35 



Tab. 56. Bergen Sommersild. 2bIII. 



,, 57. ,, Vaarsild. 2aII. (Autumn herring 



formula). 

 „ 59. Utsire „ 2aII. ( ,, ). . 



„ 90. Liimfjord. lal. 



„ 91. „ Ibll. 



„ 137. Stralsund. 2bIII. 



,, 18. Schlei. 3bII*. 



With respect to autumn herring we have the following 



groups with incorrect formulae : — 



Tab. 105. Terschelling. 2bII. (Spring herring 



formula). 



97. Mouth of the Elbe. 2bII. ( „ ). 



„ 101. „ 2bIL ( „ ). 



„ 102. „ 2bIII. 



„ 103. Helgoland. 2bIII. 



96. W. of Sylt. 2bIII. 



,, 127. Scotland, Fair Island. 2bII. (Spring herring 



formula). 

 „ 129. „ Peterhead, lall. 



50. Bohuslan. Ibll. 



93. \^arberg. lal. 



38. Bay of Kiel. 2bII. 



This last group should not have been left out of the 

 summary (Text p. 126), since they are not only herring 

 from the Bay of Kiel but the formula is the average of 

 47 individuals and not as the other three groups only 30, 

 21 and 19, respectively, and the greater the number of 

 individuals the greater the accuracy of the average 

 formula. 



Heincke himself writes, p. 126: — "Die Genauigkeit 



aller Mittelwerte einer Rasse hangt ab von der Zahl n 



der Individuen, aus deneu das Mittel gezogen ist." 



* In Tab. 193 this is given as 2. It should be 3 (see Tab. 18, p. 40. 

 Index = 2-28). 



