Thomson. — Portobelh Marine Fish-haichery . 541 



when the fish were gently pressed. No female fish were taken 

 in the trawl-nets on that occasion. He also noted at the same 

 time that no mature fish were obtainable in the harbour. 



On the 19th May, 1905, a stock of flounders was obtained 

 in the harbour by seine-netting on the edge of the banks at 

 low water, and these were transferred to one of the ponds. 

 These shallow- water fish were from 8 J in. to 12 in. in length, 

 and were not in a sexually far-advanced condition ; there were 

 about three male, to each female fish. It was noticed that they 

 lay about the deep end of the pond half buried in the mud, 

 but on the beginning of the flood tide they always commenced 

 to be more active. Towards evening also they moved about 

 more freely, and food thrown into the pond and neglected 

 during the day was always gone by the. morning. 



On the 5th June the temperature of the water in the ponds 

 fell to 0° C, but the cold did not seem to trouble the fish. On 

 the 7th June, with the temperature of the water at 1-2°, they 

 were feeding quite freely. 



On the 26th June 104 fish were in the ponds, all taken in 

 rather shallow water ; they averaged about 9 in. to 10 in. in 

 lensrth, and the females — even a few 12 in. or 13 in. fish — were 

 not nearly ripe. On the same day, while netting m from 10 

 to 15 fathoms in the channel between Quarantine and Goat 

 Islands, four large females were taken, and were found to be 

 nearly ripe. 



On the 20th July two 9 in. mature males and a 9 in. mature 

 female were placed in the observation-tank, the latter fish 

 being the only small mature female of its size noticed. They 

 were kept in confinement for forty-two days, being returned 

 to the pond again on the 31st August. During all that time 

 the female fish ate no food at all, while the males ate voraciously 

 every day. No ova were obtained. 



The surface of the pond was tow-netted every morning, 

 but no ova were obtained till the 3rd October, when about 

 a thousand were secured. More were obtained pretty well 

 every day till the end of the month, the total number secured 

 being about 180,000. 



Both species of Bhombosolea being together in the ponds, 

 it was not possible to distinguish the two kinds of ova. In 

 one case the egg averages - 82 mm. in diameter, and has only 

 one relatively large oil-globule. The egg of the other species 

 is distinctly smaller, averaging - 65 mm. in diameter, and is 

 furnished with from eight to thirteen small oil-globules. The 

 two kinds are quite readily distinguished from one another, 

 but unfortunately it was impossible to separate the eggs, or to 

 determine the species to which they belonged. This can be done 



