82 DEPARTMENT OF THE NATAL SERVICE 



counting these layers, to arrive at the age of the specimen. This theory was formu- 

 lated in 1759 by a Swede, Pastor Hederstrom, in the following manner : " Anyone 

 taking the trouble to examine a vertebra from a boiled fish, will observe certain rings 

 thereon. And as many rings as there may be, so many years will be the age of the 

 fish." 



This suggestion, however, was afterwards lost sight of, and it was not until about 

 1900 that the skeletal parts of fish were called into requisition for the purpose of age 

 determinations. The same observation was then revived by two scientists, Hoffbauer 

 (VIII) and Reibisch (XIII) and the question has since been further dealt with by 

 many others.^. 



It has been found that the different bony parts are not all equally well suited for 

 the purpose of age determinations ; in the case of the plaice, for instance, the scales 

 are difiicult to deal with from this point of view, being small, and with indistinct ring 

 formation. In this species, the otoliths and the opercular bones furnish the best 

 means of ascertaining the age. In the case of the herring, the opposite has been found 

 to be the case, the otoliths are here small, and awkward in shape, whereas the scales 

 are large, easily collected, and distinctly ringed. Although the otoliths and vertebra? 

 of the herring can be, and have been, used for age determination, the scales offer so 

 many advantages that there can, in my opinion, be no question as to choice. In the 

 following pages, the structure of these scales will be briefly described, with special 

 reference to the pattern of the annual rings. A resume will likewise be given of the 

 exact facts adducible in proof that the ring formation actually does consist of annual 

 rings, and that the scales may therefore be employed for the purpose of age deter- 

 mination. Finally, mention will be made of the methods of preparation and investiga- 

 tion which the writer's experience has shown to be most convenient. 



III.— STRUCTURE OF HERRING SCALES. 



On examining a number of herring scales, it will be found that these are thin, 

 pliable plates, differing both in size and shape according to the position on the body 

 of the fish. The scales from the forepart of the body are larger than those from the 

 caudal region, and those on the back smaller than those situate farther down, near 

 the lateral line, etc. The shape, or outer contour, differs between scales set, for 

 instance, immediately behind the gill cover and those farther back. Most scales are, 

 however, more or less, of the shape, shown in fig. 1, which is reproduced from a photo- 

 graph of a scale taken from near the lateral line, almost straight above the pectoral 

 fin. A feature common to almost all scales is the fact that the anterior portion, which 

 lies embedded in the scale pocket, presents an entirely different appearance to that 

 of the posterior portion. The former is of even contour, and appears to be finely 

 striped{, whereas the latter has a fringed contour, and lacks the fine striped pattern. 

 The two portions are divided by a line, termed the basal line (h — h in plate I, fig. 1). 

 The stripes of the anterior portion appear to fall into two distinct systems, with a 

 zigzag boundary between, this being as a rule ahr.ost perpendicular to the basal line 

 (z — c in plate I, fig. 1). At the point of intersection between this zigzag line and the 

 tasal line lies the centre of the scale (c) ; this is in most cases distinctly marked, but 

 may scarcely be seen on this figure. In addition to the fine stripes, a number of very 

 pronounced lines are seen running from the margin some distance in (r) ; these are 

 called the radial lines, from the fact that in the scales of many species of fish, the 

 corresponding lines radiate out from the centre of the scale. In plate I, fig. 1 will 

 be seen eight dark narrow lines, arranged concentrically about the middle of the 



lA list of the most important works on tliis subject will be found in Dahl (I) The assess- 

 ment of ane ajid growth in fish, Intern. Revue d. gesamt. Hydrobiologie u. Hydrographie 1909. 

 Bd. 11. 



