COLOR PATTERN OK ELYTRA. 83 



In the lineata group, e. g., the evolution of species has clearly followed 

 narrow, fixed lines in the color pattern of the elytra, and that this has resulted 

 in the production of species in certain directions only. It is in the directions 

 in which the species have evolved that we find the individual variations fluctu- 

 ating on cither side of the mode toward other closely related species until, 

 in very many cases, they overlap. It is also to be noted that individual 

 variation does not extend in other directions than those mentioned. We can 

 explain this condition only on the basis that the variation of the species is con- 

 trolled and directed by the constitution of the organism independent of the 

 conditions of its existence. The study of variation can not give us the 

 reasons for these phenomena ; it can simply discover the facts concerning the 

 direction and extent of variations and the laws which they follow. A.s to whv 

 they follow these laws variation studies must remain forever silent. Embry- 

 ology and experimentation only can offer aid in the solution of these 

 difficulties. 



In the elytra the following laws hold for the variations exhibited by the 

 color pattern : 



(1) There exist two primary sets of color-pattern markings: (a) bands 

 across the wing and (b) stripes along the length of the wing. All color pat- 

 terns in the genus consist either of stripes or modifications thereof, bands or 

 modifications thereof, or combinations of the two. Unicolorous conditions, 

 either melanic, albinic, xanthic, or rufic, result from complete fusion of bands 

 and stripes, or complete reduction thereof. 



(2) We recognize six stripes related to the veins, and five bands. These in 

 all species occupy homologous positions. (See third chapter, on the develop- 

 ment of color patterns, for causes of these conditions.) The point of crossing 

 of a band and stripe is an area of pigment production more resistant to reduc- 

 tion than any other area. 



(3) The edges of the wing are more variable than the middle, and the anal 

 edge less variable than the costal in banded species and more variable in 

 striped species. The distal end of the wing is more variable than the proxi- 

 mal. These laws are in accordance with Bateson's (1894) law that the 

 extremes of a series are more variable than the middle. 



(4) Individual variations which are limited to definite directions are either 

 along stripes, by the reduction at the extremes, by the breaking up into spots, 

 or by fusion ; or it is along bands by reduction at the extremities, by breaking 

 up into spots, or by fusion. Variation in a given species may go along one or 

 the other or both of these lines at the same time, but for a given species the 

 direction of variation is always constant 



