THE VARIATION, RACES AND MIGRATIONS OF THE MACKEREL. 2G9 



(see Table E, p. 293), and exactly 50% of the fish exhibited the modal 

 or most frequent value. 



In all other samples the distribution of values was asymmetrical. 

 The percentages for the various localities show that the modal value is 

 12 for all localities except Scilly, for which it is 13. This exception 

 is no doubt partially due to the smallness of the consignments from 

 that locality (only 74 fish in all), and is paralleled by a similar 

 phenomenon in the case of a small consignment from Brandon, April 

 16th, in which the most frequent value was 11, although 12 in a 

 subsequent and larger sample. 



For most localities the number of fish having 13 rays preponderates 

 over the number having 11. This is true for Lowestoft, Eamsgate, 

 Plymouth, Scilly, and County Kerry. But the American fish show 

 a slight preponderance of the lower value, and the preponderance of 

 this value is considerable in the case of the fish from Kinsale. In this 

 respect the Kinsale fish differ from those from all other British localities, 

 and the matter requires special consideration. 



On examining the data for the various samples received from Kinsale 

 (Table E), it is seen that the preponderance of low values was not 

 exhibited by all the consignments, those taken on September 17th, 1897, 

 and July 1st, 1898, being normal in this respect. The preponderance 

 of low values was entirely due to the samples dated July 30th and 

 September 3rd, 1897. Now these were the only samples in the whole 

 series of fish which were not examined in a fresh condition. They 

 arrived during my absence in Canada, and after measurement were 

 placed in tanks of formaline to await my return. The excess of low 

 values is so unique in these two cases that I consider myself justified 

 in attributing the difference to the effects of this re-agent, the tendency 

 of which to develop free formic acid is well known. The amount 

 of calcareous matter in the minute posterior rays is so small, that 

 its solution by the acid would be merely a matter of time. As the fish 

 remained six weeks in the formaline before examination, there can 

 be little doubt that the calcareous matter in the smallest rays was 

 dissolved in a certain number of cases to an extent sufficient, at any 

 rate, to invalidate the records. The omission of the posterior ray from 

 20% of the fishes in these samples would be more than sufficient 

 to account for the observed differences between the fresh and preserved 

 samples of Kinsale fish. 



If this correction be permitted, the variation of the first dorsal fin 

 becomes very uniform for all British localities, with the exception of 

 Scilly, an exception which is readily explicable by the inadequacy 

 of the sample from that region. The numbers of finrays vary round 

 12 as a mode, and the percentage of values above the mode tends 



s 2 



