INTRODUCTORY. 1 ■> 



Heliconid would, for instance, not appear to be possible., 

 Similarly, we find among the Vanessidte ("Red Admiral," 

 " Tortoisesbell," etc.) a very general dusky coloration of the 

 underwings, which often bears a very perfect resemblance to a 

 withered leaf, and, being thus probably advantageous to the 

 insect, may have been produced by the survival of the best suited 

 varieties; but among those Vanessidas where such aresemblanee 

 is not by any means perfect^e.y., the Painted Lady — there is 

 still the same confused mottling, which might, with but a little 

 change, be improved in the required direction. There is, so far 

 as I am aware, no case known of a Vanessa with leaf-green 

 underwings, such as are occasionally met with in butterflies 

 that frequent trees. This plan of colour would probably be 

 equally advantageous to a Vioiessa, for green leaves often 

 sprout out low down on the trunks of trees. There is here 

 apparently, if not an impossibility of modification, at least a 

 tendency for progression along the line of least resistance. 



Comparative Constancy of Colour in Genera and Families. 



The fact is that not only is coloration, with a few exceptions,, 

 constant for a given species, but it is also, with, of course, a 

 wider range of variation, constant to genera and to families. 

 There are exceptions, as there are among species ; but some of 

 these exceptions are often correlated with anatomical differences 

 which indicate that the supposed genus or family should be 

 divided. For instance, the " laughing jackass " of Australia 

 differs much from other kingfishers in colour ; and so do the 

 rest among themselves ; but this family of birds presents a 

 considerable variety of anatomical structure which argues a 

 wider separation between some of the species than has been 

 yet allowed by systematic Ornithologists. We find green to 

 be a very common colour among Parrots, Touracous, . and 



