POSITION AND DISTKIBUTION OF SAGARTIA LUCI^. 737 



Yaiicouver Island forms with those from New England and 

 Plymouth, so that we now know of three areas inhabited by the 

 species, separated from one another by wide intervals, two of 

 them by the North Atlantic Ocean and the other from these by 

 the whole width of the North American Continent. Further- 

 more, there is the certainty that the species has been " in 

 residence " on the Canadian Pacific coast for over sixty years, 

 and if the suggestion be correct that it has been secondarily 

 inti"oduced into the other two areas, the Pacific may have been 

 its original home. But Mr. Cocks's discovery of his A. chryso- 

 sj^lenium on the Cornish coast dates back seventy-five years, and 

 this lessens the probability of the introduction hypothesis. There 

 is another possibility, however, namely, that originally the species 

 had a circumpolar distribution, like Metridium senilis, Uriicina 

 felina^ and a number of other Ccelenterate forms, and that its 

 present areas of distribution are but separated remains of a much 

 larger area. If further observations should reveal its presence 

 on the Asiatic side of the Pacific the pi-obability of this 

 suggestion would be greatly increased* ; in the meantime it is 

 merely offered as an alternative to the introduction hypothesis. 



But no matter what the original home of the species may have 

 been, the evidence is clear that in each of its known localities 

 it has in recent years more or less markedly extended its dis- 

 tribution. The cause of this is also obscure. It scores largely 

 in favour of the introduction hypothesis, but is not necessarily a 

 proof of it. It may be a phenomenon in some respects and on 

 a lesser seal 3 comparable to the migration of the Colorado 

 Potato Beetle many years ago, but as to the influences that 

 determine it in three widely separated areas I have no suggestion 

 to make. 



To return now to a consideration of the systematic affinities of 

 the species. The possession of acontia marks it as a member of 

 the family Sagartiadee, a family in which several subdivisions are 

 now recognized. From the subfamily Phellinse it is excluded 

 by the thinness of the column-wall and by the fact that acontia 

 may be extruded through the wall, and it finds no place among 

 the Sagartiinse, since it has not more than six pairs of perfect 

 mesenteries, these being also sterile. This leaves only the 

 Metridiinse and the Aiptasiinpe for its reception, and it is very 

 doubtful if the separation of these two groups can be maintained. 

 For the Aiptasiinfe difi'er from the Metridiinse only negatively, 

 in the lack of a mesogloeal sphincter, and since it may be 

 supposed that they are descendants of forms possessing that 

 structure, its absence in them is due tu a process of reduction 

 and is of less importance than the arrangement of the mesenteries. 

 Fvirthermore, I have found in Aiptasia [Heteractis) lucicla faint 

 indications of a mesogloeal sphincter, and, I may add, since this 



* Since possibilities are being discussed I may suggest that perhaps the form 

 collected by Stimpson in Hong Kong Harbour and described by Verrill (1869) as 

 Sogartia Uneata maj' be the Asiatic representative of S. Incice. 



