RECORDS. 359 



at exactly similar points on the skull, namely, at the lateral 

 junction of the frontals with the nasals. 



The communication had been in part presented before the 

 Brooklyn Institute of Arts and Sciences, and before the Zoo- 

 logical Congress at Berne. 



Dr. Sumner described his experiments that were undertaken 

 in order ( i ) to determine the relative sensitiveness to asphyxiation 

 of the three commoner species of Fundulus ; (jz) to determine the 

 relative ability of these and some other fishes to survive trans- 

 fer to fresh water ; (3) to determine the minimum salinity which 

 certain salt-water fishes could withstand ; (4) to determine the 

 effect upon these fishes of gradual and of abrupt changes in the 

 density of the water. 



The results of extended biometric studies were set forth, from 

 which (i) it was shown that when a given species was subjected 

 to destructive conditions, and the mean characters of the more 

 and the less resisting individuals were compared, differences 

 were evident both in respect to type and to variability ; (2) that 

 when different methods of eHmination were employed with the 

 same species, selection had reference to different characters ; 

 (3) in the only case in which this question was tested, that the 

 selective elimination of two closely related species, under the 

 same conditions, appeared to have reference to the same char- 

 acters ; (4) that specimens of F. Jicteroclitiis inhabiting brackish 

 water of low salinity differed in all of the measured characters 

 from those living in pure salt water. (5) It was shown, never- 

 theless, by comparison with the more and the less fit individuals 

 of those experiments where fresh water was employed as the 

 eliminative agent, that the fishes inhabiting brackish water could 

 not have owed their modification to the natural selection of those 

 individuals better adapted to a life in water of a lower density. 

 (6) It was shown by comparing the mean characters of the 

 three species of Fiuiduhis and by taking into account their rela- 

 tive fitness to withstand certain conditions, that these differences 

 of type could not have been due to natural selection acting 

 with reference to these particular conditions. 



Dr. Hornaday, director of the New York Zoological Park, and 



