BULLETIN OF THE UNITED STATES FISH COMMISSION. 357 



continued experimenting, with considerable success, until the summer 

 of 1874, when, desiring to enlarge the scale of operations, he determined 

 to seek for some locality possessing greater natural facilities. After 

 carefully examining different places more or less favorable for the pur- 

 pose, the present site of tehe Northville hatchery was selected, as afford- 

 ing not only the best advantages he was able to find, but as leaving 

 little or nothing to be desired in the possession of all those natural sur- 

 roundings necessary to the perfect development of piscicultural science. 

 To this point he at once removed, and began the construction of a build- 

 ing and of ponds and raceways suitable for his purpose. 



The building is an ordinary one-story frame structure, 80 feet long by 

 nearly 30 wide, and contains an office, sleeping apartment, and tank 

 room, besides the main hatching room, which is furnished with such 

 appliances as are best calculated to do the work. At first, and for sev- 

 eral succeeding seasons, the style of incubator used was of Mr. Clark's 

 own invention, * which had been thoroughly tested by him in theClark- 



* This invention consists in the construction of a suitable building, at one end of 

 which, nearest the water supply, are tanks, containing many barrels of water con- 

 voyed through faucets from spring or lake, as the nature of the eggs to be hatched 

 may require, which passes through flannel screens, and is thus filtered from all sedi- 

 ment before entering the troughs containing the hatching boxes. These troughs are 

 about one foot (or more, as the case may be) in width and ten inches deep, each of them 

 containing a series of water-tight compartments, which contain the same number of 

 boxes of less dimensions, also water-tight, except the bottoms, which are covered with 

 finely-perforated copper or brass wire cloth to prevent the eggs or fish from escaping 

 when hatched out. 



These last boxes are filled with several screens, each containing many thousand eggs, 

 and may be of sufficient capacity to hatch an almost unlimited number of eggs. 



Over these screens, and after the eggs are equally distributed over them, there is 

 placed a finely-perforated metallic plate, B, and the whole is kept in place by a cross- 

 bar, C, fastened to the sides of the main trough. These boxes are elevated upon feet 

 to raise them from the floor of the trough, to allow a free passage of water under 

 them and to raise them above any sediment that may pass through and settle on the 

 said floor. 



The first screen that lies over the copper cloth is also raised to gain free circulation 

 to the water. The main trough must have a descent of three-sixteenths of an inch 

 to the foot, to form sufficient fall of water into each separate box to produce a mod- 

 erate current of water down through the eggs. 



This arrangement completed, the water is let in at the upper end of the long troughs 

 upon the perforated cover, which spreads it equally over the whole part of the eggs 

 below, which, owing to the declivity of the main trough and the water-tight partitions 

 in them, causes it to flow over said partition on to the next below, which produces an 

 up and down movement to the current running throughout the whole series of hatch- 

 ing boxes, making changes around and through the whole number of eggs in each 

 compartment constant while in process of hatching. 



Many more fish are hatched by this process than can be stored and cleansed from 

 their shells and other impurities consequent upon the last stages of hatching; hence, 

 it becomes necessary to add store-room and an additional process for cleansing the fish 

 when hatched out from the impurities above named. 



To remedy this trouble a series of large tanks, G H K, are erected for the reception 

 of the water as it leaves the hatching troughs. From ten to twenty days are required 

 from the commencement of the hatching season to its close, consequently a propor- 

 tionate number of fish are hatched daily. These are washed from the unhatched eggs 

 into the first receiving-tank before mentioned, and allowed to stand quietly without 

 much current to the water in which they are. The eggs thus cleansed are returned 

 to the hatching boxes from which they came. As soon as the shells from the eggs are 

 well settled to the bottom, a moderate current of water is allowed to flow through an 

 opening to the next tank below, carrying the cleansed fish with it, depositing any im- 

 purities that may yet be left with the fisli iu eaid settler; and the fish are allowed to 

 follow on with the current, passing still through another opening to the large recep- 



