the Lepidopterous Family Cossidae. 189 



The Lepidoptera have specialised in colour and wing- 

 pattern, not in structure. The great achievement of the 

 order in structure was the development from the maxillae 

 of the spiral proboscis, and this happened long since. 

 Among existing families structural evolution has had but 

 trivial results, consisting (I am writing, of course, of the 

 imagos) of little more than secondary sexual characters.* 

 On the other hand, there has been a strong tendency in 

 nearly all the families to progressive reduction in structural 

 complexity, to a progressive simplification by structural 

 loss, on parallel lines. Unless this is fully recognised 

 no progress will be attained in the true phytogeny of 

 the different groups. I will enumerate some of these 

 lines, and point out how remarkably they have been 

 followed even within the very ancient and primitive 

 Cossidae. 



(1) The proboscis and maxillary palpi have been lost in 

 the Cossidae, Psychidae (only the <$ can be brought into 

 comparison, the $ being degraded to an extreme degree), 

 Limacodidae, Lasiocampidae, Liparidae, and in other whole 

 families, as well as in many isolated genera. 



(2) The labial palpi have been lost in some genera of 

 Cossidae and quite independently in many genera of other 

 families. 



(3) The tibial spurs have been lost or much reduced in 

 most Cossidae, in the Zygaenidae, Psychidae $, and in 

 other instances. 



(1) The frenulum has become shortened and non- 

 functional in a few Cossidae, wholly lost in all the Lasio- 

 campidae, Endromidae, Uranianae, and in some genera of 

 the Drepanidae, and Bombycidae and the Geometrinae 

 subfamily of the Geometridae; in the last instance every 

 grade between full development and complete loss can 

 still be traced. 



(5) The median vein is always present in the Cossidae, 

 but in a few genera, the first step in its obsolescence, the 

 obliteration of the median cell has taken place. It is 

 completely lost or merely vestigial in most Lepidoptera. 



(6) The areole and chorda are lost in at least one genus 

 of Cossidae, in all genera of most other families, and in 

 those families that retain it, it has been lost in some, if not 



* I leave out of consideration the basal abdominal cavities, as 

 to which I have no precise knowledge. 



