yt5 BrLLETIN OF THE BrREAT' OF FISHERIES. 



Numerous aulliors have detcriiiinc<l the i.crccnta.uvs of ■•ash" in the flesh i.| 

 fishes, thouu'h luiiif, so tar as I know. ha\c had in \ irw a lonipai-ison hctwccnihi 

 fre.sh and the .salt water spceies. 



The average percentage of ash in 7 fr(>sh-water" s])c(ies analyzed Ity Atuati r 

 was 1.30; for 11 salt-water species. 1.11. 



From the determination-; of Payen and Kostjtscheff (cited liy Atwater) it. mav 

 likewis- be gathered that the averages for the .salt-v/ater tislies are suhstantiallv 

 greater than those for the fresh- water ones. 



The figures of Balland (189S) and of Milone (cited by Lichtenfeid. l'.to4) al^) 

 give average percentages of ash which are considerably greater for >alt-\vater than 

 for fresh-water species. According to Balland, the hgure for eels from fresh water 

 is 0.76, that for salt-water .specimens being 0.S7. 



An exhaustive search of the literature of this subject would perhaj)s reveal Hl:- 

 ures which would not harmonize with the general trend indicated above. But. -o 

 far as I have ))een able to learn, one important conclusion may l>e drawn from the 

 records as a whole, despite great inconsistencies — namely, that both the prop<)rti<in I't 

 ash in general, and the proportion of chlorides in partieidar. are, on tlie axcraL^i . 

 greater in salt-water fishes than in fresh-water ones. The ratio Ijetween the two can 

 not, of course, be stated with any exactness. For such a ratio to have any value it 

 would be necessity that each of the mean percentages should be based upon a veiy 

 large number of exact determinations. Leaving out of account the figures of Katz. 

 however, the mean percentage (either of chloi-ine or of ash) for the salt-water fishe-- 

 is in everj^ case less than twice that for the fresh-water ones. In most cases it is 

 nuich le.ss. On the other hand, the ratio between the amount of chlorine found in 

 sea water and that in ordinary fresh water is as several thousand to one. The fol- 

 lowing table gives the percentages of chlorine found in several of the grades of 

 water used by me in the experiments: 



New York City (Croton) water 0.0003 



Woods Hole { Fahnouth) water 0011- 



Daggetts Pond ( Marthas Vineyard )>' 00.51 



Water giving salinoineter reading about 1.001 0.")S4 



Water giving salinometer reading about 1.002 ll'.52 



Local sea water (1.023^ l.sl.so 



It is thus seen that the water of specific gravity l.oo:.' (e(iui\alent to that of 

 Ta.shmoo Pond during the present summer) contains a percentage of chlorine not 

 very far ditferent from that of the fishes (7^ di((pfii(nu.'< iuid the white perch, experi- 

 ments 104 and 108) living in it and, indeed, of the same order of magnitude as that 

 of anv of the teleosts analyzed by me. That such water can not e\en he approx- 

 imately isotonic with the body fluids of these fishes seems evident from the cryoscopie 

 determinations of other investigators.'' Nevertheless the approximate eciuality 

 between the percentage of chlorides in this slightly brackish water and in the tissues 



« I liave counted the sliad and salnnwi mnoni,' tiu- fre.sh-water fishes, the smelt among the salt-water ones. 

 ^From which cume the « liiie p--n!i n^rd m i-xpirinient 110. 



cThe figures obtained by inc d.i n..t <.i . .mr-r indi.-ate the proportions of i-hlorine in the fluids, but rather those for 

 the body as a whole. It is likewis.- ii. !..■ niiunilrird thnt the osmotic pressure of the body fluids is in part determined 



