J. W. H. Harrison 211 



pupae in response to an exposure to a period of diminishing daily tem- 

 peratures, it is clear that individuals derived from early hatching ova 

 and from more rapidly feeding up larvae would tend to be exposed 

 earliest to this optimum state of falling temperature and thus would 

 emerge earlier. On the contrary, individuals exhibiting retarded de- 

 velopment would tend to emerge so late that the rapid appearance of 

 early winters would destroy them before they could secure the perpetua- 

 tion of their species. In this fashion, by natural selection, a race 

 provided with early hatching ova and pupae would be built up ; Calluna, 

 not being a deciduous shrub, would by that very fact favour early emer- 

 gence and assist this development. 



Parallel with the acceptance of a Calluna diet another change, 

 probably connected with the difference in nutritive value of Calluna 

 and Betula to Oporabia, occurred, and that was size diminution — a 

 phenomenon that always manifests itself when larvae feed on foliage 

 less nourishing than their own special food and as happened with 

 0. autumnata larvae experimentally transferred from birch to heather. 

 But its germinal fixation as well as that of the diet habit itself as a 

 hallmark of the race is another matter. Are we dealing with genuine 

 Lamarckian effects or with a Weismannian case of parallel selection ? 

 Are insects whose germinal composition induces that small size which 

 would of necessity demand a shorter larval period than the normal being 

 selected with those undergoing selection because of their speedy feeding 

 up on other grounds ? Probably the bulk of present day opinion would 

 favour the Weismannian view but personally I am far from ruling that 

 of the Lamarckian out of court. 



The repeated failures in experiments projected to test the validity 

 of the inheritance of acquired characters savour too much of attempting 

 to determine all that occurs in a long express train whilst one compart- 

 ment flashes by. Besides, one must attach some value to evolutionary 

 experiments in bacteriology even if Sumner's^ results with mice and 

 Kammerer's^ with amphibia are to be neglected. It seems to me that 

 the experience of a hundred thousand years may sooner or later be 

 indelibly impressed on the germ plasm 'of the race. 



1 Sumner, " The Appearance in the Offspring of Artificially Produced Carental Modi- 

 fications," Am. Nat. Vol. xliv. (1910). Sumner, " Some Effects of Temperature upon 

 Young Mice and the Persistence of Such Effects in a Subsequent Generation," Am. Nat. 

 Vol. XLV. (1911). 



2 Kammerer, " Direkt induzierte Farbanpassungen und deren Vererbung," Zeit. f. ind. 

 Aist. u. Vererb. Vol. iv. (1911). Also similar papers. 



Journ. of Gen. ix 14 



