IRON AND STEEL

developments during 1910- 26 in iron and steel were improvements in processes and ¢equip- ment rather than new methods. <an increase in the per cu pita consumption, far greater than the remarkable increase of the preceding decade, forced: attention to means of securing max- imum outputs as well as to the ever-present effort to secure econ- omies. larger units of manufacture were generally favoured and so-called ‘ duplexing ”’ and “ triplexing’’ were outstanding features in steel-making. the latter part of the decade was marked also by a marvellous growth in popularity of the electric furnace, until at the end of 1922 there were 1,175 such furnaces in the world for steel-making alone, against 114 in 1910. the world war gave an artificial stimulus in general to plant expan- sion to meet the demands for ships and shells and resulted in a realignment of national capacity. in the main the following 538 analysis is devoted to the economic side of the evolution of the period. 3 iron blast furnace construction.—design was influenced by local experience of experts in the different iron-producing dis- tricts. profiles depend on raw materials—quality of coke, nature and concentration of ores. generally speaking, tendency toward greater bosh and shaft angles continued (bosh angle 75° to 80°; shaft angle 84° to 86°). the size of stack increased only in the districts treating low-grade ore, while with high-grade ore (50% to 62% iron) the 500- to 600-ton-per-day size became the standard and many old furnaces were enlarged. in the minctte district of europe, where the ore charged contained between 30% and 35% iron, the 200- to 250-ton unit became popular in all new construction. the cubical capacity of a blast furnace ranged between 40 and 100 cu. ft. per ton of pig-iron blown in 24 hours. individual parts of blast furnaces received particular attention. the hearth construction became reinforced and often cooled to avoid breakouts of the molten metal, and greater attention was given to brickwork to limit the downward destructive action of the metal. emergency tuyeres at mid-hcight of the bosh— standard in the minette district—lost their popularity and dis- appeared in nearly all new furnaces. for the handling of the material—ore, limestone and coke—mechanical clevices won increased favour owing to the enormous masses involved and to the growing scarcity of labour. in america, the simple skip hoist was almost universally adopted in new plants, while in europe the drop-bottom bucket hoist became popular and its design was ingeniously varied. it is noteworthy that the drop- bottom bucket, which was originated in america (duquesne works) and was perfected in germany during the years 1905-10, was little used in the united states ten years later. it had been supplanted by the cup-and-cone top with a rotary distributor (mckee); in europe the double cup-and-cone construction, giving low drop height and assuring less breakage of the softer coke, was preferred. these two solutions were more or less linked to the hoist system adopted. the furnace interiors were bricked up, in america of standard-shaped refractorics giving a multitude of joints, while it was customary all over europe to use special large-volume brick, shaped to reduce the joints. both systems seemed to give satisfaction to the opera- tors, and comparisons were not possible owing to the difference in the operating conditions. speaking generally, the american design with a plate lining involved less steel for construction, while the luxembourg-lorraine type of blast furnace was con- servative and substantial. blast furnace operution.—general progress was marked; greater familiarity with the chemical problems and increased mechanical equipment made operation more easily controllable, results more positive and disturbances less frequent. the use of excessively fine ore was accompanied by a high-solution loss of coke, and agglomerating of ore was recognised as desirable. this led to increasing attention to sintering fine ore mixed with flue-dust, and recharging of dust without treatment seemed likely to die out. higher temperature of the air blast was a note- worthy development, as each increase of 200° f. has been reflected by 4% to 5° reduction in coke consumption. in the minette district 850° to goo° f. blast temperatures were quite common. the blast pressure, which remained without much change, varied according to districts and the forcing of the operation from 4 to 15 lb. per sq. inch. in case of a relatively cold spell within the furnace due to overiluxing, bad coke or dropping of a hanging, the introduction of kerosene (parafiin) through the tuyeres proved a quick remedy, although it neces- sarily required care and progressive application to avoid acci- dents. the use of this cure rendered superiluous the auxiliary tuyeres at mid-height of bosh. the use of the oxygen torch to burn out the iron notch in case of metallic incrustations, extremely difficult to remove, proved a great help to the working crews. mechanical appliances in cast-houses were installed in increasing numbers to supplant hand work, especially in sand casting floors. cranes equipped with lifting magnets and pneumatic hammers elastically suspended proved a step toward the best method of iron and steel moulding, breaking and handling of sand-cast pig-iron. for all qualities of pig-iron the continuous casting machines continued to be the accepted standard. in plants adjoining steel-works transport ladles of larger size (30 to 50 tons) and designs assuring better insulation and easier skimming of slag were evolved. blast furnace gas.—continuously increasing price of coal and coke affected the economics of the blast-furnace gas and made it a by-product of great importance. the fuel value of this original waste became more and more recognised in america. the first effort, to avoid all gas losses so far as possible, led to the installation of double furnace tops, which became universal. next gas-cleaning processes were developed to remove dust impurities from the by-product fuel, thus increasing its adaptability to combustion and securing greater efficiencies in its utilisation. to facilitate operation and render supervision automatic, pressurc-regulating devices were evolved and accumulators were installed to equalise supply and provide steady outilow. gas- cleaning attracted the attention of operators in european countries greatly in need of fuel, because the calorific value of the gas counterbalanced the cost of purification. in america, the coal shortage, due to inordinate demand and dislocation of railway service, produced the same result. the first step in gas- cleaning was the installation of a dust collector close to the off- takes and the downcomers of each blast furnace. in it the coarse dust was deposited by a slowing-down of the gas flow and a sudden change in its direction. this apparatus was independent of all further cleaning methods. to separate the fine dust particles two different processes were applied: (1) wet method; (2) dry filtration. (1) by the wet method the gases were cooled by injection of water, and the dust particles, passing through the fog artificially produced, were arrested by scrubbers. if-xperience led to a subdivi- sion of the operations, called medium cleaning and fine cleaning, with