TURBINELLA RAPA. 



233 



of which it consists are bound or agglutinated together by a 

 distinct membrane, adherent to the outer margin a and to 

 the flattish surface h of the sacs. On the opposite surface, 

 corresponding to b, the capsules are free, but on the margin c 

 they are partly attached by a sort of articulation, each sac 

 being locked by a sort of process to the one next it in the 

 series. This process is seen at cl in Fig. 2, in which five 

 capsules are shown with their upper walls removed, so as to 

 exhibit the young shells in the interior. Immediately in 

 front of d there is a membranous spur or prolongation e, 

 the probable use of which will be alluded to hereafter. The 

 process of laying the eggs in an allied species, Turbimlla 

 pyrum, is so well described by Sir Everard Home, that I 

 take the liberty of transcribing his narrative from Johnston's 

 Introduction to Conchology, page 353 (Edition, 1850). "A 

 friend of mine," says Sir Everard, ' ' saw the female shed her 

 eggs; a mass, apparently of mucus, passed along the deep 

 groove in the lip of the shell in the form of a rope, several inches 

 in length, and sunk to the bottom ; this rope of eggs, enclosed 

 in mucus at the end last discharged, was of so adhesive a 

 nature, that it became attached to the rock, or stone, on 

 which the animal deposited it. As soon as the mucus came 

 in contact with salt water, it coagulated into a firm mem- 

 branous structure, so that the eggs became enclosed in 

 membranous chambers, and this connected nidus, having 

 one end fixed and the other loose, was moved by the waves, 

 and the young in the eggs had their blood aerated through 

 the membrane, and when hatched they remained defended 

 from the violence of the sea till their shells had not only 

 been formed but had acquired strength." 



The end /, Fig. 1, Plate II, would appear to be that which 

 remains free, the other being attached to a rock, coral or stone, 

 by two membranous expansions, and it is not improbable that 

 the peculiar attachment of the agglutinating membrane, 

 already described, and the membranous prolongation e } Fig. 2, 



