THK SKRITLAKID/E. 23 



homologous, and are usiiiilly siinilur in function. I shall therefore speak of tin- <>p<Mviil;i of such 

 species us X. jni/nilii as two-flapped, and will include practically all of tin- .-pecie- in which (lie 

 "collar" described by Levinsen appears. 1 I have not been able to find >pecimen- in which the 



collar presents just the appearance delineated by Levinsen. ' This may he explained by the fact 



that my specimens were in a very % 1 state of preservation and stained, while the hvdrothec:c 



figured by him do not appear to contain hydranths. and may have lieen boiled in caustic potash 

 or otherwise treated to bring out the unimpeded view of the chitinous parts, thus destroying the 

 hydranths and possibly altering the appearance of the operculum. In unstained specimens it is 

 very diflicult. if not impossible, to see the tine line marking the outline of the upper liorder of 

 the adcauliiie Hap seen in tig. .">> ,/,/. When this is not seen the structure looks verv much as it 

 does in Levinsen's figures. 



In sj)ecies with very long teeth it .appeal's that the teeth continue growing after the 

 operculum has lieen formed. If we examine a specimen of X, rinhn-'ni /i, ,;,//, if, i. for instance, 

 we will lind that the teeth project far beyond the distal ends of the opercular Haps (see lio-. 41). 

 [f the operculum originated in this species in the same manner as it dues in X. jn/,n/l/i. the greater 

 part of the growth of the teeth must take place after the operculum has been formed. Other- 

 wise the latter would reach clear to the points of the teeth. It would lie interesting to study the 

 operculum in sertularians with numerous teeth, such as are found in Australian species of 

 S,'iiii/iii'in. and to investigate the relations between the two structures. I do not know that 

 any one has investigated these species with this point in view. 



The one-flapped operculum (tigs. (it!-(iit) is common in Thunn-in, and is universally found in 

 .\l>i: tiiitii'in, I>iji/ii/niii, and S,l<ii/!>i,ijiKiN. As maintained by Levinsen. the number of teeth 

 seems to determine the number of valves. This may be understood on purely mechanical 



principles. In tl te-flapped forms the cleavage seems most easily effected where the operculum 



joins the thinner parts of the hydrothecal margin. It so happens that large groups of species 

 seem to be fairly constant in this matter, some having the hydrothecal margin thickened on the 

 adcauliiie side, as I)ij>h<txi<i and Abietinaria, and hence the opercular Hap remains attached to 

 that side. In other cases, as Thiii/irin (in part), and X /,,,///, ,<//*, the thickening is on the 

 abcauline side and the operculum is abcauline in position. There is one very curious exception 

 to the undoubted systematic value of this character, and that is in the case of Tlni!nr!<t tln/i- 

 iir!i>!<li-s, in which the species is a typical thuiarian in both trophosome and gonosome save in 

 the one matter of the opereulum, which is unmistakably adeauline in position. 



The strongest and most easily seen opercula of this type are found in the genius D!j>Jt(ti<i, 

 notably in I >. lUi/itnHx (Plate XXX, tig. fi), where this structure takes the form of a vaulted, 

 almost dome-shaped hood, which seems to be of texture as strong and dense as that of the 

 hydrothecal walls, and tits like a cap over the hydrothecal aperture. 



The three and four flapped opercula are characteristic of the genus X, liuliir, II, i and I)!rti/n- 

 <Jinl.il/in, a closely related genus, llartlaub, as before stated, claims that this kind of an oper 

 culum differs essentially from the two-Happed form. I fail to tind any very material difference, 

 the main distinction being that in X, rtnliir> H<i the margin of the hvdrotlieca is usually thicker and 

 better outlined, and the opercular structure stronger and more evident. The cause of the initial 

 rupture of the operculum may lie somewhat different in this case, although purely mechanical. 

 The margin is stronger and more thickened at the points occupied by the teeth. These latter, 

 moreover, are often, if not generally, slightly inclined "outward or away from the center of the 

 aperture. The former consideration would render the hydrothecal margin stiil'cr at the points 



occupied by the teeth, and the latter would result in greater tension across the operculum along 

 lines connecting opposite teeth. When the hydranth pushes outward for the tir>t time the elastic 



1 In order tu lir very sure that my iiittT|>ivtutiiiii (if this structure is correct I have examined a mnnlicr of stained 

 ami mounted specimens with ;,'reat rare, nsini; hi^li powers of magnification and examining speeimens in which the 

 hydranth was in various stages of contraction and expansion. In liu's. "i.'i fxi a specimen is illustrated which was 

 turned over so that lioth the front and hack views of the margin were olitaincd. The sketches \\ere made liy myself 

 with the use of the camera lucida. I also examined a number of other species in which the "collar" is found. 



''Meduser, Ctenophorer og Hydroider fra Gronlands Wstkysi, isn.'i, pi. vn. litrs. s-lt). 