apparatus protected by patents. the sludge of dust and water was removed by the application of centrifugal force, separating gas and liquid in specially designed fans or washers (theissen, brassert, etc.). the disposal of the water created a problem, as contamination of rivers is against the law in most industrial countries. in deposit ponds the settling of the impurities was incomplete and its removal a tedious manual operation; and the dorr thickener, developed in ore- concentration districts, was adopted, as assuring continuous service automatically by means of a special mud pump requiring little atten- tion. a drawback of wet systems was that the sensible heat of the blast furnace gas was absorbed by the cleaning water and lost be- yond recovery. on the other hand, it permitted the installation of smaller gas-piping and dispensed with the insulation of the lines against heat loss, thus saving appreciable capital outlay in the case of long-distance distribution. another drawback of the wet method was that recovery of the dust required driving off the water from the heavy mud in any briquetting or concentrating process attempted. (2) the dry methods of gas-cleaning had their advocates where every little economy was watched, such as retaining the sensible heat of the gas and saving the expense of water-handling in keeping the dust dry. to separate the dust out of the hot gases, filtration ap- peared to be the best process. with mechanical filtration, finely woven cloth or asbestos-fabric bags or slag-wool layers let the gas pass at low velocity but retained the solid dust, which was removed periodically by return currentsof clean gas. the principle was adopted in the beth-halberg system in europe and the kling-weidlein apparatus in the united states. the chief ways in which blast-furnace gases were utilised were as follows:— (a) cowper or tlot-blast stoves.—the absence of dust in the gas provided for rational stove design, as the complicating side issues of clogged-up passes and slagged-up checker holes disappeared, as well as the periodic waste of cooling, cleaning and warming-up of each unit. theclean gas meant a reduction of the area of heating surfaces and brick volume expressed by fewer stoves per blast furnace—three to four per furnace against four to five to years earlier. then began a systematic study of the heat-transmission phenomena within the mass of checker work. | (6) boilers —clean blast-furnace gas allowed for advantageous use in connection with steam boilers; first, through more efhcient combustion, in effect fess gas per pound of steam produced; second, higher ratings of boilers, in effect more steam per unit of boiler evaporating surface or fewer boilers for a given plant capacity; and, third, quick adaptation to any load required, in effect flexibility or ease of operation. alany efficient burners were invented and some were installed on a large scale. (c) metallurgical furuaces.—the removal of flue-dust made possible a wider distribution of the blast-furnace gas, and in europe use was made of the surplus gas with success in all kinds of furnaces, iron and steel the low calorific value coupled with the small amount of air required for complete combustion, opened fields where so-called mellow heat- ing tlames are demanded, such as core drying, mould drying, anneal- ing, roasting and ore concentrating. (2) gus engines.—the principle that clean gas was indispensable for internal-combustion engines was long recognised, but its practical application did not occur until after 1910. also a cool gas was re- garded as essential to secure adequate volumetric efficiency of each cylinder. among gas engines, the four-cycle type outranked con- siderably the two-cycle type. devices for close regulation were developed on the principle of qualitative-quantitative mixture. the built-up cylinder seemed to win greater favour than the one- piece casting. the safety of operation reached a parity with that of steam-cengines or turbines, the gas being clean. the exhaust heat of the engines, representing some 40°, of the energy, was utilised to gencrate steam, and 70°% was thus recovered in some instances. of all the various us2s made of the gaseous by-products of the blast furnace only the heating of the hot-blast stoves was universally applied. all experts agreed that 30% to 40°3 of the gases are best employed for that purpose. the surplus of 60°) to 65% was utilised for the other purposes already mentioned. in the utilisation to pro- duce blast pressure and to develop power, the battle for suprem- acy between the gas-engine using blast-furnace gas and the steam boiler using the gas to supply energy to engine or turbine remained undecided. thermal efficiencies were not the only issues at stake. in europe, the gas-engine had the firmer standing, while in america the boiler seemed to be the more in favour. even for generating the blast pressures, the competition between gas-engine-driven air com- pressors, steam-engine blowing engines and turbo-blowers had gone on without absolutely proving the superiority of any one combina- tion. varying economic conditions in each country and dificrent local considerations, as well as the purely technical aspects of the problem, were deciding factors. europe, with its skilled workmen and more stabilised market conditions, presented a background different from that of america with its fluid trade conditions and its unsettled unskilled labour. plant layout and size.—<a single blast furnace, built alone on a site, no matter how well chosen, proved not to be a logical industrial enterprise. the number of such plants existing was the result of competition, of fluctuating market conditions, and constituted an economic waste, speaking gencrally. with combined units the ac- cessory equipment became cheaper in installation cost and in terms of iron output and more efficient in operation, through flexibility and insurance against breakdown. three to six furnaces grouped in well-laid-out plants were established as an cconomic whole. to avoid the loss of the sensible heat of the molten pig-iron and to refine the metal without cooling, steel-works were logically joined to blast-furnace plants. the two separate departments were thus combined in one industrial unit, with the added advantage that the surplus of power available at the furnaces could be absurbed in the rolling- mills. electric pig-iron furnaces.—tests made at trollhattan, sweden, were conclusive only for high-grade pig-iron similar to the swedish charcoal pig-iron. since 1918, the domnarfvet works in sweden had operated several shaft-type furnaces (with gas circulation using 60 %% to 62 % of iron ore and charcoal as a reducing agent). mixtures of charcoal and coke up to 50% coke were found satisfactory. per ton of pig-iron produced, 3.400 lb. of ore (containing 61-5% fe), 120 ib. of jime and 740 ib. of charcoal were charged; 15,000 cu. ft. of gas at 240 b.t.u. per cu. ft. were captured at the top; 2,150 kilowatt-hours was the electric energy consumption per 2,000 ib. of pig-iron. the problem of using electric current for supplying heat in the blast furnace reactions had particular interest for the eastern pyrenees in france, british columbia, brazil, italy, as well as sweden and norway, where fuel is scarce and low-priced electricity might be made available. the sieel plant.