EFi'ECTS UPON FISHES OF CHANGES IN SALINITY OF WATER. 65 



Euuene Smith (1902) says of this species that it "can be ai-custouied to the [fresh 

 water] aquarium bj' gradual steps," but he does not state how gradual these steps 

 must V>e. 



E.vperiiiunt 2^. 



Woods Hole, August and September, 1904. Three ('i/2>ruio(.lo)i r((rle</iitu!< 



treated as in preceding case. Two died 3 days and one -1 days after reaching 



fresh water. (Compare experiment Ui.) 



Krp,riiiumt '26. 



Woods Hole, August and September, 190-i. Thirteen cunners {Tautvgu- 



Jahrux (uhpermis) treated as were the fishes in the two preceding experiments. 



All died within 12 hours after reaching fresh water though apparently healthy 



till this occurred. 



E.vperiineiit 27. 



AVoods Hole, August and September, 1904. Nine black-fish (Tawtogfao^i/^/.v) 



treated as were fishes in experiments 24, 2.>, and 2(1, except that an additional 



step was interposed just before the change to fresh water, which was thus 



reached somewhat less abruptly. Nearly all of these fish died within 12 hours 



and all died within 24 hours after reaching fresh water. Up to this time they 



aj)[)e:intl to be in perfect health. 

 The foregoing experiments prove, if an3-tliing, that fresh water itself is fatal to 

 the fishes under consideration, and that the degree of abruptness with which the 

 change is made is of secondary importance. There is evidence, nevertheless (experi- 

 ments 21 to 25), that a certain modification may result wherebj^ the fatal effect of the 

 iinal change is deferred. A partial acclimatization of a somewhat different character 

 is reported by Bert (1871), who succeeded in accustoming various fresh-water fishes 

 to water of half the density of that of the sea, though abrupt change to this proved 

 fatal. De Varigny (1892), likewise accustomed young eels to water containing five- 

 tenths per cent sodium chloride, though the additions of salt had to be ver_v gradual. 



ENDIKANCK IX WATEK OF LOW SALIMTY. 



It is plain from certain of the preceding experiments that some fishes which die 

 (|uickly in pure fresh water will endure, for short periods at least, water of a very 

 low degree of salinity. Special experiments were made with a view to testing thi,> 

 point. 



Evprrlmeiit 28. 



Woods Hole, September, 19o4. vSeventy-five specimens of F. JieferocUtu.t, 

 from same lot of fishes as those used in experiment 10, were changed from 

 salt water (density 1.023)" abruptly to water of density 1.004. No deaths in 

 11 days, after which experiment was discontinued. Compare with experi- 

 ment 10, for which this serves as a good control. 

 Eqx'rhjwnt 29. 



Woods Hole, September, 1904. Twenty-five E. inujaliv subjected to same 

 treatment and with same results. 



a Unless otherwise specified the density here given has been corrected for temperature— i. e., the specific gravity at 

 4° C. is given. 



B.B.r.l905— 5 



