273 



and under the microscope it appeared to be made up of crys- 

 talline plates. In the softer varieties, which are easily reduced 

 to powder, there appear crystals with well-defined edges which 

 resemble those in the published descriptions of kaolinite ; the 

 majority of them, however, being elongated, and the hexagonal 

 plates appearing more rarely. Analysis, however, shows a 

 discrepancy between the composition of this mineral and that 

 of kaolinite. Four samples were analyzed ; I. and II. are a 

 hard compact variety, and III. and IV. a compact, granular 

 variety. In some cases the results given are the mean of sev- 

 eral analyses." V. is the average composition of kaolinite. 





I. 



II. 



ni. 



IV. 



V. 



SiOa 



76.63 



80.03 



82.52 



84.54 



46.4 



A1203 



17.25 



14.31 



13.75 



11.77 



39.7 



H20 



5.89 



5.13 



4.24 



3.48 



13.9 



Na^O 



.07 











K,0 



.66 











100.50 99.47 100.51 99.79 100.00 



The high silica ratio which these analyses show appears to 

 be consistent only with the theory that this substance, notwith- 

 standing its homogeneity, is a mixture of some hydrous alu- 

 minous silicate, such as kaolinite, with free silica. It is prob- 

 ably, like the pinite, a decomposition product ; and I would 

 suggest that such a chemical composition might result from the 

 kaolinization of an acidic petrosilex. 



DEVONIAN? 



The rocks here referred doubtfully to the Devonian system 

 are confined to the Naragansett basin, skirting the crystallines 

 of north-eastern Rhode Island with a breadth of one to two 

 miles and composing the principal part of the band of uncrys- 

 tallines reaching from Rhode Island north-easterly to Braintree, 

 where they nearly, perhaps actually, connect with the deposits 

 of the Boston basin. The true stratigraphic relations of the 

 two basins at this point cannot be observed on account of the 



OCCAS. PAPERS B. S. N. H. — IIT. 18 



