bacteria (Takaichi 1999). Thus, spirilloxanthin-2-ol is the first carotenoid with chirality to be identified in anaerobic photosynthetic bacteria. The possible evolutionary and functional significance of this is as yet unknown and requires further study.
In the absence of recent or ongoing igneous/geothermal activity within the Bahamas carbonate platform there is no immediate explanation for the recorded temperature anomaly in the South Andros Black Hole. Therefore, alternative mechanisms must be considered. In formulating a convincing theory to explain the observed temperature anomaly the following points were considered: (i) the physico-chemical gradient boundaries are unusually sharp indicating little mixing, (ii) the temperature anomaly is coincident with the phototropic bacterial plate, (iii) the cells are rich in carotenoids especially spirilloxanthin, and (iv) spirilloxanthin is a known to have a relatively low ET efficiency (*30%) in antenna complexes (Frank and Cogdell 1993).
We propose, therefore, that the mass populations of anoxygenic phototropic bacteria present at 17.8 m are functioning as heat engines by dissipating excess light energy as heat. It is clear that at this depth in the water column most of the penetrating light is in the 450–550 nm region (Sullivan 1963) and this is the wavelength region where the light is absorbed maximally by the carotenoids (Hawthornthwaite and Cogdell 1991; Zuber and Cogdell 1995; Takaichi 1999; Takaichi and Shimada 1992; Frank and Cogdell 1993). The key question is whether there is sufficient energy dissipation by the carotenoids to account for the recorded increase in temperature?
In order to address this question, we have attempted to estimate the energy budget for the South Andros Black Hole as follows: the volume of the 1 m thick microbial plate at a depth of 17.8 m is approximately 4.9 · 104 m3, and within this layer the water temperature increases by 7C. Since 4.186 · 103 J of energy is required to raise 1 Kg of water by 1C the energy required to raise this water mass by 7C corresponds to *14.6 · 1011 J. Clearly, this is an over simplification since as the water is brackish/ saline, the spatial increase in water temperature is not instantaneous (Fig. 2) and the phototrophs themselves represent a sizable fraction of the total volume. Further- more, we have no data on how much of this energy input is renewed on a daily basis. However, it clear from the temperature profile of the water column (Fig. 2) that a heat engine which can put out sufficient energy must be present in order to sustain the observed temperature differential of this water mass.
Since, no field measurements are available of the total daily energy input resulting from solar irradiance at the surface of the South Andros Black Hole we have used a similar approach to calculating the daily energy input. On the basis of light attenuation with depth measurements made by Dunstan (1982) on Dancing Lady Reef, Jamaica we have assumed that 66% of the surface light irradiance is attenuated at 17.8 m depth and that all photons are ultimately transduced into heat energy. Pinkney et al. (1995) reported that in Storr’s lake, San Salvador, The Bahamas surface light irradiance values measured in mid- summer and at mid-day exceeded 2,200 lEinsteins s–1 m–2. If we assume a typical Caribbean summers day consists of 9 h of sunlight (at 2,000 lEinsteins m–2 s–1) we estimate that the energy input assuming no light attenuation cor- responds to *7.3 · 1011 J at 17.8 m depth. Factoring in a light attenuation value of 66% reduces this value to *2.4 · 1011 J. Since the purple sulfur bacteria have very low carotenoid to bacteriochlorophyll ET efficiencies (ca. 30%, Table 1) then up to 7.2 · 1010 J are potentially available to be released by the bacteria in the form of excess heat on a daily basis. On the basis of these cal- culations we estimate it would take approximately 21 days to incrementally raise the temperature of the water mass at 17.8 m depth by 7C, a timescale which we believe is not unrealistic.
A combination of unique factors may account for the origin of the bacterial anomaly that is omnipresent between the brackish and saline water. At the boundary between two water masses there is a steep salinity gradient, oxygen is depleted enabling sulfate-reducing bacteria to grow and generate hydrogen sulfide (H2S). In turn, this allows the growth of mass blooms of sulfur-oxidizing bacteria. These bacteria occupy a highly specialized niche (saline, sulfide, low light) by exploiting the remaining solar radiation that penetrates deep into the water column, and transforming it into useable chemical potential. After sufficient time (perhaps over a period of years) these bacterial populations have achieved a state of homeostasis with respect to their sulfur cycling, which is manifested by the development of a unique biological phenomenon that is the Heat Engine in the South Andros Black Hole, the Bahamas.
In conclusion, sulfur-based phototropic bacteria isolated from the South Andros Black Hole have evolved a novel ecological adaptation by a self-generated rise in water temperature thereby optimizing their growth conditions in this specialized ecological niche.
Acknowledgments We would like to acknowledge generous financial support from the Biotechnology and Biological Sciences Research Council, United Kingdom (RJC); the Centre National de la Recherche Scientifique, France (BR); the Commissariat `a l’Energie Atomique, France (BR); the European Union (AG, contract MEIF- CT-2005-00951); the Federation of European Biochemical Societies