276 L. HILLIS-COLINVATJX 



and other still-water sites of 0-8 kg m~2 yr~i. Where the lagoon is of 

 considerable size, as in many atolls and barrier reefs, this bimodal 

 division of production rates indicates that the lagoonal environments 

 probably contribute most to the mass of the reef, since the area of the 

 lagoon only has to be four to five times that of the reef ridges for its 

 total contribution to be the larger. 



Production of carbonate in lagoonal and still-water areas is 

 principally divided between coral pinnacles, Foraminifera, chemical 

 precipitation of calcite and the calcareous green algae. It is, therefore, 

 of interest to see what densities of Halimeda are required to yield a 

 significant part of the carbonate production of 0-8 kg m""^ yr~i found 

 by Smith and Kinsey (1976) for the lagoon as a whole. 



The incrassata production rate was calculated above to be 0-8 g per 

 thallus per year. For the total production of a lagoon to be provided by 

 sand-dwelling Halimedae growing on horizontal surfaces would require 

 a population density of 1000 plants m"^. Apparently the highest 

 densities so far recorded are 500 plants m"^ in a few "clearings" among 

 the Thalassia beds of the Glory Be reef. Dense populations in shallow 

 water at Glorj^ Be, Enewetak and elsewhere of an incrassata-cylindracea 

 type of thallus have about 100 plants m"^. A clearly visible growth of 

 Halimeda over the floor of a whole lagoon, therefore, would produce 

 about one-tenth of the total estimate of Smith and Kinsey. This is by 

 no means impossible and there are reports in the literature of lagoon 

 floors having what seem, from a ship-board collecting station, to be 

 dense covers of Halimeda (Emery et al., 1954). On two dives to the 

 floor of the Enewetak Lagoon in 1975, however, I found that much of 

 the visible terrain was almost devoid of Halimedae. Shallower lagoons 

 (Enewetak is from 40 m to 70 m deep), however, may have denser 

 stands. But there are also other factors to be taken into account. 



The above calculations neglect two sources of Halimeda segments 

 that may well constitute the major part of the total Halimeda contribu- 

 tion: the rock-growing opuntia-type Halimedae of the pinnacles and 

 of the deep fore-reef. Halimeda opuntia at Glory Be attained local cover 

 densities of 90%, which is roughly equivalent to an incrassata popula- 

 tion of 900 plants m~^, and, since the sprawling life-form of these 

 plants always tends to produce high cover, like the tillering patterns of 

 grasses, local populations always tend to be dense. Some of the pinnacles 

 at Enewetak support considerable populations of Opuntioid forms, and 

 the total vertical faces of the pinnacles must be very large. With over 

 2000 pinnacles in the lagoon, if the lagoon were drained the aspect 

 would be rather like the teeth of a rubber hair-brush. Without direct 

 data on segment production by these rock-growing forms (not so far 



