arrive at better alloys by research in this country. As a result, a series of new aluminium alloys for use in the wrought form have been developed, mainly as the result of extensive researches at the Nation- al Physical Laboratory. In the first place, a series of alloys containing from 1 8 to 20% of zinc were produced. These were of a type which had up to that time been regarded as incapable of being rolled or forged, but the initial difficulties of that kind were overcome by a careful study of their properties both in the foundry and the roll- ing-mill. The result was the production of an alloy containing 20% of zinc and 3 % of copper, generally known as " 3-20 " or " Alloy A." This shows a tensile strength, in the condition as rolled, of 27 to 28 tons per square inch with an elongation of 1 8 to 20% on 2 inches. It is thus a little stronger than duralumin, but also a little heavier (its density is about 3'i). It is simpler to make and use than dura- lumin, as it requires no special heat-treatment, and it is unlike duralumin not liable to be seriously weakened by a slight amount of annealing. On the other hand, this alloy is liable to be damaged in other ways if heated much above 250 C., and it loses its strength very rapidly with rising temperature, at all events above 100 C., while it is also distinctly less resistant to corrosion than duralumin. For many purposes, however, where cheapness and simplicity of treat- ment are important, and where the material is not exposed to severely corrosive conditions, the " Alloy A " (320) possesses distinct merits. This alloy has, however, been considerably improved upon by a series of alloys in which both manganese and magnesium have been added to the simple aluminium-zinc alloy. These require quenching and ageing, but after such treatment can be made to attain a tensile strength of 40 tons per square inch. They are, thus, in regard to strength for a given weight, considerably superior to duralumin and this relative value is particularly apparent under compression (buckling) tests. On the other hand, these alloys require careful protection from corrosion and their heat-treatment must be care- fully carried out. Another very important group of "wrought" alloys are those containing nickel and magnesium in addition to copper. The most important of these is one developed at the Na- tional Physical Laboratory and known as "Alloy Y," or "42-15," the latter figures representing the composition: copper 4/0, nickel 2%, and magnesium ij%. This alloy, when quenched from a tem- perature of 530 C. after previous cold-rolling, can be made to attain a tensile strength of 28 tons per square inch combined with an elonga- tion of 20 % on 2 inches ; its density is 2. 8, and it possesses two very important further properties, viz. remarkable resistance to corrosion, and a relatively very high resistance to fatigue (repetition stresses), particularly at slightly elevated temperatures. Forgings of this alloy have been successfully used as connecting-rods in high-speed internal combustion engines, and there is every reason to anticipate a constantly widening range of engineering uses.
Promising and important as are the results achieved with the wrought alloys just described, results of more immediate importance were achieved with casting alloys of aluminium. At first these were employed mainly on more or less subsidiary castings, such as crank- cases, and for that purpose an alloy containing from 12 to 14% of zinc and about 2j% of copper (generally known by the number of the British Air-board Specification as " LS ") was very widely used. Efforts were soon made, however, to employ light-alloy castings for more important parts in aeroplane machines, viz. cylinders and pis- tons. Here the value of these materials lies not so much in their specific lightness as in their high thermal conductivity. In the case of the cylinder castings of air-cooled engines particularly, this is valuable in preventing distortion arising from unequal cooling of the windward and leeward sides, while in the pistons it reduces the tem- perature of the compression space and thus increases the density of the indrawn charge, and at the same time allows of the employment of higher compression ratios. The effect of these advantages is to increase the power output of an engine of given size and weight very appreciably, while also reducing the petrol consumption. The alloys first and most extensively used were those of aluminium with copper, a 12 % alloy being particularly popular. Another widely used alloy contains 7 % of copper with I % of zinc and I % of tin, but it is now recognized that the presence of tin renders the alloy weak under shock when hot. These alloys, although initially not as strong as some of those containing zinc, do not lose their strength so rapidly when heated, so that at the working temperature of an aluminium-alloy piston (about 250 C.) they are stronger than such an alloy as " LS." Even these alloys, however, are relatively very weak when hot - they register a tensile strength of about 6 to 7 tons per square inch at 250 Centigrade. Recently, researches at the National Physical Laboratory have shown that the alloy already referred to above as " Y " containing copper 4 %, nickel 2 %, magnesium ij % is partic- ularly strong at high temperatures, even in the cast state. It is, further, amenable to hardening by quenching and ageing even in the form of castings, and when thus treated attains a tensile strength as high as 20 tons per square inch at the ordinary temperature and 13 tons per square inch at 250 Centigrade. This alloy is rapidly finding its way into extensive use and many important applications are being opened up as the result of its remarkable properties.
