70 THE CONSTITUTION OF THK GASTROPOD PROTOCONCH, 



passed within the capsule. This leads to the suggestion that, 

 in all those cases where a calcareous protoconch of but few 

 large whorls is present, the whole of thei larval stages are 

 passed within the egg. 



The converse of this, however, does not hold true ; there 

 are very many species which pass the larval stages within the 

 capsule, and which do not possess Tumid, MarmUate or 

 Eccentric apices; such is the case throughout the Pulmonata 

 and in Littornia, Furpura lapUlus, Ver/netus and many 

 Murices. I'ahidnia and many Pulmonates are viviparous, as 

 inded are some species of Jyittonna, if not all ; and, in all 

 these cases, the apex is perfectly regular. 



The abundance of nutrient material within the capsule does 

 not, of necessity, determine a tumid aj^ex ; in Purpura lapU- 

 lus Linn., and several other species, the few embryos which 

 escape from the capsule have previously eaten the many 

 others originally present in it. In Megalatractus aruaiius 

 Linn., there is sufficient food-material to support the embryo 

 whilst forming five or six whorls of (ananeanic) shell, but the 

 apex is eccentric in this species, not tumid. 



"A number of the species (of Ci/matium) have a world-wide 

 distribution, which is doubtless due to their free-swimming 

 pelagic larvae (Tryon). I am inclined to carry this statement 

 further, and to say that the distribution of a species is largely 

 decided by the size of its protoconch." The above is a quota- 

 tion from my paper on " Lotoriui/i ," and needs modification. 

 I had been struck by the fact that most of the many-whorled 

 protoconchs occurred in species of wide distribution, whilst 

 the localised species had, in general, small protoconchs. In 

 the case of the species of Ci/fnatium , the generalisation is 

 justifiable ; the protoconch is the replica of the Nectoconrh, 

 that is to say, the number of whorls in it is a measure of the 

 length of the pelagic existence of the species. Beyond the 

 genus Gi/matium, the generalisation, as expressed above, is 

 liable to lead erroneously to its own refutation. Stated so as 

 to be generally applicable, the idea which the above statement 



