500 Mr. A. Quail on 



On the surface of the segments there are minute " hairs," 

 which may be sparsely scattered as on the scaj^e, or closely 

 cover the surface of the segment as on the clavola : those 

 of the scape are perhaps more bristle-like, and remind one 

 of the minute haiis on the Hepialid wing membrane. One 

 cannot help thinking that these are the original antennal 

 " hairs," antecedent to all other hairs and scales on the 

 Lepidopterous antenna; those which we find on the clavola 

 segments everywhere, scaled or not, are, I take it, these 

 "primary" hairs, perhaps somewhat modified; they serve 

 probably tlie function of protection. 



Amongst the " primary " hairs others arise which are 

 longer and tapering, not rigid, usually fine but sometimes 

 stout with socket rims at base ; these are probably " sense " 

 hairs, and appear to be in some way associated with reduced 

 scaling. Tlie sensory hairs appear to be confined to the 

 clavola segments. One has, however, some ditSdence in 

 deciding whether some forms of "sense" hairs are not 

 bristles. There are short, pointed, socket-rimmed bristles 

 at the base of both scape and pedicel, and scattered among 

 the scales are longer bristles ; there are always bristles on 

 the clavola segments. The transition from pale-coloured, 

 fine scape bristles to red-tinted, robust, rigid clavolar 

 bristles, I have demonstrated on the antenna of Porina 

 fuscomaculata. The bristles on the antenna of one species 

 may resemble the " sense " hairs on the antenna of another 

 species, but all the " sense " hairs on any given antenna 

 differ from the bristles of that antenna. 



There appear to be good reasons for believing the 

 piimitive form of Lepidopterous antenna to have been 

 one closely covered with scales — such forms one finds 

 among the lower groups of Lepidoptera — the scales after- 

 wards replaced by "sense" hairs, but amongst the scales 

 there were semi-erect bristles, and, it may be remarked, 

 similar bristles, whatever their functions, occur on different 

 parts of the " scale-covered " wings. The bristle is probably 

 older therefore than "sense " hairs. 



The scape and pedicel retain comparative rotundity with 

 remarkable constancy throughout the Hcpialidx^ known to 

 me, with the exception of Oncoptcra (intricata), in which 

 the scape is remarkably elongate. The scape and pedicel 

 are never pectinate in the He/pialidx: 



The segments of the clavola vary greatly in number, 

 form, and dimensions in Heinalidm. The proximal joint 



