576 



SCIENCE 



[N. S. Vol. XXXIX. No. 1007 



views he expresses, it seems to me that his 

 plan of treatment of isomorphous series is 

 likely to complicate rather than simplify 

 matters. Professor Rogers proposes that min- 

 erals be named by the predominant molecule of 

 the isomorphous series, that isomorphous ad- 

 mixtures be indicated by adjectives, and that 

 chemical prefixes be restricted to artificial 

 substances. It would appear to me preferable 

 to have mineral names apply to the isomorph- 

 ous series as a whole, and, in most cases, to 

 use chemical prefixes for end-members of 

 series, whenever it is desired to discuss their 

 relations. 



The difference between these methods can 

 best be brought out by citing a few examples. 

 For instance, according to Rogers' plan the 

 term carnotite would apply only to a (theo- 

 retical) potassium end-member of the iso- 

 morphous series of double uranyl vanadates, 

 tyuyamunite to the calcium end-member. 

 Specimens in which both are present would 

 have to be called calciferous carnotite, if 

 potassium predominated, and kaliferous 

 tyuyamunite if calcium predominated, and 

 indeed an analysis would be necessary before 

 the material could be correctly named at all. 

 According to my plan the name carnotite 

 would apply to the isomorphous series itself, 

 and could then continue to be used by laymen — 

 miners, dealers, newspaper writers — to whom 

 the exact composition is of no significance, to 

 describe any sample of the mineral; while the 

 compound names calcio- and kalio-carnotite 

 would be used by the specialist discussing the 

 composition of a given specimen, or its rela- 

 tions with other minerals. 



The first plan would also do away with a 

 number of well-known, useful, mineral 

 names; — with axinite, columbite and wolfra- 

 mite for instance, because they are made up 

 of iron and manganese end-members — and, in 

 general, with every name of a mineral as soon 

 as it is discovered to be an isomorphous mix- 

 ture. It would further require the introduc- 

 tion of two new names, for the end-members, 

 in most cases. According to the second, on the 

 other hand, advance of knowledge would not 

 necessitate either the discarding of estab- 



lished names, nor the introduction of wholly 

 distinct, independent, arbitrary ones, but only 

 the addition of appropriate chemical prefixes, 

 yielding ferro- and mangano-axinite, ferro- 

 and mangano-columbite, etc. Some of these 

 compound names have already been intro- 

 duced, and have proved very useful in discus- 

 sions of the relationships of the minerals; 

 and they can cause no ambiguity so long as 

 it is definitely understood that they apply to 

 end-members of series only. Natronorthoclase 

 in the sense of orthoclase containing sodium 

 replacing part of the potassium, cobaltocalcite 

 in that of calcite containing a little cobalt, 

 etc., would, however, have to be dropped (as 

 also recommended by Rogers). 



In the few cases where the end-members of 

 a series are common, well-known and well-char- 

 acterized minerals, convenience may demand 

 that independent names be retained as syno- 

 nyms of these compound ones. Thus while 

 wolframite remains the name for the iso- 

 morphous mixtures of iron and manganese 

 tungstates, ferberite and hiibnerite may at 

 times be used instead of ferro- and mangano- 

 wolframite, since they are sufficiently charac- 

 teristic to be recognized by miners, dealers and 

 beginning students. Yet I have found that 

 practically every mineralogist, when asked 

 what ferberite is, will reply " wolframite high 

 in iron " rather than " iron tungstate." So, 

 even if the distinct names are retained in this 

 case the compound chemical ones had better 

 be used synonymously with them. 



My suggestions are then, in short, that when 

 a mineral is discovered to be an isomorphous 

 mixture of two "(or niore) components the 

 species name shall be regarded as a group 

 name, representing any and all members of the 

 series; that the end-members in general shall 

 be designated by chemical prefixes applied to 

 the species name; but that if the end-members 

 are so distinct, characteristic and well known 

 that they can be thought of independently, 

 and already have separate names, that these 

 names be retained as synonyms of the com- 

 pound, chemical, ones. If this plan should be 

 systematically applied it would simplify 

 mineral nomenclature with a minimum of 



