302 DISCOVERY REPORTS 



In December there were only six stations suitable for this comparison, but at three of them additional 

 4-hr. hauls were made, so that there is a considerable body of data over the most critical part of the 

 distance range. The whole series of observations was completed between 13 and 16 December 1931, 

 so that there is little chance of the comparison being vitiated by the time factor. 



Table 20. December data bearing on hake movements in the northern region 





I 



The results are summarized in Table 20, and show that the greatest concentration was at St. WS790, 

 85 miles offshore. The length data for males are again unsatisfactory because of the wide dispersion 

 and smallness of the samples containing the larger fish. I suspect that the males were still widely 

 dispersed, as is suggested by the low sex ratios at the richer stations. Here the female length data are 

 very interesting, the mean length at the peak station being considerably higher than at the richest 

 station of the earlier series, although 45 miles farther inshore. Thus 



Difference ^=7-2 cm 



Mean length ?, WS790 A + 6 = 51-2, 



Mean length ?, WS771 =44-0, 



\/i-^i()= 1-19 

 6-05. Strongly significant. 



0-4672, 



N' 



^=0-9488. 



od=\/{o-i\.6']2 + 0-9488) -- 

 d 7-2 



Od 1-19 



This agrees well with the view that larger females may catch up and pass the smaller ones on their 

 way inshore. But, considering the December figures alone, we find that at St. WS791 A + B the 

 females were significantly smaller than those on either side of them ; the relation of length to distance 

 offshore is discontinuous. The detailed length-frequency distributions showed that this was due to 

 a much higher proportion of immature fish, especially of about 30-32 cm. length, at St. WS791. It 

 would seem that large females are definitely heading the shoreward movement at this time, a few 

 having penetrated right inshore among the juveniles, which are perhaps almost non-migratory. Com- 

 parison with the October-November results suggests that in the interim (about 7 weeks, taking mean 

 dates) the large fish have travelled shorewards some 100 miles, while the smaller fish advanced some 

 15 miles only. 



In March also there were only six stations providing comparable data from the northern region, 

 but fortunately five of these yielded rich hauls. The remaining station, WS860, presents some 

 anomalies that spoil an agreement with our theory that is otherwise complete ; for a small number of 

 large hake were found less than 100 miles from the mainland, where a good haul of small hake should 

 have been taken. It is probable, however, that the net did not fish properly at this station. A note in 

 Gunther's hand in the original rough log reads : ' Haul disappointingly small. A few hake escaped. 

 Those present of large size— majority caught in after wing which opens suspicion that net may have 



