to the variability in my series of 

 the amount of yellow scaling, and this brings me to a point that has 

 astonished me. This is, that in Swiss and English 0. atrata, a 

 majority of specimens present a fair sprinkling of yellow scales, and 

 yet in no description of the insect I have referred to is it described 

 as otherwise than (except the apical Avhite line) absolutely and com- 

 pletely black. It is true that these non- Pyrenean examples look black 

 even on close observation, but here and there one may be seen with 

 the yellow scales visible to a slight scrutiny, and in very few are they 

 seen to be quite absent, when examined with a hand lens. Still, only 

 a few exceed, if they do exceed, the one or two Pyrenean specimens 

 that are the blackest and most free from coloured scales. In this 



