62 MADRAS FISHERIES BQLLETIN VOL. XI, 



of Noctiliica, the latter to myriads of diatoms, chiefly Rhizosoma ; 

 he held that neither exercise harmful effects upon fish life even 

 when in vast superabundance. 



This latter opinion is directly opposed by Gilchrist who states 

 that ^ :— 



" Red water is a phenomenon observed not infrequently in South 

 African seas. It consists of masses of red-coloured water some- 

 times a mile or two in extent, at other times occurring only in small 

 patches. In False Bay it may be seen usually several times 

 during the summer months, and presents a very remarkable 

 appearance, being frequently of an almost blood-red colour. It 

 consists of multitudes of Noctiliica, normally present in sea water 

 nearly everywhere. At certain times, however, they increase 

 enormously in numbers, when they can be seen to consist of minute 

 egg-like bodies, which in mass present the conspicuous red colour 

 referred to. Such crowding together of these minute organisms 

 appears to pollute the water, for when examined microscopically 

 most were found to be dead ... It has been observed that 

 fish seem to avoid the red water and fishermen do not care to 

 fish in its vicinity. It is said that mullet caught in it, decay very 

 quickly, becoming quite decomposed if left ungutted overnight. 

 One or two instances are known in which fish and other marine 

 animals have been killed apparently by such polluting of the 

 water. I am indebted to an old resident in Saldanha Bay for some 

 particulars of such an occurrence at that place. He stated that on 

 one occasion, the only one in his experience, about the year 1907, 

 the bay, which is almost landlocked, became filled with red water, 

 known locally as ' flower water.' At the time there had been a 

 north-west wind for some days previously. The fish in the bay 

 were seen floating belly upwards in a disabled condition. Some 

 of them were cast on shore in quantities at the end of the bay, in 

 such numbers that they were 'taken away in cartloads.' Even the 

 shell-fish, such as Mussels {Donax scrra probably), Klip-koes 

 {Haliotis) were killed off in large numbers, apparently on account 

 of the presence of large quantities of decaying organic matter." 



Gilchrist is also of opinion^ that when diatoms occur in vast 

 swarms, under certain circumstances they may decay and cause 

 the death of fishes by suffocation. 



Loc. cii., p. 17. ^ Loc. cit., p. 19. 



