472 



NATURE 



[September 17, 1903 



thirty years or more the depression has been closely watched, 

 and thousands of successful forecasts have been based upon 

 a knowledge of its habits. But unfortunately the travelling 

 depression cannot be said to preserve its identity in any 

 sense to which quantitative reasoning can be applied. As 

 long as we confine ourselves to a comparatively small region 

 of the earth's surface the travelling depression is a real 

 entity, but when we widen our area it is subject to such 

 variations of path, of speed, of intensity, and of area that 

 its use as a meteorological unit is seriously impaired, and 

 when we attempt to trace it to its source or follow it to 

 its end it eludes us. Its origin, its behaviour, and its end 

 are almost as capricious as the weather itself. 



Nor if we examine other cases in which a veritable entity 

 is transmitted can we expect that the simple barometric 

 distribution should be free from inexplicable variations. 

 We are familiar with ordinary motion, or, as I will call 

 it, astronomical motion, wave motion, and vortex motion. 

 Astronomical motion is the motion of matter, wave motion 

 the motion of energy, vortex motion the motion of matter 

 with energy, but the motion of a depression is merely the 

 transmission of the locus of transformation of energy ; 

 neither the matter nor the energy need accompany the de- 

 pression in its motion. If other kinds of motion are sub- 

 ject to the laws of conservation of matter and conservation 

 of energy, the motion of the depression must have regard 

 also to the law of dissipation of energy. An atmospheric 

 disturbance, with the production of rainfall and other 

 thermal phenomena, must comply in some way with the 

 condition of maximum entropy, and we cannot expect to 

 account for its behaviour until we can have proper regard 

 to the variations of entropy. But the conditions are not 

 yet in a form suitable for mathematical calculation, and 

 we have no simple rules to guide us. So far as Meteorology 

 is concerned, Willard Gibbs unfortunately left his work 

 unfinished. 



When the cyclonic depression was reluctantly recognised 

 as too unstable a creature to carry the structure of a general 

 theory Mr. Galton's anticyclones, the areas of high pressure 

 and descending currents, claimed consideration as being 

 more permanent. Prof. Koppen and Dr. van Bebber have 

 watched their behaviour with the utmost assiduity and 

 sought to find therein a unit by which the atmospheric 

 changes can be classified ; but I am afraid that even Dr. 

 van Bebber must allow that his success is statistical and 

 not dynamical. " High pressures " follow laws on the 

 average, and the quantity we seek is not an average but 

 an individual. 



The question arises, whether the knowledge of the 

 sequence of weather changes must elude us altogether, or 

 will yield to further search. Is the man in the street right 

 after all? But consider how limited our real knowledge of 

 the facts of atmospheric phenomena really is. It may very 

 well be that observations on the surface will never tell us 

 enough to establish a meteorological entity that will be 

 subject to mathematical treatment ; it may be that we can 

 only acquire a knowledge of the general circulation of the 

 atmosphere by the study of the upper air, and must wait 

 until Prof. Hergesell has carried his international organisa- 

 tion so far that we can form some working idea therefrom 

 of general meteorological processes. But let us consider 

 whether we have even attempted for surface meteorology 

 what the patience of astronomers from Copernicus to Kepler 

 did for astronomy. 



Do we yet fully comprehend the kinematics of the 

 travelling depression ; and if not, are we in a satisfactory 

 position for dealing with its dynamics? I have lately ex- 

 amined minutely the kinematics of a travelling storm, and 

 the results have certainly surprised me and have made it 

 clear that the travelling depressions are not all of one 

 kinematical type. We are at present hampered by the 

 want of really satisfactory self-recording instruments. I 

 have sometimes thought of appealing to my friends the 

 professors of physics who have laboratories where the read- 

 ing of the barometer to the thousandth of an inch belongs 

 to the work of the " elementary class," and of asking them 

 to arrange for an occasional orgy of simultaneous readings 

 of the barometer all over the country with corresponding 

 weather observations for twenty-four consecutive hours, so 

 that we might really know the relation between pressure, 

 rainfall, and temperature of the travelling depressions ; but 



NO. 1768, VOL. 68] 



I fear the area covered would even then hardly be large 

 enough, and we must improve our self-recording instru- 

 ments. 



