664 SCIENCE PROGRESS 



are lightly dismissed from serious consideration, and we are asked to nail 

 down all the properties of all the elements to simply the geometrical form 

 of practically solid atoms ! 



Briefly, the argument is that the facts of crystallisation point to " differ- 

 ences of form as the most natural explanation of this great diversity of 

 character " among the elements, since the electronic " conception of atomic 

 structure appears to be at variance with our knowledge of crystal structure, 

 and fails to assist us with the many problems of organic chemistry by which 

 we are still confronted." To the element carbon, with the structure of the 

 compounds of which the book mainly deals, the author has assigned the 

 form of an irregular tetrahedron of special dimensions (the carbonoid), and 

 from such a fundamental unit he has arrived at models to represent many 

 well-known organic compounds. But in spite of the ingenuity displayed 

 in the construction of these models — and it cannot be denied that they 

 have been built up with much care and are beautifully reproduced in the 

 tigures — it must be confessed that many of the arguments on which they 

 are founded are deplorable. If there is one thing in this world "at vari- 

 ance with our knowledge of crystal-structure," surely the author has hit 

 upon it in his theory of atomic form. We cannot believe that he has ever 

 seriously considered the long-established treatment of geometrical crystal- 

 lography or the recent analyses of well-known crystal-structures which have 

 been accomplished by the method of X-rays. Otherwise he would never 

 have allowed himself to write : " The simplest form of the Diamond is a 

 tetrahedron, not the irregular tetrahedron of the carbonoid, but one that 

 could be used for either an octahedron or a cube ; the base is an equilateral 

 triangle, and the three other sides are right-angled triangles " : or, " Can 

 a space-lattice be built up to correspond with the structure of the Diamond ? 

 This can be done, but the number of carbonoids employed is so enormous 

 that we have been compelled to resort to the use of hexagonal cards to 

 represent the Benzene nucleus." But the book is full of similar statements. 



May we point out to the author that Bragg's analysis of the diamond 

 structure has shown conclusively that the carbon atom does not resemble 

 the carbonoid but possesses a trigonal axis, and that the fact that benzene 

 crystallises in the orthorhombic system with two molecules per cell is almost 

 final evidence that, in crystals at least, the benzene ring does not possess 

 hexagonal but only a fourfold symmetry at most ? 



In the words of the author, " much difficulty and uncertainty surrounds 

 the theory of Atomic Form when applied to other elements," and " any 

 attempt to apply the Theory of Atomic Form to the elements generally 

 brings us face to face with problems of some dif&culty." 



W. T. A. 



The Origin oJ Spectra. By Paul D. Foote and F. L. Mohler. [Pp. 250, 

 with 46 figures.] (New York : The Chemical Catalogue Company, 

 1922. Price $4.50.) 



All students of spectroscopy are aware that in the last few years a great 

 deal of important work on ionisation and resonance potentials has been 

 carried out, but many will owe to this book a full realisation of the extent 

 and significance of these investigations. From the reputation of the authors 

 in this field one would have anticipated no less, but actually they have 

 achieved much more than a summary of researches of this kind. In fact, 

 the treatment of the subject is nearly as broad and comprehensive as the 

 title would lead one to expect, and the authors give us so much that is new 

 and suggestive that their apology for the omission of band spectra and 

 Zeeman and Stark effects from their programme will be very readily accepted. 

 Chapter I contains a very useful summary of the present position of the 

 quantum theory of spectroscopy. In several respects this is an advance 



