82 



J. E. SHELBOURNE 



technical details and results will beg:in to la Mer during 1962. This paper deals 

 appear in the Journal du Conseil Perma- only with the design and operation of our 

 nent International pour I'Exploration de closed and open sea- water circulations. 



CLOSED CIRCULATIONS AT LOWESTOFT 



The 50-gallon system 



In the absence of running sea water at 

 Lowestoft, we were obliged to start fish- 

 rearing studies in static conditions and 

 closed circulations. It soon became evi- 

 dent that small static tanks were unsuit- 

 able for survival beyond the early larval 

 stage. By 1957, a closed circulation with 

 limited physicochemical control of water 

 conditions had been designed. It incorpo- 

 rated a fundamental principle of tropical- 

 aquarium technique — the use of photo- 

 synthesizing plants to control the CO2 con- 

 centration and stabilize the pH, as well 

 as to remove some end products of protein 

 metabolism and to add oxygen. 



The pH can be regulated by chemical 

 means — the addition of lime, for instance 

 (Cooper, 1932), or sodium bicarbonate 

 (Breder and Smith, 1932) — but these 

 methods involve changes in the balance of 

 salts. The COo may also be removed by 

 vigorous aeration (Downing, 1958) ; cod 

 larvae, however, are known to be harmed 

 by oxygen supersaturation (Henly, 1952). 

 Nitrogenous end products of metabolism 

 cannot be controlled by purely chemical 

 or physical means without affecting the 

 survival of fish in circuit. 



The design of our first circulation is 

 shown in figure 1. It has three main com- 

 ponents: (a) a 30-gallon covered reser- 

 voir, (h) a covered, temperature-con- 

 trolled water bath housing two moulded- 

 glass tanks (incubators) each of 8 gallons' 

 operational capacity, and (c) a polythene 

 header tank containing illuminated fronds 

 of Enter omovpha intesfhialis attached to 

 stones. A small centrifugal pump under 

 the control of a mercury float switcli inter- 

 mittently transfers water from the reser- 



voir to the header. The pump volute is 

 best constructed from stainless steel or 

 high-density polytliene. The header tank 

 can contain any sessile green alga able to 

 withstand leng-thy immersion in water of 

 high salinity. Fronds and substrate must 

 be thoroughly washed to remove mud and 

 debris before use. Tungsten lighting can 

 be used, but a fluorescent source is prefer- 

 able because it produces less heat. With 

 flexible switching arrangements and a lit- 

 tle practice, we found it possible to adjust 

 the light intensity manually, giving a 

 stable pH of 8.1 for days at a time. 



Sea water flows by gravity from the 

 header into the incubators at a controlled 

 rate after passing through a nylon mesh 

 screen to hold back plant debris. In 

 circulations of this size, a drip feed of 4 

 liters per hour per incubator is a practical 

 rate of flow for pelagic eggs and larvae 

 at temperatures up to 10° C. The opti- 

 mum irrigation rate may differ from spe- 

 cies to species ; clumped demersal eggs are 

 likely to benefit from a rapid flow. Slow 

 rates of flow can be adjusted by control- 

 ling the number of drops falling per unit 

 of time past a "window" (perspex tube) 

 fitted into the inlet system, whilst periodic 

 variations in dropping rate are minimized 

 by setting the header float switch to op- 

 erate within narrow changes of water 

 level. 



Incubator design is a very important 

 factor. Plaice and most other pelagic 

 marine fish larvae are delicate creatures, 

 not adapted to live in constant contact 

 with surfaces. Tanks must therefore be 

 simply constructed, with smooth internal 

 surfaces and no unnecessary inclusions. 

 Two closely apposed surfaces can act as 



