I030 BULLETIN OF THE BUREAU OF FISHERIES. 



that passes through the screen, and affords a desired current for the young fish, 

 which do not hke to be inactive. 



For this purpose a second shghtly larger tube is slipped over the stand- 

 pipe and, by means of a clamp collar supported on three legs, is held with its 

 lower end just above the bottom. The current thus produced can be regulated, 

 greater force being obtained when desired by lowering the outer tube and thus 

 preventing the full outflow of water. This raises the level in the trough, and 

 the difference between the level in the trough and the stream which can escape — 

 namely, the height of the inner standpipe — makes the pressure to force the 

 water up from the bottom and carry with it the refuse matter in the trough. 

 It is not in a trough such as this that there will be ill-smelling bottom water. 



The apparatus does not, as one might think, act merely as do communi- 

 cating vessels. With a flow of 2 to 3 liters per minute I obtain a difference of 

 level of 3 to 4 centimeters, according to the elevation of the outer tube. By 

 regulation of this tube, which is very simple, both surface and bottom water 

 will be renewed. If only the lower water were emptied an oily layer would 

 form at the surface and act as an insulator between the air and the water. 



As the aquarium of the Trocadero is supplied by sluices, in the flow of 

 which there might be the same fluctuations as in the river, I have provided 

 an emergency overflow to balance any sudden rush of water. 



When it is desired to remove some of the fish from the trough the whole 

 apparatus may be removed, the mouth of the outlet in the bottom of the trough 

 being closed with an ordinary cap or plug. 



A SUCTION APPARATUS FOR CLEANING HATCHING AND REARING TROUGHS. 



This device is designed for use in the removal of dead eggs or fry, remnants 

 of food, or any undesirable substance that may be found in the troughs. The 

 use of the usual metal or wooden tweezers, or perhaps long pins, too often 

 causes the eggs to burst, thus spreading infection from their decomposed 

 contents. Little glass pipettes are used, taking one egg at a time. But this 

 often escapes and, falling to the bottom of the trays, is left to give rise to 

 Saprolegnia. 



To meet these difficulties I have used a pipette with a rubber bulb attached. 

 The tubes vary in diameter, according to the sizes of the different species of 

 eggs, and are 0.25 meter long, being slightly bent at the entrance to the bulb. 

 At the outer end is a ring of blue glass to guide the eye of the operator. With 

 the aid of this form of pipette 15 or 20 eggs may be taken up to be thrown out 

 of the trough at one motion of the operator. 



Upon this appliance as a basis I su' equently devised a second means of 

 cleaning by combination with the siphon principle. The later apparatus 



