SCIENCE. 



[N. S. Vol. IX. No. 234. 



Students, and even older chemists, who 

 find difficulty in recalling the atomic 

 weights of common elements will scarcely 

 welcome the proposition to give to each ele- 

 ment another factor, though in the case of 

 antimony this objection disappears. 



This alphabetical arrangement of the ele- 

 ments prevents carrying out one of the 

 prime objects of classification, namely, the 

 grouping of related matters ; thus 

 02J0 = Cobalt, 0500 = Nickel, 

 0690= Sulphur, OnO = Tellurium. 



The natural group Ba, Ca and Sr, have re- 

 spectively the unrelated numbers 0150, 

 0320 and 0680. Surely the elements might 

 have been arranged systematically, so that 

 related bodies would have contiguous sym- 

 bols. 



Annexed to the table of elements are in- 

 structions for sub-dividing entries and the 

 following paragraph : " Specific entries re- 

 lating to the halogens collectively shall be 

 arranged in Division 0350 uuder Halogens.^' 

 This mixing of a word-heading with nu- 

 merical symbols is a weak feature to be 

 again noticed. 



The instructions for entering titles in 

 sub-divisions of 0100 include tlie following 

 paragraph : " Entries relating to com- 

 pounds, which in the Slip Catalogue bear 

 the number and symbol of the dominant 

 element, together with the symbol of the 

 secondary constituent, or dominant second 

 constituent, shall be printed in the sub- 

 division of their second constituent." If 

 we understand aright this rather obscure 

 sentence, it provides for writing on slips 

 according to one rule and for printing them 

 in book-form according to another rule ; 

 sodium chloride would appear, therefore, 

 under the symbol for sodium in the written 

 slips and under chlorine in the printed 

 volume ! 



A second paragraph provides that " refer- 

 ences to hydroxides, acids and salts shall 

 be entered under the oxide, and corre- 



sponding sulphur compounds under the 

 sulphide." 



A third paragraph reads as follows : " {d) 

 In each sub-division the entries shall be 

 arranged in such order that those relating, 

 a, to the history or origin of the substance 

 come first, and following these, in the order 

 mentioued, those relating, /5, to its prepara- 

 tion or manufacture ; y, to its structure, or 

 of a theoretical nature ; <5, to its interactions 

 or use ; £, to its compounds — these five sev- 

 eral sections being denoted by the letters a, 

 /?, r, <5, ^." 



Passing without comment this non-pars- 

 able English (which occurs elsewhere in the 

 report), the scheme introduces another 

 arbitrary feature, Greek letters for specific 

 subjects, which is an admission that the 

 numerical plan is found insufficient ; though 

 it need not be, provided decimals were 

 used, a plan which does not seem to be con- 

 templated by the Committee. The sugges- 

 tion is even made that " it would be possible 

 to carry the analysis still farther by means 

 of symbols, such as <,'', z, and so forth, to 

 indicate physical properties, crystalline 

 form, etc." The writers of this review ven- 

 ture to suggest that when the Greek alpha- 

 bet is exhausted the Hebrew will come in 

 bandy. 



This mixture of numerical symbols with 

 word-headings is again resorted to in di- 

 vision ' 0930 Operations in inorganic chem- 

 istry,' where it is suggested that "entries 

 shall be made under significant headings, 

 such as dissolution (sic) and solvents, 

 crystallization i distillation, * * * oxi- 

 dation, electrolysis, furnace operations, 

 etc., arranged alphabetically." 



To organic chemistry the symbol 1000 is 

 assigned, under which all entries shall be 

 arranged that relate to the subject gen- 

 erally ; substitution derivatives of the com- 

 pounds included in each of the numbered 

 divisions — especially haloid and allied de- 

 rivatives — shall, as far as possible, be en- 



