488 REPOET OF COMMISSIONER OF FISH AND FISHERIES. 



(eight to nine gallons;) and on the bottom of this gravel, sand, and char- 

 coal are heaped up so as to constitute a filter. A purified water runs 

 from this reservoir by a stop- cock situated underneath it, and falls into 

 troughs placed like steps, which may be multiplied at pleasure. This 

 arrangement is entirely similar, as we see, to that which M. Coste had 

 already chosen ; but M. Millet has added an improvement, which, we 

 hasten to say, the learned professor of the College of France has at once 

 adopted in his turn. 



However pure running water may be, it always bears with it and 

 deposits at the bottom which it covers foreign particles, which, if they 

 rested upon the eggs, would finally surround them with a sort of slime 

 favorable to the development of byssus and mold. To meet this 

 objection, M. Millet thought of suspending the eggs a little below the 

 surface of the water. M. Vogt* had already taken the precaution to 

 place them in a muslin bag, permeable on all sides, which he threw into 

 the lake after having fastened it to a stake, or kept it in place by a large 

 stone. Starting upon the same principle, M. Millet has arrived at a 

 surer and more complete result. He places the eggs upon sieves, which 

 little rods, sliding on the edges of the tubs, hold at the desired height. 

 This skillful experimenter has successively employed sieves of various 

 substances, of hair, of silk, of willow, &c., and 'has finally given the 

 preference to galvanized metallic cloth, which have more solidity and 

 durability, do not spoil, are easil}' cleaned by the help of a brush, and 

 are only very rarely attacked by sea- weed. 



The expense of outfit of such an apparatus is quite insignificant. 

 The working consists merely in filling the reservoir every morning and 

 evening, in moving the sieves once a day, and taking away the eggs 

 which may become opaque. For many years the eggs of trout, of sal- 

 mon, of the umber, &c., have been developed in this way, and hatched 

 in considerable quantities in the same apartment which the experimen- 

 ter occupies at Paris, in the middle of the rue Castiglione. 



When the process can be carried on in the water of a stream itself, 

 of a lake, or of a pond, M. Millet recommends the employment of double 

 sieves of metallic cloth, which may be kept at a suitable height by the 

 help of floaters, and which follow all the changes of the level of the 

 water. For the species which spawn in sleeping water, he lines the 

 double sieve with aquatic plants, or limits himself to placing the eggs 

 in large shallow tubs 'with plants, which prevent the water from corrup- 

 tion. When the fecundated eggs are to be transported to great dis- 

 tances, M. Millet advises placing them in a flat box, in quite thin layers, 

 between two wet cloths. In this state he has sent them to Florence, 

 where they have reached the hands of M. Yaj and the Professor Cozzi, 

 after a journey of twenty or twenty-five days, and have not failed to 

 hatch soon after. The use of moist linen is preferable to that of aquatic 



* Euibryologie des Salmones; Histoire Naturelle des Poissons d'eau douce do 

 L'Europe ceutralc, by L. Agassiz, p. 16, 1843. 



