butterflies in habits. They are far more fixed and still, in their daylight 

 perching, and they sit -jiatly applied to their perches, instead of standing out 

 from them perpendicularly. Thus they are fit subjects not only for higher 

 particularization and greater minuteness of pattern, but for direct, near, flat- 

 surface-picturing costumes, from which the element of pei'Speciive--pictunng 

 is largely excluded. Naturally, this rule has exceptions. There are ground- 

 perching moths — grass moths — for instance, covered with representations like 

 the Rocky Mountain Ptarmigan's (Figs. 40 and 41 of Chapter VII), of dry 

 grass blades crisscrossing over a background of shadow. Such a moth is 

 shown in Fig. 134, A. Perching among slender grasses some inches above 

 the earth, a moth like this has often a background full of diverse distances, 

 of little 'vistas' — ^and it is patterned accordingly. This, indeed, is not an 

 uncommon tj^e of coloration among moths, and there are many variations 

 on it. Kindred pattern-schemes occur also among woodland moths — both 

 terrestrial ones, which perch on fallen dead leaves, and those that are wont 

 to sit in trees, on green or withered foUage still unf alien. (See Fig. 135.) 

 But the vast majority of moths habitually sit pressed close against a flat or 

 flattish surface, which wholly hides their undersides, and prevents their hav- 

 ing any more distant background, except in extreme side view. The aspect 

 of this flat surface is pictured on the shielding upper sides of the moth's 

 fore-wings, and on its head and thorax, — or, in the case of species which sit fully 

 outspread, on both fore and hind wings, and the entire upper surface of the 

 body, — often with an almost microscopic particularity of detail. Only the 

 slightest reduction from the real size of the imitated details is needed in this 

 picture-pattern, for the space between picture and model is almost nil. In- 

 deed, though we may pretty safely assume, in view of Nature's prodigal per- 

 fectionism, that the moth's pattern is thus as it were 'focused' for the very 

 closest average unison with its background, the difference and its adjustment 

 are often too slight for man's eyes to recognize with certainty. But there 

 can be no doubt that the coloration of these flatly-applied moths is on the 

 whole ohliterative rather than mimetic; that, in other words, such a moth 



235 



