26 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 7-1 



quantities of magnetite, measurable by the thousands of pounds. 

 The iron chloride vapors were continually swept out by the con- 

 tinuous passage of steam, reacting at the mouth of the fumarole 

 with this steam to form magnetite and hydrochloric acid. As the 

 fumarole decreased in temperature the magnetite thus formed was 

 reattacked by the acid and the deposit of iron ore eventually dis- 

 appeared. While there seems little evidence to support such a 

 hypothesis for Cerro Mercado, yet the occurrence of the iron min- 

 erals there in tuffs, as at Katmai, and the physical resemblance of 

 the Katmai magnetite with much of the martite ore of Mercado 

 is striking and such a hypothesis may well be borne in mind. 



An origin by the replacement of the inclosing rocks by iron bear- 

 ing solutions is suggested by a number of features of the iron de- 

 posits of Cerro Mercado and is the hypothesis most favored by the 

 writer. The impression that one first receives upon examining the 

 ore bodies is that they are deposited by replacement but close con- 

 sideration of the matter shows that all points are not entirely clear. 

 This, however, resolves itself down largely to lack of definite 

 knowledge concerning the chemical and physical behavior of iron 

 bearing solutions. 



The direct replacement of the brecciated country rock by iron 

 ore has not been observed by the writer but the replacement of the 

 fragments by pyroxene is common and the close relationship of the 

 pyroxene to the iron ores makes it probable that the ore minerals are 

 a result of this same process- There has been observed in specimens 

 from the locality called Labores de la Cueva a replacement of the 

 breccia fragments by calcite which is in turn replaced by magnetite. 

 Further the schlierenlihe masses of ore in the tuffs at Socavon de la 

 Cueva can only be explained as a replacement of the tuffs by ore and 

 not by any means of injection. The puzzling feature of this occur- 

 rence, however, is that in a mass as susceptible to chemical change as 

 a tuff, there is no appreciable effect accompanying the introduction of 

 the iron ores. 



The alteration of the wall rock to pyroxene must certainly have 

 been brought about by aqueous solutions but whether these solutions 

 were the iron-bearing solutions or solutions given off by a magnetite 

 mag-ma there is no way of determining. The amount is less than one 

 would expect if a large volume of solution had passed, yet a large 

 amount of porphyry has been completely altered to montmorillonite 

 although not pyroxenized. Wliat the relation to the extensive 

 silicification of the rhyolites at the north side of the hill has to the 

 ore-bearing processes is still to be determined. 



The nature of the solutions that may have carried the iron and 

 replaced the wall rock is a matter of speculation and will not be dis- 



