More accurate conclusions will be reached after the above -described 

 experiments have been repeated at another suitable opportunity using a 

 thermocouple „ — Accordingly it can be presumed with a fair degree of 

 certainty that fishes, or at least sardines^ do not have a heat or cold 

 center. In other words, we can affirm that it is actually impossible for 

 sardines to adjust their body temperature to a level which differs from 

 the temperature of the envirorjner^ 



b) Resistance to high and low water temperatures 



The experiments described in the preceding section have shown how 

 the surrounding water temperature controls the body temperature of the 

 sardine s but the sardine is a living creature and therefore has its own 

 particular limits of water temperature within wnich it can exist, The 

 author performed the following experiments to find these limits and the 

 resistance of bait fish to sudden changes in water temperature. 



The experimental tanks, as shown in Figure 8, were two wooden boxes 

 100 cm long by 60 cm wide by 60 cm deep. One was used for the experiment 

 and the other for the control. Both had an opening for the water supply 

 pipe in the middle lower portion, and a drain 10 cm from the upper edge. 

 Mercury thermometers were inserted at two opposite corners, One of 

 these tanks was connected by a siphon to a reservoir through a temperature 

 regulator which could cool or heat the sea water to the proper temperature. 

 The control tank was connected directly with the reservoir without passing 

 through the regulator. 



With such equipment a suitable supply of sea water was constantly 

 drawn from the reservoir and supplied to both tanks. At the same time 

 20 each of sardines and anchovies were transferred as speedily as possible 

 from the net live=pound. The resistance to high and low temperatures was 

 tested three times at each temperature with the following results, 



1) Resistance to high temperatures 



To put it in different words, this experiment was to find out how 

 much of a sudden rise in water temperature is required to kill bait fish 

 which are living at a normal temperature, According to Table 14, when 

 sardines which at the time of the experiment were living in sea water of 

 about 22°C were transferred directly into an experimental tank having a 

 water temperature of 28°C there was only 10 = 15$ mortality in the course 

 of three hours. Table 15, however^, shows that wnen sardines living under 

 almost the same conditions were transferred directly to an experimental 

 tank having a water temperature of 30°Cj, there was within 1 hour and 30 

 minutes a mortality of 65 /"sic. Should be 95, J - 100$, a result almost 

 equivalent to total destruction. 



With anchovies which were taken at the same time as the above-mentioned 

 sardines and which after capture were kept in the same net live-pound 



21 



