no. 1637. ON THE FORMATION OF GEODES— BASSLER. 139 



ently always have a fossil as their basis of formation. Under the (lis 

 cussion of mineral veins, Professor Branner, in his Syllabus of Ele- 

 mentary Geology (p. 226), cites geodes as examples of enlargements 

 of fractures by the expansion due to the crystallization of minerals 

 in the incipient crevice. On page 228 of the same work he figures 

 crinoid stems broken asunder, describing the figures as "Geodes 

 formed in the stem of a crinoid. The deposition of quartz began in 

 the hollow stem which was finally broken asunder." As may be 

 observed from later remarks, the observations of the writer agree 

 with Professor Branner's explanation save in the place of deposition 

 of the silica. The subject of the force of growing crystals has been 

 discussed by Merrill in 1895 " and by Becker and Day in 1905. 6 



The majority of geodes in the Knobstone group may be traced 

 directly or indirectly to a crinoidal origin for the simple reason that 

 these strata are often crowded with fragments of this class of organ- 

 isms. Probably next, in order as a geode maker is the common 

 brachiopod Athyris lamellosa, but no class of fossil is exempt from 

 replacement by silica when the proper conditions obtain. Fragments 

 of crinoid columns particularly are so abundant that series showing 

 the complete development of geodes are readily obtained. These 

 crinoidal columns are made up of a series of segments or " buttons " 

 pierced centrally by a rounded or five-sided hole or canal. Professor 

 Branner believed that the deposition of silica started in this canal 

 and continued until the column was fractured by the expansive 

 force of crystallization. This is a very reasonable assumption, but 

 none of the many unfractured crinoid columns before the writer 

 shows a trace of silica in the central canal. These crinoidal columns 

 are particularly subject to fracturing in the shale because of* this cen- 

 tral canal. The shape of the canal, as well as the five-sided symmetry 

 prevailing in the class, usually causes five longitudinal fractures 

 to form when the crushing is direct (see Plate XX. figs. 2 and 3). 

 If pressure is exerted obliquely upon the column, a slipping of the 

 segments upon each other is likely to occur (Plate XX, fig. 4). 

 although fracturing may also accompany this slipping (Plate XX, 

 fig. 5). All of the geodes originating in crinoid columns can be 

 traced back to such fractured stems, and, also, only such fractured 

 specimens occupying a position of ready access to water shown in the 

 sketch on page 130 were geodized. Numerous fragments similar to 

 the originals of figs. 1-6 on Plate XX, can be dug up from the im- 

 pervious shales and no signs of silicification observed, but in the 

 water horizons almost all the fractured specimens show some trace 

 of silica. It therefore seems apparent that the deposition of silica is 



a On the formation of stalactites and gypsum incrustations in caves, Proc. 

 T T . S. Nat. Mas.. XVII, pp. 77-sl. pis. n-v. 



6 The linear force of growing crystals, Proc. Washington Acad. Sci., VII, pp. 



283-288. 



