266 WALTER E. CARREY. 



different tests made with animals taken from the same medium. 

 In all these experiments it was necessary to sacrifice several ani- 

 mals to get even a minimum amount of blood for making determi- 

 nations, so that the results were on the whole unsatisfactory ; still, 

 they suffice for the conclusion that only a slight, if any, change 

 is induced by a change in the osmotic pressure of the external 

 medium. The animals are " homoiosmotic." 



Fiinditlus lieteroclitits is a hardy little fish well suited to this 

 form of experimentation, although the quantity of the blood is 

 too small to admit of making cryoscopic determinations. It was 

 found that if care was taken to select individuals which were not 

 injured in catching, about eighty per cent, lived in fresh water for 

 six weeks when the experiment was discontinued. This is as 

 high a percentage as can be kept alive in the sea water aquaria of 

 the laboratory. When placed in external media of concentra- 

 tions varying from fresh water to sea water concentrated to 

 twice its normal strength they showed the same hardihood. It 

 is reasonably certain that the osmotic pressure of the blood does 

 not change to any marked degree for an examination of the 

 blood corpuscles did not show either laking in the dilute media 

 or crenation in the media of higher concentration. The integu- 

 ment and gills are therefore impermeable. Loeb u - 16 has found 

 that Fnndnlus embryos will live in distilled water and in sea 

 water to which 5 per cent. Nad has been added. 



The view that the membranes of these fish are completely per- 

 meable (Brown) 25 is not tenable, at least concerning adult Fun- 

 dulits, as is shown by the following series of experiments which 

 were repeated often enough to assure the verity of the results. 



A large number of healthy specimens were selected and about 

 one half the body surface denuded of scales by gentle scraping 

 with the edge of a scalpel, or the skin was removed over an area 

 of one square centimeter on each side ; then they were divided 

 into three lots and placed respectively into fresh water, sea water 

 diluted with an equal volume of distilled water, and normal sea 

 water. Of those kept in fresh water in every experiment from 

 eighty to ninety per cent, died within twenty-four hours while all 

 died in less than thirty-six hours. In normal sea water the fish 

 suffered a similar fate although death did not intervene so soon. 



