406 Be ABN: 
locality and as nearly as possible at the same time, measuring the length 
of each individual fish and then plotting the measurements in the form 
of a graph or curve. 
The following curve [cf. Hjort and Petersen (11), Plate IV] in which 
8046 Cod, caught off the East Coast of Iceland in July, 1904, are graphi- 
cally represented (Fig. 2), illustrates the method. Each fish was measured 
and the numbers found at each centimetre are plotted. It will be seen 
at once that the fish group themselves around certain definite lengths. 
The smallest sized fish in the collection was 3 cm. long, and of this size 
there were 6 fish. At 4 cm. there were 65, and at 5cem. 189. The num- 
bers then begin to fall, there being only 43 at 6 cm., 8 at 7 em., and but 
one fish at 8em. Then they rise again until another maximum occurs 
at 12. em., with 139 fish. At 17cm. the number has fallen to 12, after 
which another rise occurs until at 22 cm. we have 107. In this way the 
fish fall naturally into the six groups O-V. This division into size groups 
is due to the fact that the spawning season of the Cod in each year is 
a limited one, extending over only two or three months at the beginning 
of the vear—January, February and perhaps March. By July the fish 
born at that time will have reached about 5.cm. in length. The next 
maximum at 12 cm. represents the fishes born the year previously, and 
the difference between 12 cm. and 5 cm., 1.e. 7 cm., expresses the growth 
in length during the year. Then at 22 cm. we have the fish which have 
completed two full years, these fish in July when the samples were taken 
being two and a half years old. 
Since all the fish were caught in 1904 we are able to say in what year 
each group was born, the V group, with a maximum frequency at 88 cm., 
being Cod born in 1899. 
Now, although this method is satisfactory for the early years, the 
distinction between the groups becomes much less marked as the fish 
grow older, until finally the different year-groups run into one another 
and become indistinguishable. Fortunately, however, other methods 
have been discovered which enable us to attack the problem of age and 
rate of growth with even greater precision. 
In the vear 1898 Hoffbauer (13) published a paper in which he showed 
that the age of a fresh-water carp could be determined by an examina- 
tion of the markings on the scale. In 1902 Stuart Thompson (22) published 
au account of some work carried out at the Plymouth Marine Laboratory 
in which he showed, for the first time, that a similar method was applicable 
to sea-fisheries, his researches having been made upon fishes of the Cod 
family the Gadidee—especially on Whiting and Pollock. 
_ * The number of fish at 45 cm. has been taken at forty-seven instead of ninety-seven 
given by Hjort, which appears to be a misprint. Forty-seven agrees with what is shown 
iv the graph and is in accordance with expectation, 
