BARNES AND McDUNNOUGH: CATOCALA 35 



The type of the species came from Texas; we have a few specimens before us from this state, all of which show the 

 unbroken band and would therefore fall under the term westcotti which was described from material from Illinois and 

 Wisconsin. We have not examined these types and have no material from this region so cannot say if Beutenmuller's 

 conception is correct or not; we presume that it is, as Grote himself regarded westcotti as a mere form of amestris. Apart 

 from the type localities, we have a single specimen from Nebraska and Snow has recorded the species from Kansas; 

 Beutenmuller secured the larva? on Robinia in the Black Mountains, North Carolina, one of these specimens serving for 

 the figure given on plate XII, figure 7. Apparently the species has a wide range throughout the Eastern States, although 

 on account of its rarity it has seldom been recorded. 



Group XIV 



{Andrew sia Grote) 



Early stages unknown. Male claspers slightly asymmetrical; apex of left clasper projecting slightly beyond the 

 thinly chitinized ventral area. 



Messalina, the only species in the group, has usually been placed close to arnica on account of the lack of a median 

 dark band on the secondaries; a study of the male claspers, however, shows that its affinities are rather with the illecta 

 group than with arnica; the early stages will probably shed light on the correct position. Hampson separates the genus 

 Andrewsia on the dorsal tufting of the abdomen, which he states is lacking in messalina; we can, however, see no appreci- 

 able difference between this species and some of the smaller species of the illecta group. In fact, if a well-marked speci- 

 men of messalina and a pale specimen of abbrematella be compared, it can readily be seen how a further reduction in the 

 maculation of the primaries of the latter species and an elimination of the median band of the secondaries would produce 

 messalina; we might even venture to predict that the larva of messalina will be without lateral filaments. 



Catocala messalina Guenee 



Plate X, fig. 20; PI. XXII, figs. 34 and 35 (claspers). 



Catocala messalina Guenee, 1852, Hist. Nat. Spec. Gen. Lep., VII, p. 107. 

 Catocala belfragiana Harvey, 1875, Bull. Buffalo So c. Nat. Sci., II, p. 281. 

 Catocala jocasta Strecker, 1875, Lep. Rhop. Het., p. 107. 



The illustration is ample to identify this striking and aberrant form. It is decidedly rare, apparently being most 

 common in Texas; it occurs, however, in Kansas and a single specimen has been reported (Ent. News, X, p. 283) from 

 Montgomery County, Virginia, so that its range will probably be similar to that of the preceding group of species. 



Group XV 

 Early stages unknown. Male claspers symmetrical, rounded apically. 



The two species included here, gracilis and andromedee (tristis), are apparently related, to judge by a comparison of 

 the claspers, but a knowledge of the early stages will be necessary to definitely place the species. 



Catocala gracilis Edwards 

 Plate IX, figs. 7-9; PI. XXII, fig. 7 (clasper). 



Catocala gracilis Edwards, 1864, Proc. Ent. Soc. Phil., II, p. 511. 



Catocala similis Grote (nee Edwards), 1872, Trans. Amer. Ent. Soc, IV, p. 17. 



Catocala gracilis var. sordida Grote, 1877, Can. Ent., IX, p. 170. 



Catocala proeclara Holland (nee Grote and Robinson), 1903, Moth Book, p. 269, PI. xxxv, fig. 7. 



This small species shows considerable variability in the coloration of the primaries. The typical form (PI. IX, fig. 7) 

 has a broad brownish streak along the inner margin; pale forms occur without this streak and again other specimens in 

 which the whole wing is suffused with dark shades, this latter being the form sordida Grote. A study of the figures will 



