BR. .1. I'N'I. NAT. HIST., 7: 1994 



is not noted, but progeny from a number of crosses suggest that half-melanics and 

 melanics are controlled by separate single loci, with their respective alleles both 

 dominant to the typical alleles (Majerus, unpublished). Jones (1993) questions 

 KettlewelFs suggestion that melanism in E. bistortata is industrial in origin, noting 

 the occurrence of melanic individuals from a number of rural regions. 



In both A. monoglypha and H. furcata, Kettlewell (1973) reports the existence of 

 ancient, non-industrial melanics. In neither species has the inheritance of the colour 

 pattern variation been analysed, but the continuous variation in A. monoglypha is 

 likely to be under polygenic control, with the darkest form, f. aethiops Tutt (not present 

 in the Juniper Bottom samples), possibly being due to a single gene showing incomplete 

 dominance (Kettlewell, 1973). In H. furcata preliminary analysis suggests at least Rve 

 gene loci are involved, three of these respectively controlling ground colour, flecking, 

 banding, and two affecting melanism (Majerus, unpublished data). That the genetic 

 control of melanism in the two species which do not show sharp morph-specific 

 frequency changes is more complex than that in any of the species which do show 

 this phenomenon, raises the possibility that the mode of inheritance of melanism, 

 as well as its time orgin, may be important in the evolution of morph-specific habitat 

 preferences. If the mechanism by which a particular type of behaviour (such as 

 actively preferring a particular habitat) becomes associated with a specific phenol ype 

 is because the genes controlling the behaviour and the phenotype are tightly linked, 

 as suggested by Howlett (1989), such linkage may only be possible if melanism 

 is controlled by one or two gene loci. In species where melanism is controlled 

 polygenically, it may be that morph-specific habitat preferences and morph-specific 

 resting site preferences can only evolve if choice is made through a mechanism such 

 as KettlewelFs (1955b) contrast-conflict hypothesis of resting site selection. That is 

 to say if a moth's behaviour is dependent on its own perception of its phenotype. 



Acknowledgements 



We wish to express our thanks to the National Trust for permitting us to trap in 

 Juniper Bottom and John Bebbington, the staff of Juniper Hall field centre and the 

 Field Studies Council for providing facilities at Juniper Hall. Liam Cox, Sarah Hunt. 

 Clare Mason, Catherine Martin, Bryony Green and Clare Dornan assisted with moth 

 trapping. June Hunt typed the manuscript. Tamsin Majerus corrected the manuscript. 

 Carys Jones is supported by an SERC Quota Award. Michael Majerus is the recipient 

 of an NERC grant. 



References 



Aldridge, D., Jones, C. W., Mahar, E. & Majerus, M. E. N. 1993. Differential habitat selection 



in polymorphic Lepidoptera in the Forest of Dean. Entomologist's Rec. J. J ar. 105: 203-214. 

 Howlett, R. J. 1989. The genetics and evolution of rest site preference in the Lepidoptera. Ph.D. 



thesis. Cambridge University. 

 Jones, C. W. 1993. Habitat selection in polymorphic Lepidoptera. Ph.D. thesis. Cambridge 



University. 

 Jones, C. W., Majerus, M. E. N. & Timmins, R. 1993. Differential habitat selection in 



polymorphic Lepidoptera. Entomologist 112: 118-126. 

 Kearns, P. W. E. & Majerus, M. E. N. 1987. Differential habit selection in the Lepidoptera: 



a note on deciduous versus coniferous woodland habitats. Entomologist 's Rec. J. 1 ar. 99: 



103-106. 

 Kettlewell, H. B. D. 1955a. Selection experiments on industrial melanism in the Lepidoptera. 



Heredity 9: 323-342. 



