190 



MAkKKTAHLi: URITISII MARINE FISHES 



on one occasion on May 2Sth a number calculated to be 107,000. 

 These specimens were 14 to 20 mm. long ;.; to ! inch; and were 

 still very transparent, but the first dorsal and the ventral fins had 

 appeared. These larvae (Fig. 98) were still very slender in propor- 

 tion to their length, and were remarkable for the great size of 

 the air bladder. In the middle of June the little fish were i inch 

 to i ._,".. inch long, and had nearly reached the form and character 

 of the perfect smelt. They were still however transparent, and 

 without the scales and silvery coat. 



Even in August the young smelts, now I ./- inch to it 

 inch, although more pigmented and less transparent, had not 

 acquired -their scales and silver} 1 garment. The development of 

 these last characters of the perfect fish takes place in Septem- 

 ber and October, when the young are 2 inches to 2-j inches 



FIG. 97. Larva of the Smelt, } inch long, 6 days old ; after Ehrenbaum. 

 Fi<;. 98. Larva of Smelt, f inch long ; after Ehrenbaum. 



long. It will be seen from this that the transformation of the 

 young smelt is more delayed and more gradual than in the case 

 of the spring herring, whose change to the form and character of 

 the adult is completed in the Baltic at the end of July. 



In August the multitudes of smelt fry of the year become 

 large enough to be caught by the large stow-nets (stccrtlianioi) 

 which are used in the Elbe. But these are employed not as a 

 delicacy for the table as whitebait is in London, but only as bait 

 for the eel traps. 



After October the smelt, young and old, descend towards the 

 sea, and return towards fresh water at the end of February and in 

 March. In the Elbe Ehrenbaum found that while the smallest 

 ripe fish were 4 inches long, others which were as large as this 

 were immature. He also observed that in the first half of the 



