THE METAL CARBONYLS.

When the metal carbonyls were first discovered, their properties were startling because they seemed to violate nearly all the previously recognized generalizations of chemistry. Even to-day the existence of the carbonyls is not particularly emphasized in elementary courses of chemistry because it is rather hard to reconcile them with the first presentations of the generalizations of chemistry. Nevertheless, as the student progresses deeper into the knowledge of chemistry it becomes desirable to include the knowledge of the carbonyls both because they become more comprehensible when viewed in the light of Werner's system of coordination and because they themselves contribute to the comprehension of the Werner theory. As long ago as 1931, Reiff in his discussion of cobalt nitrosyl carbonyl recognized the correlation between the effective atomic number and the volatility of carbonyls. A more recent study of charged Werner coordination complexes, that is, of complex ions, has shown a similar role of the effective atomic number. We are standing on fairly firm ground when we point out the correlation between E.A.N. and the volatility of the carbonyl complexes and the existence of complex ions. Be it noted that we have made no postulates as to the arrangement of the electrons in quantum levels. In the inert gases the outer principal quantum group is supposed always to contain eight electrons. In the carbonyls and other Werner complexes there is no compelling reason to suppose that the electrons in the coordinating layer, be this layer of eight, ten, twelve or sixteen electrons, are not all at the same energy level. Although we have confined our discussion almost exclusively to the property of volatility, the carbonyls are very interesting from the standpoint of several other properties, for example, magnetic susceptibility and dielectric constant. Enthusiasts in the interpretation of such properties try to draw conclusions as to the condition of the electrons, sometimes they become so dogmatic as to seem really.to believe in the actual existence of the condition they postulate. As Professor Smith said, "Theories come and theories go, but facts live on forever." The facts of chemistry are so multitudinous that we would be utterly helpless to use them had we not means of correlating them. Any postulates which reach beneath the surface of the directly observable to give a mechanism to correlate the facts are helpful. But a scientist without a sense of humor is pretty hopeless. Who in his right mind can regard as other than absurd the idea that an electron pair can simultaneously occupy positions in two atom shells to make up the supposedly necessary number in each atom? In fact, is not the electron itself a pretty ridiculous figment of the imagination? To be sure, we recognize the electron as a discrete entity with certain very definite properties, but in the light of comparison with any mechanism within our comprehension is not the electron perfectly impossible? By all means let us use a postulate which allows us to make a useful classification of facts, but never let us lose the sense to see how utterly ridiculous the postulate will look to one who has not like ourselves grown attached to it. We do not expect ever to discover the ultimate reason for things, but we do expect ever to make progress in correlating and classifying the facts which we have already discovered and shall continue to discover. Dogmatic belief in ridiculous postulates retards this progress.