130 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 
Concomitantly with this variation in mass of flesh per unit 
length (in the plaice) there appears to be a decrease of water 
in the flesh, and (in the herring) a decrease of water, an increase 
in oil, and a slight decrease in proteid. 
The temperature factor an integrative 
one. 
The sea-temperatures considered here are those recorded 
daily at Carnarvon Bay Light Vessel, which is placed at a station 
not very far from the herring fishery region, and is such that 
the sea-temperature is not affected to any considerable extent 
by “ accidental ”’ factors like tides, or the influence of the land : 
these data represent very well the periodic changes of tem- 
perature in the sea. The periodicity is, of course, mainly annual, 
but there are significant differences in the amplitude of this 
annual wave from year to year—another period of several 
years, in fact, being superposed on the annual one. The maxi- 
mum of sea-temperature for the period 1907-1916 occurred at 
the end of August and was about 14-6°C. The minimum 
occurred about the middle of March and was 7-6° C. 
The effect of sea-temperature upon the metabolism of the 
herring is, however, a cumulative one, that is, we must correlate 
the variations in fat-contents and time of spawning with the 
total quantity of heat received during the period of the matura- 
tion of the gonads. In other words, we must consider, not the 
slope of the sea-temperature curve but rather the integral of 
the curve between the limits represented by the beginning 
of the fishing season and the time of spawning. The tempera- 
ture function must be integrated. Now the function itself is 
a rather complex harmonic one, and the technical integration . 
is difficult, but we can approximate to its value by simply 
summing the 10-daily means of sea-temperature: we find 
>" 0, where t, and ft, are respectively, the beginning of March 
ta p Ns 8 8 
and the end of September, and @ is the mean sea-temperature 
