126 COIvORATlON IN LEJPTINOTARSA. 



more apt to show these variations than are bands. Each genus or family follows these 

 general laws of variations, but each displays modifications peculiar to itself; these 

 modifications being sometimes of so marked a character as to constitute an exception 

 to the general rule. Such wide departures from the rule are exhibited by broken rows 

 of spots on the fore wing of Hesperidse. Also by double spots in rows of otherwise 

 single spots on the hind wing of Hesperidae. In both of these cases the variation is 

 commoner in the middle than at the ends of the row. Like kinds of markings are more 

 apt to fuse than are unlike. For example : Rows of spots often fuse with other rows 

 of spots, or bands with other bands ; but fusions between rows of spots and bands are 

 rare. Fusions are commonly mere contacts, which are terminal for at least one of the 

 fused rows. Each pair of fused markings is more apt to come in contact at their ends 

 than at their middle. A definite relation obtains between the number of markings on 

 the fore and hind wings. Also the markings on either wing tend to appear in definite 

 order of frequency which is peculiar for each genus or family, but which, when plotted, 

 gives a fairly regular frequency polygon, with one or two maximum points. The above 

 statement applies to the number of spots in a row, and the length of bands, as well as 

 to the number of rows of markings on the wings. These relations are too definite and 

 orderly to be due to accident, and appear to be the expression of a natural law. 



Among the other orders of insects I have shown that the color patterns on 

 the body of Coleoptera, Orthoptera, Hemiptera, and Hymenoptera are com- 

 posed of segmentally arranged spots, originating upon important sclerites or 

 over muscle attachments. I consider the color pattern to be a segmental one, 

 showing in successive segments a repetition of the same spots and color areas. 

 Enteman has come to a similar conclusion concerning coloration in Polistes, 



PHYI^OGENY OJ? COI.ORATION. 



The first thorough and consistent attempt to study and trace the phylogeny 

 of insect coloration was made by Darwin, who developed around this subject 

 to a considerable extent his theories of sexual selection, protective coloration, 

 warning color, etc. Natural selection and segregation were for this author 

 sufficient to account for the phenomena of insect coloration. The researches 

 of Darwin, Wallace, Bates, Weismann, Poulton, Hasse, and others upon the 

 coloration of insects, especially that in Lepidoptera, have developed a school 

 which stoutly maintains that the brilliant colors of many insects, especially 

 those in males, have been developed by a process of sexual selection. They 

 likewise hold that mimicry, protective resemblance, etc., are due to selective 

 influences alone, and Weismann has invented some ingenious theories to 

 account for the rise of these phenomena by means of selection, natural, sexual, 

 or otherwise. 



Strongly opposed to the views of Darwin, Wallace, Weismann, and their 

 followers is the theory promulgated by Eimer, who, in 1889, conducted an 

 extensive research upon coloration in the Papilionidse, and in 1898 published a 

 still more extensive work along the same line. Eimer and his followers hold 

 that the color patterns of insects are only slightly influenced by natural selec- 



