December i6, 1915] 



NATURE 



439 



a community of origin being thereby probably indi- 

 cated. M. P. Carrasco now points out {Comptes 

 rendus, vol. clxi., p. 631) that there is an additional 

 member of this series, the next earlier term being the 

 red line at A 63742. This line, unknown when Prof, 

 Nicholson was pursuing his investigations, was the 

 most important feature of the coronal spectrum as 

 l)hotographed at the late eclipse (August 21, 1914), and 

 M. Carrasco was one of the fortunate observers who 

 obtained a record of the line. 



Areas of AnsoRniON Markings on Sectrohelio- 

 GRAMS. — The results of the limb prominence observa- 

 tions made at Kodaikanal Observatory during the first 

 half of the current year are given in Bulletin No. 47. 

 The mean daily area is the largest since igo8, and 

 shows an increase of 59 per cent, over that for the 

 preceding six months. Prominences seen projected on 

 the disc as absorption markings on spectroheliograms 

 taken in the light of the Ha reversal are being specially 

 recorded with the grating spectroheliograph for 

 measurement. Mean daily areas in millionths of the 

 sun's visible hemisphere and mean daily numbers for 

 the five half-years 1912-13, and January-June, 1915, 

 are published in the above bulletin. For the second 

 half of 1913 the areas were sixty, whilst for the first 

 lialf of the present year they were 1375-6. The dis- 

 tribution in latitude of these markings shows the pro- 

 minence maximum between 50° and 60°, together with 

 a pronounced maximum about 30°, due to prominence 

 in spot latitudes, the equator being avoided, as in the 

 case of sun-spots. 



THE CORROSION OF METALS. 



ON December 8, under the presidency of Sir Robert 

 Hadfield, the Faraday Society held a discussion on 

 the corrosion of metals. Of the seven papers con- 

 tributed, only two dealt with the more general aspects 

 of this very important question. Three were con- 

 cerned with the corrosion of iron, and some of the 

 numerous steels which find application in modern 

 industrial life; the remaining two had reference prin- 

 cipally to marine condenser tube alloy, 70 : 30 brass. 

 As it turned out, there was almost no discussion on 

 the fundamental characteristioi of corrosion pheno- 

 mena. During the first hour instances of corrosion 

 among iron alloys came under review, the remainder 

 of the evening being devoted to a consideration of 

 the corrosion of copper-zinc alloys. 



As Dr. Rideal pointed out in his printed contribution 

 to the more fundamental aspects of the question, "the 

 phenomena observed in the corrosion of metals are to 

 he found scattered among the earliest records of man- 

 kind, and in consequence of the universality of the 

 subject we have received a heritage consisting of a 

 jumble of facts and theories." The first report by 

 Dr. Bengough to the Corrosion Committee of the In- 

 stitute of Metals consists mainly of a critical examina- 

 tion of the views held with regard to the cause, or 

 causes, of the corrosion of marine condenser tube 

 alloy, the general conclusion being that the evidence 

 is so conflicting that no particular view can be re- 

 garded as at all firmly established. The committee, 

 therefore, in planning the experimental investigation 

 decided that there was nothing for it but to begin at 

 the very beginning and take nothing for granted. 



Dr. Rideal's definition of corrosion is as follows : — 

 "Corrosion may be said to result from an irreversible 

 chemical change proceeding with a small velocity and 

 taking place on the common surfaces between two or 

 more phases, the products of which change are con- 

 tinually removed from the sphere of action." More- 

 over, it takes place generally on the surfaces of phases 

 which are electrically conducting, a fact which lies 

 at the base of the now generally accepted electrolytic 

 NO. 2407, VOL. 96] 



theory of corrosion. This theory requires the presence 

 of minute cells operating on the surface of the corroded 

 me^al or alloy. As yet, however, there is no informa- 

 tion as to the number of such cells or the rate at which 

 they work. 



Dr. Desch's contribution dealt with the influence of 

 physical and mechanical factors in corrosion, an aspect 

 of the subject the importance of which is by no means 

 always adequately realised. Although the process of 

 corrosion is probably in all cases initially one of 

 chemical solution, the physical heterogeneity of the 

 metal or alloy has a considerable influence on its 

 nature and velocity. More especially has this to be 

 considered when it is remembered that many of the 

 commonly used industrial alloys are in a "strained" 

 condition, and contain, as Dr. Beilby has shown, 

 films of amorphous material on the surfaces of slip of 

 the crystals. Such films have been demonstrated to 

 be more electropositive than the crystalline material, 

 a fact which determines the course of corrosion of 

 cold-worked metals in particular. The increased 

 corrodibility of such alloys is no doubt also partly to 

 be ascribed to the energy produced by work being 

 stored up in these films. 



The Cumberland process for preventing the corro- 

 sion of metals immersed in liquids, of which a demon- 

 stration was given at the close of the discussion, is 

 based upon a recognition of the galvanic nature of 

 this phenomenon. It consists in introducing a higher 

 counter-electromotive force to that causing the corro- 

 sive action. A continuous current working at 10 volts 

 is supplied to the anode, consisting of pieces of iron 

 suspended in the liquid and insulated from the vessel 

 being protected. It is claimed that this system has 

 been in use in all types of steamships and in many 

 large power plants, and that it is applicable to any 

 metal in contact with water or any other corrosive 

 liquid. H. C. H. Carpenter. 



VISCOSITY OF OILS. 



THE Institution of Petroleum Technologists is one 

 of the most recent of our technological associa- 

 tions. Founded in 1913, to advance the study of 

 mineral oils from the various points of view of the 

 chemist, the geologist, the engineer, the prospector, 

 and the financier, it has shown from the first a iiealthy 

 vitality and the promise of a vigorous future. At a 

 general meeting held on November 16, at the Royal 

 Society of Arts, the institution had the pleasure of 

 listening to an illuminating address by Dr. Glazebrook 

 on the viscosity of oils in relation to the rate of flow 

 through pipes. The tests described had been under- 

 taken at the request of the Admiralty, and permission 

 had been given for their publication. The results of 

 the investigation showed that the ordinary law of 

 viscous flow, VjW = 2r\l()gd^, holds good in the par- 

 ticular case postulated so long as the critical velocity 

 which is given in the expression pVd/»j = 25oo is not 

 exceeded. 



Many experimental diflRculties were met with in the 

 actual measurements, which were carried out in the 

 engineering department of the National Physical 

 Laboratory by Mr. Pannell. The sm.all variations in 

 pressure were measured by means of a sensitive mer- 

 cury tilting gauge, and the quantity of oil passed per 

 minute through the pipes was measured on an Avery 

 weighing machine. Thermal changes were eliminated 

 by jacketing and careful electrical heating. 



In general excellent agreement was found between 

 the calculated and observed values of P/V through a 

 wide temperature range. 



Part ii. of the research was occupied with the deter- 

 mination of the physical contents of the various oils 

 which were used. The densities call for little com- 



