22 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



filtering surface to any desired extent, by simply widening and length- 

 ening the horizontal stratum of sand which does duty as the filter of 

 the sea water and acts as a barrier to prevent the escape of the em- 

 bryos. A description of the working of the apparatus will make it 

 much better understood. 



When the trunk A is put in place (which should be done before the 

 water is let into a freshly excavated pond, and also before the water 

 is let into the trench from the sea-end), it should be securely placed in 

 position and the earth tightly rammed in along the sides so as to pre- 

 vent any sea water from finding its way into the pond, except such as 

 passes through the filtering diaphragm. It is also unnecessary to in- 

 sist that the trunk be constructed in such a way that it will be practi- 

 cally water tight, and not liable to leak between the planks or at the 

 corners. The wire-cloth, sacking or gravel, and sand having been got 

 into piace, and wlien complete forming a stratum having a total thick- 

 ness of five or six inches, the operator is ready to cut away the barrier 

 at the sea-end of the trench to let in the water. 



If then the trunk A has been let down into the trench deep enough, 

 the sea level at low tide ought to be somewhat above the upper edge 

 of the board e. The water will then, as the tide rises, flow back over 

 the sand as far as the board i, and will percolate through the diaphragm 

 into the space I, under the latter, and so find its way into the pond. 

 After a df»y or so the pond will be filled with sea water, which has prac- 

 tically been filtered, and filtered more or less effectually in proi)or- 

 tion to the thickness of the stratum of sand constituting the dia- 

 phragm. After the pond has once been filled, with the rise and fall of 

 the tide in the open water, the level of the latter and that in the pond 

 will be constantly changing 5 in other words, when the tide is ebbing 

 the water level in the pond will be higher than that of the water out- 

 side, as in fact represented at wl and tl in Fig. 1. Under these cir- 

 cumstances there will be a supply of water flowing out, through the 

 under division I of the trunk A, up through the sand and out over its 

 surface, through the outlet O under the gate G. After the ebb tide is 

 over and flood tide begins these levels will be reversed and lol in the 

 l^ond will be lower than tl in the open water, and under those circum- 

 stances there will be an inflow of sea water into the pond through the 

 diaphragm instead of an outflow, as in the condition of the water levels 

 during ebb tide. Under such conditions there will be four alternating 

 periods during every twenty-four hours of inflow and outflow, lasting 

 we will say four hours each, not reckoning the nearly stationary inter- 

 vals between tides or during slack water. This almost constant partial 

 renewal of the water will unquestionably maintain the water inclosed 

 in the pond or i)onds, by means of diaphragms, in a condition fitted to 

 support oysters colonized therein, provided its density is not too great 

 or too slight, and if there is also some microscopic vegetation present. 



It will be readily understood from the preceding description how it 



