KELLI IGG 



] THE COLORS OF BUTTERFLIES AND MOTHS 209 



on each scale. There would thus be 12,500 of these striae to the 

 inch. On the transparent scales from Morpho the stria were .0015 

 mm. to .002 mm. apart; on opaque (pigment bearing) scales from 

 the same specimen the stria- were from .0007 to .00072 mm. apart, 

 or at the rate of about 35,000 to the inch. 



If we examine a long series of scales brushed off from different 

 parts of a wing of moth or butterfly, we can always note a series of 

 gradating forms running from slender hair-like form to typical short, 

 broad, flat scale. The significance of this, when we come to inquire 

 about the origin of scales, is plain. Scales are unusual structures 

 among insects; besides the moths and the butterflies, only few beetles, 

 the mosquitoes, and fish-moths, and a few other scattering insects 

 have them. But all insects have hairs. Hairs are structures common 

 throughout the class. And it is certain that scales are derived or 

 developed from hairs. They are a specialized, a highly modified 

 sort of hair. ( )n the lower, the more generalized moths, the hair- 

 like scales are the more abundant. The wings show a thick inter- 

 mixing of loose, fluffy hair-scales or scale-hairs with more typical 

 scales irregularly arranged. In the higher Lepidoptera, the special- 

 ized sort of hairs, namely the scales, compose almost exclusively the 

 wing-covering, and these scales are arranged in the specialized uniform 

 shingling manner previously described. But even on the wings of a 

 butterfly all the gradations from hair to scale can be found by going 

 from base out to discal area of the wing. These gradation series 

 vary in character in different families. In some the hair becomes a 

 scale by shortening and broadening, keeping its free tip entire; in 

 others the hair splits distally and then each branch splits again, and 

 so on, while the base is continually shortening and broadening so 

 that the scale form finally reached is a fingered or deeply-toothed one. 

 But in all the series the final result is that from a long, slender, sub- 

 cylindrical hair is evolved a short, broad, flattened, little scale. A 

 study of the actual development of an individual scale on the form- 

 ing wing of a butterfly during the pupil or chrysalid stage confirms 

 the hypothesis of the evolution of the scales. In the growing devel- 

 oping wing the scales begin as hairs, arising by the extension of cer- 

 tain hypodermal cells in the wing-membrane which gradually change 

 in the few or many days of pupal development into typical scales. 



We have studied now with some care the general character of the 

 scale covering of moths and butterflies, and the actual structural 

 make-up and the origin of the individual scales. And we have learned 



