HVDROIDA II J27 



thus brought down tlie number of species to 8 or 9, which figure, as will be seen from the following, 

 must be yet further reduced. Before passing- on to special treatment of the different species, however, 

 certain remarks as to the nature of the variations will not be out of place. 



The form of the colony in all young Sertularia specimens is pinnate, and with the except- 

 ion of a very few species such as for instance Sertnlaria mirabilis^ it appears to be a general rule that 

 the stem sooner or later assumes a spiral coil, while at the same time the broad plane of the branches 

 changes from the perpendicular to the horizontal, perpendicular to the axis of the stem. The cause of 

 this spiral coiling it is difficult to determine, but the feature itself cannot be accepted as a good 

 specific character (cf. Broch 1912 p. 13I More importance should be attached to the winding of the 

 coil, whether right or left; the only northern species which is dextrorse is the Sertnlaria Fabricii. 



A peculiarity in all spirally coiled species is the fact that their branches are more or less 

 markedly dichotomically ramified; as long as the colony still preserves its pinnate shape, this second- 

 ary ramification is less pronounced, but in spirally coiled forms it is more so. 



The stem ma}' be more or less prominent in one and the same species, now darker, now 

 lighter in hue, as a rule .slightly zigzag, more rarely quite straight, though this cannot be taken as a 

 specific character. The distance between the branches proceeding from the stem likewise varies within 

 one and the same species, and this feature has considerable influence on the habitus of the colony 

 generally. The stem is in all species segmented; in those investigated, the internodia had as a rule 

 one, more rarely two, and in quite exceptional cases three branches; where only one branch is present, 

 it proceeds from the basal part of the internodiuni. It is not altogether impossible that the frequency 

 of branches froni the internodia may prove of some value as a specific character. 



The hydrothecse are subject to considerable variation within one and the same species. The 

 divergent free distal part may diverge more or less from the branch, and it is somewhat different in 

 the pinnate and spiral parts of the colony. In the pinnate portions, the symmetrical plane of the hy- 

 drotheca always coincides with the broad plane of the branch; in the spirally coiled part, on the other 

 hand, we find that the hydrothecje in most of the species tend more or less towards unilateral arrange- 

 ment on the upward side of the branch. The aperture, how^ever, varies very little indeed, and Nut- 

 ting's statement (1904 p. 70) as to great variation in the cliaracter of the aperture in Scrtularia te- 

 iiera can onh- be ascribed to insufficient care in the investigation. Owing to this tendenc\' to unilateral 

 arrangement, we normally find that the hydrothecse on these branches are seen in oblique jirojection. This 

 has led numerous investigators to regard certain species, such as in particular Scrfiilaria argentea^ as 

 furnished with asymmetricallv developed lateral teeth, whereas careful investigation shows that this 

 is not the case. — The proportion between length of the hydrotheca and the free projecting part seems 

 to furnish a useful specific character. 



The gonothecte are, as already pointed out (1909 p. 172) extremely variable in Scrfularia 



species, and their developmental stages have more than once been used as specific characters. It is 



simply astonishing to see how far from critical the investigations have often been with regard to this. 



The grown gonothecse vary greatly in one and the same species. In species where they liave no 



spines, they may be round in section, or angular, with a varying number of angles; this is also noted 



by Jaderholm (1909 p. 94) but he has not noticed that one colony may have both round and angular 



l6* 