—the usefulness of mixers as an important adjunct of the steel-making plant was universally recognised, as numerous installations attest. their field was established in equalising quali- tatively the successive outgivings of the blast furnaces and in de- sulphurising the molten metal. to accelerate the removal of sul- > phur, less than 0:5 °% of manganese proved most helpful. the shape of the mixer that gave best results was the simple cylinder rotating on its axis. the most popular size proved to be 1,000 to 1,400 tons’ containing capacity. simple oil or gas burners without regenerating chamber in the united states, with pre-heating checkers sometimes in europe, completed the equipment. in germany, a 2,000-ton- capacity mixer was reported built, but only after considerable dis- cussion as to its size. the mixer was, of course, brought into being for receiving metal from the blast furnace and delivering to the ladle for transport to the steel plant as needed. slag that floats on the top of the bath must be skimmed olf from time to time. converter plants.—no noteworthy development took place in the acid-onerating (bessemer) converter or in the basic-operating (thomas) converter for making steel. the 20- to 25-ton-capacity vessel remained nearly universal. a 40-ton size was proposed in 1918 by a belgian engineer. as between europe and the united states, the hydraulic tilting mechanism of the former did not give way to the electric drive of the latter, nor did the gas-engine blowing units succumb to the turbo-blowers of american practice. 939 open-hearth plants —wiaithout radical change in type, sizes of open-hearth furnaces increased up to and above 100 tons’ capacity, but the tendency was toward fully controllable sizes. the practice in the united states settled to 8o to 100 tons and in europe 40 to 50 tons. volumes of checker chamber increased to get better so-called flywheel effect. greater attention was paid to port and head con- struction to lengthen life, and toa reinforced roof. reversing valves were marketed for reducing how resistance and waste of gas. waste heat boilers were more generally installed, but not universally adopted because of their interference with concentration and general e!ec- trification of plants. marked superiority or inferiority was not shown for the tilting construction of furnace when tested by use beside the stationary type. natural gas disappearing in america, producer gas was the more generally adopted fuel. powdered coal was tried with some success but without proved superiority; one difficulty was that checker chambers got clogged by ashdust. by-product tar and crude oil proved fuels weli suited for the purpose if available at low price and in large quantities. electric steel furnaces.—the clectric stecl furnace for refining and melting iron and stcel developed to a surprising extent in the decade 1910-20 in size and in number of installations. it proved ideally suited for quality products and, high-grade materials, because no complication through fuel medium exists and because atmosphere and temperature are attainable practically at will, from 500-ib. capacity, single furnaces were built to 40 tons, with most of them of 5- to 8-ton capacities, (for electric furnace statistics, see [ron a ge, jan. £1921.) of the 960 known electric steel furnaces in existence in jan. 1921, 356 were in the united states, 150 in england, 100 in germany, 60 in france and 43 in canada. of the total, 308 were heroult arc furnaces, 102 rennerfelt induction furnaces and 90 greaves-etchclls furnaces. the electric furnace was adopted for making metal mixtures, ferro-alloys, special steels of high qualitv in large amounts—strict repetition being possible in an absolute positive way. a disadvantage was that the metal is not at rest but always in motion, through electric or magnetic influences. though agitation was often desirable, the action hampered the separation of the slag and the rising of impurities out of the molten mass. ua remedy for this was repeated skimming of slag and careful super- vision. one unusual utilisation of the electric furnace was the making of pig-iron out of scrap, especially in the united states, to supply deficiencies in the amount of low-phosphorus pig-iron, particularly in the manufacture of ordnance. it amounted to a synthetic recon- version of steel into pig-iron. fine coke was added to the slagged refined scrap for carburisation, and the method promised to be com- mercially feasible in regions having electric power but little local fuel available and situated so that delivered pig-iron was high in price. for dceoxidation in the refining process and for recarburisation, ferromanganese and spicgeleisen retained their popularity in spite of high prices under erratic market conditions. in europe pre-heating, often pre-melting, of the addition was the current practice, to save in the amount needed and to accelerate effects. in america the wasteful method of cold additions prevailed. ferrotitanium, with carbon or carbon-free, was used, as well as ferrosilicon and aluminium in small quantities. some steel plants made additions in the ladle, others finished the operation in the furnace. steel-making operations—the outstanding feature of stcel- making operations was the recognition of splitting the refining process into two phases, or the two-slag method, to increase production and to lower production costs. the efforts of bertrand-thicl and talbot recognised in effect this principle; and duplexing and triplexing were only operating variations of the same principle, to remove the im- purities of the pig-iron stepwise in the furnaces best suited for each purpose. thus sulphur and manganese pass out in the mixer; silicon and part of the carbon in the converter; the rest of the carbon and phosphorus in the open-hearth furnace; additions were made and alloys were added in the electric furnace. the plant necessitated considerable equipment, but it secured case of operations, exact control of results and macle possible quantity production. below 1,600 tons per 24 hours, savings in operation were regarded as hardly possiple, as in slack market periods overhead expense was too large. iuring the world war about ro duplexing plants, refining in an acid converter and finishing in basic open-hearth furnaces, were built in the united states under the pressure created by an ammunition famine. electric duplexing plants (meaning melting and preparing in open-hearth furnaces and finishing in electric furnaces) were built in large numbers, offering a special-quality product on a large scale. a number of new independent efforts were made to produce stcel direct from the ore without the interpolation of the iron blast fur- nace, but none could be said to have been proved feasible on a scale beyond that of the laboratory. the shaping of steel.