A. W. WATERS—REPORT ON THE BRYOZOA. 245 
Farella repens and Farella elongata have no sete, but I find the collar is 
well developed in these species. 
The gizzard of this groap of the Ctenostomata was not quite correctly 
described by the earlier authors, and the only exact figures of the structure 
with which I am acquainted are those of Calvet*. In cross-sections there 
are from 30-50 teeth, varying much in the formidable terminations, which 
stain much darker than the rest. The ends in Zoobotryon pellucidum may 
be spear-headed, hammer-headed, two-pronged, or reflex uncinate. 
There are two sets of larger teeth opposite one another, and it is these that 
are most noticeable, when the animal is alive or in whole mounts. Sections 
(Pl. 24. fig. 15) show that the muscular wall of the gizzard has sometimes 
separated from the teeth, and then the contents of the teeth are attached to 
the muscular wall and are more or less drawn out of the teeth. I have found 
identically the same separation of the contents in Bowerbankia imbricata, 
Adams, and B. pustulosa, Soland. We cannot therefore speak of this as a 
nucleus of the teeth. Calvet { mentions a conical cavity enclosing a proto- 
plasm “ pauvre en granulations, et un noyau dont le situation est variable.” 
The gizzard of Cryptopolyzoon, Dendy, is quite different to any other 
known gizzard, and as Dendy described it there are “two relatively large 
chitinous teeth ”—“ squarish in shape and flattened ”’—* planted within the 
muscular mass.” 
This being so diametrically different from other gizzards, I was anxious to 
make sections, and the British Museum authorities kindly allowed me to 
prepare some sections from their preserved material. The zoarium is covered 
with sand-grains and the walls of the zoccia are thick chitin, so that even 
after much labour only moderately successful results were possible. However, 
the gizzard was found to have very thick and powerful muscles surrounding 
the two teeth (Pl. 25. figs. 1, 2, 3), which present two flat surfaces to one 
another, with slight projections near the border. These gizzards occur near 
the “‘ cardiac portion of the digestive system,” and I distinguish them from 
other gizzards as being of the grindstone type. It will be noticed in com- 
paring my figures of Cryptopolyzoon and Zoobotryon Ke. that the gizzard and 
the zocecia of the first are small (PI. 26. figs. 1, 2, 3). The zowcia are placed 
close together, sometimes partly attached as in Amathia, and from the 
chitinous wall there are slight projections ending in a disk-like knob which is 
attached to the grains of sand. The number of zocecia is very large in each 
node of C. Wilsont, Dendy t. We must beware of judging without sufficient 
examination whether there are gizzards in any species, and it is naturally 
* Bryozoaires Ectoproctes Marins, pl. 7. figs. 4, 8. 
+ Bry. Ectoproctes Marins, p. 230. 
¢ Arthur Dendy, “On the Anatomy of an Arenaceous Polyzoon,” Proc. Roy. Soc. Vict. 
vol. i. (1888) n. s. p. 1, pls. 1, 2, 3; “ Cryptopolyzoon, an Kmendation in Nomenclature,” 
Zool. Anzeiger, vol. xxiii. (1900) No. 620. [Cryptopolyzoon Wilson, Dendy, also occurs 
from Knysna, 8. Africa (fide Kirkpatrick). | 
