ELEMENTS CONCENTRATED ON THE SEA BOTTOM AND AUTHIGENIC MINERALS

Certain elements such as calcium, carbon, silicon, iron, and manganese, carried to the sea in solution, are precipitated by organic or inorganic processes and are thereby concentrated on the sea floor. Readily recognizable minerals such as calcite and aragonite are produced in some instances, where in others the concentration is detectable only by chemical analyses which reveal the abnormal content of such elements as iron, manganese, and even the radioactive elements. The interaction of sea water and inorganic debris gives rise to certain characteristic minerals such as palagonite and phillipsite and it is probable that clay minerals are formed from the disintegration of volcanic ejecta (Bramlette and Bradley, 1940). The origin of glauconite which is a common constituent of terrigenous sediments is as yet uncertain, one theory being that it is built up from simple substances, the other that it is a product of the chemical weathering of biotite on the sea floor.

The factors affecting the deposition of calcium carbonate, both as calcite and aragonite, have been described elsewhere. Dolomite has been reported as a rare authigenic constituent of pelagic sediments by Correns (1937). Silicon in the form of dissolved silicates or as colloidal silica is one of the abundant constituents of river water. It is commonly believed that much of the silicon is immediately deposited on entering the sea by precipitation or coagulation when the river water mixes with sea water (Twenhofel, 1939, p. 373). However, there is no evidence for the presence of inorganically precipitated silica in recent marine sediments and the deposition of organic remains must be sufficient to balance the supply of dissolved and colloidal silicon compounds to the sea. The factors governing the deposition of organic siliceous remains have not been investigated thoroughly, but a relatively high content of dissolved silicates

---

within the euphotic zone and a high content in the entire water column exist in those areas where the sediments contain large amounts of siliceous remains (chapter VI).

Manganese is one of the chemical elements which is present in red clay in greater concentration than it is in igneous rocks (table 111, p. 992). In marine sediments it is chiefly present as oxides, largely manganese dioxide (MnO₂), and may exist as finely divided grains, coatings over shells and inorganic material, as a cementing matrix, and in nodules and concretions. Manganese nodules, having diameters of 10 cm or more, are conspicuous constituents in dredge and trawl samples and have aroused great interest since the Challenger investigations showed their rather wide distribution. Although the nodules are characteristics of the deep-sea sediments, manganese in its other forms is commonly found also in terrigenous sediments. The manganese concretions and nodules which are found most abundantly in the South Pacific and Indian Oceans contain variable proportions of manganese dioxide, ferric oxide (limonite), and clay. The average contents of MnO₂ and Fe₂O₃ are about 29.0 per cent and 21.5 per cent, respectively. Many individual analyses are given by Murray and Renard (1891) and by Andrée (1920). The nodules generally show laminations of different shades and textures and have a roughly concentric structure with a nucleus of pumice, volcanic glass, rock, or organic material such as a shark's tooth or an earbone of a whale.

---

Although the large nodules have aroused great interest, it is probable that they do not represent the bulk of the manganese present in most sediments. Correns (1937) in his studies of the Meteor material from the Equatorial Atlantic found that the content of manganese was conspicuously higher than in igneous rocks but he was unable to detect any manganese grains. Ten representative red clays from the Atlantic (Correns, 1939) contained an average of 1.2 per cent MnO₂ on a CaCO₃-free basis. The noncalcareous portion of 19 typical globigerina oozes contained 0.5 per cent MnO₂. Additional data are given in table 119 (from Correns, 1939). The dark color of the red clays of the Indian and Pacific Oceans has been attributed to their manganese content, but comparison of the Carnegie data from the Pacific (Revelle, 1936) shows that the red clays of the North Pacific are lower in manganese than those of the Atlantic.

Table 111 (p. 992) shows the average of 10 North Pacific red clays to be 0.83 per cent MnO₂, that is, approximately one half that of the Atlantic samples. In contrast to the samples from the Atlantic the MnO₂ content is much higher in the calcareous deposits of the Pacific Ocean from which 20 samples containing more than 30 per cent CaCO₃ contained on an average 5.5 per cent MnO₂ when calculated on a CaCO₃-free basis (Revelle, 1936). The great difference between the manganese content in the calcareous oozes of the Atlantic and the Pacific and the fact that there is no correlation with the percentage of calcium carbonate indicate that the deposition is not closely related with the accumulation of calcareous material.