Cobalt and Lead. Developments in the remaining metals are mostly of a minor nature and cannot be referred to in detail. Men- tion should, however, be made of the progress made in connexion with cobalt. Its use in steel and in certain special alloys has already
been mentioned, but it has also been shown to give a more adherent and more durable electro-plate coating than nickel, and it is impor- tant to note that its resemblance to nickel is not nearly so close as was previously supposed. In regard to lead and its alloys, a remarkable development has been that of alloys with the rare-earth metals, particularly calcium and barium. These confer a remarkable degree of hardness on lead, and a special alloy of this kind, known as " Ulco," is finding application as a bearing-metal. A substitute of this kind was called for as the result of the very high prices attained by tin under war conditions, but the permanent value of the materials has yet to be established.
PHYSICAL METALLURGY. Side by side with, and to a great extent furnishing the basis for, the development in the treat- ment and use of metals and their alloys, there has been a very great and important development of metallurgical science, particularly in the direction of what has been called " Physical Metallurgy." The mass of work which has been published on this subject is so great that even an approximately exhaustive bibliography would occupy more than the space available. Only a few outstanding features of the progress achieved can there- fore be briefly mentioned.
A very large amount of work has been devoted to the further and more detailed study of the constitution of alloy systems. Al- though somewhat rough preliminary determination as of the equilib- rium diagrams of most binary alloy systems had been previously made, a number of these have been revised and rendered more ac- curate. In ferrous alloys, the iron-carbon system has received much further study, particularly in regard to the critical points of iron itself. Important work at the Bureau of Standards, U.S.A. (Bur- gess and Crowe), has firmly established the three well-known critical points, Ai, AS, and A 3 , and has shown that previous attempts on the one hand to discredit the very existence of A 2 (Carpenter) and on the other to show that it was a double point (Arnold), were based on experimental error. On the other hand, German investigators (Ruer, Hanemann) have established the existence of a higher critical point, which in pure iron occurs at a temperature very close to 1,400 Centigrade. In connexion with the critical points, considerable attention has been devoted to the whole question of allotropy. A Dutch school of investigators (Cohen) have sought to show the existence of numerous allotropic transformations in many metals, but their conclusions are based on extremely slight evidence derived from determinations of minute irregularities in density changes. On the other hand, the Japanese school (Honda) seek to show that the Aj transformation in iron is not allotropic in character and this view is confirmed, to a certain extent, by strong evidence that the passage through this point does not involve any change of crystalliza- tion evidence which has recently been confirmed by the X-ray analysis of the crystal structure of iron and steel at various tempera- tures (Westgren). The matter, however, rests entirely upon the precise definition of allotropy which is adopted. In addition to the iron-carbon system, the iron-nickel, iron-chromium, the manganese- carbon and nickel-carbon systems have been carefully investigated. No attempt, however, appears as yet to have been made to attack the detailed study of the equilibria of such important ternary systems as iron-nickel-carbon, iron-manganese-carbon or iron-nickel-chro- mium, no doubt on account of the length and difficulty of such an investigation. In non-ferrous alloys, considerable attention has been given to the alloys of zinc, a portion of the ternary system copper- aluminium-zinc (alloys rich in zinc) having been very fully worked out (Haughton, Bingham). The allotropy of zinc itself has also been very thoroughly studied (Benedicks, Bingham) and the reality of the transformations well established. Great advances have been made in the knowledge of the equilibria of several of the important alloy systems in which aluminium is the predominant metal. The ternary systems aluminium-zinc-copper, aluminium-iron-silicon and alu- minium-magnesium-silicon (Hanson, Gayler) have been very fully worked out so far as the alloys rich in aluminium are concerned. For the representation of the results of such investigations a new type of model has been devised (Rosenhain), in which the various equilibrium surfaces are represented by systems of wires coloured to indicate the phases concerned in each transformation. The study of the aluminium-magnesium-silicon system has proved particularly important, owing to the light which it throws on the age-hardening properties, which are found in many aluminium alloys containing magnesium. It has been shown that the magnesium in these alloys is present as a definite compound Mg 2 Si, which is more soluble in solid aluminium at high temperatures than at the ordinary tem- perature. Quenching such an alloy from a temperature just below its solidus retains the compound in solid solution and in this state the alloy is soft. Gradually, however, at the ordinary temperature, and more rapidly at slightly higher temperatures, this super-saturated solid solution deposits the excess of dissolved compound in an extremely finely divided condition and this process is accompanied by gradual hardening of the alloy. This process is strictly analogous to that which can be brought about in certain alloy steels which can be rendered (or kept) completely " austenitic " (homogeneous solid solution) by quenching; they are then soft and ductile, and do not
