304 Part ITI.—Twenty-fourth Annual Report 
the spring herrings spawn when the bottom water is the coldest of the 
year and the autumn spawners when it is the warmest of the year. 
The effect of the difference on the rate of development in the eggs must 
be considerable. From the Table given above it may be inferred that the 
eggs of the spring fish take from about 25 to 30 days to hatch in 
February aud March, while the eggs of the autumn spawners hatch in 9 
or 10 days, or approximately in a third of the time. One must therefore 
add these numbers to the date fixed for the maximum spawning in order 
to determine the period when most of the larval herrings appear in the 
water. In the case of the spring fish the period is about the middle of 
April, and in the case of the autumn fish it is about the second week of 
September, 
The advantage to the autumn herring in respect of the quicker develop- 
ment in the egg is, however, compensated by the difference in the tem- 
perature to which the young growing herrings are soon exposed. With the 
spring fish the temperature is a rising one, favourable to growth, for many 
subsequent months, while with those hatched in autumn, the temperature 
is a falling one, especially at the surface, and thus less favourable to 
growth 
How long it may take in some instances for the young of the spring 
herring to reach even a small size is shown by an experiment of Mr. 
Harald Dannevig. On 13th—14th March he fertilised some herring eggs 
and brought them to the Marine Laboratory at Aberdeen, where they 
hatched on 9th April, after an “incubation” period of 26 or 27 days. 
The larve then measured 8mm. in length. The yolk was not absorbed 
until 9 days later, and 23 days after hatching the post-larve measured 
only 10mm. In this case 49 or 50 days elapsed (viz., 13th-14th March— 
2nd May) before the young herring measured 2-ths of an inch, thus 
offering a great contrast to the results of Meyer, but in the latter case the 
temperature during the experiment was not the same as at Aberdeen. 
Much stress cannot be put upon the experiment, but Mr. Dannevig 
is skilful in rearing post-larval fishes, as his success with the plaice 
shows. 
The difference in temperature between the bottom and surface water, 
above shown, no doubt explains the fact that the larve in spring soon 
seek the upper layers, while in autumn they do this to a very small 
extent, the later larvee, as Masterman says, appearing never to leave the 
bottom, but to migrate shorewards at once, without an intermediate 
journey through the mid-water and surface layers. 
The young herring when it is hatched is of a length ranging from 
52mm. to 8°8mm. (4-2 inch), and the yolk disappears in from three or 
four days to a week ; traces may be found in larve measuring over 9mm. 
It may be mentioned that the larva of the sprat is 3-3°7mm. when 
hatched, and, according to Ehrenbaum, the length when the yolk is 
absorbed (in about a week) is 477mm, They may thus be easily 
separated in some collections. 
6.— A Criticism or Mryer’s CoNCcLUSIONS AND OBSERVATIONS. 
Before dealing with the collections of herrings given in the Tables 
appended to this report, it may be well to consider some general results 
as to the growth of fishes deduced from my researches with a number of 
other species, and how they bear upon the statements concerning the 
growth of the herring above quoted, and more especially those of Meyer 
and Jenkins. 
