33 



have known anything of Spuler's paper, and describes the '' Haft- 

 feld " as a new structure, under the name of " spiny area." So far, 

 however, as I am aware, Mr. Cooley first pointed out the cor- 

 relation of the spiny area with another on the metathoracic dorsum, 

 and showed that the use of these structures was to hold the wings 

 in position when they were closed. 



There is no doubt that the spicules of the " spiny area " or 

 " Haftfeld " are survivors of the spicules of the Neuroplera, as we 

 see them already differentiated to some extent in the Micropteryges. 

 They exist also in certain Phryganidcc. Spuler and Cooley agree 

 in the distribution of the Haftfeld amongst the Lepidopterous 

 families ; they exist, in fact, everywhere, except in a few families and 

 exceptional species in others that usually have them. They exist 

 in all the families that close the wings over the back when at rest, 

 not in those that do not. 



Those that want them are chiefly the SphingidiC, Salurnidip, 

 Bonibycldce, Geoinet/ idee, Ceratocampidic, Endromidic, Bnphidce, 

 F/atypterygiidce, and Pterophoridce. Many Sphinges fold their wings 

 over the back, and ought by rule to have a haftfeld. Probably it is 

 absent owing to an ancestry related to Satiirnidic and represented 

 in Sphingids by Smeri/ithidie, which clearly have no use for the 

 structure. They are also wanting in the butterflies. 



Cooley says they exist in the Boinbycidcc, but I imagine he 

 applies this term to the LasiocampidcB, which have them. 



In Meyrick's Notodontiiia all are without " haftfeld " except 

 Notodontidcc and Polyplocidcc ; this adds another item to my argu- 

 ment against JVotodonta being associated with this group. As to the 

 Polyplocidie, they seem out of place here, yet they have many 

 characters in common with the rest of the group, especially the form 

 of the egg, as well as the neuration. This haftfeld, however, must 

 add another straw to the indications that they are somehow related 

 to the Noctuids (Caradrinina). In making use of this as of any 

 other character, we may note that, since it exists in all the lower 

 groups, families that have no haftfeld must be derived from 

 forms that possessed it, but that no family without it can be 

 ancestral to one that possesses it. 



I may refer here to a modification of the scales that often occurs 

 in a region close to the " Haftfeld," sometimes over a considerable 

 diffused area, sometimes on very defined and circumscribed spots, 

 in the latter case often suggesting patches of Androconia; but as they 

 occur in both sexes this is not so. Achcrontia atnpos, from its large 

 size, exhibits these patches both on upper and under wings very 

 plainly. Their use is perhaps not difficult to guess. They consist 

 of very strong, smooth scales, with often inflexed tips and packed 

 closely together, and are well calculated to resist the wear and 

 tear of the wings against each other, and especially to obviate 

 friction between them. 



They exist also in many butterflies, as in sundry Vanessids. 



3 



