April II, 1889] 



NATURE 



555 



tution. In 1845, the scheme for the creation of the Queen's 

 Colleges was launched, and their Presidents and Vice- 

 Presidents were appointed as a preparatory step to the 

 building of the Colleges. Andrews at this time received 

 his appointment as Vice-President, and he was the first 

 Professor of Chemistry, both which offices he retained till 

 the failure of his health in 1879. His whole life-time was 

 thus— with the partial exception of his medical practice 

 from 1836 to 1845— occupied in scientific teaching and in- 

 vestigation ; and from the early age of fifteen, when he 

 published in the Philosophical Magazific a paper on the 

 action of a blowpipe flame on other flames, he never 

 ceased to devote himself to original research. 



He was never in a hurry to rush into print, but took 

 care to be accurate and thorough in an investigation 

 before announcing his results. Regularly day by day he 

 was to be found at work for hours in his laboratory, 

 patiently conducting with his own hands every detail of 

 his elaborate observations. 



The most important of his researches relate to heat of 

 combination, the nature and properties of ozone, and the 

 transition of such substances as carbonic acid from the 

 gaseous to the liquid state. But besides these, we find 

 in the present collection brief papers on a variety of 

 subjects. One contains an account of experiments on 

 the conducting power of flame for electricity, showing 

 that the current from a single cell could be transmitted 

 through a circuit part of which consisted of an alcohol 

 flame. Another describes the attainment of a very high 

 vacuum by a good ordinary air-pump aided by the intro- 

 duction of carbonic acid into the receiver and the 

 absorption of the last traces of this gas and of aqueous 

 vapour by caustic potash and sulphuric acid. Another 

 gives a comparison of the conducting powers of different 

 gases for heat, as shown by their cooling action on a 

 platinum wire kept incandescent by a current. In all 

 these subjects he was early in the field, and obtained 

 results much in advance of those obtained by his 

 predecessors. 



An instance of his careful criticism is afforded by one 

 of the latest papers in the volume— a lecture on recent 

 improvements in magneto-electric machines— in which he 

 points out that Paccinotti's machine, if the inventor's 

 original description of it can be relied on, makes its 

 contacts (for collecting the currents from the ring; in the 

 wrong places. The criticism is certainly justified by the 

 figure and accompanying description which are repro- 

 duced from Paccinotti's paper ; but from inquiries which 

 we made at the Paris Electrical Exhibition, where the 

 machine was on view, we believe the fault was in the 

 description and not in the machine itself. 



As regards ozone, Andrews appears to have been the 

 first to establish the following points : — 



(i) That the peculiar substance obtained by the action 

 of the electric spark on oxygen is identical with that 

 obtained in the electrolysis of water, and with that 

 obtained in the slow oxidation of phosphorus. 



(2) That it is not a compound body, but is oxygen in an 

 altered or allotropic condition. 



In subsequent experiments, with the assistance of Prof. 

 Tait (who was at that time Professor of Mathematics in 

 Belfast), he compared the amount of contraction produced 

 by the partial conversion of oxygen into ozone with the 



amount of ozone thus obtained as tested by chemical . 

 action, and hence deduced the density of ozone. 



His researches in various branches of the subject of 

 heat of combination were spread over many years, and 

 were, for the most part, conducted at a date when an 

 accurate thermometer for measuring small differences of 

 temperature could only be obtained by making it for 

 oneself. The subsequent researches of Favre and Sil- 

 bermann, wherever they differed to any large extent from 

 his, have since been shown to be erroneous, and his 

 results agree fairly well with the latest and best 

 determinations yet obtained. 



But his permanent fame will rest mainly on his dis- 

 covery of the continuous transition which can be made 

 from the gaseous to the liquid state of a substance, or from 

 the liquid to the gaseous. The main result which he 

 established is best set forth by the geometrical illustra- 

 tion employed by his colleague, Prof. James Thomson. 

 Let the volume, pressure, and temperature of a given mass 

 of carbonic acid be represented by the three rectangular 

 co-ordinates of a surface ; volume being represented by 

 height, while the pressure and temperature co-ordinates 

 are horizontal. The surface will resemble the side of a 

 mountain which is precipitous in one part, but in another 

 part furnishes a gradual ascent by which the summit 

 can be reached. The ground-plan of the precipice is 

 the curve of boiling-points, and the height of the pre- 

 cipice at each point of the curve represents the increase 

 of volume in passing from the liquid state to the gaseous. 

 As the temperature increases, the precipice diminishes 

 in height, and finally runs off to nothing at a point 

 whose horizontal co-ordinates are the " critical tempera- 

 ture" and "critical pressure." At higher temperatures 

 there is no boiling-point, and in place of the precipice 

 there is a gradual ascent, by means of which the precipice 

 can be rounded and the summit attained. Starting from 

 the ground below the precipice (that is, from the liquid 

 state), the ground above the precipice (tl at is, the 

 gaseous state) can thus be attained without any kind 

 of discontinuity. 



Cagniard de Latour and Drion, who preceded Andrews 

 in this field of research, failed to obtain this result, 

 because their method of experiment placed only one 

 independent variable at their disposal. The substance 

 was inclosed in a sealed tube, and there was no way of 

 altering its pressure except b/ altering its temperature. 

 Andrews used a screw plunger, which enabled him to 

 increase and diminish the volume independently of tlie 

 temperature. By making simultaneous m3asures of 

 volume, pressure, and temperature for various values of 

 the two independent variables, he was able to map out 

 the surface, and Prof J. Thomson constructed a wooden 

 model of it from the data thus obtained. 



In subsequent researches he investigated the effect of 

 mixing various quantities of nitrogen with carbonic acid, 

 and found that the critical point was largely shifted by 

 su^ch admixtures. 



Shortly before his health broke down, he devised an 

 ingenious apparatus for making successive additions of a 

 known column of mercury to the pressure of a gas confined 

 in a long tube, hoping in this way to be able to test de 

 partures from Boyle's law at very high pressures without 

 employing a column of mercury of unwieldy length. In 





