KKFECTS Ul'ON FISHP^S OF CHAN(iES IN SALINITY OF WATER. 61 



winter HoundiT (Psiiijhijth'unmcctrK niinfi-'icdiniK) .survived coiisideratjly longer, as 

 appears from experiment Jr<S, while the tomcod {Micnitindux iomrnil), or at legist 

 some individuals, continued to live indefinitely after the chant^e. It is to \w remem- 

 bered, however, that the tomcod is an anadromous fish. 



Loeb (lltUO, p. ?y?A) has stated that - FutuJuhix can he thrown from sea water 

 into distilled water without any considerable swelling, or w ithout any visil)le injurious 

 etlccts.'' It is possible that Loeb would not regard death as a " visilile injurious 

 etlect." In four experiments with F. hiteroci if ux {the, species studinl by Locli) I 

 found that death resulted after an interval of from less than a day to ;! days. 

 Ten fishes at a time were put into se\eral liters of distilled water. The latter was 

 aerated in all cases but one, though this precaution was quite needless, .since even a 

 greater numbi^r of fishes may be kept in an eipial quantity of sea water which is 

 neither changed nor aerated. The fishes in my experiments came not from fiilb 

 strength sea water but from the supply system of the New York Aquarium (at tliis 

 time density = 1.015). Whether or not " swelling" was evident before dcatii 1 (bd 

 not determine, but the dead fishes were found in one i-ise to ha^e gained more than 

 lo per cent in weight. 



SYMPTOMS I'HECKDINU THE DEATH OF SAET-WATEK FISHES IN FKESH WATER. 



FiiikIiiIiix liiti'VocJ'duti^ as stated al)Ove, took little or no food when in fresli water, 

 sometimes tasting and then rejecting it, while those in the control tank crowded 

 eagerly to get it. Tiiey were ordinarily far more sluggish in their movements than 

 the ■ normal fishes, but at times, especially when disturbed, they were subject to 

 peculiar parox3sms. A fish so affected would start suddenly' upon an erratic course 

 around the tank, swinmiing in a zigzag or circular path and ending in a state of 

 tetanus, the body rigid and twitching slighth'. For a period the fish woidd remain 

 nearly or quite motionless, commonly somewhat flexed, the mouth open and the gill 

 covers widely extended. After a few minutes respiration was resumed and tlie fish 

 gradually assumed a normal aspect. Sometimes my approach to the taidc was 

 sufficient to provoke these paroxysms in one or more fishes. 



These same phenomena I have likewise observed very plainly in /'. diajihanm^ 

 and in one instance in F. inajdlix. Most marine lishes, however, die nuich more 

 rapidly than either of these, and (juite different symptoms are manifested. Scup 

 {Stenotoinim chrysops), when placed in fresh water, settle at once to the bottom, and 

 sooner or later "keel over," though this is generally deferred for some time. Res- 

 piration is at first nearly or quite normal, l)oth as to rate" and manner. Later it 

 becomes irregular, the normal rhythm being interrupted at intervals by spasmodic 

 gulps accompanied by general agitation. The tish from time to time darts violently 

 about the tank, as if in an effort to escape, sometimes even leaping from the water. 

 Tiiese frantic movements of the body recur at intervals, even after respiration has 

 ceased. The rate of respiration does not decline gradually, but the ihytlnnic move- 

 UKMits give place rather al)ruptly to convulsive ones, which soon ceas(> altogether. As 

 already stated, death occurs in from 1 to 3 hours after transfer to fresh water. 



te ot respiration being determined by means of a stop watch. These individ- 