—rolling-mills (used if the demand for a product is large and if its shape lends itself to a continuous process, like rails, angles, plates, bars, etc.) and the forge-shop (if the shapes to be produced are complicated, short in length, unsuited for the rolling-mill), both change the shape of the metal heated at high temperature, about 2,000° to 2,300° fahrenheit. both require finishing departments to straighten, shear or bundle the rolled product or to clean off the fins, rough off the unevenness of the forging operations, and they may need annealing and pickling facilitics to improve the quality of the product. a special process of milling the top and bot- 540 tom of rail blooms, to remove cracks and roughness from the semi- finished steel and also surfaces decarbonised in the heating furnaces, was put into use at the lackawanna mills in america and resulted in a reduction in the number of finished rails classed as seconds. in the period 1908-26 the development of the rolling-mill was influenced, first, by the great manufacturing principles of concen- tration and specialisation, and, second, by the electrification of the motive power. concentration demanded large production in one sl ft tt ee see r rrr ttt pree all i ee 1 *! nz 5 : » a thousands of metric tons nnw hhh zf elen simwieln aran naa va aav im vaauaasav = adu “hee aa 7000020000 azania aviv. 1913 1915 1916 1917 1918 oe i920 1921 1922 1923 1924 1925 v4 united states german franc . er lt cothers - fic, 1.—diagram showing the production of pig-iron in the prin- cipal producing countries, for 1913 and 1915-25 (monthly average). note: “others” include luxembourg, belgium, the saar, sweden, canada, poland and the union of socialist soviet republics. unit and suitable equipment to attain that aim; in other words, mechanical devices in preference to hand operation. specialisation was applied to the shape to be rolled as well as to the mill used for production. standardisation of rails, beams and angles, the reduc- tion of the number of profiles, and the simplification of shapes were consequences, as well as the installation of mills for specific purposes. the application of these rational principles was accelerated by the use of the electric motor. the advantages were recognised about 1905, but the next 15 years brought their practical realisation. the numerous little steam-engines disappeared and the electric motor revolutionised the handling of the material by cranes and overhead trolleys as well as the mill accessories, like tables, skids, transfers, etc. the first step was the creation of central power plants where electricity was gencrated either in turbine or gas-engine generators, preferably with the help of the surplus gas from the blast furnace. many steel plants in 1920 were equipped with 20,000- to 40,000-k.w. power stations. the second step was the development of speed re- ducing devices made necessary by the high speed of electric motors, the advance of the gear-cutting industry and the advent of spiral- type teeth, single or herring-bone, and the development of new types of teeth giving less wear, more rolling surface, and, later, the use of special hardened, heat-treated stecls were eagerly taken up by the designers of mill machinery to increase the quality of their product. reduction gears transmitting up to 5,000 [1.p. came into daily use, and. the ratio of 10 or 12 to i in one reduction gave satisfactory serv- ice. the third step was the development of specd-regulating de- vices, especially in connection with alternating-current motors, to secure efficient operation for variable conditions, the fourth step was the solving of the load problem of large, intermittently operating motors, reversing their direction of rotation by means of the motor- flywheel set advocated by the austrian engineer ilgner in connection with suitable controllers of which the ward-leonard system was the prototype. much work and inventive genius were concentrated on these difficulties to bring about in less than 15 years the high effi- ciency and great safety of operation by electricity of steel mills. the development of the rolling mills is still progressing. for driving continuous mills electricity was preferred to the steam-engine. the uniflow steam-engine found favour up to 3,000- h.p. units, but installations were few. where the effort was to bring about a complete electrification of the plant, steam-engines proved unpopular, possible economy being counterbalanced) by complica- tion of maintenance and other administrative considerations. for iron and steel the reversing drive of large units, requiring loads up to 20,000 or 25,000 it.p., the electrical drive was not necessarily in the ascend- ancy. the high initial cost of the motor generator flywheel set with direct-current motor directly connected to pinion and mill was made the chief argument against universal adoption of electricity, and numbers of old plants were reluctant to change their somewhat obsolete steam equipment. in america only a few engines as large as 25,000 to 30,000 h.p. (weirton and lukens) were installed in new work, and in england large vertical engines were built. the electrical units, however, increased considerably in number and size and considerable progress was rcalised in the matter of manocuvring capacity, in standardisation of winding and accessories, in records of output and efficiency of running. in 1925 the electrical industry was aggressively working on betterments, while the partisans of the steam-engine rested on past laurels. parity existed on all points except first cost of installation. even for smaller mills of the revers- ing type electrical drives were built, such as 24-in. mill (mark, indiana harbor), 26-in. mill (atlanta, ga.). two-high reversing plate-mills also disputed the field with the three-high type in the me- dium-sized equipment, and latest universal-mill installations were of 11) tttttcm 6500 ll rett eee 6000 — “reece ccong 5000 s . 3 __ominn ssn nt ns 4 4500 ps et fe = == sos a esr es 8 a0 \ n = ae = he . | meelse/\= fier ao eeaiaeag “=o mee a see iai) united states = oeriiony [{]] france united kingdom cj others fic. 2.—diagram showing the production of crude steel in the principal producing countries, for 1913 and 1915-25 (monthly average). note: “others” include luxembourg, belgium, the saar, sweden, canada, poland, italy and the union of socialist soviet republics. the reversing type electrically driven. the largest mill for plates, 192 in. wide, at lukens, pa., which was completed in 1918, was of special design, reversing and steam-driven. (zron age, jan. 2 1919.) the concentration of production in large well-balanced plants with adequate resources and sales organisation permitted a logical subdivision of the rolling-mill programme with a subsequent reduc- tion of production cost. the smallest bloom section, 6x6 in., had a tendency to grow toward 8x8 in., as some experts claimed that 50 sq. in. was the economic limit of the range of a large mill. to reduce the time required for roll changing, complete spare housings, com- pletely mounted, came to be good mill practice, these being dropped on the shoeplates by the cranes. in some specialty plants sometimes two or three housings were changed together. furnaces.—vhe devclopment of soaking, reheating and annealing furnaces was influenced by the increasing price of fuel and consider- able efforts were made to boost heating efficiencies. in europe gas- firing with recuperation and regeneration of waste heat was the favourite, especially because producer-gas firing was much used and remarkable results achieved. in america continuing shortage of natural gas for industria! purposes in regions such as the pittsburgh district made a substitute market for crude oils, coke-oven gas and powdered coal. especially since 1915, the use of powdered coal developed to a surprising degree for all kinds of metal-heating appli- cations. mechanical stokers were evolved in numerous designs to dispense with hand labour and to control by mechanical contrivances the combustion of coal] in an efficient way. iron and steel sai afiil equi pment.