= = = - eae Ee 
SEA-FISHERIES LABORATORY. 1B RS: 
through the “ shoulders,” trunk and tail parts of each indivi- 
dual were made. These were placed in a weighed and numbered 
weighing bottle containing a small Soxhlet extraction thimble 
having a number written in pencil on the outside, and a small 
cotton-wool plug. In the cases of tinned fish a paste was made 
from parts of each fish, the latter bemg taken from the tin 
and placed on filter paper so as to absorb the oil or sauce 
adhering to their surfaces. The weight of the wet flesh was 
found by difference. The water was estimated by drying for 
at least 48 hours in a steam oven (temperature 98° C.). 
Absorption of the oil and water by the thick walls of the 
thimble materially aided the process of drying. 
Some general remarks on the water estimations may be 
made : | ; 
(1) The percentage in fish taken from the same shoals 
and at the same time varies with the length (and thus the age) 
of the sprats. Thus :— 
Devonshire Sprats. 
9-10 cms. long, 68°0 %; 13-14 cms., long 60°1 % of water. 
Sprats from Menai Straits. 
8-10 cms. long, 76°5 %; 10-12 cms. long, 74°6 % of water. 
Morecambe Sprats. 
6-8 cms. long, 75:0 9%: 8-10 cms. long, 72°8 %: 10-12 cms. long, 
71:2% of water. 
Whitebait, which are either herrings or sprats, and vary 
between 45 and 80 mm. in length, had from 78-4 to 80-4 % of 
water, although the fish were collected during the summer 
and autumn months, when the percentage of water should 
have been minimal. 
(2) The percentage of water plus the percentage of fat: is 
approximately constant, thus :— 
Devonshire Samples. 
Water and fat = 79°5 to 82°4, though water varies between 60°] 
and 67°1, and fat between 15°1 and 22°3 %,. 
