Hyatt.] 78 [March 5, 



a cap of uncolored cells and ectoderm ciliated, the interior is solid 

 and continuous in the cap, and in the centre is a group of 

 colored cells as if the endoderm were represented. 



We have also a figure of Halichondria distorta with pre- 

 cisely similar form, though younger. The ectoblast cells are 

 not ciliated, and there was also a continuous perfectly trans- 

 parent mass filling the interior, and rising up posteriorly as in 

 Keller's larva of Esperia. We did not define the cells of the 

 interior, but simply noted its homogeneous transparency. The 

 figure of the later stages of this species showed the usual 

 larva of Halicondria with the central mass of spicules. In 

 Chalina oculata, we also found a very similar though older larva. 

 This was at first supposed to be an intrusive Calcispongian, but 

 its size and the peculiar smooth and hyaline, structureless aspect 

 of the posterior region indentified it as a true chalinoid larva of 

 the same age as Keller's Esperia. It was exactly oval and had 

 the cells of the anterior larger part ciliated. The resemblance 

 to the amphiblastula of Calcispongiae showed that closer com- 

 parisons could be made between them and Carneospongia than 

 had yet been attempted. 



A gastrula was observed and figured in Halisarca Dujardinii hav- 

 ing two layers, the outer with the usual cylindrical cells and the 

 inner of smaller more or less compressed epithelial cells. Haeckel 

 is quoted as having figured a gastrula of Halisarca in his papers 

 on the Gastrula theory, but we have failed to find the figure in 

 the Jenaishche Zeitschrift. Our gastrula had the aperture rather 

 large, but in other respects exactly resembled the double-layered 

 planulae and gastrulae figured by Haeckel in his Calcispongiae. 

 Schultze (Zeitschr. Wissen. Zool. vol. xxviii, f. 20, pi. 4) figures 

 what may be the beginning of a permanent gastrula in Halisarca 

 lobularis having cylindrical cells, and below in the same figure, 

 which includes several larvae, there is a planula, becoming filled 

 with mesoblast cells. These figures are exceedingly interest- 

 ing because the embryo has no endoblastic cap or collar 

 as figured by Schultze, and also, is advanced to a compara- 

 tively late stage in which the endoblast cells become converted 

 into ciliated cylindrical cells before becoming invaginated. 

 Dr. Sollas (Quart. Journ. Micr. Sci., vol. xxiv, p. 609) has 

 described gastrulae in Hal. lobularis as habitually occurring, and 



