228 REPORTS ON THE STATE OF SCIENCE. — 1920. 



case of every substance examined.^* Further than this, it is well known that 

 certain substances exhibit more than one absorption band in the visible or ultra- 

 violet, and it has been found that the frequencies of each of these absorption 

 bands are exact multiples of one and the same frequency characteristic of that 

 substance in the infra-red. It follows, therefore, that when a substance shows 

 more than two absorption bands in the visible or ultra-violet there must exist a 

 constant difference between the frequencies of consecutive bands, and this 

 difference must equal the fundamental infra-red frequency. This has also been 

 proved to be true. 



The application of the Planck theory has led to the discovery of relationships 

 between the frequencies of the absorption bands shown by a substance, relation- 

 ships which are of considerable importance because they form a quantitative 

 basis of molecular frequencies. It is not possible here to give the mathematical 

 development of Planck's theory, and the theory is only mentioned because it led 

 to the discovery of the relation between the frequencies. 



It is advisable at this point to discuss in some detail what is meant by the 

 frequency of an absorption band and also the influence of a solvent upon that 

 frequency. It is common knowledge that in many instances under high resolving 

 power an absorption band is found to possess a structure. The most common 

 phenomenon is when an absorption band consists of a series of sub-groups. In 

 this case one sub-group always exhibits a maximum absorptive power, and those 

 on either side exhibit decreasing absorptive power the farther they are situated 

 from the principal sub-group. Then, again, it is generally found by the 

 examination of the vapour of the substance that each of the sub-groups is 

 resolved into fine absorption lines, and that the arrangement of these lines as 

 regards their intensity is analogous to that of the sub-groups themselves. There 

 is always in each sub-group one line of maximum intensity, and the other lines 

 are arranged in series of decreasing intensity with regard to this central line. 



Now when a substance is cooled to low temperatures it is found that its 

 absorption bands become narrower, this being due to the suppression of the 

 outermost sub-groups. With further fall of temperature more and more sub- 

 groups disappear, and finally there is left only the principal line of the principal 

 sub-group. This absorption line persists even at the lowest temperatures yet 

 reached. It is perfectly evident therefore that this single frequency is truly 

 characteristic of the molecules, and that the other frequencies which make up 

 the breadth of the band are due to some cause connected with the temperature 

 of the molecules. There is, of course, no necessity to cool a substance to low 

 temperatures in order to recognise the true molecular frequency, because this 

 frequency is always that one for which the absorptive power is the gi'eatest in 

 the absorption band. In the quantitative relationships given above it is this 

 true molecular frequency which is referred to. 



It is perhaps not out of place to refer to the confusion that has arisen from 

 time to time from carelessness in nomenclature in dealing with absorption 

 spectra observations. The term ' band ' is applied to the whole region covered bv 

 one set of as.<;ociated groups or sub-groups. In the literature the word band 

 has been used when a sub-group of a band is meant, and thus considerable 

 confusion has been caused. 



The next point to be dealt with is the variation in absorption caused by a 

 solvent, a fact that is of material importance in connection with the Quantitative 

 relations between the molecular frequencies exhibited by a compound. Hartley 

 was the first to observe the difference in frequency of a particular absorption band 

 according to whether a substance is examined in the vapour etate or in solution 

 in a solvent, and he noted that there is always a small shift towards the red in 

 passing from vapour to solution. There are, in fact, two different effects of a 

 solvent npon the absorption spectrum of a substance as observed in the vapour 

 state. One of these has already been mentioned, namely, the appearance of an 

 entirelv different absorption band when the substance is dissolved. In this 

 case the vapour exhibits a molecular frequency which is one multiple of the 

 infra-red frequency, whilst the solution exhibits a molecular frequency which 

 is another multiple of that infra-red freouency. In the case of some comnounds 

 ''t Vas been shown that by tbo use of different solvents a number of diffprent 



fpultipka of the Infrgrred fundamental are called into play. 