There are two possible sources of the mangenese in deep-sea deposits. Murray and his associates concluded that it was a product of the submarine weathering of volcanic material. However, volcanic rocks do not contain very large amounts of manganese and in order to build up the concentrations found in red clays and in the noncalcareous fraction of globigerina oozes, it would be necessary to have large amounts of silica, aluminum, and iron leached out and redeposited elsewhere, since there is no accumulation of these substances in solution. According to Washington (1920), the average managanese content of the oceanic volcanics, expressed as MnO₂, is between 0.05 and 0.18 per cent, the lower value being for the islands of the Atlantic and the higher for those of the Pacific Ocean. Consequently, a five- to tenfold concentration of manganese would be necessary if the clays were formed from this material. The manganese carried to the sea in solution forms an extremely small fraction of the dissolved solids and is rather rarely determined. According to Twenhofel (1926, p. 406) the concentration generally varies between 0.5 and 5.0 mg/L, which is greater than the amount present in sea water, namely, 0.001 to 0.01 mg/L. Murata (1939) has studied the total and exchangeable manganese content of river muds, namely, particulate material which is carried to the sea. The average total manganese in five river muds was 0.18 per cent, as MnO₂, and the exchangeable manganese in these river muds was found to range from 0.03 per cent to 0.07 per cent as MnO₂. From this it may be concluded that managanese is carried to the sea both in solution and as a constituent of the sedimentary debris.

---

---

Manganese is most soluble in acid solutions when in reduced state. Hence, environments where such conditions prevail will tend to leach the manganese out of the solid material. Conversely, alkaline and oxidizing environments tend to precipitate the manganese in the very insoluble form of manganese dioxide. Whether or not bacteria or other biological agencies are directly involved in the solution or precipitation of manganese is not definitely known (ZoBell, 1939).

On reaching the sea the dissolved and exchangeable manganese is probably precipitated as finely divided MnO₂ suspended in the water. If this settles in terrigenous environments where reducing conditions prevail, it again may pass into solution. The most favorable conditions for accumulation are those where the oxidation-reduction potential is relatively high as on topographic highs and in pelagic deposits where the conditions are always oxidizing, and in the presence of even slight amounts of solid CaCO₃ where the pH must be relatively high. Hence, once the manganese is deposited in such an environment it is not likely to pass into solution again. This explanation does not account for the formation of the large manganese nodules which, according to Murray and Renard, may occur in clay which itself is relatively low in manganese. However, it is known that manganese accumulations tend to form on similar material, somewhat analogous to crystal growth, and possibly this may be one of the important factors. The difference in the manganese content of the sediments of the equatorial Atlantic and those of the Pacific Ocean is a problem which is yet unsolved.

Iron (table 111, p. 992) is present in red clays in concentrations somewhat greater than in the igneous rocks and terrigenous deposits, although the increase is not as marked as for manganese. As pointed out above, the manganese nodules contain a large proportion of iron oxides. Authigenic iron compounds may be found as the oxides, principally limonite and as sulphides. Table 119 contains data on the iron content of the Meteor samples, which, computed on a CaCO₃-free basis, is rather erratic, and shows no correlation with the content of organic remains. The average iron content, calculated as Fe₂O₃, is 7.12 per cent, which, as in the case of manganese, is less than the average for the red clays. This Correns attributes to the higher iron content of the deposits of the Pacific and Indian Oceans. Examination of the Carnegie data shows that the samples with less than 30 per cent CaCO₃ calculated on a CaCO₃-free basis contain 7.0 per cent iron calculated as Fe₂O₃, whereas the calcareous deposits computed in the same way contain, on an average, 15.6 per cent

---

Fe₂O₃. Therefore, as in the case of manganese the Meteor data and those of the Carnegie give conflicting results. Correns’ data show no correlation between iron and manganese, but those of Revelle show a remarkably good correlation for the calcareous deposits. For the Pacific calcareous deposits the ratio of iron to manganese, calculated as Fe₂O₃ and MnO₂, is approximately 2.8:1 on a weight basis.

Two sources may supply iron to the marine sediments. These are: (1) the weathering of the oceanic volcanic material, which is relatively high in iron as indicated by the analyses of Washington (1920), which give about 10.8 per cent iron, calculated as Fe₂O₃; and (2) the considerable amounts which are carried to the sea in solution and as a constituent of the mineral debris. As the latter material may be relatively low in iron it may have a diluting effect and the dissolved iron must be the important source. Acid and reducing conditions tend to dissolve iron compounds, while alkaline and oxidizing conditions tend to precipitate the higher oxides. A mechanism somewhat similar to that described for manganese may therefore apply to the iron. The most favorable environment for the accumulation of iron oxides will be in oxidizing sediments containing calcareous material. Iron is precipitated as sulphides in stagnant environments where hydrogen sulphide is produced. Bacteria may be directly involved in the solution and precipitation of iron oxides and sulphides (Twenhofel, 1939).

