DYSSTKOJIA CONCINNATA, STEIMI. 179 



granite boulders, />. tntnrata and /'. citiatu prefer the branches and 

 trunks of trees. 



Concinnata is very variable in colouring, but has a different facies 

 from either tnoicata or citntto, a difference better brought out by long 

 series than by isolated specimens. 



It agrees with trumata in shape. The wings lack the narrowness 

 and slight hardness of surface seen in citrata. 



The shape of the outermost line of the broad central band is a 

 fairly reliable feature when large numbers are examined, and in this 

 too it agrees with tnnicdta rather than with citrata. In the latter this 

 tends to make a smooth sweep from the costa to a projection outwards, 

 which is usually double-toothed and sometimes deeply indented. 



Traniata and rourinnata generally show one or even two sharp 

 projections near the costa, and the large elongation outwards shows 

 three teeth, which are often ill-defined. 



The dappled appearance of the central area and the well-marked 

 pale submarginal spots on the hindwings are usually co isidered the 

 most characteristic features of concinnata. In a considerable number 

 of individuals which I have examined they were always present. 

 Both features are often found in Shetland citrata (var. /iijt/i'niixsata, 

 Milliere), the pale submarginal row of spots in the hindwings are some- 

 times well developed in tnoicata, and exceptionally the central area 

 in this species may be as dappled as that of c(>nci)inata. The central 

 area in some specimens of concinnata is greatly suffused with black or 

 chesnut scales, but there is never the uniformity of colouring which 

 is the rule in similarly coloured tnoicata. 



The line on the underside of the hindwing is shaped like that of 

 triincata, with a deep indentation between veins 5 and 6, but the line 

 is very thick. In truncata this line is usually tj^uite thin, especially in 

 the lighter Scottish forms, and in the darker ones there is a general 

 suffusion of the whole wing with dark scales, which makes it even less 

 conspicuous. In the plate the indentation in tnonata is not sufli- 

 ciently clear. 



Thickening of the line is rarely met with in truncata, but is con- 

 stant in concinnata, and the thick indented line appears to me the 

 single character which is most reliable in differentiating it from its 

 allies. 



In the most melanic concinnata there is no general darkening of 

 the underside of the hindwing, though thert; may be a good deal of 

 black scaling in the basal area. 



The shape of the line in citrata is usually quite distinct, being 

 sharply angled, and without the indentation. The difference is well 

 shown in the plate. (Pit. vii.) 



The central spot in cincinnata is, on the average, larger than in 

 the other species, and the wing surface is more glistening. 



In such protean species where variation runs on such closely 

 parallel lines, too much reliance must not be placed on a single 

 point ; all must be considered to arrive at a correct determination. 



I have failed to find any constant structural characters in antenna? 

 or elsewhere except in the genitalia, by which I can separate one 

 species from the others. 



In outline, markings and colouring, the Arrau species comes nearer 

 to truncata than to citrata. 



