ANEMONES 395 



Genus Metridium Oken. 



" Metridiinae with a smooth body wall and numerous tentacles 

 which extend over nearly the whole of the oral disk, which is greatly 

 expanded and frilled ; the upper border of the column is swollen to 

 form a thickened ring, above which is the delicate capitulum." 

 (Haddon, 1S9S.) 



METRIDIUM DIANTHUS Ellis. 



Localities. — Kodiak, Yakutat and Sitka, in great abundance. Also 

 coasts of Europe, east coast of North America from the Arctic to Cape 

 Hatteras and west coast of same from Alaska to San Francisco. 



McMurrich described M. dianthus in great detail in 1901, and 

 gave cogent reasons for uniting M. marginatum and JSf. Jii)ibriatu7n 

 with Ellis' species. I shall therefore limit myself to an examination 

 of the variations in Metridium. 



VARIATION IN METRIDIUM. 



As is well known, M. dianthus possesses three conspicuous color 

 types : brown, salmon or orange, and white. According to a sugges- 

 tion by McMurrich, the brown may be the fundamental type, from which 

 the others are derived by direct transformation. In support of this 

 view he states, firstly, that the smallest individuals of his material are 

 brown ; secondly, that white and salmon polyps may exhibit blotches 

 of brown, as though in process of changing their color, although white 

 polyps never appear to be blotched with salmon, and vice versa. 



Embryological evidence which would be conclusive on this point, 

 is as yet wholly wanting. My own observations on great numbers of 

 Metridium in Oakland Harbor, Calif., indicate that this change of 

 color is not a widespread phenomenon. I have found white and 

 salmon polyps of exceedingly small size quite commonly. Whether 

 these have come from the Q^<g or from fragments of the pedal disk is a 

 question which applies with equal force to the brown type. That all 

 the very small polyps one finds do not arise from this fragmentation 

 seems to be a necessary assumption concerning an animal which re- 

 produces sexually as well. Recent observations made for me by my 

 friend John M. Willard, of Oakland, Calif., show no difference in the 

 relative proportions of the color types among the very smallest and 

 among the average individuals. This fact is significant. The pro- 

 portions were : salmon, 50 percent, white, 45 percent ; brown, 5 

 percent. Special pains were taken to avoid errors referable to basal 



