1*2*2 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



Quite at the end of liis work (p. 128) Heineke writes : — 

 >v lliernack bekimden die von Petersen gegen mich ins 

 Eeld gefiikrten Messungen von Herbst und Friihjahrs- 

 lierigen aus den danischen Gewassern, die ieli in den 

 Tab. LXXXI. bis LXXXIX. s. 105 bis 1G\X wiedergegeben 

 habe, einen unzweif elhaft sicheren Unterschied zwischen den 

 beiden Saisonrassen, indem bei den Frukjakrskeringen 

 die Mittelformen 2bII., bei den Herbstheringen 2aII. und 

 2al. vorkommen." 



When these Tables are examined it is seen that Heineke 

 is incorrect; Table 82 gives for 15 autumn herring from 

 'Varberg in the Kattegat the average formula 21>II., that 

 is to say the average formula of spring herring. Further, 

 in tke examples cited above it is seen that the formula? of 

 Heineke do not kold good for the herring of tke Danish 

 waters, for example, in tke case of Liimfjord (two groups) 

 and Yarberg. 



To return to Heincke's summary (Table 193) : — 



Spring Herring. 



Tables with correct average formula 2bII. 15 i.e. 65 %. 

 „ other „ „ 8 i.e. 35%. 



Autumn Herring. 



Tables with correct average formula 2a T. or 2aII. 



17 i.e. 607 % 



. other „ „ 11 i.e. 39*3 '".. 



Again, we have 30 body measurements of full herring 

 from the Grreifswalder Bodden given in Table 138 (p. 151), 

 which herring were captured in November, 1891. 



In Table 139 (p. 152) we have the skull measurements 

 of 20 of these herring given. 



In the summary of body measurements (pp. 196-199), 

 the 30 Greifswalder herring are not given, but in tke 



