324 



DISCOVERY REPORTS 



I-200-I 



HOO- 



LU 

 <S) 

 < 



LU 



•000- 



line) joining the points has no real meaning. It is useful as a guide to the eye, and is a freehand 

 approximation to the curve expected, judging by the way this relationship varies in other fishes for 

 which we have better data. It is quite clear, however, that whatever curve were fitted it would show a 

 point of inflexion between a length of 25 and 30 cm. It is therefore probable that these fish first 

 attain sexual maturity at about that length. 



The seasonal variation in ponderal index of Salilota more than 25 cm. long, for the period December 

 to the end of March 1932, is shown in Fig. 34. This shows a steady rise, steeper latterly. We have 

 just seen that most of these fishes are probably mature, 

 so that we have here a strong suggestion that spawning 

 takes place in late spring or early summer. 



Several series of length frequencies of this species gave 

 strong modal indications of the younger year classes. 

 The percentage frequencies of the pooled results, in the 

 southern and intermediate regions, over stated periods, 

 are shown in Fig. 35. The implications of these are fairly 

 clear: the autumn results are most helpful, for we were 

 then fortunate enough to capture sufficient of the young 

 fry (about 5 cm. long) in the accessory nets attached to 

 the back of the trawl, for these to appear as a mode of 

 equal strength to that formed by the (presumably) I-group 

 fish at 16 cm. This indicates a growth of some 11 cm. 

 during the first year of life, and there is also just a hint 

 of a possible submode at 26 cm. suggesting 10 cm. as the 

 second annual increment. The main mode of the January 

 figures, and of the winter (June-July) figures is clearly 

 due to I-group fish as represented by the 16 cm. mode 

 in autumn. It will be seen that the shift to the right of 

 6 cm. of the I-group mode between midsummer and 

 midwinter is compatible with the growth-rate suggested 

 above, for growth is usually more rapid during the second 

 half of the year in fishes living in cold or temperate lati- 

 tudes. In conjunction with the evidence afforded by the ponderal indices we may therefore say that 

 S. australis is a fish of rapid growth, probably reaching maturity in the third year of its life. It may 

 be well to state that rapid growth rates are known in other small Gadidae, notably in Gadus merlangus 

 (Hartley, 1940, p. 48, where an even higher growth rate than that postulated for Salilota is explained 

 by the absence of a winter fast in young fish frequenting estuaries). 



No time could be spared for investigation of the feeding of S. australis. One was observed to have 

 eaten large isopods, and it may be permissible to guess (from its colour pattern) that it is more of a 

 bottom feeder than Micromesistius. 



Physiculus marginatus (Giinther). This small species was taken at the following trawling stations, 

 all in our southern region : 



0900- 





DECEMBER JANUARY FEBRUARY MARCH 



Fig. 34. Seasonal variation of ponderal index observed 

 in Salilota australis, 1932. 



WSys 

 WS817A 



18 of 5-7-3 cm. 

 I of 18 cm. 



WS820 

 WS821A 



1 of 16 cm. 



2 of 16-3 and 17 cm. 



Norman (1937, p. 54) gives 22-5 cm. as the length of the largest specimen known to him, and 

 southern Chile and Magellan Straits as other known localities. As far as we know, therefore, the 

 species is too small and too scarce to be of practical use to man. 



