THE IRON-BEARING MEI^IBER. 205 



presence is indicated by these rhombic crystal sections. The mao;'netite, 

 when present, is conmionly rhombic in outline. Such rliomljic outlines 

 may, of course, be produced by random sections of the ordinary mag'netite 

 octahedra, but the question has suggested itself as to wliether it is not 

 possible that these rhombic magnetite sections express in their shape the 

 outlines of carbonate crystals. In some cases such magnetite crystals witli 

 similar shaped sections of carbonate and hematite are found in the same 

 rock. 



In one direction these rocks may, then, be traced into those of the tirst 

 type. In the reverse direction by concentration of the iron oxides and by 

 the development of a concretionary and brecciated character this phase 

 passes into the second phase of the second type of rock. 



In the ferruginous cherts the iron oxides are often concentrated more 

 or less into regular bands, Ijut besides these bands there are many oval or 

 spherical bodies of iron oxide, so that the specimens are best described as 

 cherts containing bands and shots of ore. These shots occur in cavities in 

 the cherts and often they do not entirely fill them. In such cases the iron 

 oxide is usually lined with crystals of quartz. The structure is, upon a 

 small scale, that of a geode, the ca^nty of which has a laver of iron oxide, 

 and within this quartz crystals. Even when the cavities are completely 

 filled with iron oxide the same similai'ity to a geodic structiu'e is apparent. 

 It would seem that the cavities formed at some stage in the development 

 of the rock (perhaps by a solution of a part of the iron carbonate at the 

 time another part was oxidized, or else by solution of silica), were subse- 

 quently partly or completely filled ^vith iron oxide, after which, if space 

 remained, followed silica. (PI. xxiii. Figs. 1 and 2.) 



In this phase of rock the arrangement of the constituent particles is 

 often closely similar to that which has been described on a preceding page 

 as characterizing the chert of the Limestone member ; that is, it often j)re- 

 seuts a more or less peifect concretionary arrangement, but in the limestones 

 and cherts this is at times exceedingly vague. However, the concretions 

 are so numerous as to he one of the most important characteristics of this 

 phase of rock. (PI. xxii, Figs. 1 and i'.) The concretionary structure affects 

 both the iron oxide and chert, although it is most clearly made out by 



