found most frequently on young female fish. 

 Few male spawners fell into this category, in- 

 dicating that they were more highly colored than 

 females. Females were predominantly silver 

 when young and became yellowish, olive, or red 

 with age. Males were mainly yellow or red. 

 The olive-brown color phase was predominantly 

 found on male fish. 



Fish from streams in close proximity 

 were more similar in frequency of occurrence 

 of certain color phases than fish from streams 

 far apart (table 4). Fish with silver -white col- 

 oration were most common in Pelican, Cub, and 

 Clear Creek runs, indicating a large proportion 

 of young spawners. The yellow phase was most 

 abundant in Chipmunk and Grouse Creeks, fol- 

 lowed by fish with red coloration. Arnica Creek 

 had about equal amounts of silver-white, yellow, 

 and red-pink fish. This wide range and varia- 

 tion in color among spawning populations of 

 cutthroat trout in Yellowstone Lake makes it 

 difficult to describe the coloration of a repre- 

 sentative specimen from this subspecies . In 

 fact, coloration of most cutthroat subspecies 

 can be duplicated by individual fish from Yellow- 

 stone Lake, which suggests that the use of color 

 as a major factor for separating cutthroat sub- 

 species needs reevaluation. 



Discussion and conclusions 



As mentioned previously, coloration and 

 spotting differences have often been used a? 

 major distinguishing features between subspecies 

 of cutthroat trout, and the presence or absence 

 of hyoid teeth has frequently been the deciding 

 factor in calling a specimen a cutthroat, cut- 

 throat-rainbow hybrid, or rainbow trout. Hence, 



natural variations in these characters within 

 the same subspecies are of considerable im- 

 portance . 



On the basis of the amount of spotting 

 below the lateral line, the four samples in this 

 study ranged from an overlap of 1.20 (p»0.006) 

 to 81.60 (p=0.408) percent. On hyoid teeth counts, 

 the samples ranged from 2.50 percent overlap 

 (p=0.013) to 78.00 percent (p=0.390). The sig- 

 nificance of these differences is apparent when 

 the usually accepted levels of £ values for separat- 

 ing species and subspecies are noted. Ginsburg, 

 as quoted by Royce (1957) suggested that a p = 0. 1 

 would commonly be found between species, 0.3 

 between subspecies ,0.4 between races , and up 

 to 0.5 between varieties. Mayr et al. (1953 sug- 

 gest that p = 0. 1 is the usual level of subspecific 

 difference. Hubbs and Hubbs (1953) feel that a 

 better level of subspecific difference is p = 0.25. 



Considering spotting below the lateral 

 line by. GinsDurg s definition. Sedge Creek fish 

 would be a separate species from the Yellowstone 

 Lake and Greston hatchery stocks; Bear Creek 

 fish would be a different species from the Greston 

 hatchery stock. All except Yellowstone Lake and 

 Greston hatchery fish would be classified as dif- 

 ferent subspecies . For hyoid teeth , Sedge Creek 



fish would be classified 

 as a different subspecies 

 from Yellowstone Lake 

 and Greston; Bear Greek 

 would be different from 

 Greston; and Greston 

 fish would be a differ- 

 ent subspecies from 

 Yellowstone Lake. Even 

 on the level of sub- 

 specific difference 

 (p =0. 1) selected by 

 Mayr et al. , Sedge 

 Greek fish would be a 

 distinct subspecies from 

 Yellowstone Lake and Greston fish, and Bear Greek 

 fish would approach subspecific differences from 

 Yellowstone Lake and Greston populations. Ac- 

 cording to the levels of difference set by Ginsburg 

 and by Hubbs and Hubbs, Greston hatchery fish 

 were able to alter sufficiently in four generations 

 to be classified as a different subspecies from 

 their ancestral stock in Yellowstone Lake. 