Phosphorus, as P₂O₅, (table 111, p. 992) is found in approximately the same amounts in the red clays as in igneous rocks. However, Correns’ data (table 119) indicate that on a CaCO₃-free basis there is an increase in P₂O₅ with the carbonate content indicating that the phosphate may form about 0.1 per cent of the calcareous skeletons. The supply of dissolved phosphate and its organic cycle in the sea have been discussed elsewhere. Apparently the regeneration of phosphorus utilized by plants is relatively complete, but it is being added to the sea as dissolved phosphate as well as a constituent of the minerals, chiefly apatite. The supply of dissolved phosphorus may be balanced by the removal in skeletal structures, some of which are extremely high in phosphorus. Furthermore, a certain amount may be deposited as a constituent of the detrital organic matter. Authigenic phosphorus compounds are not found in the pelagic deposits, but in certain nearshore localities phosphorite, Ca₃(PO₄)₂, forms a cementing material which accumulates in nodules and crusts. These phosphate nodules are first discovered by the Challenger off the Cape of Good Hope and have subsequently been found in many coastal regions on the edge of the continental shelf and on topographic highs. Dredging off the southern California coast has shown phosphorite to be the most abundant type of rock collected.

The material collected off the coast of southern California (Deitz, et al, 1942) contained phosphorite nodules ranging in size from small oolites to

---

masses weighing about 50 kg. The phosphorite was usually found at depths less than 1000 m, although some was obtained from the slopes at greater depths. Phosphorite is commonly associated with abundant calcareous remains and glauconite. Pelagic foraminifera and benthic remains may be found and also the teeth and bones of fish and marine mammals.

The phosphorite nodules have a characteristic smoothly rounded surface with the upper surface having a glazed unweathered appearance. Freshly broken surfaces are light brown to black and the outer surface is generally somewhat darker, owing to a thin coating of manganese oxides. Microscopic and chemical analyses show the principal mineral to be collophane. This may form virtually pure masses or act as a cement for phosphatic oolites and inorganic and organic material some of which may be more or less completely phosphatized. Most of the nodules show banding, indicating that the accumulation has been discontinuous. The calcium phosphate on the average forms about 67 per cent of the nodules, with calcium carbonate the second most abundant constituent. Sediments in the vicinity of phosphorite deposits contain approximately the same amounts as given above for pelagic deposits.

The phosphate nodules appear to be forming at the present time and several hypotheses have been advanced to explain their mode of formation (Twenhofel, 1939, Dietz et al, 1942).

Barium sulphate concretions have been found in recent marine sediments only in three widely separated localities, namely, off the west coast of Ceylon (1235 m), near the Kai Islands in the Dutch East Indies (304 m), and near Catalina Island off the coast of California (800 to 650 m). The concretions, ranging in weight from a few grams to about 1 kg, are generally of irregular shape, sometimes tubular, with concentric banding. They contain from about 60 to 82 per cent barium sulphate, which forms a cementing matrix, and contain mineral grains as well as organic remains. The bulk density of the barite concretions is high, averaging 2.70 and sometimes exceeding 3.0. Emery and Revelle (1942) examined the California material and consider that the only explanation which will account for their characteristic properties and their restricted occurrence is that the concretions are formed in the mud by the interaction of hot spring water high in barium with the sulphate present in the sea water.

The radioactive elements, particularly radium, in marine sediments have received considerable attention because of the unusually high radium content of pelagic sediments and because of the possibility of using the changes with depth in cores as a means of establishing the rates of deposition. Existing data have been assembled by the principal workers in the field, namely, Evans and Kip (1938), Evans, Kip, and Moberg (1938), Utterback and Sanderman (1938), Föyn, Karlik, Pettersson, and Rona (1939), and Urry and Piggot (1941).

---

The radium content of marine sediments varies between 0.1 X 10⁻¹² g/g and values more than one hundred times as large. Low values are found in coarse-grained terrigenous deposits, whereas the high values occur in noncalcareous pelagic sediments. Bottom samples collected in the Pacific Ocean by the Carnegie were examined by Piggot and the average radium contents of the various types of sediments are given in table 120. (Revelle, manuscript.) Nearshore sediments, such as those studied by Utterback and Sanderman (1938), average about 0.3 X 10⁻¹² g/g of radium.

