Till. M.KTt l.AKIDJi. 



45 



In answer to this it must In- urged that a condition, not a theory, confronts u-. and it 

 appears to lie impossible to Invak up the family Sertularidse into Croups tliat do not intergrade 

 to some degree. Reducing the number of genera will not help us. a statement that will lie 

 continued by again turning to Le\ insen's attempt, in which he recogiii/ed hut live of the nineteen 

 genera used by Marktanner-Turneretscher. Investigation shows that these the genera inter 

 grade just as vexatiously as do the twelve included in my scheme, and the five genera are 

 individually much more unwieldy and ditlicult to manage than are the twelve. Thus there is 

 nothing lost in the delimitation of genera, and much gained in convenience when the larger 

 number is utili/.ed. 



Here again the plasticity of the Hydroida as a group is the cause of much of our dilliculty. 

 It seems that these lowly animals have not yet crystalli/ed into definite and unvarying forms to the 

 extent that is found among the higher mela/.oa. and the result is that both specitie and generic 

 boundary lines are crossed in various directions, much to the perplexity of the systematist. as I 

 have elsewhere attempted to illustrate. 1 In this case it seems to me that we must abandon the 

 idea that a genus is untenable so long as there is any intergradation with other genera, and take 

 the position that a genus is simply u group of closely related species that are related more nearly 

 to each other than to members (1 f other similar groups, and that the genus can be good, both in 

 practice and in theory, even if certain species do intergrade in s.nne individual characters with 

 species of other genera. We must remember, moreover, that all genera would intergrade with 

 other genera, were a complete record accessible, and that generic distinctions must nece-sarih be 

 an expression of the gaps in our knowledge rather than of natural boundary lines. 



I have taken the position indicated above in this work, and frankly confess that some of the 

 genera used intergrade with others. These intergradations will be pointed out with care when 

 the several genera are discussed later. 



Distribution of jAmerican SertuLa/ridsB, 



'C. C. Nutting. Aililivss "1 tlie rliiiiniian nf the Sei'timi nf /,i mingy ainl \ irr-|iv,'>i.lriii nl llir Ami-riraii A.nria- 

 tion fnr tlie Aitvaiiceiiii'iit nf Si-icurc. Srii'iii-i-, .Taimary '.'. I'm:;. ],, 9, 