—roller tables, skids, transfers, cooling-beds and tilters were perfected and installed in increasing numbers owing mainly to the elficiency and handiness of electric motors. variable- speed drives gained in favour. straighteners, saws, punches and shears were made in conformity with the availability of clectric power. hydraulic devices were pushed in the background and elec- tric drives supplanted engines in the field of these mill accessorics, metallographic progress ——metallographic knowledge spread in the period ro10-25 from the university laboratory to the steel mill, outgrowing the narrow circle of students to become the helpmate of the operator. pyrometers or temperature re- corders and the scientific control of temperature came as a matter of course in numbers of steel mills. the variety of alloy steels offered for practical uses was due to the theoretical investigations of the metallograph. chromium and vanadium, nickel and cobalt, tungsten and molybdenum added their special properties to stecl, and improved heat-treating methods enhanced these qualities with a skill and positiveness hitherto unknown. beyond the realm of iron and steel making, properly regarded, but coming within the purview of the industry, the remarkable development of the period was the heat treatment of metals. it went hand-in-hand with the study (by means of the micro- scope, and thus of the photomicrograph) of grain structure and the transformations which take place in the so-called solid solu- tions, according to the degree of heating and cooling given to the metal. practical applications of the investigations of the scientist were numerous, and the history of the temperature experience of a given metal product going into an article of commerce was accepted as equally important with the chemical constituents, for two pieces of steel, identical chemically, may be made to behave physically very differently according to their crystalline state. (see metallography.) other developments which must be briefly enumerated were: efforts to test a material’s fitness or agreement to specifications with- out destroying it, as by x-ray photographs (sheets), or by magnetic analysis (by noting changes in permeability of an article of constant cross-section, as a rille-barrel, wire or steel rail, by moving it througn a magnetic field); these, however, were not definitely of commercial dependence, pending further investigation; success in making large chains of cast-steel links followed by heat treatment, such as anneal- ing; efforts to cast in centrifugal moulds, such as cast-iron pipe by introducing molten iron into the rotating mould; commercial re- covery of potash from blast-furnace mlue-dust deposits at the base of hot stoves; making iron pipe by an clectrolytic process of de- positing iron on a rotating cathode in a ferrous-chloride electrolyte; commercial manufacture of a stainless steel having 10°, to 13 “6 of chromium, which, harry brearley in england discovered, gave amazing resistance to corrosion, so that it became the base of an im- portant cutlery industry and offered a satisfactory material for rifle-barrels, turbine blades and steel articles subject te both eresion and corrosion; additions to the numberless varicties of alloy steels, largely to secure some desired physical characteristic for specific needs, such as increased tensile strength in terms of lighter members of a fabricated steel product; elements like cerium and zirconinm entered the ferro-alloy circle, but a delineation of the various alloys and of their definite fields of usefulness was not completed. production.—the accompanying table shows the relative pro- ducing capacity of the leading industrial nations of the world. the figures are of actual production as well as capacity to pro- duce. the world’s steel-making capacity was put, in 1926, at 116,000,090 tons in round numbers. one-half was credited to the united states, which could make more than four times as much as great britain and more than three timesas much as germany. the united states’ percentage of the world’s pig-iron-making facilities was somewhat overt 46%. more than one-third of the total annual output of iron ore in the world came from the united states, and of the american production 85% came from the lake superior district. the lorraine ore-fields supplied about 25° of the world needs, 80% of its output going to france and germany. world statistics of the production of ore, pig-iron, and crude and finished forms of steel are obtainable from the national federation of iron and steel manufacturers, london. tron and steel of the world capacity to produce largest production ever made stecl ingots ae pig-iron and castings united states 49,000,000 56,000,000 canada . 1,300,000 1,900,000 mexico . 300,000 300,000 north amcrica 50,600,000 58,200,000 great britain 12,000,000 12,000,000 germany 15,000,000 17,000,000 france 11,000,000 9,500,000 belgium. 3,500,000 3,500,000 luxembourg 2,800,000 2,250,000 holland 150,000 es russia . : 3,500,000 3,500,000 poland . i 1,200,000 1,800,000 rumania 350,000 250,000 austria . j 600,000 joo0,000 hungary $00,000 400,000 czechoslovakia 1,300,000 1,500,000 ttaly : 600,000 1,600,000 spain. ‘ 600,000 600,000 sweden . 1,000,000 750,000 europe . 54,000,000 55,850,000 japan 1,200,000 i, 500,000 china 950,000 400,000 india 800,000 350,000 australia 500,000 400,000 south africa - 50,000 the world 108 050,000 116,750,000 | pig-iron year steel year 40,361,146 1923 45,430,000 1925 1,107,000 1918 1,695,000 ig18 10,481,917 19t3 9,553,715 1917 19,309,172! 1913 18,935.089' 1913 8,323,000 192 7,289,700 1925 2,781,000 192. 2,780,000 192 2,547,861 1913 2,053,100 1925 4,637,3004 1913, 4,769,200° 1913 1,031,123 1913 1,648,533 1913 384,0002 ig2 541,000? 1924 295,000° 1923 283,046? 1923 i,050,0002 1924 1,350,000? 192: 197,000 19i7 1,311,000 1917 497,726 1916 470,241 1917 828,969 iqi7 o14, 111 1916 780,000 1919 845,036 1920 257,048 1920 120,000 1922 613,627 1923 215,465 192 416,050 1q2 284,669 4 192 .. = 27,064 1923 77,536,000 iqi3 80,308,000 | iqi7 figures in tons: gross, for england, united states, etc.; metric, for france, germany, ete. 1includes both luxembourg and alsace-lorraine; 1913 pig-iron production of luxembourg (2,547,861 tons) and of alsace-lorraine (3,864,000 tons), deducted from the reported figure, leaves 12,897,311 tons tor germany. similarly, for steel, deducting the 1,336,263 tons of luxembourg and 2,286,354 tons of alsace-lorraine, leaves 15,312,472 tons for germany. 2 austria-llungary in 1916 produced 2,380,000 tons of pig-iron and 3,278,000 tons of steel. 3 includes production of poland. 4qne plant only. 542 bibliography.—w’. h.