The radium content of granites is about 1.6 X 10⁻¹² g/g and that of basalts even less. The higher radium content of deep-sea marine sediments must therefore be due to a concentration taking place on the sea floor. Radium has a relatively short half-life period of about 1600 years. If radium alone were deposited on the sea floor where the rate of sedimentation is less than 1 cm in a 1000 years the amount present in the lower parts of cores, a meter or more in length, would be extremely small. Although there is generally a decrease in radium with depth, the distribution (cf. Urry and Piggot, 1941) shows that other radioactive elements which disintegrate to form radium must be deposited. Of these ionium and possibly uranium are of importance.

Various mechanisms by which radium and the other radioactive elements may be concentrated in marine sediments have been suggested. Coprecipitation of radium with calcium carbonate and of uranium with iron and manganese compounds has been offered as possible mechanisms, but such hypotheses are not substantiated by the available data. Revelle (1936) pointed out the correlation between the radium and the decomposable organic matter in the Carnegie samples and suggested that organic activity might be the effective precipitating agency. Subsequently (Revelle, manuscript) he has developed this hypothesis and shown a correlation between the radium content and the organic siliceous remains

---

in the sediments (table 120) and from this concludes that dissolved radium is extracted from the water by the diatoms and radiolarians and is carried to the sea floor in these siliceous remains. In red clays the siliceous skeletons have been largely dissolved and the organic matter destroyed, leaving the radium behind, whereas in certain deposits high in siliceous remains the radium content may be low because of the diluting effect of the inorganic debris.

Glauconite is a green, blue, or brown hydrous silicate of potassium, magnesium, aluminum, and ferrous and ferric iron. The formula for glauconite can be written R₂′O, 4(R″O,R₂‴O₃),10SiO₂,4H₂O, where R′ represents the univalent bases and R″ and R‴ the di- and trivalent bases. It is found in recent marine sediments and in many marine sedimentary rocks and it is considered to be a substance formed only in the marine environment. The distribution and properties of glauconite in both recent and fossil deposits and the various theories which have been advanced to explain its formation have been reviewed by Hadding (1932), Takahashi (1939), and Galliher (1939). Analyses given by Hadding and Takahashi show that although there is a considerable range in chemical composition there is no essential difference between the recent and the fossil material. However, as Takahashi has pointed out, the chemical composition can be varied considerably by treatment with neutral salt solutions as well as with acid and alkaline solutions. As the X-ray spectra are not materially different, it is probable that the exchange of cations is somewhat analogous to base-exchange in clay minerals.

Glauconite occurs in sediments as irregular grains between 0.1 and 1.0 mm in diameter. The surface is often polished and sometimes cracked and the grains have no definite internal structure. When present in sufficient quantity they give a greenish color to the sediment and the terms green sand and green mud were originally restricted to sediments containing glauconite. The glauconite grains often represent casts of foraminiferal shells and, in some cases, the shells are found completely filled by glauconite. Hadding has recorded glauconite present as interstitial material in porous organic and mineral substances and also as an incrustation.

Glauconite occurs in recent terrigenous sediments off continental coasts at depths ranging from a few meters down to about 2500 m. According to Galliher, it is more abundant adjacent to land areas where plutonic and metamorphic rocks are exposed. Although glauconite is found in many parts of the world it is abundant only in rather restricted localities where the environment is appropriate for its formation or accumulation, where deposition is not rapid, namely, upon the continental shelf or isolated topographic highs near the coast.

The mode of formation of glauconite is not yet definitely established. The older theories postulate the formation of an aluminosilicate gel which

---

later absorbs iron and potassium from solution. This general theory, that glauconite is a complex substance built up from relatively simple compounds upon the sea bottom, has been accepted in various forms by most writers upon the subject. Galliher (1935a,b) has advanced the hypothesis that glauconite is formed by the submarine weathering of biotite. In studies of recent marine sediments from Monterey Bay, California, Galliher was able to find a transition in distribution and in physical and chemical composition between fresh biotite near shore and glauconite in somewhat deeper water. According to him, the transformation involves the oxidation of part of the iron, loss of aluminum, and hydration, which results in changes in structure such as swelling and cracking. Galliher considers that during the metamorphosis the material increases in volume by ten- to twentyfold and that it is relatively spongy. Part of the characteristic form of glauconite grains he attributes to the swelling, but he also considers that a certain amount of abrasion and molding may arise from the activity of mud-eating animals which tend to shape the spongy material in their digestive tracts. Whether either of these hypotheses affords the correct explanation of the formation of glauconite awaits further investigations.