EFFECTS UPON FISHES OF CHANGES IN SALINITY OF WATER. 103 



The testimony of a number of investigators seems, then, to be in full accord 

 upon certain main points, which may be provisionally accepted as proved. Fred- 

 ericq (190i) has classified the three sorts of aquatic organisms, relative to osmotic 

 conditions, as follows: (1) Molecular concentration and salt content both approxi- 

 mately the same for the [vascular fluids of] animal as for the surrounding water 

 (marine inveitelirates); (2) molecular concentration the same, but proportion of salts 

 less — the deficiency being compensated for by organic matters in solution — (elasmo- 

 1)ritnchs); (8) both molecular concentration and salt content very different from those 

 of external water (teleosts, both marine and fresh water; fresh-water invertebrates). 



It is with the third class that we are especially concerned in the present paper. 

 The molecular concentration (hence the osmotic pressure) and the salt content are 

 both verv different from those of the surrounding- medium. It seems never to have 

 been fully appreciated that there is even hei'e a certain correlation between the inner 

 and the outer fluids, both as regards osmotic pressure and salt content. But reference 

 to the various cryoscopic determinations shows that not all teleosts have blood of the 

 same osmotic pressure. Rodier found the latter to range between z?=— 0.62'-' and 

 /J = —0.80^ in LopMu>i alone. What is more significant is that the blood of fresh- 

 water fishes has been found to possess, on the average, a consideralily lower osmotic 

 pressure than that of marine fishes. The mean figure given by Fredericq (1904) for 

 two marine teleosts is /:/= — 0.80°, that for three fresh- water species is about — 0..53°. 

 Of course in the case of these fresh-water fishes, the osmotic pressure of the blood, 

 though lower than that of marine fishes, is nevertheless very much higher than that 

 of the fresh water in which they live. Satisfactory determinations of both fresh 

 and salt water individuals in the case of species inhabiting both have not, so far as I 

 know, been made. As noted above (p. 96), Balland found that the flesh of eels from 

 salt water contained a considerably greater percentage of ash than those from fresh 

 water. Similar difl'erences in chlorine content were obtained by myself in the case of 

 several species, and in general it has been shown that the percentage of salts is greater 

 in salt-water fishes than in fresh-water ones. It must be borne in mind, however, 

 that these determinations of salts were made for the flesh of the fishes, while those 

 of osmotic pressure were made upon the blood. So far as I know the only recorded 

 comparison of the salt content of the l)lood of fresh and salt water fishes was made 

 by Quinton (p. 95). A considerably higher percentage was indicated for the marine 

 forms. If the figures of Quinton are correct, it is likewise to be noted that the per- 

 centage of chlorides in the blood is several times a- uii'at as in the flesh. Of course 

 the osmotic pressure of neither is entirely ilrpindiiit \ipon the percentage of salts, 

 but maj' depend upon organic matters in solution, as was found in the case of elasmo- 

 branch blood. Fredericq (1901) has attempted to determine the osmotic pressure of 

 the solutions contained in the various tissues of fishes and invertebrates, partly Ijy 

 extracting the soluble ingredients by boiling, partly by noting the changes of weight 

 in strips of tissue suspended in solutions of varying concentration. Reasons for 

 doubting the value of the latter method have already been given (p. 81). 



Where a correlation is found to exist between the osmotic pressure of the l)ody 

 fluids of an animal and that of the surrounding water, the question arises. How is 

 this correlation maintained^ For various invertebrates, it seems to have been gen- 



