EFFECTS ITPUN FISHES OF OHANGES IN SALINITY OF WATER. 59 



of salt wiitiT entirely prevented the li:iniitiil eticctts. No trace of fmious was to he 

 obsei-\i'd and nu surface lesions of any sort were diseoveral)le. Lack of some or all 

 of the saline ingredients of sea water seems to have been responsible for the death 

 of the iishes. Perliaps the most curious feature in these results is the prolonged 

 period of resistance shown by some of the tishes in each experiment. This suoyests 

 cei-tain of the I'esults of Bert (see p. 67). 



Some observations made in New Yoriv City during the past winter and spring 

 are interesting by way of comparison with the preceding results. In these later 

 experiments the tishes came from marshes of bi'ackish water, and were kept for some 

 days or weeks prior to the experiments in water of a salinity slighth' over half that 

 of water from the open sea. The temperature at the time ranged from 2 to Itl C, 

 that at Woods Hole having ranged from 20- to 22 ' C. These experiments were 

 complicated by the appearance of fungus {Sai>roJegnia), which invariably devastated 

 the tanks within a week or two after F. ht'ti-rcclitas was placed in fresh water. 

 Even here, however, it was noted that a certain proportion of tliose which died 

 showed no traces of Sitjirahyniii. 

 E.rp,rn,i,n1 f,. 



New York, November and December, 1904. Two hundred and eleven F. 

 hrfi^fdcl/tiix transferred to fresh water. Thii'teeu died within the first 15 da^ys 

 before the disease appeai'ed; 82 per cent of the remainder died during the 

 next 16 days, after which the experiment was discontinued. 

 A certain proportion of these tishes died without being atiected by the fungus, 

 but it is of course impossible to determine what the death rate would have been if 

 the Sd.prolegnia had not appeared. It nmst be i-ememl)ered that fishes which are 

 already weakened by other causes are more ready victims to the fungus. It was 

 found in experiment 41, c, and others that the addition of even a small percentage of 

 sea water served as a decided check to the appearance of the parasite. It is my 

 belief that this preventive action of the sea salt was due not so much to a toxic etlect 

 upon the fungus as to a beneficial physiological effect upon the tislies. resulting in a 

 greater power to resist the parasite. 

 F.rperhnent H-. 



It was here observed that the fungus, when once it had gained a footiiold upon 



the fishes, continued to thrive after the latter were transferred to water of 



density 1.005, while few traces appeared upon tisiies which were trruisferred 



in a healthy condition to water of this density. 



In some cases (experiment 41, r) water of a nuich lower degree of salinity acted 



as an efl'ective check. 



It is evident from experiment 13 (which was rejjeated with similar results) that, 

 under the conditions existing at the time, fresh water proveii nnuii less rapidly fatal 

 to Fnididus heterocUtKs than in the Woods Hole experiments, and this in spite of 

 the fungus. The factor responsible for this diffei'ence may have been either (1) the 

 lower temperature, or (2) the previous history of the fish (life in diluted sea water), 

 or perhaps the two combined. It nevertheless seems (juite i)robabie, although no 

 definite control experiments were made, that e\ en here the fresh water had a dele- 

 terious efi'ect. 



