J. W. H. Harrison 223 



us. With the inhibition of birch and alder by the larches and pines the 

 insect, if it had to survive, had to accept substitutes either in one or 

 both of these. This it found in larch, upon which it subsists exclusively 

 in the western part of the wood. Its birch instinct has not, however, 

 been lost ; about half a mile within the wood where a little island of 

 birches lies surrounded by pine and larch, larvae can be beaten from 

 either birch or larch. Neither of the first two differences is germinal, 

 the other two stand upon a vastly different plane. Both are germinally 

 fixed and both remain unaffected by any amount of rearing under a 

 changed environmental complex. 



In investigating the change in the modal state of the wing pattern 

 one must guard against the assumption that any extension of the range 

 of normal variation has occurred ; what has happened, in reality, is a 

 very definite contraction of that range of such intensity as to^ throw 

 the modal condition into a position giving the two races facies so 

 diverse as to suggest an apparent discontinuity. This, examination 

 proves not to occur, for the alteration in mean proceeds directly from a 

 preferential preservation of the darker, suffused, feebly marked forms 

 and an accompanying elimination of the paler and the banded genetical 

 strains existing in birchwood 0. autumnata not only on Eston Moor but 

 on Waldridge FelP, Co, Durham; in Cos. Tyrone and Fermanagh, Ire- 

 land ; and Cos. Perth and Kincardine, Scotland. 



To emphasise the fact that we do not here encounter any develop- 

 ment of new varietal forms or range, just prior to writing the previous 

 sentence I deliberately and without bias took all of my long series of 

 pinewood insects and likewise those from the birchwood, and having 

 examined them individually, assigned each to what I should have con- 

 sidered its point of origin had they been placed before me for judgment 

 as being certainly from either pine or birchwood but exactly which un- 

 known. So treated I should have allocated eleven birchwood individuals 

 (10 males and 1 female) out of 71 (15-5 °/„) to the pinewood and six pine 

 specimens (5 males and 1 female) out of 144 (=4-16 7o) to the birch. 

 In connection with this test, in the case of these last examples a re- 

 markable fact was apparent when I placed them alongside similar indi- 

 viduals derived from the birchwood. I found that they resembled the 

 darker examples of their class, and further it was forced upon me that 

 considered simply as Cleveland insects, with particular history unknown, 

 they would have been deemed representatives of the alderwood in 

 Lonsdale — -a. totally unexpected result of a casual test. Whence it will 

 1 For the v^etation on Waldridge Fell, see Fig. 12, p. 239, 



