[21] ARTIFICIAL PEOPAGATION OF OYSTEKS. 761 



serted iu ajar or tank, A, which is hermetically closed. The eud of the 

 tube B should pass through the cover </ g. E is the overflow pipe, fitted 

 with a sliding extension, D, so that the overflow may take place at any 

 desired height. F is the air pijje ; C, the stream of water entering 

 the funnel from a small vessel. The apparatus requires but very little 

 water and little or no head to sui)ply all the aeration necessary for lab- 

 oratory purposes. The water running from the recei)tacle H, through 

 the nozzle C, is made to impinge against the sides of the funnel so as to 

 give a rotary motion. A small glass ball or vial jDartially closes the ujjper 

 end of the tube B, and the column of water in the tube is thus a broken 

 one, the intermediate spaces being filled with the air carried in by the 

 suction. As the tank A fills with water the air is forced out of F. The 

 overflow of superfluous water is regulated for difierent depths, or rather 

 for different pressures on the end of F, by raising or lowering the mov- 

 able arm D. The power of the pump is of course proportional to the 

 length of the tube B. Four feet fall of water furnishes a sufficient sup- 

 ply for most laboratory purposes. 



A second simple pump is shown in Fig. G. A A are two casks having 

 stop-cocks at m, n, o, p, s, t. The cocks s and t are connected by rub- 

 ber hose ; from m and n hose runs to a point, ^, whence the main air- 

 supplying hose is led to the aquaria. Both casks are fitted with slings 

 and tackles for raising and lowering them, and the pump is worked in 

 the following manner : The cask a being filled with water, the cocks s 

 and m are closed and the cock o opened ; the cask is then raised six or 

 eight feet. The cock p of cask A' is closed and cocks n and t opened. 

 Then, by opening cock s, the water flows from s through t into A', forces 

 air out of n to fc, and thence to the aquaria. When the upper cask is 

 empty, it is lowered and the other one raised, and the process reversed. 

 A 40-gallon cask will supply air for eight to ten hours, and with a 

 small luff-tackle can be raised easily by two persons. 



I have made these sketches and accompanied them with the descrip- 

 tions not because of any great merit or originality the apparatus pos- 

 sess, but because others may find their use of benefit and be saved some 

 trouble in experimenting with methods and appliances having the same 

 end in view. 



