April 28, 1899.] 



SCIENCE. 



611 



■ber of the elements is limited, and why- 

 there are not as many kinds of matter as 

 there are different wave-lengths of light — all 

 these seem to belong as yet to a scientific 

 dreamland rather than to the realm of 

 legitimate research, yet their solution, if 

 possible at all, will be accomplished only by 

 the labors of the inorganic chemist. 



Let us now turn to the more special con- 

 sideration of the questions of the constitu- 

 tional formulas of inorganic compounds. 

 The more conservative organic chemists 

 have always been careful to state that the 

 so-called structural formulas are reaction 

 Jormulas merely, that is, that they are not 

 intended to express the actual relations of 

 the atoms in the molecule, but are merely 

 convenient schemes for rendering possible 

 reactions visible to the eye. Probably most 

 chemists regard them as more than this, as 

 actual diagrammatic representations of the 

 way in which the atoms are combined. 

 The formula of marsh gas, for example, 



H 



I 

 H— C— H, 



is regarded as more than a visualizing of its 

 chemical properties ; it implies that the 

 carbon atom is an actual physical link be- 

 tween the hydrogen atoms, which are com- 

 bined directly with the carbon but not with 

 each other. Stereochemical formulas are 

 confessedly more than reaction formulas, 

 and the steric conception of the so-called 

 double and triple union asserts that these 

 actually exist in the sense the words imply, 

 and are not merely names for unknown 

 conditions. 



Many of the simpler organic structural 

 formulas uuquestionably^have an enormous 

 mass of evidence in their favor, but many 

 others we must be on our guard against 

 taking too seriously, and must for the pres- 

 ent regard as nothing more than reaction 



formulas. That we can regard any of them 

 as well established is due, more than to 

 anything else, to the almost invariably con- 

 stant tetravalency of the carbon atom. 

 Unfortunatelj', the valency of many of the 

 elements entering into the composition of 

 inorganic compounds appears to be ex- 

 tremely variable and uncertain, and this 

 has greatly impeded the study of the struc- 

 ture of these bodies. The inorganic chemist 

 has been far too prone to assume that the 

 structural theories of the organic chemist 

 are of universal applicability, and, having 

 once for all attributed a certain valency to 

 an element, has been often content with 

 devising sti-uctural formulas which have no 

 better claim to recognition than that all the 

 so-assumed bonds are ' satisfied.' At other 

 times a particular valency has been as- 

 sumed for no other reason than that it en- 

 abled him to contrive a formula for the 

 special case under consideration. The books 

 treating of such matters frequently exhibit 

 wonderfully ingenious inorganic structural 

 formulas which are wholly devoid of a rea- 

 sonable amount of experimental evidence 

 and which are, therefore, often nothing but 

 pure rubbish. With many inorganic chem- 

 ists, formula worship has degenerated into 

 fetishism. Let us consider a few examples. 

 For nitric acid, one of the simplest and 

 most familiar inorganic compounds, several 

 constitutional formulas may be written, in 

 which the hydrogen is directly united to 

 the nitrogen or sepai-ated from it by 

 one or two oxygen atoms, and in which 

 nitrogen may be either tri- or pentava- 

 lent. Some of these are given in the 

 books as if they were gospel truth. Briihl, 

 who has investigated the question by phys- 

 ical methods, suggests that the hydrogen 

 atom is not directly united to any part 

 of the NO, radical, but is rotating around it 

 and possibly combined with each oxygen 

 atom in succession, a view approaching that 

 of Werner. There are at least five formulas 



