bo 



NATURE 



[September 17, 1914 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice ii 

 taken of anonymous communications.'] 



The Age of a Herring. 



The very remarkable conclusions which Dr. Johan 

 Hjort has drawn from his study of the scales of 

 herrings are discussed, over well-known initials, in 

 Nature of August 27. Your contributor is evidently 

 convinced of the truth of Dr. Hjort 's deductions, and 

 this opinion is shared, I believe, by very nearly all 

 those biologists who make a study of fishery questions. 

 Nevertheless, even in face of this general consensus of 

 opinion, I am unable to persuade myself (on the 

 evidence at present in hand) of the validity of Dr. 

 Hjort 's conclusions. The leading statements made by 

 Dr. Hjort and his friends are these : — (i) That the 

 age of each individual herring, or the year in which 

 it was born, is at once made known by the number 

 of rings upon its scales ; (2) that, ipso facto, the shoal 

 can be at once analysed into its component "year- 

 classes"; and (3) that herring spawned in 1904 

 (according to such evidence) were so prodigiously 

 numerous that year after year they have seemed to 

 dominate the Norwegian herring-catch, and have con- 

 stituted from 77 per cent, to 54 per cent, of all the 

 Norwegian "spring herring" during at least the five 

 years from 1910 to 1914. 



To attempt to deal in a few words with all Dr. 

 Hjort's careful work and laborious arguments is, of 

 course, out of the question ; but let me try to say as 

 shortly as possible why I find it hard to follow him. 



The whole argument depends on the primary 

 assumption that the age of each herring is marked 

 upon its scales, year for year, ring for ring. This is 

 still a hypothesis. That there is no little circum- 

 stantial evidence in its favour I readily admit ; but it 

 has never been proved, and is only acceptable so long 

 as the deductions to which it directly leads are them- 

 selves intrinsically acceptable. 



Now if we sample at random a shoal of herrings, 

 and analyse them acording to their scale-rings into 

 so-called "year-groups," we find that these numbers 

 ran^e themselves with great apparent regularity in a 

 unimodal skew-curve : just as the same fish group 

 themselves also, according to size, in a unimodal but 

 more normal curve. Here is one of Dr. Hjort's own 

 samples, reduced to percentages as he himself gives 



analogy is not very remote. Miss A. L. Massy has 

 lately made a study of oyster-shells, using oysters of 

 known age from the hatchery at Ardfry. I extract 

 from her data the following table : — 



Analysis of a Sample of 341 Three ^ear-old Oysters, 

 according to the Number of Shell-rings. 

 Number of rings ... 2 3 4 5 67 



Number of oysters ... 16 169 94 48 12 2 

 That is to say, in the case of the oyster the number 

 of shell-rings does not correspond to the years of the 

 oyster's age, but, on the contrary, is subject to varia- 

 tion according to regular and simple law. And the 

 variability of the shell-rings in a sample of oysters all 

 of identical age is in a very close degree similar to 

 that of the scale-rings in a sample of herring drawn 

 from one and the same shoal. The simple and crucial 

 question is, then, whether in a shoal of herrings the 

 evidence tends to show that all the fish are of one 

 and the same age, like Miss Massy's oysters, or 

 whether the shoal is a mixture of many generations,^ 

 as Dr. Hjort asserts. 



When we come tc Dr. Hjort's final deduction, we 

 find it illustrated, as your contributor tells us, by the 

 following table, showing the manner in which suc- 

 cessive "year-groups" of herring are said to have 

 combined to constitute the Norwegian adult herring 

 population of 19 14 : — 



Spring Herring, 19 14 (2205 Fish Examined). 



population consisted, to the extent of 54 per cent, of 

 its total numbers, of herrings born in 1904, we must 

 also believe that the remaining 46 per cent, was con- 

 stituted of some eleven other successive yearly broods, 

 all so nicely graded as to their proportionate amounts 

 that these latter form an all but smooth curve, rising 

 up to and falling away from the modal group of ten- 

 year-old fish. Any normal population, as of men, 

 consists, of course, of individuals of all ages, less and 



" Age-groups," or Number of Scale-rings, in a Sample of 6T,i, Herrings from the Dogger Bank (1910). 



Number of rings 23 4 5*6 78 9 j« ii 12 13 



Presumed year of birth 1908 1907 1906 1905 1904 1903 1902- 1901 1900 1899 1898 1897 



Per cent, number of fish 0-2 84 14-2 21-3 18-9 14-3 9-3 57 5-5 1-3 08 0-3 



_ Now, to say the least of it, it seems to me statis- 

 tically improbable that a dozen separate generations 

 of herrings, spawned in as many years, should have 

 entered into the formation of the composite shoal in 

 these curiously and regularly graded proportions. On 

 the other hand, were we to assume that these herrings 

 were all of the same age and origin, then we should 

 have at hand an easy explanation of the facts, viz. 

 that just as the individual herrings vary in a normal 

 fashion about a certain modal size, so do they also 

 vary, in the number of their scale-rings, about a 

 certain modal number. 



The rings on an oyster-shell are not just the same 

 things as the rings on a herring's scale, but yet the 

 NO. 2342, VOL. 94] 



less numerous the greater their age. If there be, as 

 among herrings there well may be, large and irregular 

 fluctuations in the annual birth-rate, then the descend- 

 ing age-curve of the population would be irregular, 

 one or more particular ages becoming unduly pro- 

 minent. But that Is a totally different thing from the 

 population-curve which Dr. Hjort presents to 

 us. He tells us that the 1904 group was, and has 

 remained, so predominantly abundant as to constitute 

 for several successive years from one-half to more 

 than three-fourths of the whole stock of adult herrings ; 

 not merely that that year-group was more abundant 

 than any one of its immediate successors or predeces- 

 sors, but that it outnumbered all its contemporaries. 