-hatfield, cast iron (1912); d. carnegie and 5. g. gladwyn, liquid steel (1913); j. h. hall, the steel foundry (1914); c. a. edwards, the physico-chemical properties of steel (1916); h. m. howe, the aletallography of steel and cast iron (1916); a, sauveur, the metallography and ileat treatment of iron and steel (1916); r. moldenke, vhe principle of iron founding (1917); j. e. johnson, the principles, operation and products of the blast furnace (1918); c. c. gow, the electro-afetallurgy of steel (1921); l. aitchison, engineering steels (1921); r. forsythe, 7he blast furnace and the afanufacture of pig iron, revised ed. (1922); f. w. harbord, and j. w. hall, the afetallurgy of steel 7th ed. (1923); a. stansfield, the electric furnace for iron and steel (1923); ul. j. skelton economics of iron and stecl, 2nd ed. (1924); f. t. sisco, the mann- facture of electric steel (1924); z. jeffries and r. s. arcner, the science of metals (1924); h. j. gough, the fatigue of metals (1924); j. m. camp and c. b. francis, the making, shaping, and treating of steel, 4th ed. (1925); e. heyn, trans. from german and augmented by m. a. grossman, physical metallograbhy (1925): sir r. a. hadfield, afetallurgy (1925). cw. w. mm.) see electro-chemistry; electro-mectallurgy; internal combustion engine; metallography; metallurgy; rustless steel; turbine, steam. irrigation engineering (sce 14.841).—during the pres- ent century development has continued at an accelerated pace; one of the many causes was and is the desire to meet the world’s demand for cotton, particularly the more valuable long staple varieties. burrages and dams.—where areas to be served were large enough, barrages have been built; where rivers have a very irregular discharge, dams have been erected to impound water from the flood seasons to augment the supply in drier periods. from the barrages or dams high-level gravitation canals lead the water to the soil. where the barrage system is not possible, modern water-lifting appliances serve new lands, or in many cases replace in areas already cultivated both the hand-worked water-lifting appliance known in egypt as the skadoof and the cattle-driven water-wheel. irrigation is thus serving an ever increasing area throughout the world in regions where suitable soil, climate and plain exist in combination. distribution.—with these extensions, in places approaching the limits imposed by the volume of water available, the prob- lem of equal distribution becomes important. this, in turn, depends on an accurate estimation of the volume. measure- ments were formerly made by means of data obtained from noting the speed of objects floating with the stream at various depths. in the last 30 years, current meters replaced floats and give more regular results, but it was not definitely known whether these results indicated actual volumes. afcasurements.—this doubt has now been settled through the full-scale measurements made since 1905 at aswan by which the nile discharge has been calibrated to within an error of 1°% of the actual volume passing during all but the very highest levels, and it is hoped in time these will be measured also. current meter discharges taken simultaneously with the measurement at the dam, repeated a few times and the mean taken, gave very similar results and thus proved the accuracy of current meters. the data obtained from small scale models can be depended upon to within a small percentage of error. work in ecgyypt the cultivable area to-day in southern or upper egypt, from aswan to cairo, is about 2,500,000 ac. of which 2,200,000 were cultivated in 1910, 1,000,000 ac. being under the perennial form of irrigation and 1,200,000 ac. under the older basin system. in the delta there were about 3,000,000 ac. cultivated, all under the perennial system, out of a total cultivable area of about 4,800,000 acres. the area of all egypt is, however, over 500,000,000 acres. thus only about 7,300,000 ac. of this vast extent can be culti- vated from the waters of the nile. all the remainder must con- tinue in its present inhospitable state unless a climatic change occurs. in 1902 the completion of the original and low level aswan dam permitted 1,000,000,000 tons of water to be stored annually from the season of excess waters in the river, to be released again in the following spring and summer. in the few years following -1902 the irrigation engineering need for an immediate increase in the volume of water available for sununer irrigation became pressing if the area then cultivated was to be annually assured of an adequate supply in really low years, and if some new areas not hitherto cultivated were to receive water at all times. the work of storing more water at aswan by heighten- ing and thickening the original masonry dam and increasing its capacity to 2,400,000,000 tons, was completed in 1912. the need of existing areas for all the water available was accentuated by the phenomenally low flood of 1913, which was followed as a natural consequence by a poor spring and summer supply in 1914. the 1913 flood was estimated to be the lowest for 1506 years and except for the existence of the various barrages and the dam across the river, would have spelt famine as well as financial disaster to the country. the 1914 river, even when increased by the whole of the stored water from aswan, was only just sufficient to ensure the safety of the cotton crop of that year. additional supplies of summer water are needed for some of the still uncultivated areas in the delta, and the conversion of further areas in upper egypt from basin to perennial irrigation can only be satisfied by still greater control of the nile. the construction of a dam on the white nile near khartoum is now being taken in hand, forming another, though not the final, stage in the provision of water for :gypt. the very least quantity of water storable each autumn by the new dam will be greater than that now annually held up by the aswan dam, and like it, the water can be used in the following spring and summer in egypt. with a minimum of 3,000,000,000 tons thus provided, an additional 1,000,- 000 ac. of egypt’s waste lands may be given summer water. further developments—further works must, however, be undertaken in the sudd region of the sudan and even beyond it, before the final stage of an egypt cultivated to its capacity, with an adequate supply of water to meet all her requirements, is reached. ‘these works will most probably consist of training the waters through the sudd region to prevent waste and the erecting of a dam at lake albert. the sudd region lies on the white nile and commences about 500 m. south of khartoum and is in reality a vast plain through which the white nile meanders for 250 m. with its water either at or just above ground level. as a consequence the whole forms a great reedy marsh interspersed with open lagoons; into it flow the ilood season waters of the white nile; they evaporate them- selves away to such an extent that no great variation of volume is normally experienced at its outlet. the lost waters could, in great measure, be stored in lake albert and let out in such a manner as best to suit the requirements of egypt. conservation from evaporation when passing through the sudd region can be secured by confining the flow to a definite channel. the increased acreage which it will then be possible to cultivate in egypt will bring the total area up to its maximum of about 7,300,000 acres. developments in the sudan previous to 1910, except for a few thousand acres served by pumps, all the crops grown in the sudan along the main nile from wady halfa to khartoum and the white nile from khar- toum to the sudd region were sown on land naturally inundated on