64 



Appendices to Fourth Annual Report 



standard. As stated above, the standard generally adopted (I shall have 

 to refer later to some necessary departures from this rule) was that of the 

 length from the anterior end of the lower jaw to the base of the caudal 

 fin ; and to avoid confusion, this has been called length to end of 

 scales, as distinguishing it from length to tip of tail. The means by 

 which is indicated the position of the fins, length of head, &c., in 

 works treating of fish, has varied generally in two directions. Either 

 the dimension is stated as a part of the total length, i.e., employing 

 the total length as the unit, or the dimension itself is employed as 

 the unit. Thus, for instance, the length of head of a 270 mm. 

 herring would be described as being (say) 'ISS of the total length (or, 

 roughly, it would probably be stated as -1), or it would be described 

 as being contained in the body 7*5 times ; the former result, of course, 

 being found by dividing the particular dimension by the total length, the 

 latter by dividing the length by the dimension. The common way 

 appears to have been to obtain the latter result as a whole number only, 

 and turn it upside down, so as to form a fraction of the body length. 

 This latter is convenient, and for general purposes useful enough ; but 

 when comparisons of what may be rather minute differences, and where 

 considerable exactitude are required, it may lead to important errors. For 

 as the dimension descriptive of any character decreases in amount, the 

 apparent variation by this scale becomes greater and greater ; I shall have 

 to refer to some of these effects later on. Professor Heincke in his paper 

 has noted this error, and has, in dealing with such a small figure as the 

 length of fins, changed his index of comparison, and calculated according 

 to the proportion — Total length : 1000 : : Length of fin : a; (the index) ; but 

 he has not employed this method for finding the head index — a dimension 

 of about the same magnitude. There is, however, such a difficulty in 

 comparing the amounts of variation even where large dimensions are dealt 

 with, such as the distance of dorsal and anal fins from tip of jaw, that for 

 that reason, as well as for uniformity, I have adopted the more correct 

 method throughout ; my equation, however, reading — Total length : 1 : : The 

 dimension : so that in place of getting a whole number, as Heincke 

 does, I get that decimal fraction which at once conveys to the reader the 

 proportion which the dimension bears to the whole length. 



The various characters of the fish and their variations having been thus 

 reduced to a regular standard for comparison, the method in which this is 

 to be made has to be considered. In this preliminary paper I have treated 

 dimensions, variations, &c., which were found to exist on the fish sent 

 from the various Fishery districts in a more general manner than is 

 essential, if we wish to reach a fairly dependable conclusion as to whether 

 these variations are merely individual, or may fairly be called those of 

 'race' or 'variety.' I shall try to indicate the general characteristics 

 and their variations which are to be found among our herring, the presence 

 or absence of any one or more of these if fairly evident among the herring 

 of particular seasons or localities, and generally the extent of the actual 

 variation and the prevalence of any degree thereof. I leave for further 

 investigation the more intricate work of ascertaining to what extent 

 particular or generally localised variations may be found among those fish 

 to which they are not peculiar, and what combinations of the several 

 characteristics, as subdivided by their variations, may be common to all 

 the herring, or any part of them. 



General Size. 



It may here be stated that the herrings subjected to examination were 

 samples of the ordinary takes of the fishermen at various places around the 



