4 MB. E. T. WATSO^' 0!s THE HESPERIID^. [Jau. 17, 



already established ; so that it will be fouud that there are many 

 species noted below for which new genera have not been erected, 

 but which have been placed in that geuus to which they seem most 

 closely allied. In addition to the collection of the British Museum, 

 free access has been afforded me to the valuable collection of 

 Messrs. Godman and Salvin, to whom my best thanks are due for 

 their courtesy and kind assistance. 



The system of numbering the veins has been adopted in the 

 descriptions for brevity and clearness, and, as this system is not in 

 universal use, the veins in the first figure of neuration have been 

 numbered to exemplify the method. 



Before 1874 no serious attempt had been made to arrange tlie 

 genera of the Hesperiidae in natural groups, but since that time 

 several arrangements, though in most cases only relating to a limited 

 fauna, have been proposed. 



The only suggested arrangement which seems to be perfectly 

 natural is that proposed by Scudder in the Bulletin of the Buffalo 

 Society of Natural Science ', and afterwards worked out more com- 

 pletely in his 'Butterflies of New England.' Though it is only for 

 the Hesperiidae of New England that this arrangement is fully 

 worked out, yet, on examination, it has been found, with certain 

 modifications, applicable to the Hesperiid genera of the world, and 

 has accordingly been adopted in this paper. In this arrangement 

 Scudder divides the genera of New England Hesperiidae into two 

 groups, which he names respectively Ilesperidi and Pamphilidi. 

 These two divisions are based to a very large extent on the secondary 

 sexual characters of the male imago, the egg, larva, and pupa supply- 

 ing subsidiary characters ; these latter, however, are, as pointed out, 

 of a slight and ill-defined character, and would be inapplicable 

 generally, since, in the great majority of the genera, little or nothing 

 is known of the earlier stages. The male characters are, however, 

 sufficient in themselves to enable the majority of the genera to be 

 readily assigned to the respective groups, and where no secondary 

 male characters exist the neuration or habits supply the necessary 

 indication. 



Mabille has further amplified this arrangement of Scudder's in a 

 paper* on the Hesperiidae in the Brussels Museum, wherein he 

 further subdivides the main divisions and assigns additional genera 

 to their respective groups. These further subdivisions have unfor- 

 tunately been only very partially characterized, owing, as M. Mabille 

 himself states, to his investigations not being completed. 



In the allocation of many genera I have found it necessary to 

 entirely diff'er from M. Mabille's conclusions ; for instance, nearly 

 all those genera which Mabille includes under his subdivisions 

 "Ismenini" and " Tciffiadini," and assigns to the Astyci=i Pamphilidi 

 (Scudder), should, in my opinion, be transferred to the Hesperidi 

 (Scudder), with which their habits and neuration better agree, and 

 Mabille's "tribe" Pyrrliojiygini be erected into a group of equal 



1 Bull. Buff. Soc. Nat. Sc. vol. i. pp. 19>-196 (1874). 

 ^ Anu. Soc. Ei]t. liclgo, vol. xxi. p. 12 ct scq. (1878). 



