22 University of California Publications. [| ZooLocy 
combination of strictly opposite branches, smooth margined hy- 
drotheeae, absence of opereula, as well as the intrathecal origin of 
the gonangia. 
Analysis, however, does not reveal the strength which is 
elaimed for this structure. In the first place, combinations are 
of little value unless the characters selected for combination 
vary independently of each other, which is obviously not true 
of margin and operculum, as Nutting is aware. In the second 
place, it is well known that at some stage in the development of 
all hydroids. a perisarcal membrane blocks the exit of the 
hydranth from the hydrotheca, and that this membrane becomes 
the one-, two-, three-, or four-parted adult operculum, according 
to the character of the margin, or may be wanting altogether. 
Among sertularians with smooth round margins, it is often deli- 
eate, and is commonly lost. In Sertularella formosa, according 
to Nutting, it is usually wanting, but occasionally appears as a 
‘‘thin membrane stretched like a drumhead across the aperture.’’ 
In Sertularella hartlaubi, according to the same authority, the 
operculum is ‘‘in some cases an adeauline flap, in others appar- 
ently an irregularly ruptured membrane stretched straight 
across the aperture like a drumhead.’’ Sertularella halecina (a 
Synthecium aceording to Nutting) possesses a thin drumhead 
operculum before the hydranth emerges for the first time, but 
is non-operculate in the adult. Such facts only lead inevitably 
to the conclusion of Hartlaub (:00, p. 8) that the absence of an 
operculum is of no taxonomic consequence. In the third place, 
Nutting does not appear to insist that Syntheciwm shall exhibit 
the opposite branching which his definition demands, when he 
places S. halecina in that genus. It is possible to assume that 
he was heedless of the mode of branching in this species, but 
this assumption is hardly applicable to the alternately branching 
Synthecium alternans Allman. It is more probable that Nut- 
ting included XS. halecina in spite of its branching. Yet in thus 
escaping the responsibility of removing it from Synthecium to 
Sertularella or to an entirely new genus, he abandons opposite 
branching as a distinguishing mark of Syntheciwm. It would 
appear, then, that there are but two instead of four characters 
on whose association the genus is really based: the smooth round 
