of the Fishery Board for Scotland. 



159 



parts ripe and measured 22.j inches. On the other hand, among those 

 examined during the spawning season he found females immature as 

 large as 36 inches.* 



IV. — The Influence of Temperature on the Growth of Fishes. 



In one of my previous papers dealing with the growth of fishes I 

 referred to the important influence which the temperature of the water 

 exercises over the rate of growth, and gave examples from the observa- 

 tions made on certain species, and particularly the plaice in its younger 

 stages when inhabiting the sandy beaches. f It was shown also that the 

 haddock and whiting and other forms increase in length much more 

 rapidly in summer than they do in winter, but from the want of a series 

 of periodic observations on the temperature of the deeper offshore water 

 in the various months throughout the year, it is not yet possible to bring 

 the observations on growth into exact relation with the temperature 

 variations in the water. 



It appeared to me that some results of interest might be obtained 

 directly by keeping fishes in the winter in water which was artificially 

 heated, and comparing their growth with other fishes kept under similar 

 conditions but in water at the ordinary temperatures. This has been 

 done for over five months with the results described below. 



Four different lots of fish were kept in separate and similar tanks, 

 which may be distinguished as No. 1, No. 2, No. 3, No. 4. Each of 

 the tanks is of concrete with the front and back of plate glass, and the 

 light from windows in the tank-house passes through them, but not very 

 strongly. The tanks are of uniform dimensions, measuring 5| feet in 

 length by 4 feet 4 inches from back to front, and the depth of water 

 during the experiments in Nos. 2, 3, and 4 was 25 \ inches ; the volume 

 of water in each of these tanks was therefore about 315*5 gallons, or 1433 

 litres. The other tank, No. 1, owing to a defect could not be filled so 

 full, and in it the depth of water was 15 inches, the volume being thus 

 about 185*5 gallons, or 842*5 litres. Tanks Nos. 1 and 2 were supplied 

 from the ordinary supply pipe to the tank-house, the water thus having 

 approximately the same temperature as the sea water on the beach. 

 Tanks Nos. 3 and 4 were supplied with sea water from the same pipe, 

 but it was first passed through a heating arrangement by which its 

 temperature was raised. Considerable difficulty was at first experienced 

 in raising the temperature of the water in these tanks sufficiently high. 

 It was soon discovered that the use of oil heaters was insufficient, and the 

 method adopted was to utilise an ordinary slow-combustion stove for heat- 

 ing the apartment, upon the top of which was placed a common galvanised 

 iron hot-water tank, such as are used for supplying hot water, of forty 

 gallons capacity, and around it was placed an iron jacket with a space 

 between in which was enclosed the smoke pipe from the stove. This 

 arrangement has answered very well and with comparatively little atten- 

 tion or extra cost. 



No arrangement was employed for the mechanical regulation of the 

 temperature, which varied considerably from time to time, as shown in 

 the tables, falling generally during the night ; but a little experience in 

 firing enabled the variation to be to some extent controlled. The tem- 

 perature was taken every few hours daily, and the supply of hot or cold 

 water regulated accordingly, and maximum and minimum thermometers 

 were also used in order to ascertain the extreme nightly range. 



*Journ. Mar. Biol. Assoc. III., Special Number p. 377, 380, Ibid. III., No. 1, 79. 

 t Twentieth Ann. Rep., Pt. III., pp. 335, 342. 