its low banks by the annual rise of the nile. on the blue nile, rainfall alone was depended on to fertilise the crops grown. about 1909 one of the uncultivated areas below khartoum was turned into a perennial irrigation farm and was among the first to produce cotton in the sudan, by means of water pumped from the nile. soon after that date the possibility of growing cotton in the gsezira, that huge tract which lies in the fork of the blue and white niles, became apparent. here some 5,000,000 ac. form a gently sloping plain stretching along between both rivers for about 200 m. from khartoum. from this point southward the plain is interspersed with small isolated granitic hills which, as a line south-east is tra- versed, increase in number and height. of the 5,000,000 ac. plain probably 3,000,000 will form the maximum area cultivable. on the remaining 2,000,000 ac. near khartoum the soil is said not to be so good, being of a more sandy nature. two experimental farms of a few hundred acres cach were set down in the cultivable area and proved after a few years’ trial the suitability of the soil for cotton growing. the summer climate, how- ever, was found to be too trying, although below khartoum and in egypt cotton is entirely a summer crop. cotton sown in mid-july in the gezira and picked in the following spring gave a return onan average of joo ib. per ac., a figure cqual to the normal i-gyptian production. this result was obtained at a season of the year when there is usually an abundance of water in the river. the cotton plant in the gezira is in the soil from mid-july, when it is sown, to final picking in april, or a period of abcut nine months. irrigation engineering in exceptional circumstances a final watering may be required as late as april 15, though normally the last watering is expected to be given by march 31. long before either of these dates egypt, not- withstanding her great reservoir at aswan, in low river years requires all that the blue nile can supply. it became neccessary, therefore, to devise the best method of lifting water from the blue nile on to the plain and of storing a sufficient volume of it to meet the demands in the gezira in those months when egypt requires all that flows in the river. the decision arrived at was to build a combined dam and weir at a point 5 m. south of sennar, where a narrow belt of gabbro rock, which scarcely rises above the level of the plain on either side, runs across the river. the construction of the combined dam and weir was proceeded with and completed in july 1925, at a cost of about £6,000,000, the canal system bringing the total up to about {9,000,000, the dam can store about 600,000,000 tons of water for use in the critical period, which extends from jan. to march and possibly to april, when egypt requires all that passes down in a very low year. the main canal leading from the dam is some 60 km. long before reaching the point where branch canals spread out from it on to the land to be irrigated. the area judged suthcient to form a commercial proposi- tion in view of the cost of the works was 300,000 acres. [{t 1s now believed this area can be considerably increased without endangering egypt’s supply. certainly if an amicable agreement can be arrived at between the two countries, greater areas can be cultivated in the gezira in all other years than the phenomenally low one on which the 300,000 ac. scheme was based. new projects—when still greater areas have to be brought into cultivation, and it is possible that 3,000,000 ac. may even- tually be cultivated in the gezira, either new storage works must be built or such further control of the nile be organised as will permit egypt to reduce her demands on the volume in the blue nile. special storage works for the sudan can probably be best built by permission of abyssinia. at the exit of lake tsana in that country at the source of the blue nile, by storage in the lake, the annual flood waters can be controlled and about 3,000,000,000 tons of water conserved for use at a later period, thus permitting a great increase in the area cultivable in the gezira. | there are possible alternatives, such as the building of a mini- ature aswan dam somewhere near roseires. here a regulation of the river will not be quite such a simple matter as at lake tsana, as the flood waters at roseires are heavily laden with silt. "these must be allowed to pass and water only be abstracted at a subsequent but still high stage when little or no silt js still present, otherwise egypt would be deprived of this valuable material. if collaboration with egypt in works in the sudan is adopted, then the white nile dam will be made as large as is consistent with safety so as to store as much water as can be secured from the flood waters. whatever capacity be given to the white nile dam egypt must still provide herself with more water than even it can hold, and some regulation of the sudd region must of necessily be one of the features of such a provision, coupled with a huge dam at the source of the white nile and the mouth of lake albert. besides the gezira plain the sudan has other irrigation areas where great improvements have taken place, as at tokar and kassala. tokar.—the river barakat rises in the rainy season in the abys- sinian hills and rushes as a chocolate-coloured thick stream on to the tokar plain where it eventually spreads out into a thin film which is sucked up by the thirsty soil. the plain, however, is much greater in extent than the water can cover, although some finds its way to the sea as recorded in 1921. great anxiety is continually causecl by the liability of the flood to break away into areas not hitherto cultivated or, even if cultivated, so far from tokar as to be inconvenient for transport. in recent years designs were studied for the purpose of exercising some control over the direction of flow. these have been carried out and have so far been successful. the total area served varies greatly from year to year. in 1921-2 about 55,000 ac. were flooded, though not all were cultivated. kaassala,—the river gash rises in the abyssinian foothills and becomes of considerable volume in the flood season by the time it reaches kassala. here it spreads itself over the plain in such a manner that none of its waters ever reach the nile or its nearest tributary the atbara. some control has been exercised over the flow to keep the water annually to definite areas. in a really good year with the supply of water at a maximum, 50,000 ac. of cotton might be cultivated. owing to the poor flood of 1925 only 11,000 ac. have been sown. 543 progress in india irrigation gives valuable aid in the fight against those periodic famines which always happened after rainfall failures, and it also causes an increase in production through its extension into new and suitable areas hitherto lying fallow. during 40 years, 1r885—1925 in particular, developments have steadily progressed. 10,500,000 ac. were irrigated in 1878-9; 19,250,000 ac. at the beginning of the century, and 28,000,000 ac. in 1923-4. addi- tional works now under construction will add 2,500,000 acres, new schemes are contemplated which will add a further 4,750,000 acres. when completed they will bring the irrigated area of british india up to about 36,000,000 acres. these figures exclude the water supplied from the punjab canals to 650,000 ac. in the native states. the sutlej valley project, mainly for the benefit of the native states, will increase the total by a further 3,250,000 ac., making in the proximate future a grand total of about 40,000,000 ac. of irrigated land in all india. afileage of channels.