CHEMICAL COMPOSITION OF FOOD-FISHES. 
813 
The factor used for computing the quantities of flesh and its constit- 
. uents after dialysis was obtained by dividing the percentage of protein 
in the not dialyzed by that in the dialyzed flesh. In the James River 
specimen the figures were 10.634-8.79=1.21 (1.2093). The percentages 
in the dialyzed specimen multiplied by this factor gave the absolute 
quantities after dialysis corresponding to 100 before dialysis, in the 
case of the protein we have 8.79x1.2093=10.63, implying no change, 
in now, the protein, instead of remaining constant during dialysis, suf- 
fered loss, the correct figures for absolute gain and loss should show a 
loss of protein ; in other words, the factor, to be correct, should itself 
be smaller. The smaller factor would give smaller weights of flesh and 
flesh constituents after dialysis, which would reduce the computed gain 
of flesh and of water and would make the loss of water-free substance, 
protein, fats, extractives, and ash greater. That is to say, if the body 
lost nitrogen during dialysis the actual gain of total weight of flesh 
and the gain of water of flesh must have been less, and the loss of 
water-free substance, protein, fats, extractives, and ash must have been 
greater than the figures in the table imply. 
If, on the other hand, the body gained nitrogen during the dialysis 
the actual gain of total flesh and water must have been greater, and the 
actual loss of water- free substance and its constituents must have been 
less than the computations imply. It is extremely improbable that the 
oysters could materially increase their store of protein from any food 
they could get from the fresher water. The increase of protein in the 
liquids was apparently at the expense of that of the flesh, as above 
explained. On the whole, then, it seems reasonable to assume a very 
small loss of protein. Unless, therefore, there is a decided error in the 
assumptions made, the conclusion is unavoidable that the actual losses 
of water-free substance and of each of its ingredients could not have 
been less, and were probably somewhat more, than the figures in the 
table for absolute changes in dialysis represent. 
So far as I can learn, the change in weight of oysters in the ordinary 
process of floating is fully as large or larger than in the cases here re- 
ported. According to such data as I have been able to obtain, oysters 
ordinarily increase from one-eighth to one-fifth in bulk in “ floating,” 
the latter proportion being common with ‘‘good, fat oysters.” The 
increase in weight, which would correspond very nearly to the increase 
in volume, was computed in the above cases to be, in one case, one-fifth 
(21 per cent.), and iu the other, between one-seventh and one-eighth 
(13J per cent). That the oyster absorbs the more water the salter it is, 
is fully supported by all the experience of which I have accounts and 
accords entirely with what would be expected in dialysis. 
Without entering into the theory of osmose, which is abundantly dis- 
cussed in the text books, or speculating as to how much of the changes 
in the flesh was due to osmose and how much to secretion, it will suf- 
# 7 
flee here to briefly recapitulate the results observed. 
