ANNIVEESAET ADDKESS OP THE PEESIDENT. 79 



simple explanation to suppose that it had been formed out o£ the 

 augitic or hornblendic constituents of the deposit. I am by no 

 means certain that some of the grains are not really pseudomorphs ; 

 but at the same time the amount of glauconite in some beds of the 

 Greensand is far too great to allow us to suppose that it was to 

 any considerable extent formed directly, in situ, from augitic or 

 hornblendic sands. In some cases I am inclined to think that the 

 grains were formed elsewhere, and drifted as sand into their pre- 

 sent position. On w^ashing different specimens of Greensand I 

 was also much struck with the manner in which the glauconite 

 seemed, as it w^ere, to replace yellow ferruginous mud, as if it had 

 been to some extent formed out of it. On the whole, then, we 

 have evidence of considerable variation in the conditions ; but yet 

 all the facts agree in showing that glauconite was formed by con- 

 cretionary aggregation at the time of deposition, or very shortly 

 afterwards, from material closely related to decomposed augite, 

 hornblende, or olivine. 



There is far better evidence of the direct formation from augite of 

 green minerals more or less closely allied to glauconite, in the green 

 slates of the English lake- district. I may here say that, according 

 to my own observations, the augite and olivine of our older volcanic 

 rocks occur altered into at least three different kinds of green 

 material. One of these corresponds closely, if not absolutely, with 

 glauconite, and is composed of laminae ; another is composed of 

 fibres ; and the third is either amorphous or belongs to the regular 

 system, unlike the others, not depolarizing. All these three sub- 

 stances also occur in what must have been cavities in the rock, into 

 which no doubt they have been introduced by the agency of water. 

 The fragments of augite in the Cumberland green slates have been 

 changed, and the cavities filled, in the same manner. On the whole, 

 the fibrous substance is the most abundant, and is characterized by 

 having its positive axis of depolarization not strictly parallel to the 

 line of the fibres, as though the mineral might be some variety of 

 asbestos. It fills the cells of fragments of pumice, and also larger 

 cavities, formed, like those in peperino, by the evolution of gas 

 amongst the unconsolidated mud, or by contraction during con- 

 solidation. Since the identification of the true nature of frag- 

 ments of pumice is a matter of much interest, I give fig. 7 (p. 80) 

 as an illustration of their general character, which represents one 

 magnified about 100 linear. It must have been just like those so 

 common in deep -ocean mud. The original walls of the cells are 

 about -y-y(7o ^^ ^^ i^^^ thick, and are now partially, if not wholly, 

 devitrified ; and the empty cells are filled with the green substance 

 already named, just as the cells of roraminifera in more recent 

 rocks are filled with glauconite. 



Though there are several quite distinct kinds of these green 

 minerals, which possibly differ in composition, yet they form a 

 very definite group; all occur under similar circumstances, and 

 often intimately associated. On comparing different specimens of 

 slate we can see that those in which their amount is greatest are. 



