3 io THE WILDERNESS OF THE UPPER YUKON 



habiting regions where the mountain ranges are broken, 

 having the character of complex groups separated by 

 wide valleys, tend to vary. 



No positive conclusions can be drawn from the known 

 facts about the colors of sheep until science has revealed 

 more knowledge of the causes of color variation. I know 

 of no case among the large mammals of this continent 

 analogous to the colors of these sheep. 



According to the theory* of "Protective" or "Conceal- 

 ing" coloration a theory developed and elaborated with 

 great detail, and universally applied, by Abbott Thayer in 

 his book on Concealing Coloration in the Animal King- 

 dom the colors of sheep must result from Natural Selec- 

 tion acting through long periods of time, continually exter- 

 minating sheep possessing a color easily observed by their 

 enemies, and preserving the sheep having colors which 

 conceal them from their enemies. Mr. Thayer insists 

 that this has resulted specially in animal coloration that is 

 most obliterative, for enemy or prey, at the moment when 

 the enemy is about to seize its victim the last rush, the 

 last spring, when sight alone is necessary for success. 



* I do not use the word theory as applicable to the facts of Mr. Thayer's 

 analysis of animal colors in their relations to certain backgrounds. But, as a 

 cardinal point, he insists that animal coloration is principally obliterative against 

 the background when "animals are on the verge of catching or being caught." 

 Since the exact conditions of attack, capture or escape between enemy and prey 

 are in many cases unknown, many of the matching backgrounds, which Mr. 

 Thayer describes as those in which the attack takes place, must be assumed. It 

 is not, however, proved that such are the backgrounds which oppose the attack, 

 and, therefore, I use the word "theory" to include the facts together with the 

 unproved assumptions. Writing subsequently on the subject, Mr. Thayer ob- 

 serves that his book might have been called "An expert's presentation of ex- 

 amples of consummate resemblances between animals' costumes and certain of 

 their backgrounds." To such a definition the word "theory" would not apply. 

 C. SHELDON. 



