36 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



the centre of the ship, which was kept constantly full for 

 supplying baths, lavatories, &c, and for flushing purposes. 

 The tap I used was an emergency one, rarely required. 

 The tank is filled by a pump worked by the engines, so 

 that it is in constant action while the ship is steaming. 

 The sea- water enters by an aperture, provided with a 

 valve, and covered by a grid of three-quarter inch mesh, 

 placed in the ship's bottom, about 8 ft. above the keel 

 and 14 ft. below the surface of the sea (fig. 2). 



Over the tap I tied two nets (B and D), 

 the coarser meshed one inside, so that as the 

 water ran through, the larger organisms were 

 screened off and retained in net 

 B, while the smaller passed e<>ar$« T*t i««ii* 

 through and were caught by Fi , c „ a oicU ». 

 D, the outer net. The ship's 

 carpenter fixed up for me a ^^^— 



shallow tub placed underneath fr — 



the tap, and into which I T u 



my nets dipped, while. ssrt^ f f-f» _—A fr^ MM _ 

 from a hole near the deck 



bottom of the tub a >[Vts wTa^fort^^o^. 

 hose pipe was led to Fig. 3. 



the nearest scupper to convey the waste water overboard 

 without flooding the deck (fig. 3). 



On the starboard side of the ship an overflow pipe from 

 the top of the tank discharged over the side a little below 

 the level of the main deck. I used this pipe by tying the 

 two nets, A and C, over its open end, the coarser one 

 being inside the finer (fig. 4.) 



The supply of water pumped into the tank is so much 

 in excess of what is normally used that, during the whole 

 voyage, water was pouring freely from the four-inch over- 

 flow pipe. When the ship is rolling, however, the di§- 



