ÖFVERSIGT AF K. VETENSK.-AKAD. FÖRHANDLINGAR 1894, N:0 10. 577 



has the most turbot-like lateml line is quite natural; but 

 that one of the brill-like speciinens, N:o 4, conies next is 

 rather stränge. Of these five specimens the most turbot-like, 

 N:o 1, and the most brill-like, N:o 5, have similar lateral 

 lines. This shows how in hybrids the variation sometimes is 

 entirely free. 



The position of the anterior nostril points to an inter- 

 mediate character in all the five specimens and also shows a 

 progressive development throughout. In a true turbot the ante- 

 rior nostril is on a line with the first ray of the dorsal, in a 

 true brill on a line with the third ray. In N:o 1 it is on a 

 line with the second ray or nearly so (possibly a little in front 

 of the same), in N:o 2 on a line with the second ray, the same 

 in N:o 3. In N:o 4 its situation is betweeu the second and 

 third rays; and the same in N:o 5. 



A genuine turbot has at most 64, a genuine brill at least 

 73 rays in the dorsal fin. If we look at the above-given table 

 we shall find that N:o 1 is above the limit of the turbot and 

 N:o 5 below the limit of the brill; the three other ones are in- 

 termediate forms, but N:o 3 and N:o 4 approximate more to 

 the brill. 



The number of rays of the anal fin is in a typical turbot 

 at most 47, in a brill at least 54. In this respect as well N:o 

 1 and N:o 5 come within the limits of the parent-forms (N:o 1 

 — turbot; N:o 5 — brill), and the other ones are intermediate, 

 but N:o 2 and N:o 3 are nearer to the turbot, N:o 4 to the 

 brill, just as their exteriör appearance indicates. 



About the other fin-formulae there is not much to say, 

 except that N:o 3 and N:o 5 have a different number of rays 

 in the pectoral of the blind side and eye-side. In N:o 3 the 

 pectoral of the eye-side has 13 rays, thus one more than in 

 any of the parent-forms. 



If we now look at the table of measurements, it will 

 give US some interesting points of view. The percentages given 

 are mostly intermediate between those of the parent-forms. But 



