THE VARIATION, RACES AND MIGRATIONS OF THE MACKEREL. 279 



Table X., sh&iving frequency of High and Low Values of the 



Second Dorsal Fin. 

 (Percentages.) 



Undoubtedly the two groups which agree most closely according to 

 the above table are those from Kinsale and Kerry. For these groups 

 the frequency of the mode is the highest recorded, and is practically 

 identical in the two cases (85"3 and 85*4 7o). The two lowest frequencies 

 of the high values (39 and 4*9 7o) are also found in the same groups. 



The figures do not appear to justify any further amalgamation of 

 the groups ; but it is worth noticing that the North Sea group 

 approximates closer in its values to the Irish samples than does either 

 of the others — a feature which is again exhibited with respect to the 

 dorsal finlets. 



We have already seen that, in respect to spottiness and the number 

 of finrays in the first dorsal fin, the Irish fish approach most nearly 

 to the theoretically primitive condition. Now the primitive number 

 of rays in the second dorsal fin would appear almost certainly to have 

 been 12, partly on account of the high frequency of this number 

 throughout the samples, and partly on account of the fact, which is 

 established below, that when this number is exceeded, the number of 

 finlets tends to be reduced, and when the number of rays is reduced 

 below 12, the number of finlets tends to be increased. This correlation 

 implies a primitive constancy in the number of rays in the posterior 

 dorsal fin of the ancestral mackerel prior to its subdivision into second 

 dorsal and finlets. The total number of rays in the ancestral continuous 

 fin was probably 17, which became subdivided into 12 rays for the 

 primitive second dorsal fin and 5 rays for the finlets. If this view 

 is correct, it is to be remarked that in the high frequency of the modal 

 number of rays in the second dorsal fin, the Irish fish again display 



