Aiiicricdii Fishrrics; Sociefij 171 



and arc kept in a ]ar<;e artiiic-ial ])ond liigh above the k'vel of 

 tlie sea. The pond aeeoniniodates l,oOO cod averaging 7 or 8 

 jwunds in weight: dnring the season of 1905 there were 503 

 fisli on hand, of which only one-tenth were males. The pond 

 has a sheet-iron roof, and no dii-ect light reaches it. It is 15 

 feet deep, witli a false l)otfom of wooden slats, through which 

 the debris passes to the real bottom, wh(>nce it may be drawn off 

 by opening pipes. 



C^) Tlie In-ood cod are fed regularly from the outset and 

 remain in sound condition, although they lose weight. The 

 best food in lean fresh herring, but fat herring, salt herring, and 

 other fish may be used. The darkened pond prevents the loss of 

 sight which is so common in fish confined in shallow live-cars in 

 the o])en air. 



C'^) Tlie cliaracteristic and very important ])oint in the 

 operations is that the cod are not handled at all, but are allowed 

 to spawn naturally. Spawning takes place from 9 p. m. to 1 

 a. m., and practically all the eggs are fertilized. The eggs come 

 to the surface, and pass out of the pond through a broad, 

 shallow overflow into a flume which leads to a special house, 

 where the bottom of the flume is covered with silk-mesh screen 

 on which the eggs are left as the water passes through. The 

 eggs are gently scooped up and transferred to buckets or tul)s 

 of salt water the density of which is gradually reduced, so that 

 the eggs will sink while the oily refuse remains at the surface 

 and may be poured off. The eggs are next measured, 450,000 

 lieing allowed to a liter (our figures are 337,000 to a quart, or 

 about 346,000 to a liter— a discrepancy of 104,000, or 33 per 

 cent, which tlie fish culturists must explain), and are then 

 placed in the liatching boxes, 1 to li/> liters to a box. The 

 largest number of eggs taken in one season was 1,440,000,000; 

 in 1905 the number was 3(53,000,000, an average of something- 

 over 800,000 per fish. 



(4) The hatching a])paratus is essentially the same as our 

 automatic ti(hil box, most of the ditt'erences in consti'uction l)e- 

 ing in favor of ours, in my judgment, although of course Cap- 

 tain Dannevig is firmly convinced that his is the most efi:ective. 

 The hatching season begins in February and continues till May. 

 Tlu' maximum number of eggs that may be incubated at one 



