VT TNTBODTTrTTON. 



may not vary, or we may not have sufficient material to form a correct 

 opinion. My own experience tends frequently to the latter view, and 

 I have repeatedly noticed that species, presenting the most fixed and 

 constant characters in a certain locality, and which I have learned to 

 look upon as most constant in character, offer a very considerable 

 range of variation when obtained from another district, where a 

 different environment, or abnormal conditions have tended to develop 

 latent energies in the direction of variation. There are species, how- 

 ever, in which one rarely meets with the slightest divergence from what 

 we have learned to look on as their normal forms GonopJiora derasa 

 (in Britain), Cymatophora fluctuosa, Cuspidia strigosa, Tapinostola elymi, 

 Neuria saponarice, Chortodesbondii,Toxocampapastinum, and Xylophasia 

 lithoxylea are examples of this kind in different genera, in which the 

 species are normally inclined to vary. Leucania litoralis, too, rarely, 

 if ever, varies except in shape. It then varies in a most remarkable 

 manner in the direction of the pointed wings of Meliana flammea. 



In some species, we find a great amount of constancy in the indi- 

 vidual specimens of each sex, but considerable variation between the 

 sexes themselves. This sexual dimorphism is fairly common among 

 the NOCTU^E, although the sexes generally exhibit variation inter se, 

 and at the same time show a remarkable tendency to meet in some 

 particular form or forms. This tendency is exhibited in a very striking 

 manner by Agrotis cinerea,A.lunigera,A. valligera, A. sujfusa, Nonagria 

 geminipnncta, N. arundinis, N. neurica, Aporophyla australis, and very 

 many others. Of those species which are sexually dimorphic, and in 

 which the sexes are more or less variable in themselves, but in which 

 there is very little tendency for the forms of one sex to overlap those 

 of the other, may be mentioned the following : Chortodes arcuosa, 

 Ccenobia rufa, Hydrilla palustris, Viminia myricce, Neuronia popularis, 

 Eusina tenebrosa, and Noctua umbrosa among the more constant, and 

 Cuspidia tridens, C. psi, Nonagria cannce, and Agrotis puta among the 

 rather less constant species. 



We may now consider that most extreme form of variation, poly- 

 morphism, and mention some of those polymorphic species that show 

 endless variation in both sexes, where the different forms are equally 

 abundant in either sex, and where not only modifications of colour, but 

 modification of the size, direction, and position of the stigmata and 

 markings, give extreme forms which bear very little, if any, superficial 

 resemblance to each other. Among the NOCTUA these polymorphic 

 species are probably more abundant than in any other group. To 

 judge how far these species do vary, a very large quantity of material 

 is needful, and immense series of such species as Apamea didyma, 

 Agrotis tritici and -Tceniocampa instabilis are necessary to get the faintest 

 idea of the very great range of variation which these insects present. 

 Although these species are, perhaps, the most variable British NOCTU^E 

 we have, yet it must be confessed that Miana bicoloria, M. strigilis, 

 Crymodes exulis, Hydrcecia micacea, BryopMla glandifera, Noctua festiva 

 (and conflud}, N. xanthographa, Pachnobid hyperborea, Tceniocampa 

 gothica, Ortkona suspecta, Anchocelis pistacina, Dianthcecia carpophaga, 

 Charceas graminis, Agrotis segetum, A. cursoria, A. corticea, Eelo- 

 tropha leucostigma, Xylophasia polyodon and others offer a sufficient 

 range of variation to satisfy the most exacting. These species exhibit 