Then, again, have we arrived at the extremity of our 

 knowledge of the surface circulation of the atmosphere? 

 We know a great deal about the average monthly distribu- 

 tion, but we know little about the instantaneous distribu- 

 tion. It may be that by taking averages we are hiding 

 the very points which we want to disclose. 



Let me remind you again that the thickness of the atmo- 

 sphere in proportion to the earth's surface is not unsatis- 

 factorily represented by a sheet of paper. Now it is obvious 

 that currents of air in such a thin layer must react upon 

 each other horizontally, and therefore we cannot a priori 

 regard one part of the area of the earth's surface as 

 meteorologically independent of any other part. We have 

 daily synoptic charts for various small parts of the globe, 

 and the Weather Bureau extended these over the northern 

 hemisphere for the years 1875 to 1879; but who can say 

 that the meteorology of the northern hemisphere is in- 

 dependent of that of the southern? To settle that primary 

 question we want a synchronous chart for the globe. As 

 long as we are unable to watch the changes in the globe 

 we are to a certain extent groping in the dark. A great 

 part of the world is already mapped every day, and the 

 time has now arrived when it is worth while to consider 

 what contributions we can make towards identifying the 

 distribution of pressure over the globe. We may idealise a 

 little by disregarding the local peculiarities without sacri- 

 ficing the general application. 1 have put in the exhibition 

 a series of maps showing what approximation can be made 

 to an isochronous chart of the globe without special effort. 

 We are gradually extending the possibility of acquiring a 

 knowledge of the facts in that as in other directions. With 

 a little additional enterprise a serviceable map could be 

 compiled ; and when that has been reached, and when we 

 have added to that what the clouds can tell us, and when 

 the work of the Aeronautical Committee has so far pro- 

 gressed that we can connect the motion of the upper atmo- 

 sphere with the conditions at the surface, when we know 

 the real kinematics of the vertical and horizontal motion 

 of the various parts of a travelling storm, we shall, if the 

 universities will help us, be able to give some rational, 

 explanation of these periodic relations which our solar 

 physics friends are identifying for us, and to classify our 

 phenomena in a way that the inheritors of Kepler's achieve- 

 ments associated with us in this Section may be not un- 

 willing to recognise as scientific. 



SECTION B. 



CHEMISTRY. 



Opening Address by Prof. W. N. Hartley, D.Sc, F.R.S., 

 F.R.S.E., President of the Section. 



The ofttimes laborious method of investigating the re- 

 lationship of substances by ascertaining how one form of 

 matter can operate upon another, in other words by chemical 

 reactions, has of late been supplemented by the examin- 

 ation of their physical properties, and has been extended 

 to compounds, both organic and inorganic. In several 

 directions this has led to results of very uncommon interest. 

 Accordingly I propose to offer a brief account of twenty-five 

 years' experimental work in that branch of chemical physics 

 which deals with the emission and absorption of rays of 

 measurable wave-length, and to review its present position 

 chiefly in relation to the theory of chemistry, indicating 

 where it may be usefully and profitably extended. 



According to Davy (" Chemical Philosophy," vol. i., 

 1812, p. 211), Ritter observed chemical action on moist 

 chloride of silver to be different in different parts of the 

 spectrum, slight in the red, greater towards the violet, and 

 entending into a space beyond the violet where there is no 

 sensible light or heat. .W'ollaston discovered that chemical 

 action was exerted by refracted rays in a region where they 

 were of a higher refrangibility than any rays that were 

 visible. Young showed that the invisible rays are liable 

 to the same affections as visible rays. Hence we have the 

 beginnings of spectrum analysis in its chemical relations 

 to terrestrial matter, in the infra-red, the visible, and the 

 ultra-violet regions. 



