CHEMISTRY IN THE OCEANS 591 



fractions. 5-AIn02 is composed of disordered sheets of manganese 

 dioxide, whereas the manganite crystals possess a double layer- 

 lattice structure in which ordered layers of manganese dioxide 

 alternate with disordered layers of manganous or ferric hydroxides 

 or basic salts. The disordered layer is the site for the accumulation 

 of the metal ions whose enrichment in these materials has been 

 pointed out previously. In nodules with amounts of iron, greater 

 than can be accommodated in the disordered layer, the mineral 

 goethite appears. Buser (personal communication) points out 

 that the appearance of goethite in the minerals suggests that 

 particles of fresh iron hydroxide, or oxides of lower molecular 

 weight, are initially deposited, which possess the potentiality of 

 transformation to goethite. 



The rates of accretion of these minerals on surfaces or during 

 formation of the nodules is extremely slow. The "Horizon" 

 nodule, a large concretion dredged from the North Pacific with a 

 longest dimension of about 1 m, was analyzed by the ionium- 

 thorium method and was found to amass its solid phases at a 

 rate of somewhat less than 0.01 mm per thousand years or a rate 

 of about one atomic layer per day. This is probably one of the 

 slowest reactions occurring in nature in which a measure of the 

 rate of the reaction can be ascertained. It should be emphasized 

 that such rates are not necessarily continuous, as often layers of 

 clay or detrital materials are found between the slowly accreting 

 ferromanganese lamina. 



The rather exotic nature of these minerals has inspired a number 

 of hypotheses as to their mode of formation. The early workers 

 were more concerned with the immediate origin of the manganese 

 and associated elements, rather than any physicochemical con- 

 siderations. Such sources of manganese as Foraminifera, volcanic 

 debris, and subsurface springs were postulated. Further, both bio- 

 chemical and inorganic mechanisms to deposit the manganese and 

 iron oxides were invoked. Of special interest today is the thesis of 

 biological oxidation, first proposed by Dieulafait (1883), taken up 

 again by Dorff (1935), and recently revived by Graham (1959). I 

 shall return to this problem somewhat later. 



A satisfactory hypothesis must account not only for the chemi- 

 cal and physical characteristics of the materials, but also for the 



