234 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 



present species have their front edges in a nearly straiglit line, slightly notched 

 medially, and extend about four times the width of tiie shell, or some 20 nmi. 

 "When folded preparatory to complete retraction, they are contracted about one- 

 fourth of their extent. 



The parapodia are roughly triangular with the anterior margin longest, the 

 lateral margin distinctly trilobate and passing imperceptibly into the posterior 

 lamina, which is slightly reflected over the reflected anterior edge of the ventral 

 plate. In no case was this lamina seen to extend over tlie convexity of the ven- 

 tral plate, as it is asserted to do in the Mediterranean species. If tlie dimensions 

 of the figures of the animal in the plates of the " Bonite " are taken from the fully 

 expanded animal (which I strongly doubt), they are hardly half as large as those 

 of our Pacific variety. But the contractibiiity of these delicate organs is so 

 excessive that in a dead or exhausted specimen the real extent of the fins is never 

 shown, much less in one which has been subjected to a preservative fluid. The 

 appendages of the mantle which extend from the lateral slits are short and also 

 trilobed, but not deeply ; in the Mediterranean form they are represented as 

 entire or obscurely bilobate, and longer. The color of the living animal is trans- 

 lucent yellowish, with two large, sharply defined dark violet spots on the para- 

 podia, and a paler violet of some of the viscera shining tlirough the shell. In the 

 Mediterranean form the violet is represtnted as gradually fading out toward the 

 edge of the parapodia, but in the Pacific variety the violet area is dark up to its 

 extreme edges and not graduated in any way. The anterior edge is mesially 

 indented ; beliind and above it extend two small tentacles (figure Ic) minutely 

 swollen at their distal ends. The right-hand one is about twice as long as the 

 left one. The mouth is axial in its longest diameter, narrow, with raised margin 

 slightly pointed in front. 



In captivity in a bowl of fresh sea water the individuals kept up swimming for 

 some time. "When exliausted, tlicy drew in the parapodia and slowly sank to the 

 bottom of the bowl. After a rest they would resume their activity, and several 

 of them lived for three days, the water being frequently renewed. 



The most nearly related species, so far as the soft parts are concerned, to our 

 Pacific variety is C. gibbosa Rang, as figured by D'Orbigny in the "Voyage dans 

 I'Anierique Meridionale, but this species has a shell of quite different form. 



Hyalaea trunrata Krauss (1848), wliich appears to be the same as H. cumingi 

 Deshayes (18/7), seems to me from authentic s])ecimens to be a distinct species. 

 It is reported from the South Atlantic in latitude 40° S, the Cape of Good Hope, 

 and tlie Indian Ocean. 



The very great discrepancy in size of extreme mutations of the species of Cavo- 

 lina has led to the separation of some of them as distinct species, but we iuive iu 

 such genera as Cypraca and Marginella other instances of such discrepancies be- 

 tween individuals of tlie same species. Individuals of a single swarm are usually 

 nearly identical in size, according to my observations. It is tlicrcfoic not improb- 

 able that tlie discrepancies observed arc due to some conditions of food supply or 

 temperature which have affected a ])articular swarm during its period of adules- 

 ccQCC, which is presumably very short. 



