TRANSACTIONS OF SECTION C. (i07 



sandstone of the Rosebrae division, I'.O.R. sandstones = 'Orio per cent, of ^InO... 

 (4) In small spots or liniuji: minute cavities, and evidently following carbonate of 

 lime, in Triassic rocks at Spynie and Lossiemouth, MnO„ = -035 per cent. (5) In 

 veins or lining joints, occasionally parallel to and some distance back from the 

 joint-plane. An example from Bishopmill, U.O. U., gave MnO, 1-27 per cent. + MnO 

 •27 per cent. (6) Along the upper surface and occasionally irregularly diffused 

 through tlie interbedded clayey bands of the Rosebrae division. (7) Uniformly 

 diffused through the sandstone in the same way as the much more frequently 

 occurring ferric hydroxide. Seer, at Newton, Millstone, and Cloves quarries in the 

 U.O.R. (8) In some organic remains in the same iormation a scute of Bothrio- 

 lepis major gave MnO 2-38 per cent., of Pmmmosteus taylori -83 per cent. (9) 

 As a brown or blackish staining on the casts of organic remains. (10) As illustra- 

 tive examples are cited fragments of cherty limestone in the local boulder clays, 

 with their interstices filled with MnO.j, the carbonate of lime having been totally 

 removed, and a specimen of ' black ' sand from under boulder clay on the Banffshire 

 coast, MnOj 6-58 per cent. + MnO '48 per cent. 



Experiments made by allowing a dilute solution of manganese sulphate — thirty 

 grains to the gallon — to drip slowly on various rocks and sandstones, showed that 

 common chalk and sandstone containing carbonate of lime were darkened in colour 

 within twelve hours. Sandstones without carbonate of lime were not darkened. 

 If the specimens were first moistened with dilute ammonia, caustic soda or potash, 

 or the carbonated alkalies, darkening to a degree took place very rapidly. Free 

 ammonia was found to exist in every specimen of sandstone from the area ex- 

 amined, and to be particularly plentiful iji Newton sandstone. An acid reaction 

 was obtained in some of the ' black ' spots at Newton, due, it was believed, to the 

 oxidation of sulphur, which was also present. Others gave a marked alkaline 

 reaction. The presence of ammonium chloride was also demonstrated in a number 

 of the sandstone specimens. In the presence of ammonia and ammonium chloride, 

 manganese is only precipitated after it is peroxidised, but peroxidisation is rapidly 

 effected in the presence of free ammonia or other free alkali. Though the solu- 

 tion of ammonia and ammonium chloride in the sandstones is no doubt very dilute, 

 it is probably, as compared with the solution of manganese in the infiltrating 

 water, relatively strong. 



Presuming the access of oxygen — which may be taken for granted in the case 

 of porous rocks like sandstones — the explanation of the precipitation of manganese 

 dioxide simply resolves itself into accounting for a preponderance of alkalinitv at 

 the special points of precipitation of that substance. Analyses show that'the 

 manganese areas contain excess of lime, magnesia, and alkalies, compared with what 

 obtains in the surrounding sandstones. This may have been a cause of increased 

 alkalinity, and hence of the precipitation of the manganese dioxide. On the other 

 hand, it may simply be a concomitant of the precipitation, and due to the same 

 cause. The alkalinity in some of the manifestations enumerated has undoubtedly 

 been due to carbonate of lime (]), (2), (4), (8), (9); to the ammonia arising 

 from the decomposition of organic remains (1), (8), (9) ; to the presence of carbo'^ 

 nated alkalies (3), (5), ( G), (7) ; and to free ammonia of the sandstone water after 

 the total precipitation of the ferric hydroxide in a higher zone of the sandstones (7). 

 The conditions that are necessary for the precipitation of manganese dioxide are 

 the presence of alkali or alkaline substance in excess, soluble manganese com- 

 pounds in transit, and facility for oxygenation. 



There is nothing particularly unique in the precipitation of manganese dioxide. 

 It is simply an extension or continuation of the same action as determines the 

 precipitation of ferric hydroxide, and a general separation is effected between the 

 two substances by the fact that the iron compounds fall out before the manganese 

 as the infiltrating water containing them encounters further and further supplies of 

 alkali. The general distribution of the two substances in the Elgin sandstones 

 illustrates this natural method of separation, the rocks impregnated with secondary 

 infiltration of ferric hydroxide in a general way occurring in a zone overlying 

 those impregnated with manganese dioxide. 

 . In the author's opinion the manganese nodules of the deep-sea deposits owe 

 their origin to the operation of the same or similar causes. 



