MISER AND FAIRCHILD: HAUSMANNITE 



able oxygen. By referring to a few published analyses of psilo- 

 melane, the average ratio for the analyses cited in table 2 is 

 seen to approach fairly close to the above ratio (1.5). 



TABLE 2 

 Oxygen Ratios of Psilomelane as Shown by Published Analyses 



No. Locality 



I Ilmenau, Germany 



2 Silver CliflF, Colorado 



3 Romaneche, France 



4 Schneeberg, Germany 



5 Balaghat, India 



6 Kajlidongri, India 



7 Batesville, Arkansas 



Average 1.41 



1. R. A. F. Penrose, Jr. Manganese — its uses, ores and deposits. Arkansas 

 Geol. Svirvey, Ann. Rept. 1890, i: 146. 1891. 



2. F. W. Clarke. Data of geochemistry. U. S. Geol. Survey BuU. 616: 534. 

 1916. 



3. Idem, 534. 



4. R. A. F. Penrose, Jr., op. cit., p. 146. 



5. Dana, System of mineralogy, third appendix, p. 39, 191 5. Sample of Hol- 

 landite. 



6. Idem. Sample of Hollandite. 



7. R. A. F. Penrose, Jr., op. cit., p. 147. 



The analyses of the samples of the mineral in question from 

 the Batesville district show the following calculated compositions: 



Psilomelane 

 (Mn2Mn05) 



Sample No. i 13 per cent 



Sample No. 2 31 per cent 



Hausmannite 

 (MnaOi) 



87 per cent 

 69 per cent 



Oxygen 

 ratio 



.65 

 •30 



The analyses show that silica was absent in No. i and that 

 there was only o.i per cent of it in No. 2. These analyses have 

 been confirmed by testing for gelatinous silica 12 other samples 

 from 9 different localities. No silica was found in most of them 

 and hardly more than a trace was found in the others. On the 

 other hand, similar tests were made on 5 samples of braunite 

 from as many different localities in the Batesville district., and 

 they all yielded fairly large quantities of gelatinous silica. The 

 test for gelatinous silica is, in fact, the easiest method for dis- 



