FLIGHT OF INSECTS. 317 



perhaps generally be owing to the presence of some 

 colouring material : but the more delicate hues are 

 probably the result of the optical effect of the striae 

 on the surface ; and in some cases they result from 

 the thinness of the transparent plate of which they 

 consist ; for I have observed in several detached 

 scales, that the colours they exhibit by transmitted 

 light are the complementary colours to those which 

 they display when seen by reflected light. 



The forms of these scales are exceedingly diver- 

 sified, not only in different species, but also in 

 different parts of the wings and body of the same 

 insect ; for the surface of the body, generally, as 

 well as the limbs, and even the antennae in some 

 species are more or less covered with these scales.* 

 Fig. 164 exhibits some of the more usual shapes 

 as they appear when viewed with high magnifying 

 powers. 



Each scale is inserted by a short pedicle into 

 a small tubular sheath, which is itself implanted 

 into the membrane of the wing, or root; each 

 overlaps the adjoining scales ; and the whole arc 

 disposed in rows, with more or less regularity ; one 

 row covering the next, like tiles on the roof of a 

 house. t This imbricated arrangement, together 

 with the marks that are left on the membrane of 



* In the posthumous work of Lyonct, which has lately appeared, 

 nearly the whole of six quarto plates are crowded with the delinea- 

 tions of the different forms of the scales found in the Bombyx 

 Cossus. 



t The scales on the abdominal rin^s of the Lepisma are of two 

 kinds; one set being arranged in rows, as usual, and the others, 

 which are shaped like a heart, being inserted by a pedicle projecting 

 from the middle of the base between and over the former, so as to 

 fasten each firmly in its place. 



