115 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



quate the classification. In the present paper characters have 

 been freely drawn from two important additional prov- 

 inces, namely those of geographical distribution and of habits. 



In general, characteristics from these provinces are not 

 sufficiently well known to permit of their use to any great 

 extent. Confessedly, also, such characters, especially those 

 from habits, are often less tangible and far less convenient 

 than physical characters, but as information is accumulated 

 they must certainly find larger place in taxonomic considera- 

 tions. 



In view of the aberrant nature of Phenacomys longicaudiis 

 there would seem to be some argument for the erection of 

 a new genus for its reception. It seems to the writer, how- 

 ever, that a principle given expression by several authors, 

 notably Osgood (U. S. Dept. Agric, Bureau Biol. Surv., N. 

 Amer. Fauna 28, 1909, pp. 24, 25), and Sumner (Science, 

 June 18, 1915, pp. 899-902), should be recognized, namely, 

 that in a classification which is inevitably critically analytic, 

 the synthetic phase should not be forgotten. The multiplica- 

 tion of genera for the purpose of emphasizing group dif- 

 ferences which are comparatively slight would seem to be 

 unwise. 



In the present instance, particularly, the chance that the 

 fundamentally close relationship of Phenacomys longicaudiis 

 to the genus Phenacomys will be overlooked is greater than 

 the chance that its differences therefrom will not be ap- 

 preciated, so it has seemed wisest to accord it subgeneric 

 rank only. 



The chief Characteristic of our knowledge of mammalian 

 life-histories is its incompleteness. It is estimated that of 

 even our best known species the life-history material avail- 

 able is only five to twenty-five per cent of what it should 

 be. These considerations emphasize the obligation imposed 

 upon the investigator to put on published record such facts 

 regarding habits as he may discover. 



While it is to be hoped that the study of morphology will 

 be no less vigorously prosecuted in the future than in the 

 past, it would seem desirable that the study of psychological 

 predilections and associational relations be much more em- 

 phasized in the future than in the past. Habits and associa- 



