ON THE CORRELATION OF PATTERN IN RHOPALOCERA. 



177 



of forms of Ai/lia tan : this is followed by a more detailed account (in 

 German and French), by Dr. Standfuss, of crossing of mutational and 

 other forms, their relative fertility, etc. 



The first 21 plates in the volume are portraits of eighteen Lepidop- 

 terists, all well-known to English entomologists, the most interesting 

 being, perhaps, the six first, Boisduval, Herrich-Schaefler, Rambur, 

 de Graslin, Guenee, and Milliere, and the two last, Reverdin and 

 Oberthiir. The remaining plates we have already referred to as fully 

 as space permits, being by M. Culot, praise of them is superfluous. 



On the Correlation of Pattern and Structure in Rhopalocera with 

 special reference to the Ruralidae. {With aeroi idate.s.) 



By G. T. BETHUNE-BAKER, F.L.S., F.Z.S., F.E.S. 

 {T]te subject of a paper read before the British Associntion.in Birmingham in 1913.) 



It is many years ago since Schoyen's discussion on Lijcaena arnus 

 and L. aegon (as the two species were then called) took place, and I 

 only mention it now because it was his papers that impelled me to 

 vmdertake what had till then been a more or less spasmodic investi- 

 gation, viz., a thorough systematic study of the sexual armature of 

 butterflies, and especially of that group of butterflies to which those 

 two species belonged. I felt that we ought not to depend solely upon 

 pattern for the dift'erentiation of closely allied species, and I therefore 

 at once set about making microscopic preparations of all the Palsearctic 

 species of the Ruralidae. This naturally led on to a much wider field 

 of research, extending beyond the Rhopalocera and also beyond the 

 Lepidoptera. The taxonomie value of these organs gradually pressed 

 itself forcibly upon me, until to-day I regard them as necessary to the 

 correct grouping of the Puiralidac, and probably (I do not say certainly) 

 of other families of Rhopalocera, if not of the Lepidoptera as a whole, 

 so that in any cases of doubt after the neuration, I should first 

 investigate the sexual organs. 



Long continued study of these organs gradually brought to light 

 the fact that a marked change of form in them was also accompanied 

 by a change in pattern in the species and the genus. From the very 

 beginning I had learned that there were small specific difl'erences, but 

 it was only a wide experience that could show their value from the 

 taxonomie point of view, and it is this view, especially, that I want to 

 lay before my readers, viz., that so far as the linralidae in their widest 

 sense are concerned, it is a fact that change of structure is accompanied 

 by change of pattern, or, vice versa, change of structure accompanies 

 change of pattern. 



As my first instance may I give one of the species already mentioned, 

 Plebeiiis argns (the type of the genus riebeiim), and compare with it an 

 allied genus Celastrina, whose type is aryiolus. The difi"erence in 

 colour is at once apparent, whilst the pattern of the underside is very 

 diverse. 



These distinctions are followed by an equal change in the male 

 armature. P. anjus (PI. xiv., fig. 1), has the longish clasp, the gently 

 projected (backwards) girdle, the tegumen strongly bifid and very 

 narrow in front with strong falces or hooks, and with the -apical hood 

 fairly broad. In C. argiolua (PL xiv., fig. 2), the clasp is totally 



