of the Fishery Board for Scotland. 



295 



APPENDIX F. — No. VIII. 



REPORT as to VARIETY among the HERRINGS of the SCOTTISH 

 COASTS. Part II. By J. Duncan Matthews, F.R.S.E. 



In the previous part of this paper I dealt with the variations occurring 

 in the principal external characters of the herring as each appeared sepa- 

 rately, that is to say the variations of each character were ascertained for 

 the herring generally, but without respect to the conditions of variations 

 in any or all of the characters. It is clear that the normal condition of 

 any one character, or the prevalence more or less common of certain forms 

 of it as it was found in the herrings as a class, would not be increased when 

 it came to be arranged, in terms of its constant combination with certain 

 other definite characters, especially if these others were subject to variation. 

 In ascertaining therefore how far what may be called the normal and extra- 

 ordinary forms or conditions affecting the various characters are found in 

 combination on the herrings, the percentage of the latter given in the first 

 part of this paper are almost sure to be reduced. While this does not 

 affect the absolute conditions in which the various characters are found, it 

 may have considerable influence in determining whether the individuals 

 concerned are to be considered as of different races or species ; and this is 

 always strengthened as the number of characters included in the combination 

 is increased. 



If any great variations of two or more characters were found to exist iu 

 the fish examined, it might be expected that we would find particular ex- 

 tremes of these generally common to certain of the fish, and this would be 

 a good guide in determining their relationship ; but while my first paper 

 showed that the more or less variable common ground of certain characters 

 held as a rule a somewhat different position in the summer as opposed to 

 the winter herrings, the extremes of the variations found were not so very 

 great in themselves, or so invariably present in the proper fish as to settle 

 definitively whether or not they involved distinct races of the herring pre- 

 senting them. 



The further examination of the characters already dealt with, but taken 

 in combination with one or more others, while expectedly reducing some- 

 what the value of the percentages of herring presenting them as given in 

 my former paper, do not seem to so seriously affect the results there re- 

 corded as to alter the probability of the conclusions arrived at ; but it must 

 be remembered that so far as they go they do slightly tend to reduce the 

 probability of the distinction of species considered to exist between the 

 winter and summer herrings. On the other hand this reduction strengthens 

 the opinion that there is no real racial distinction between the herrings of 

 different localities around our coasts in these respective seasons. 



The only combinations which in the comparatively narrow variations 

 recorded have much interest, or indeed are common enough to be of any 

 value, are those of two characters. When the combination of certain 

 definite conditions in more than two characters on the same fish is looked 

 for, the number of individuals having what may be called the normal state of 

 2 P 



