828 



SCIENCE 



[N. S. Vol. XLIV. No. 1145 



coupled traits coexist in the same community 

 under conditions which preclude all possibility 

 of ascribing one of the traits to an indigenous, 

 the other to an immigrant culture. This 

 being so, what justification is there for ad- 

 vancing such an interpretation in any case, 

 unless the assumption can be supported by 

 specific evidence? Obviously, the easier it is 

 to explain a phenomenon in one of two ways, 

 the more vigorous must be the proof if one 

 of the two alternative explanations is selected. 



After all, then, there is a close similarity 

 between Rivers of the Melanesian Society and 

 Graebner of Die Melanesische Bogenkultur, 

 The former author takes special pains (II., 3, 

 seq.) to assert his complete independence of 

 Graebnerian method. That the author's posi- 

 tion is in part justified, has been shown before. 

 But in one respect the relationship of the two 

 systems is unmistakable. Both authors utilize 

 diffusion not as a process to be demonstrated 

 but as one to be assumed for the purpose of 

 hypothetical culture building. To be sure, 

 what Bivers builds is altogether different from 

 that which is built by Graebner, but the prin- 

 ciples according to which the different parts 

 of the structures are welded together are the 

 same in either case. 



Before closing it will be well to refer to 

 Dr. Bivers's own definition of his method. 

 We read: 



This method has been the formulation of a work- 

 ing hypothetical scheme to form a framework into 

 which the facta are fitted, and the scheme is re- 

 garded as satisfactory only if the facts can thus 

 be fitted so as to form a coherent whole, all parts 

 of which are consistent with one another (II., 

 586). 



The method, thus formulated, is, as a method 

 of historical research, self-condemnatory. It 

 may well be applied in the shaping of those 

 hypothetical conceptual systems which are in- 

 troduced by the theoreticians of the exact sci- 

 ences for the purpose of providing a simplified 

 description of the data of experience in a par- 

 ticular field. It does not matter how the 

 vortex looks (or whether it looks at all), if only 

 the functions of the ether can be readily de- 

 rived from it. It may not be of importance 



whether the atom exists or not (with apologies 

 to Lord Kelvin), but if it furthers a successful 

 formulation of the facts of chemistry (a task 

 in which of late it has conspicuously failed), 

 its conceptual existence is vindicated. Not so 

 in history. It has been said, with some truth, 

 that for an understanding of society it is less 

 important to know what has occurred than 

 what may have occurred. But surely this does 

 not apply to the study of history as such, nor to 

 ethnology, in so far as its task is historical. 

 Here the search is altogether for what has oc- 

 curred, although the knowledge of what may 

 have occurred can serve as a useful guide in 

 the search. In the domain of ethnology, more- 

 over, our knowledge of what has occurred will 

 have to be increased many times before we 

 can safely trust our intuitions as to what may 

 have occurred. 



To repeat, then, Dr. Bivers has labored 

 fiercely against heavy odds, he has reopened an 

 old and much trodden field; his work empha- 

 sizes once more the amazing cultural com- 

 plexity of those southern seas; it is rich in 

 subtle psychological analysis and happy formu- 

 lation of theoretical principles ; it also abounds 

 in ingenious hypotheses of great prima facie 

 plausibility. But we can not endorse this " his- 

 tory " as a model of ethnological method, for 

 a history surely it is not. 



A. A. GoLDENWEISER 



Columbia University 



SPECIAL ARTICLES 



LOBSTER MATING: A MEANS OF CONSERVING 

 THE LOBSTER INDUSTRY 



During the summer of 1914 the writer, 

 working under the auspices of the Biological 

 Board of Canada, attempted to rear lobster 

 fry to the crawling stage, using the now famil- 

 iar apparatus of the Bhode Island Commis- 

 sion. The site chosen for the repetition of 

 the celebrated Wickford experiments was St. 

 Mary's Bay, Digby Co., Nova Scotia. The 

 attempt proved a complete failure due chiefly 

 to the extreme cold water (50° F. to 60° F.) 

 and to the extensive development of diatoms 

 which soon closed up the mouth parts of the 

 fry and caused an exceedingly high death rate. 



