334 University of California Publications in Geology [Vol. ll 



The following analysis is given by Griswold as an average of sev- 

 eral typical novaculites : 



SiO = 



99.50% 



AL0 3 



0.20 



Fe 2 3 



0.10 



CaO 



0.10 



MgO 



0.05 



K.O 



0.10 



Na.O 



0.15 



Loss 011 ignition 



0.10 



Total 



100.30 



The specific gravity is 2.64. The hardness is that of quartz. 



Under the microscope, according to Griswold, these rocks are 

 found to consist principally of nearly pure quartz, arranged in a very 

 fine grained mosaic. No fibrous structure is apparent, with the ex- 

 ception of a few spots of fibrous chalcedony which are seen in the 

 more impure novaculites. No isotropic silica is found. The thin sec- 

 tions show an occasional cavity .05 millimeter in diameter. These pos- 

 sess rhombic outlines and are regarded as left by the solution of cal- 

 cite crystals. 



In addition to the typical novaculites, impure varieties are found. 

 The novaculites are associated with sandstones and shales. Some of 

 the shales are highly ferruginous and some contain notable amounts 

 of manganese. 



The thin bedded novaculites are often highly contorted. Plate 5 

 in Griswold 's report represents well bedded, siliceous shales which 

 show a wedging out of siliceous beds and also of the partings between 

 them, resembling greatly the bedding seen in the Franciscan chert. 



The origin of these rocks has been attributed to various agencies. 

 Griswold believed that the novaculites represented accumulations 

 of very fine grains of pure quartz, deposited on the sea floor at some 

 distance from land, which later became consolidated by the permea- 

 tion of silica. 



Derby believed that the novaculites were replacements of calcare- 

 ous sediments by silica. Rutley held a somewhat similar idea, regard- 

 ing the novaculites as silicified dolomitic limestones. 



Branner 89 suggested that the novaculites were metamorphosed 

 cherts. He thought it possible that they might have been derived 



8s Ann. Rep. Geol. Surv. Arkansas, vol. 1, p. 49, 1888. 



