COOKE : ON GEOGRAPHICAL DISTKIBrilON. 105 



province, by the singular toothed Acanthina ( Calcar) and by Concholepas, 

 both of Avliich genera appear to have a wide but as yet undetermined 

 range on the coasts of AVest America. 



Similar attention might be paid to Nerita and also to Littorina, due 

 legard being had to tlie fact' that, in the latter case, some species 

 which live in tlie neighbourhood of high- water mark are viviparous, 

 while others, which live at a lower tide-level, produce a trochosphere 

 or veliger embryo. The exact distribution and economic habits of 

 such widely spread species as mauritiana, Lam., aspera, Phil., the 

 group which centres round scahra, L., niilicD-is, Quoy, and malaccana, 

 Phil., would repay investigation and might bring out some interesting 

 facts. 



We still continue to speak of the distribution of marine Mollusca 

 under the headings of 'districts' or 'provinces' or 'regions', or 

 whatever name we choose to employ, and indicate the fact tbat 

 certain wide areas or stretches of adjacent coast-land are characterized 

 by the occurrence of certain genera and species, as contrasted with 

 the phenomena observed in the case of other geographical areas. 

 This method of subdivision is convenient, but it needs careful 

 handling. The results of deep-sea dredging during the past few 

 decades have accentuated the fact that these subdivisions apply 

 solely to the Mollusca of the shore or of shallow water. And we 

 must be careful to recollect that in scarcely am^ instance is it 

 possible to draw a hard and fast line between one 'region' and 

 another. On the contrary, adjacent regions seldom fail to overlap. 

 On the west coast of America, for instance, the Magellanic region 

 overlaps the Peruvian, and the Peruvian the Panamic, and the same 

 is the case Avith the regions further north, the Californian and tlie 

 Aleutian, while the Aleutian in its turn graduates into the north 

 circiimpolar region. All that we can allow ourselves to mean, when 

 speaking of the limits of a region, is that at a certain point on the 

 map we are able to say that the characteristic fauna of that region 

 occurs infrequently, or is beginning to be sensibly replaced by a fauna 

 characteristic of another region. 



Some regions, owing to special geographical facts, may be more 

 sharply defined than others, at one or at both extremes. If we were 

 asked to cite the sharpest break in existence between one marine 

 fauna and anotlier we should lay our finger on Cape Hatteras, at 

 which point a vast number of prominent tropical species find their 

 northern limit. But how can we name a point of separation between, 

 say, the Californian and Panamic, or between the Panamic and 

 Peruvian regions ? The main but not the only factor in determining 

 the limits of a region is the surface temperature of the sea-water, 

 as distributed by ocean currents. 



The truth is that the present state of our knowledge, as regards 

 the geographical limits of this or that fauna, is singularly defective. 

 Large portions of coastline remain at present unexplored, and it is 



' W. M. Tattersall, quoted by B. B. Woodward in Proc. Malac. See, vol. viii, 

 p. 282, 1909. 



