22 BULLETIN 16 5, UNITED STATES NATIONAL MUSEUM 



follows: (1) Zooecia of articulation, the 2 large cells at the base of 

 the segment. These are convex and their opesium is large, ellipti- 

 cal elongated. (2) Radicular zooecia, cells of the base of the seg- 

 ment; they are calcified, convex, perforated in the middle by a 

 round or elliptical orifice. (3) Normal zooecia, distinct, very long, 

 enlarged distally. The mural rim is thin and rounded; the gym- 

 nocyst is very narrow and bears an avicularium; the opesium is 

 terminal, large, oval. (4) Regenerated zooecia, cells with a double 

 mural rim and a small median opesium. The avicularium is tri- 

 angular, with pivot, salient, oblique to the zooecial plane. 

 Measurements. — 



„ . (Zs = 0.8 mm. ~ . [ho = 0A mm. 

 Zooecia 7 nOK Opesium 



[fe = 0.25 mm. ^ [fo^O.14 mm. 



Sttmcture. — The zoarium is always flabellate ; it begins with 1 cell 

 and ends with 5 or 6; the cells grow in size from the base to the 

 top. The first cell is attached either to a stolon or to another 

 segment by a horny or chitinous joint. 



The two premier zooecia at the base of each segment are very 

 broad (0.3 mm) ; their gymnocyst does not bear an avicularium; 

 their opesium, almost circular, is not terminal, and its diameter 

 corresponds exactty to that of the base of the segments. We can 

 therefore suppose that these are cells of ramification and that their 

 opesium serves only for the passage of the chitinous joints binding 

 the two segments, as in Cellaria and Tubucellaria. Ramification 

 occurs, then, only at the base of the segment and never at its summit. 

 The extremity of the branches terminates on a basal lamella, as in 

 all the other Membranipores. The articulation is therefore totally 

 different from that of the other jointed species. The zoarium can 

 increase in width but not in length. It does not then fear engulfing, 

 which makes us believe that the initial stolon arising from the 

 ancestrula must have crept on some floating alga. 



These two initial cells give rise to the radicular cells. They 

 develop at first as ordinary cells, for a trace of the operculum 

 remains visible, but later they become calcified, and their opesium 

 then allows only radicular fibers to pass. They are numerous and 

 grouped at the base of the segment but they appear sporadically 

 higher up among the normal cells. Small avicularia always exist 

 on this kind of cell, but we do not know whether their function 

 continues after the transformation of the normal cell into the 

 radicular one. 



We have observed zoarial fragments composed only of radicular 

 cells, which causes us to believe that the radicells serve also to bind 

 the segments together, as in the genus Canda, to give greater co- 

 hesion and more resistance to the bushy colony. 



