November 3. 192 1] 



NATURE 



311 



tion of carbonate of lime in sea-water, and on 

 chemical changes in muds, etc. 



Murray's last oceanographic expedition was a 

 four-months' cruise in 1910 in the North Atlantic 

 with Dr. Hjort in the Michael Sars when in his 

 seventieth year. He was killed in a motor 

 accident in March, 1914. 



There is, no doubt, other Edinburgh work in 

 connection with oceanography, such as that of the 

 Fishery Board for Scotland, which should be men- 

 tioned, and other names of those w^ho are still 

 happilv with us and at work, such as the inde- 

 fatigable arctic and antarctic explorer. Dr. W. S. 



Bruce, the leader of the Scotia expedition, and the 

 founder of the Edinburgh Oceanographic Labora- 

 tory ; but in this brief record of the past it has 

 been possible only to deal simply with the his- 

 torically connected work of the three great 

 pioneers of the nineteenth century — Edward 

 Forbes, the dredger of the shallow waters ; 

 Wyville Thomson, the explorer of the deep seas ; 

 and John Murray, who may be regarded as the 

 founder of modern post-Challenger oceanography 

 — in demonstrating the effect of Edinburgh men 

 and ideas and work in advancing our knowledge 

 of the science of the sea. 



Absorption Spectra. 

 By Prof. E. C. C. Balv, C.B.E., F.R.S. 



\X WITHOUT doubt the study of absorption 

 * ^ spectra, more particularly those of organic 

 compounds, has given rise to great interest, owing 

 to the possible connection between absorption and 

 chemical constitution. The work of Hartley, 

 Dobbie, and others showed that in certain cases 

 it was possible to determine the constitution of 

 substances from observations of their absorption 

 spectra. It is not surprising that, as the result of 

 this work, a school of thought was founded on 

 the basis of a direct correlation between the atomic 

 structure of a molecule and its absorption spec- 

 trum. On the other hand, Hantzsch, who m a 

 great number of papers has maintained the opinion 

 that the absorption curve is an index to constitu- 

 tion, has travelled far beyond the original point of 

 view. He found that changes in the absorption 

 spectrum of a compound are observed when no 

 change is possible in its primary valency struc- 

 ture, and in interpreting his results Hantzsch has 

 invoked the aid of the secondary valencies of the 

 atoms. 



It is now known beyond any question of doubt 

 that one and the same substance under different 

 conditions can show different absorption spectra 

 n the visible and ultra-violet. It is also known 

 :hat, whilst a change in the primary structure of 

 the molecule might possibly be accepted as an ex- 

 planation of this in a few instances, the large 

 majority of these variations in absorption cannot 

 in any way be thus accounted for. Attempts were 

 also made to interpret absorption spectra by oscil- 

 lating linkages, such, for instance, as the equi- 

 librium between the enolic and ketonic forms of 

 ethyl acetoacetate or the oscillation that may be 

 accepted as taking place within the benzene ring. 

 This suggestion was very soon negatived when it 

 was found that many substances in which no such 

 oscillation is taking place exhibit well-marked ab- 

 sorption bands. For example, it might be pos- 

 sible to explain the ultra-violet absorption band 

 of acetone bv attributing it to the equilibrium 

 CH3— CO— CH3=CH3— C(OH)-CH2; but hexa- 

 methylacetone, in which no such change is pos- 

 sible, exhibits the same absorption band as 

 acetone. 



NO. 2714, VOL. 108] 



The only other possible variable from the point 

 of view of the chemist is the secondary valency, 

 and this has been advanced by Hantzsch and 

 others as the explanation of absorption in the 

 visible and ultra-violet, the well-established differ- 

 ences in absorption being accounted for by differ- 

 ent distributions of the secondary valencies. No 

 physical explanation, however, has been offered of 

 the assumed correlation between the secondary 

 valency and absorptive power, a matter of great 

 importance, since a theory cannot hold good 

 unless some physical basis can be found for the 

 phenomenon of the absorption of radiant energy. 



In all this work the study of absorption spectra 

 of organic compounds has been restricted to the 

 visible and ultra-violet regions of the spectrum, 

 and, indeed, only to that portion of the ultra- 

 violet "which is transmitted by a quartz spectro- 

 graph working in air. This is unfortunate, since 

 the absorption bands exhibited by the compounds 

 in that vast region known as the infra-red and in 

 the extreme ultra-violet are ignored. Then, again, 

 many inorganic substances show absorption bands 

 which are exactly similar to those on which 

 Hantzsch founds his valency formulae of organic 

 compounds, and obviously an identical explanation 

 must be found for each class of compound. It is 

 not too much to say that all the above theories 

 have been based on insufficient data. 



If the complete system of absorption bands 

 shown by a compound over the whole spectrum 

 is examined, it is found that the central fre- 

 quency — namely, the frequency for which the ab- 

 sorptive power is the greatest — in any visible or 

 ultra-violet band is always an exact multiple of 

 the frequency of an important absorption band 

 in the short-wave infra-red. Then, again, this 

 infra-red frequency is itself an exact multiple of 

 the central frequencies of well-marked bands in 

 the long-wave infra-red. This integral relation- 

 ship is of great importance, since it can readily 

 be proved that the central frequencies are truly 

 characteristic of the molecules, the subsidiary fre- 

 quencies associated with them being probably due 

 to the atoms and groups of atoms composing the 

 molecules. Again, the changes in absorption ex- 



