URBICOLA COMMA. 167 



perfectly with European comma, and, on the other hand, the 

 existing differences appear to be of too little importance, and, above 

 all, not sufficiently constant to make it possible on the strength of 

 these to declare the American forms specifically different from comma. 

 The latter is, in a high degree, under the influence of various external 

 life-conditions, and is, as both the American and the Asiatic forms prove, 

 a species varying in different directions. Whether any one of these 

 local forms has already sufficiently established itself to be able to rank 

 as a distinct species, others, who are equipped with more abundant 

 material, will be able to decide with more certainty than myself." 



CommentiDg on these conclusions, Edwards remarks (op. cit., pp. 

 147 et seq.) that the differences in the underside spotting of manitoba, 

 Colorado and nevada, as regards shape and arrangement, are constant. 

 As to the spotted fringes said to be constant in European comma, they are 

 sometimes present, but not in all the forms under review, in Colorado 1 

 $ and 1 $ , in manitoba 2 ? s were found with them, whilst no nevada, 

 Columbia, or idaho had them, but Edwards' material (like Speyer's) was 

 very insufficient (scarcely half-a-dozen specimens in some instances). 

 He considers the occasional presence of the fringe-spots in the American 

 forms of this group, may be sufficiently accounted for on the theory 

 that the European, Asiatic and American forms are of co-ordinate 

 value, and inherited these spots from their common ancestor ; in the 

 American they have disappeared, but occasionally the character is 

 recovered by reversion. He says that Scudder's types are distinct 

 enough in the case of manitoba, Colorado and nevada, and, he thinks, 

 should be considered as so many species. Commenting on the 

 fact that Speyer notices that in all these forms the two sexes 

 seem to differ much more strongly than in comma, the differences 

 noted, Edwards says, appear to be constant. Edwards enters into a 

 criticism of Speyer's remarks on sylvanoides, Scudd. ( = Columbia, Scudd.), 

 and compares the form with a $ and $ comma, falling back at last 

 almost entirely on differences of underside colour, and differences in 

 colour between the sexes (two markedly variable features in comma), as 

 entitling it to specific rank. Without going further into the question 

 of the actual value of the American forms, all of which, by the way, 

 Dyar drops in his List of North America?! Lep. (1902), as varieties of 

 comma, L., a conclusion with which our limited knowledge brings us into 

 complete agreement, and, without attempting the difficult task of 

 comparing them with the European forms, for which study we are 

 altogether lacking in material, we give the following descriptions of 

 the described American forms : — 



a. var. manitoba, Scudd., " Mem. Bost. Soc. Nat. Hist.," ii., p. 351, pi. x., 

 figs. 8-11, pi. xi., figs. 7-8 (1874) ; Speyer, " Can. Ent.," xv., p. 143 (1883) ; Edw., 

 " Can. Ent.," xv., p. 147 (1883) ; Scudd., "Butts. New. Engl.," pp. 1646 et seq. 

 (1889) ; Lyman, " Can. Ent.," xxiv., pp. 57-59 (1892) ; Dyar, "List North Amer. 

 Lep.," p. 50 (1902). — This species is the most nearly allied to P. comma of all Ameri- 

 can species ; the size of the two is the same, and the upper surface of the ? varies in 

 each species to the same extent. There are no constant features of distinction in 

 the upper surface of the g , although, in P. manitoba, the hindwings are usually 

 devoid of the appearance of the spots on the under surface, which generally are 

 faintly but exactly marked upon the upper surface in the European species. The 

 under surfaces of the two species also resemble each other closely, and it is a little 

 difficult to define in words the distinctions which are apparent. The ground 

 colour inclines more to deep green in P. comma, and the mesial bent band of the 

 hindwings is rather more uniform ; in the g it seldom departs from a certain 



