TROPICAL CLUVIATES AND BIOLOGY — CARTER 437 



tion did not usually last for more than a few days at a time whereas 

 in the Chaco swamps it might be unbroken for weeks. I believe that 

 the reason for this difference is the greater frequency of rain in Guiana 

 and the less heating of the surface by day. The African swamps (pis. 

 3 and 4) were thickly covered with papyrus and the deoxygenation 

 in them was again extreme. Everywhere, except in lakeside swamps 

 near the open water of the lake, even the surface water contained no 

 measurable oxygen and the content of carbon dioxide was high. In 

 these swamps the plankton was as sparse as in the Chaco swamps and 

 the larger fauna mostly air-breathing. The African lungfish Protop- 

 terus and several air-breathing teleost fishes live in these swamps. 



Thus, it appears that deoxygenation is a general condition in shallow 

 and stagnant tropical waters, and this is borne out by the fact that 

 air-breathing adaptations are found in the teleost fishes of similar 

 habitats in many other tropical regions. In temperate countries 

 deoxygenation does occur in shallow and stagnant waters occasionally 

 during long periods of hot summer weather, but it is unusual. In the 

 lake of the botanic gardens at Cambridge all the fish died some years 

 ago in a hot spell ; I believe that they were killed by stratification and 

 consequent deoxygenation of the water, which is muddy and nearly 

 stagnant. Normally in temperate climates, heating of the surface in 

 the daytime is not strong enough to prevent overturn at night. But 

 we have very few examples of work on such waters even in temperate 

 countries, and still fewer on tropical waters. It seems to me that 

 more accurate knowledge of the conditions in which overturn occurs 

 in natural waters would be valuable. It might be expected that at 

 high altitudes in the Tropics, where the temperature is lower, condi- 

 tions more like those in temperate waters would be found, and Beadle 

 [5] has found that in some papyrus swamps on the shore of Lake 

 Naivasha in Kenya at about 6,000 feet the water was 50 percent satu- 

 rated with oxygen. On the other hand, I have found in an open pool 

 at Kigezi in Uganda at a similar altitude (5,579 feet) apparently 

 permanent stratification and complete deoxygenation of the lower 

 water. The reason for the difference is not apparent. Clearly more 

 work is needed, and this should be both theoretical and in the field. 



Equally unexpected results have been given by work on the sulfur 

 content of tropical fresh waters in Africa. Beauchamp [6] pointed 

 out that the sulfur content of many African fresh waters is very low ; 

 in several lakes it is not above 3 parts per million (mg. per liter). He 

 suggested that lack of sulfur is a limiting factor in the growth of the 

 aquatic fauna. The subject was further investigated by Hesse [7], 

 working on Lake Victoria. He found that the lake water contained 

 0.5 to 2 p.p.m. total sulfur and less than 0.5 p.p.m. sulfate. In contrast 

 the aquatic vegetation contained a normal sulfur content (average 



