46 



scarcely perceptible in some, as in Patella limhata, Phil., a 

 kind of plexus where the odontophore is very much twisted 

 and convoluted, so as scarcely to be drawn out without breaking, 

 but this I imagine serves some purposes of digestion, very 

 different from the action of what we usually call a stomach. 

 These, however, are matters of my own opinion only, which I 

 have not been able to bring to the test of any physiological 

 experiment. 



The odontophore is easily drawn out of all gasteropoda 

 examined by me, except the species just mentioned. la 

 Patellae and Littorinidse (winkles) it is always long. In our 

 Risellas and Littorinas it is of enormous length, but lies, in 

 this case, in a simple coil immediately behind the mouth as a- 

 silky siliceous thread. In Chitons it is a closed tube with 

 teeth all round it as in Haliotis ncevosa. 



Professor Forbes (Brit. Moll.) has remarked that the 

 character of the teeth and their arrangement is very constant 

 in the various genera. I have found that it varies also for the 

 species in Tasmania, at least in the order I am now dealing 

 with, and I hold it to be a very valuable test as to specific 

 difference. Indeed it is a test where all others failed, because 

 shells are often so corroded as to obliterate marks on which 

 specific differences are chiefly erected. This is exclusively the 

 field of the microscopist, but I am convinced, not only that it 

 is a wide and valuable field for investigation, but that until it 

 is* cai-efuUy explored we shall have no solid system of 

 conchology resting upon a secure scientific basis. The teeth 

 in all the Pafelke and Acmece have raised double edges or 

 jjoints which curve, succeeded again by a smaller double edge 

 or point. Thus each set is curved back from its attach- 

 ment to the odontophore at its lower side. As far as my 

 investigations have gone there is a general correspondence 

 between the pattern of the odontophore and the organs of 

 respiration. In Patellae it is of one type, and Acmese of 

 another, though the resemblance is very close. In Siphonaria, 

 however, which is pulmoniferous, we shall see in the course 

 of this paper that it has a dentition quite uniform with the 

 land and freshwater mollusca. This can hardly be called an 

 anomaly, though it points out a singular fact rather aoverse 

 in my mind to the theory of natural selection. Here we have 

 a marine animal with a shell differing but slightly from the 

 commonest of our marine types and apparently living under 

 the same conditions, yet organised to breathe air and salt 

 water, and with a dentition exactly similar to pulmoniferous 

 mollusca living under totally different conditions. Of course a 

 double inference may be drawn from this as from similar facts, 

 but thoy point in my mind much iu the stronger way to an 



