STUDIES IN MARINE ECOLOGY. 1 87 



in some abunbance and by the larger number of peculiar species. 

 They are not closely related to other habitats if one excepts the 

 conditions under the rocks which scarcely form a different 

 association. Its main difference is that burrowing and exposure 

 are more limited than on the open flats. The animals common 

 on the flats are particularly absent from wharf pilings showing, 

 as would be expected, that these habitats have little in common. 



If quantitative data were available, the distinctions here 

 found would doubtless stand forth more plainly. The fact that 

 a beginning can be made in indicating relationships by the 

 methods used when no relationships appeared from an analysis 

 of bare check lists emphasizes the need of quantitative studies in 

 animal ecology. 



This need was first recognized by Forbes (1907) who devised 

 a mathematical formula for determining the existence of an 

 association. In 191 1, as the result of studying seasonal succes- 

 sion in ponds, I concluded that qualitative work gives insufficient 

 basis for exact conclusions. Shelf ord in 191 5 repeats the formula 

 of Forbes, and Michael (16, 21) has done more than anyone else 

 in America in showing the fundamental need of quantitative 

 investigations in ecology and in developing formulae to enable 

 one to study associations on a quantitative basis. 



The work here presented is of course only quasi-quantitative 

 in character but the greater clearness obtained indicates that 

 much of the muddle of animal ecology may be cleared by the 

 further development and the application of quantitative methods 

 in field researches. The problem of the ecologist studying littoral 

 distribution is not so hard as that of the plankton student where 

 as Michael says (1921): "Granting the equivalent of the oak 

 tree or pine tree association, the marine ecologist finds difficulty 

 not only in describing it but even in finding it. Since he cannot 

 directly witness such an association, he is compelled to rely on 

 indirect evidence furnished by tow-net or similar apparatus. 

 In other words his only recourse is to measured magnitudes and 

 application of mathematical logic thereto." For the exact deter- 

 mination of such relations the methods here used are almost as 

 gross as are the ordinary qualitative observations in trying to 

 solve the relationships existing between littoral associations. 



