[ 127 J 



XIY. 



THE NEWLY-HATCHED LAEVA OF EUPHYLLIA. By A. C. 

 HADDON, Eoyal College of Science, Dublin. Plate XI. 



[ReadMAECH 19, 1890.] 



In the early part of August, 1888, I collected some living specimens 

 of the coral EuphyUia rugosa (Dana), which is abundant on the reefs 

 in Torres Straits. One of these specimens gave birth to a number 

 of free-swimming larvae. I was unable to keep these alive for 

 very long, and I never met with the larvae again. As we know so 

 little about the development of the Madreporaria a brief descrip- 

 tion of but a single stage of one species will have some value. 



The larvae were about 1'5 mm. in length, and of an elongated 

 oval or slender pear-shape (PL xi., figs. 1 and 2), and actively 

 swam about, owing to a uniform ciliation of their ectoderm. The 

 yellowish endoderm could easily be seen through the colourless 

 ectoderm : it was marked by twelve pairs of longitudinal olive- 

 brown lines. Occasionally they floated with their aboral end 

 uppermost, as in fig. 3. After a time they very generally assumed 

 a flattened or cake-like form (figs. 4 and 5). The contour of the 

 endoderm was crenulated by twenty-four sulci, each sulcus corre- 

 sponding to an interval between the vertical brown lines. The 

 ectoderm in this stage was also crenulated, the grooves being- 

 opposite to each pair of brown stripes. The ectodermic and the 

 deeper endodermic furrows correspond with the mesenteries ; the 

 aboral disc was apparent as a small round whitish patch [ah. d.). 

 Finally, they often rested on their aboral aspect on the side of the 

 vessel, and then they had a conical appearance (fig. 6), the upper 

 portion of their bodies being frequently marked by two annular 

 constrictions (fig. 7). I could not keep them alive after this 

 stage. 



Sections made through larvae killed at the commencement and 

 at the close of the period during which I kept them alive show so 

 little differences in structure that the latter may be neglected and 

 all the forms described as belonging to one stage. The stage is 



SCIEN. PROC. R.D.S. VOL. VII. PART III. M 



