266 DISCOVERY REPORTS 



Corethron criophilum. Panthalassic here, where most were seen inshore or on the outer shelf in spring. Of its 

 generally oceanic character in most phases there can be no doubt. 



Fragilaria granulata. Panthalassic? Recorded here only from the outer shelf or even offshore, never overlapping 

 with the next species. Not recorded elsewhere except off South Africa. 



Fragilaria Karsteni. Very definitely neritic and with a spring maximum, but also quite abundant during the 

 autumn survey. Recorded elsewhere only off Cape Province (Karsten (syn. F. capensis) Boden). 



Asterionella japonica. An important inshore and outer-shelf species, especially in autumn when it was one of the 

 few other species to rival the dominant chaetocerids in abundance, and partly responsible for visible discoloration 

 of the water at some stations close inshore. 



Thalassiotlmx longissima, though never very abundant, was relatively important offshore at both seasons, rather 

 more so in spring than during the autumn survey. 



Thalassionema nitzschioides. An inshore species, very local in this area, chiefly in spring (or early in the succession). 



Nitzschia closterinm. Tychopelagic, a littoral species fairly frequent in plankton close inshore, but never abundant 

 here. 



Nitzschia delicatissima. Panthalassic, mainly offshore with an autumn maximum here. 



Nitzschia longissima. Never abundant, but very characteristic of the inshore plankton here, especially on the 

 spring survey. Possibly littoral, tychopelagic. 



Nitzschia seriata. Panthalassic, abundant and important at both seasons studied. Most important offshore in 

 spring, but more abundant inshore. 



Comparing these ecological notes with those of other workers, principally Gran, Lebour, Braarud 

 and Gran, Cupp and K. R. Gaarder, as referred to in the sources of Table 28, it appears that a large 

 majority of the species considered important or typical in the Benguela area were disposed therein 

 just as one would expect from their recorded distributions elsewhere. Minor modifications of previous 

 ecological characterizations based on these results, and on earlier work in the southern ocean (Hart 

 1934, 1942), are suggested as follows: 



Thalassiosira subtilis should probably be regarded as panthalassic rather than oceanic. Though 

 cosmopolitan in the open oceans, it frequently attains greater abundance near land. Gaarder's (1951) 

 note that though widespread in the Michael Sars material from the North Atlantic, the species was 

 observed most abundantly close in to the Azores, accords with this view. 



Conversely several of the dominant chaetocerids of the rich inshore waters, though undoubtedly 

 neritic in the main as recorded elsewhere, showed a more or less pronounced panthalassic trend ; since 

 lesser numbers of them flourished sufficiently far offshore to dominate the sparser communities 

 encountered there. This trend was most marked in Chaetoceros Lorenzianum, but was also shown by 

 the three most abundant of all the diatoms met with: C. compressum, C. constrictum and C. airvisetam. 



The inshore or neritic species C. sub secundum exhibited a striking seasonal difference from its known 

 disposition in the northern hemisphere, where most observers have recorded a strong maximum in 

 spring or early summer. On our two surveys it was relatively much more abundant in autumn. 

 Possibly we missed an earlier maximum through lack of full seasonal coverage, which would show up 

 any repetition of the succession at a shorter interval of time. Equally we may here have a real dif- 

 ference due to the ' conditions of existence ', biological as well as physical. In the northern hemisphere 

 the species has been regarded as arctic-boreal or boreal, but both Gaarder's results and our own show 

 that it can flourish in much warmer temperate waters than the earlier records suggest, where it must 

 find itself among a very different and much more varied phytoplankton community. 



The holoplanktonic cosmopolitan solenoids: Rhizosolenia alata and R. hebetata.f. semispina would 

 seem definitely better described as panthalassic rather than oceanic. Their dominance at certain 

 sparsely populated oceanic stations is unquestioned, but at the same time they were often found in 

 greater numbers among the rich Chaetoceros plankton close inshore. The same applies in a less marked 

 degree to R. imbricata (mainly the variety Shrubsolei; the larger type phase was rarer and 'more 



