DEPARTMENT OF MARINE BIOLOGY. 229 



"Sample from the Marquesas Lagoon, southwest quadrant, both water and oolitic mud, 

 show the presence of very large numbers of B. calcis, ranging from 3,000 to the cubic centi- 

 meter upwards. No other organisms have been found in these samples, possibly due to 

 the unusually large number of B. calcis. 



"Following the procedure outlined by Drew, we have been able to cause a rapid precipita- 

 tion of calcium in culture media containing calcium acetate or other organic salts of calcium. 

 No precipitation was evident from the organisms growing in natural sea-water or in syn- 

 thetic sea-water, unless carbon dioxide was introduced either from a gas generator or by 

 growing (in association with B. calcis) some bacterium which would produce carbon dioxide 

 and supplying to the culture flask small quantities of sugar or some similar carbohydrate. 



"We have notyet definitely determined whether the calcium carbonate produced during 

 the decomposition of calcium acetate is caused directly by the breaking up of the calcium 

 acetate or whether it is due largely to the formation of ammonium by B. calcis, together with 

 the absorption of carbon dioxide from the air. In our laboratory, where the supply of 

 carbon dioxide is necessarily at all times rather high in the atmosphere, it is probable that 

 this plays an important role." 



As one of the three processes whereby calcium carbonate may be precipi- 

 tated by bacterial agencies, Dr. Kellerman states i"^ 



"The associative action of mixed cultures of bacteria, one species which forms traces of carbon 

 dioxide and one of which forms ammonia either by decomposing some proteid or by reducing 

 nitrates to nitrites and to ammonia, gives rise to ammonium carbonate. This ammonium 

 carbonate reacts with any calcium sulphate which may be in solution according to the formula 



CaS04+ (NH4) 2 CO3 = CaC03+ (NH4) 2SO4 

 It is obvious that the carbon dioxide necessary for this reaction may be produced by plant or 

 animal catabolism as well as by bacterial fermentation." 



In a mixed culture of this kind Dr. Kellerman suspended a collodion sack 

 containing calcium sulphate, thus maintaining a nearly uniform concentration 

 of calcium sulphate in solution in the water. Grains of calcium carbonate 

 bacterially precipitated under these conditions are spherulitic, with a dark 

 nuclear and a lighter, fibrous outer zone. Many grains formed by the pre- 

 cipitate from calcium acetate clearly show zonal growth. 



In June 1913, I began a series of experiments to produce by inorganic 

 agencies calcium carbonate granules which would exhibit zonal structure. 

 The procedure followed was similar to that used first by Rainey. A small 

 amount of gum arable was dissolved in sea-water to which ammonium car- 

 bonate was added. The abundant precipitate of calcium carbonate, which 

 results by the chemical reaction indicated in the quotation from Dr. Keller- 

 man, is nearly all spherulitic in form. After standing for a month or 

 more the liquid was decanted from the precipitate, sea-water containing gum 

 arable was poured on it, and ammonium carbonate was added. The result 

 of such periodic precipitation was to form concentric shells of calcium car- 

 bonate around the previously formed spherulites. In order to make the 

 zonal growth more evident a number of grains formed by the process 

 stated were stained with waterproof red ink and introduced into a solution in 

 which calcium carbonate was subsequently precipitated. This experiment 

 brought out with great distinctness the zonal arrangement of the successive 

 shells. 



Considerable attention was paid to the geology of the vicinity of Golding 

 Cay and of Cocoanut Point, Andros Island, and of New Providence Island. 

 The dune origin of Driggs Hill and Golding Cay, Andros Island, was demon- 

 strated. The marine origin of the area of flat land underlain by oolite was 

 proved in 1912. Marine fossils were found in limestone underlying the flat 

 on the landward side of the dune ridges (now consolidated oolite) from Cocoa- 

 nut Point to NicoUstown, also Andros Island. Marine fossils were found not 

 only along the sea front, but also in Grantstown Plain, south of Lake Kil- 



'Journal of the Washington Academy of Sciences, vol. 4, No. 14, p. 400, August 19, 1914. 



