126 DEPARTMENT OF THE NATAL SERVICE 



d. From Cape North southwards along the Atlantic coast of Canada. 



The material from these waters embraces seven samples of mature herring, and 

 four of immature fish, covering a range from Sydney, Cape Breton, to Gloucester, 

 Mass. In going- through the scale preparations from these samples, it was strikingly 

 noticeable that the scales were considerably more difficult to read than those of the 

 specimens from the gulf of St. Lawrence. The first and second winter rings were often 

 found to be faint, thus rendering less distinct the contrast between the true winter 

 rings and the secondary rings, which latter were of frequent occurrence in these 

 scales. The rings beyond these may be said to be fairly distinct, but the outermost 

 ones, on old scales, were again difficult to discern owing to the fact that the fine ridges 

 on the surface of the scales were often irregularly developed, a symptom of senility 

 in herrings. Another feature, which will be more important in connection with the 

 growth investigations, was the frequent irregularity in the proportion between longitu- 

 dinal and lateral growth of the scales. I have made no measurements in this respect, 

 but it could often be seen that the scales had during the last few years grown more in 

 breath than in length, thus rendering the distance between the winter rings relatively 

 greater, at the two sides of the scale, than in the anterior portion. 



The pecularities here noted in the scales of herring from these Atlantic waters 

 tend to give the investigator an impression that the fish in question must have lived 

 and grown up under conditions widely different from those which obtain among the 

 herring in the gulf of St. Lawrence; true it is no easy matter to tabulate and sys- 

 tematize such features, as distinctness of winter rings or frequency of secondary 

 rings, the appearance actually presented is, however, none the less remarkable on 

 examination of the scales. In estimates (and growth measurements) this means 

 increased labour, and greater uncertainty in the results obtained. 



With regard to these difficulties, there appears to be some difference between the 

 samples from the northern parts of the waters in question, and those from Nova 

 Scotia, the scales of herring from the more southerly localities being less easy to read 

 than the others. The material is too limited, however, to i>ermit of any definite 

 statement in this respect, as much depends upon the quality of the scale preparations. 



The difficulties are most felt in dealing with older fish, for the younger specimens, 

 up to 8 years old, it may safely be said, that a fairly accurate age determination can 

 be arrived at, with continued investigations, also, and by great care to make the scale 

 preparations as perfect as possible, a similar degree of accuracy may be attained in 

 the case of the older fish. Fortunately as matters stand, the material from Nova Scotia 

 includes only one sample with many old fish; in the others, young specimens pre- 

 dominate. 



As regards the northern samples (North Sydney, Main-a-Dieu and Grand Narrows) 

 we have the further disadvantage that they were taken during May and June, i. e., at 

 a time when the herring in these waters may be expected to be commencing their 

 summer growth. And, as we here lack the support afforded by study of the quite young 

 fish, and as the samples themselves do not form any series in point of time, it is diffi- 

 cult to assign the specimens to definite year-classes. In table 9, therefore, where the 

 results of age-determinations for these three northern samples are shown, the speci- 

 mens have only been arranged in the age-group, or as one might say, " summer-zone 

 groups ". As far as it is possible to judge, however, the great majority of the speci- 

 mens in these samples have not yet begun their summer growth. 



