324 KKPORTS ON THE STATE OF SCIENCE. 



It has been noticed that the areas of melanism are generally in those 

 large manufacturing districts which have a humid atmosphere or heavy 

 rainfall, and hence it has been assumed that smoke and moisture, aided 

 by natural selection, have produced the phenomenon. Mr. Tutt has 

 argued the case from this standpoint at great length in his pamphlet 

 ' Melanism and Melanochroism in British Lepidoptera,' No doubt it is true 

 that melanism is almost confined to the western side of Britain — that is, 

 the side most strongly influenced by the Gulf Stream ; and also that it 

 is most prevalent in the manufacturing, and consequently smoky, dis- 

 tricts of our island. But if smoke is an essential, how are we to account 

 for the numerous and marked examples of melanism in the Hebrides and 

 the Shetlands and Orkneys, where there is no smoke ? As we have 

 already seen, extreme melanism occurs there in many species which in 

 the melanic area of South-west Yorkshire are not at all affected by it, 

 and vice versa. And if in these districts it is caused by humidity only, 

 why do we find that in the fen and marshy districts of Norfolk and 

 Cambridgeshire, whei'e species absolutely live in fog and damp, and a 

 humid atmosphere as well, there is practically no melanism ? I know 

 Mr. Tutt maintains that in the fen districts there is plenty of melanism, 

 but I can only say that, with a lai'ge experience of fen collecting, I have 

 never seen it. That there are plenty of dark specimens of such species as 

 Chilo phragmitellus among the pale ones T admit, but such are not at 

 all on ' all fours ' with true melanism as we understand it in Yorkshire. 

 Again, if smoke and humidity cause melanism in Yorkshire, how is it 

 that in the melanic area some species which we know have a latent 

 tendency to melanism are not there affected by it ? That the latent 

 tendency to melanism in Acro7iycta menyanthidw is strong we are sure 

 from the fact that it is developed in extreme form at Selby and York, 

 independent of smoke ; but although it occurs right on the spot where 

 numerous other species, including one in the same genus, have succumbed, 

 it remains there absolutely untouched, though its habits in both areas are 

 the same. 



An almost analogous case is that of Nochta glareosa in the Shet- 

 land Isles. There is no smoke there, and still, in apparently similar 

 localities, and with the same habits, melanism occurs strongly in the 

 species. Yet in South Yorkshire, although occurring in plenty with 

 many of our most melanic moths, it retains practically untouched its pale 

 character. 



Then, supposing that smoke and humidity do cause melanism, we 

 have to face the question as to hovj they cause it. The obvious answer is, 

 of course, that it is natural selection in the first place, followed up by 

 heredity. But surely, if natural selection operated on the few chance 

 specimens which were a little darker than the type, we should have seen 

 a gradual darkening in all melanic species ; whereas we are pretty 

 certain that in many no such thing has occurred. And it is just here, I 

 think, where our great difficulty lies. Afterthe first dark specimens, if 

 in sufiBcient numbers, hei^edity is quite sufiicient to account for a rapid 

 increa.se, if we allow that the darkening is in any way a distinct advan- 

 tage to the species ; for we have proved over and over again by breeding 

 that the progeny of melanic parents have an inherent tendency to become 

 still darker with each successive generation. ; 



I have not alluded to a theory I have seen argued, to the effect that 

 melanism may be a reversion to the original forms of the various species, 



