HVDROIDA II J,-, 



appearance at the top. And at the same time, a marked change takes place in the branches, which 

 now turn 90° on their axis, so that their original broad plane becomes horizontal; further, a displace- 

 ment of the hydrotheca towards the upper side of the branch also sets in, and the typical unilateral 

 arrangement usual in the species now becomes apparent. Soon the bases of the hydrothecse form a 

 single row, and onh- the oral part is turned or bent out alternately to either side of the median plane, 

 as a last trace of the biserial structure. Gradually also, secondary ramification takes place, the primary 

 branches throwing out secondary branchlets, directed alternately to either side in the primar}' broad 

 plane of the branch. In the more finely built colonies, the main branch has three, or more rarely 

 four hsdrotheca; between two branchlets; in colonies of coarser build, on the other hand, the number 

 is only exceptionally as low as three, and as a rule we find four or more lu'drotheca- between the 

 origins of two secondary branches. 



The features here mentioned are of importance in several respects. In the first place, the\- 

 show that Hydralliiiaiiia colonies agree entirely in their mode of development with the remaining 

 Sertulayiida: having spiral stem, as far as this has been studied and described; the}' prove also, that 

 the Hydrallmania colony, which at the first glance appears so entirely distinct, can be easily traced 

 back to the t}'pical colony in Scrtiilaria tcncra or Thujaria tliitja. In the second place, the features 

 in question reveal a striking resemblance between finer colonies of Hydralhi/auia falcata and Hydrall- 

 mania dishius Nutting from Puget Sound ; the likeness is so great that we are tempted to note them 

 down without further consideration as synonyms, or rather, to regard the specimens from Trondhjeni 

 Fjord as belonging to the latter species. It would, however, be highly remarkable that this Pacific 

 species should occur here and not elsewhere in the Atlantic. And closer study now shows that the 

 Trondhjem Fjord type is not isolated, but is related, through ever}- possible transition type, to the 

 close and robust form which is predominant in the open sea, and which has always been taken as 

 the type for Hydrallmania falcata. It is interesting also to note that several colonies in the material, 

 as for instance some from Berufjord, have in among the lower normal branches one or two others of 

 the Scrtularia type. Elsewhere in the material we also find small colonies from several places agreeing 

 with the young form from the Trondhjem Fjord, only of coarser Iniild. 1 



Nutting (1904 p. 126) notes as a distinctive character between Hydrallmania dislans and Hydr- 

 allmania falcata that the latter species lacks hvdrotheca on the stem at the base of the branch. 

 This is, however, exceptional. The h}'drotheca on the stem at the base of the branch may at times, 

 especially in older colonies, be lacking, but in most cases it is to be found, and is always present in 

 the somewhat smaller colonies. Nor can the difference mentioned by Nutting as existent in the 

 gonothecse be accepted as a specific character; it lies well inside the limits of variation observed in a 

 single fairly luxuriant colony of Hydrallmania falcata. That I nevertheless refrain from definitely 

 withdrawing Hydrallmania distans is due to the fact that my collections include no material of the 

 species in question, and I am thus unable to determine whether possibly some characters not mentioned 

 in Nutting's diagnosis might be found, sufficient to warrant specific distinction. The other American 

 Pacific species also, Hydrallmania franciscana (Trask), calls for thorough investigation, its raison d'etre 

 being likewise highly problematical. The available drawings and data for the two species are not 



enough to decide the question. 



18 



Tile IngoIf'Expedition. V. 7. 



