BIOGEOGRAPHICAL BOUXDARIES 203 



a laboratory rather than any biological character. Those of you 

 who feel this is an exaggeration should compare Walford's (1958) 

 picture of the world distribution of marine laboratories with 

 Steuers' (1933) picture of the distribution of the copepod A?w?nalo- 

 cera patersoni. The two charts are almost indistinguishable. 



The dispersal of communities on the land, in lakes, at the edge 

 of the sea, and on the sea floor is often limited by physical bar- 

 riers. Because of this, terrestrial or benthic animals and plants 

 may usually be assigned, with some precision, to a particular type 

 of fauna or flora. Frequently, terms may be chosen which describe 

 the factors limiting distribution; for example, brackish, littoral 

 intertidal, or desert. 



Because there are so few barriers to their dispersal, pelagic 

 communities are not so easily classified and it is difficult, if not 

 impossible, to use the word "community" with any precise mean- 

 ing. I shall use it rather loosely and not necessarily with the im- 

 plication of interdependence of the animals and plants described 

 in this way. 



Throughout this paper, the word "distribution" has been used 

 to refer to the geographic location of gradients of abundance. 

 Many biogeographers would use distribution to describe the 

 geographic range or limits within which a species is normally 

 found. In ecology, however, distribution and abundance cannot 

 be separated; as Andrewartha and Birch (1954) have said "they 

 are but the obverse and reverse aspects of the same problem." 



These problems of classification and nomenclature are intensi- 

 fied by one of the most interesting characteristics of pelagic organ- 

 isms. To quote Ekman (1953), "At the boundary between two 

 faunal regions the conditions of flow may on certain occasions 

 bring about a change in the composition of the plankton and 

 certain plankton species are thus indicators for various types of 

 water." 



With these difficulties in mind, it is highly desirable that dy- 

 namic biogeographical studies shall be based on standardized 

 methods of sampling applied over long periods of time. Such con- 

 sistency is useful in defining the limits of distribution of species 



