SEA-FISHERIES LABORATORY. 119 
which may, but usually does not represent the composition 
of the proteid constituents of the “ flesh.” Now, the quantity 
of mineral matter present in the “ flesh ” of fresh and preserved 
fish is rather variable: it is difficult to sample the muscle- 
substance of a Clupeoid without including some small bones 
and the hermetically-sealed sprats, herrings, &c., had to be 
sampled as they were likely to be eaten, for many people are 
not fastidious enough to remove the skin, backbone, &c., of 
the sardines that they consume: these products attained to 
a more important place in dietaries during the last five years 
than hors doewvre did formerly. The fish from tins were, 
therefore, made into a paste, bones, skin and all, and this paste 
was sampled. Again, a very variable percentage of salt is added 
in the course of preservation in different ways, thus kippers 
may contain about 3-5%, bloaters up to about 10%, pickled 
herrings from 6 to 12%, and red herrings as much as 12%. 
Hermetically sealed sprats, herrmgs and pilchards contain 
from about 3 to 5°% (though sometimes the proportion was 
found surprisingly low). _Brine-frozen fish, which were supposed 
to be “ fresh,” may have had from 3 to 9% of salt in the tissues. 
Quite fresh, untreated herring and sprat “ flesh’ contained, 
as a rule, about 1-5°% of non-volatile mineral matter, some of 
this being certainly derived from the bones. 
The fat being extracted, and the mineral matter being 
estimated by ignition, the dry, fat-free, ash-free substance 
present in 0-500 grm. of the dry residues was then calculated 
and the nitrogen found by the Kjeldahl process was expressed 
as a percentage of the dry, fat-free, ash-free substance. This 
is the ““N ” of the detailed tables of analyses and its values 
are summarised in the following statement, where some results 
taken from last year’s Report are included. 
