CLASSIFICATION OF BUTTERFLIES BY THEIR ANTENNAE. 123 



for it outside the Lepidoptora. The primary antenna within the order, 

 is one with scales and hairs uniformly distributed over the whole 

 antenna. 



In Micrupteryx (calthdla) a very curious specialisation occurs that 

 takes it out of the direct line of antennal evolution, but the distribution 

 of hairs and scaling is still quite impartial as to ventral or dorsal 

 arrangement of either. In Eriocrania [jmrpurdla) the uniform distri- 

 bution of hairs and scales over the whole antenna is unquestionable, 

 and this is the primary type, from which the further development of 

 antennal structures and their varied arrangement on the surface of the 

 antenna as evolution proceeds, must be traced. The same arrange- 

 ment is preserved, with little change, in the lower Neolcpuloiitera, 

 Nepticulids, Cochlidids, Lithocolletids {(iraciiaria, &c.), so far at 

 least as regards an universal clothing of scales to the antennae, and in 

 most of these it is necessary to remove the scales in order to view the 

 sense hairs. In CucJilididcw the arrangements are most interesting, 

 since, wherever pectination occurs, the underside of the pectinations 

 have hairs only, but wherever there are no pectinations the scales are 

 universal, with the hairs amongst and beneath the scales, perhaps a 

 little more abundant ventrally. It might be difficult to say whether 

 this totally scaled antenna is the result of the pectinations disappearing, 

 or less probably because they have never appeared. It is worth 

 noting that the pectinations occur by basal origin, and so the distri- 

 bution of hairs is parallel to that in the long antennae of the Adelids. 



Yponyuu'uta presents an interesting and instructive stage in the 

 passage to the ordinary form of Phahsnal antenna. Each joint of the 

 antenna has two rings of scales, but the basal ring is wanting vent- 

 rally, and its place is occupied by hairs, which are here collected 

 together. Flntella, Swammcrdama, and Gelechia are at much the 

 same stage of specialisation. 



The Hesperids, in preserving a very complete coating of scales, are 

 only illustrating their relationship to the earliest Lepidoptera, and not 

 as Dr. Jordan supposes, evidencing the highest development of scale- 

 covering. Dr. Jordan points out several points of relationship to 

 Hi'iiialidai', as in the arrangement of the bristles, but has been misled 

 by taking the Hepialid antenna as a primary, instead of a highly 

 evolved, form. In some respects it is also a degenerate antenna, the 

 irregularity of the segmentation into joints is evidence of this, though 

 Bodine {I.e., p. 28) regards it as evidence of evolution in progress. In 

 this respect, however, it is exactly similar to the antenna of the 

 Talaeporidae {Tsychidac), in the $ s of which it is a constant feature in 

 association with the gradual disappearance of the antennae with the 

 other appendages. The scalelessness may be, probably is, another feature 

 of the same degeneration, but it must be remembered that the 

 Hepialids, though Juyatae, and so far to be regarded in comparison 

 with any frenate form as more primitive, are, nevertheless, very high 

 above the Kriocranmh, and, though on a different stem, very far above 

 many Xeolcpidoptcra, and, were the whole of once probably existing 

 Jugate families still in evidence, it would probably not look at all im- 

 proper to draw a parallel of the Hepialids amongst the Juyatae, with 

 the Saturniids amongst I'ombycids, or the Danaids amongst Papiliunidae 

 as evidence that the highest families tended to have scaleless antennae. 

 Dr. Jordan mentions also an Arctiid, which would give an example at 



