1894.] 



Meteoric Stone found in New Zealand. 



143 



solution. Hence, determinations of the sulphur and phosphorus in 

 the mercurialised residue were made, so that the iron and nickel 

 which had passed into the mercuric solution from the affected troilite 

 and schreibersite might be allowed for. The mercurialised residue 

 was next treated with hydrochloric acid, and the solution analysed ; 

 the silica was extracted from the undecomposed residue by sodium 

 carbonate solution containing some caustic soda, and the small 

 amounts of bases simultaneously extracted were likewise determined ; 

 the undecomposed residue itself was then separately analysed. 

 Further, various determinations of the alkalies were made. 



The microscopical examination made by Professor Ulrich having 

 suggested no character or constituent different from those previously 

 met with in meteoric stones, the interest of the chemical investiga- 

 tion was rather in the study of an analytical method than in the 

 numerical results to be obtained. As pointed out by Dr. Friedheim, 

 in the memoir already referred to, recent analyses of fragments of 

 the Alfianello stone show enormous variations : they are probably in 

 great part due to the incompleteness of the separations of the con- 

 stituent minerals from each other. 



As a result of the observations made in the course of the analysis 

 of the Makariwa stone, the following points may be emphasised : 



1. It is advisable to first effect, as far as practicable, a magnetic 

 separation of the mineral constituents : otherwise, owing to the 

 malleability of the nickel-iron, the powdered material may be neither 

 sufficiently fine, nor sufficiently homogeneous as a mixture. Further, 

 without this separation, the appreciable proportion of the iron-rust 

 may escape observation. 



2. The solution of mercuric ammonium chloride, suggested by 

 Dr. Friedheim, is very satisfactory in being without appreciable 

 action on any other constituent than the nickel-iron, if used as 

 directed. 



3. If the attracted material be treated with acid without previous 

 extraction by means of mercuric solution, the composition deduced 

 for the nickel-iron may be completely wrong, owing to the solution of 

 iron-rust simultaneously with the nickel-iron. 



4. After reduction of the rust in the unattracted material by 

 heating to low redness in hydrogen, and after subsequent treatment 

 with the mercuric solution, no troilite or schreibersite will be left in 

 the residue if the operations are sufficiently prolonged. There is an 

 appreciable effect on the silicate portion during the operations. 



5. The enstatite can be completely freed from olivine by three 

 extractions with hydrochloric acid (sp. gr. 1'06) on the water- bath 

 probably by two. 



6. The Deville-Cooke method is very advantageous for the separa- 

 tion of the small quantities of aluminium and chromium from the iron. 



