296 DISCOVERY REPORTS 



The anomaly among the males mentioned above is almost certainly due to inclusion of an ex- 

 ceptionally large proportion of spawners of the smallest length class at one of the rich March hauls, 

 with inadequate sampling in the earlier periods. The curious anomaly of 31-40 cm. females showing 

 recovery by March, while smaller and larger fish did not, is probably due to inadequate sampling also. 

 As it stands one would think it might indicate that some of this small group of mainly immature fish 

 completed their cycle before their older sisters, and we know that the reverse is more generally true. 

 However, there were indications of early shoreward movement among females of that length class in 

 193 1-2, derived quite independently from this data, and it might be that our samples of it happened 

 to include a majority of some year class which had got out of phase with the norm for the species, 

 through exceptional conditions affecting them while still younger, or some such cause. In fishes with 

 such a prolonged spawning season as hake non-conformity of this type is bound to occur from time 

 to time, but to prove the point is quite another matter, and quite impossible with such limited data 

 as ours. 



The mere fact that mainly immature length classes showed a seasonal variation parallel to that for 

 older fish need occasion no surprise, for Hickling (19356) has established that a seasonal cycle fore- 

 shadowing the full sexual cycle of the adult takes place in immature European hake. The point is that 

 the adult cycle should show the greater range. Here I believe that it is the limitation of our data in 

 time that is at fault, and not insufficient sampling. 



We have considerable evidence that larger females of M. hubbsi tend to move inshore earlier than 

 smaller ones, and it is probable that they are normally the first to spawn (a feature well known among 

 European hake). Now a generalized cycle such as the curve for females in Fig. 24 would only hold 

 strictly for fish of the mean or 'average' length considered, the 41-50 cm. length class. Had we been 

 able to obtain adequate figures for the whole year, it is probable that the whole cycle for the larger 

 fish would have been found to be centred earlier in the year. If so, means for the end of October are 

 not early enough to show the full extent of the annual variation in condition of these larger fish. 



It is also possible that the anomaly shown by the females is not entirely due to the unavoidable 

 limitations of our data, although I think there is little doubt that it was collected too late in the year 

 to show up the peak period for the 51-60-70 cm. fish. In the first place, some M. hiibbsi under 40 cm. 

 long are mature. This would tend to increase the range of seasonal variation in mean values for the 

 smaller length classes far above the range observed in European hake of similar length. In European 

 hake, an altogether larger species, mature fish of such small size have only rarely been found. These 

 were among a peculiar localized stock in the Clyde basin (Hickling, 1930^', pp. 52, 53 and Table Via). 

 A further likely source of discrepancy is that the sexual activity of the largest size groups may be 

 reduced, or at any rate less regular, than in younger mature females. In European hake a reduction of 

 sexual activity in the largest length groups, associated with reduced growth rate, was postulated by 

 Belloc (1922, p. 40). Some concrete evidence in support of this view, relating to female hake of 

 90-100 cm., is given by Hickling (19306, p. 29). 



The condition factors of a few very large female M. hubbsi (over 90 cm. long) are extremely in- 

 teresting. Unfortunately we only caught fish of this size in the southern region, so that the results 

 are not strictly comparable with those previously tabulated ; but it can be seen that all but the smallest 

 of these very large fish showed very high values for i^ at a time of year when all the other females 

 showed reduced values owing to spawning. These data are insufficient to be conclusive, but I believe 

 that the oldest and largest female hake have reached a state of suspended sexual activity, and no 

 longer show the seasonal variation in condition characteristic of younger fish (Table 15). 