—in 1900-1, 39,142 m. of cnannels were in operation; by 1920-1 this had increased to 55,202 m. or an average addition of about 800 m. of channels per annum. the annual revenue return is between 7°% to 8% on the capital in- vested in government irrigation works. the following table shows the acreage of crops matured during 1923-4 by means of government irrigation systems compared with the total area under cultivation in the several provinces of india:— ' ho yl 2 bp oar 25 sor| be eee “ec gol ges om) us a tho v tle 4 by | ee | seeny 8 caa= oo ole 8:0 ne = ax wos) 3 £.o : net area ~ 5 ww w@w [se oe) "a oe tre) is ar 26 och 5 xo | 6.95| dok + aay ge oy wh = — — so |ase¥l age = s ba | oer | ec uae e2 | $5 |s638) b25 oe 5% ge) bie te a oe eal _ |lakhs| lakhs acres acres nr of of | rupees| rupees madras. . | 36,424,000] 6,891,000] 18-9 | 1,207 { 3,550 bombay deccan | 39,000,000 418,000 | 1:0 r81 538 sind : . 4,134,000] 3,427,000, 82-9 479 | 1,054 bengal . . | 22,806,000 93,000] of 22 78 united provinces 35,011,000] 1,979,000| 5:6 | 1,577 1,348 punjab 26,731,000 | 10,207,000 | 38:2 | 2,543 [ 5,505 burma 13,857,000] 1,730,000| 12°5 363 812 bihar and orissa 24,665,000 954,000} 3:9 627 622 central provinces 17,427,000 438,000} 255 483 281! north west frontier province 2,583,000 359,000 rajputna . ; 281,000 16,000 baluchistan. 286,000 26,000 total 223,205,000 | 26,538,000 thus nearly 12° of the net area cropped was irrigated by government works. the annual value of the crop so irrigated amounted to 14 times the capital outlay expended on the works. new schemes; afadras presidency.—(1) the improvement of the, at present, somewhat irregular supply of water in the can- vey delta canals will add about 301,000 ac. to the 1,000,000 ac. already served. (2) the vengalapuran scheme will provide water for 50,000 acres. (3) the kishna river storage will serve about 1,750,000 acres. the lower bhavari project will command 110,000 acres. bombay presidency—although only 1% of the land cropped was irrigated in the bombay deccan district, its annual average of irrigated land is now about 418,000 acres. the bhavari river works and the nira right bank canal scheme will increase this area by 35,000 acres. sind.—the rainfall in sind averages only about 6 in. per 544 annum. in one year out of five it is almost certain to be less than 2 inches. sind, however, has the indus carrying a supply of water which is capable of meeting her needs if properly utilised. in 1926 inundation canals supplhed what land is irrigated. there are no means of regulating the river level artificially and continuing the supply when the river falls, consequently these canals obtain a full supply only when the indus is in flood. in the cold weather when the river is low, only the most fortunately situated obtain any supply at all and a minimum of 600 tons per sec. run to waste. ' the cultivable area of the province is about 14,000,000 acres. the indus is estimated to be capable of irrigating 12,000,000 ac. in the hot weather and 8,000,000 ac. in the cold season, or say 20,000,000 ac. in all of crop. in 1923-4 the net area cropped was 4,134,000 ac. of which 82-9 °4 or 3,427,000 ac. was irrigated; this is the highest per- centage of irrigated to cropped area in the whole of india. the province naturally divides itself up into three main tracts, upper, central and lower sind, and three large barrage projects are con- templated at an estimated cost of {50,000,000 to deal with them. the main project now in hand consists of a barrage across the indus just below sukkur where the river passes through a deep gorge. from the barrage, seven canals (four on the left bank and three on the right) will run southwards and extend facilities for cul- tivation to the enormous area of 5,500,000 acres. 2,000,000 ac. represent the existing inundation irrigation area and will be given in addition an assured perennial supply by the new canals. the barrage when completed will measure 4,725 ft. between the faces of the regulators on cither side. it will have a masonry floor across the river, above which an over-bridge of 66 spans, each of 60 ft., will be constructed. from this over-bridge shutters 60 ft. long and 18} ft. high can be lowered on the floor when the river falls, thus pounding up the water to the level required in the canals; conversely, when the river rises, they can be lifted and the whole of the waterway of the bridge will be left free for the passage of the annual floods. these floods will be greater than those passing any other regulating structure in the world, at the maximum they will be nearly double the nile flood. the supply required by the whole system of canals will vary throughout the year from 22,600 cu. ft. per sec. in jan. and feb. to 46,000 cu. ft. per sec. from june to september. it is estimated that the annual net revenue after paying working expenses will be 103°, on the capital expended. this is the return from water rates alone, but a further large increase in general revenues may be reckoned upon from the 3,500,000 ac. of waste land which will be brought under cultivation. bengal.—the percentage of 0-4 of irrigated to cropped area in bengal is the lowest in the whole of india, the net area cropped being 22,806,000 ac. of which only 93,000 ac. are irri- gated. the average rainfall is about 55 in. per annum. <a project for a canal from the damodar river is sanctioned which will irrigate 196,0co acres. several minor works are being considered. punjab.—with the exception of sind no portion of india is so favourably situated as regards its rivers, or so unfavourably as regards its rainfall, as the punjab proper, expecially the tract between the jhelum and the sutle)—the greater portion of this district having less than 15 in. of rainfall per annum and much of it less than ro inches. as a consequence great developments have already taken place in the irrigation scheme of the province. in all 10,207,000 ac. were irrigated in 1923-4. further large works are contemplated which will increase the total and reduce to a minimum the waste areas. the united states one of the larger projects is that connected with the colorado river which serves two distinct areas. the upper is at a consid- erable elevation above sea level and covers 4,500,000 acres. of this about 1,500,000 ac. are now irrigated. in the lower basin there are 960,000 irrigable acres, of which about 670,000 are irrigated. when the works are complete the colorado river will be controlled to a very large extent and thus reduce the dangers arising from excessive floods. an interesting feature of u.s. irrigation works is the series of very high dams which have been built to control and conserve river supplies. the elephant butte dam has a height of 306 ft. and the reservoir has a storage capacity of 3,200,000,000 tons of water. so far this is the largest reservoir in the world. the shoshone dam has a height of 328 ft. and has a capacity of 550,- 900,000 tons of water. the roosevelt dam has a height of 280 [t. irrigation engineering and the reservoir has a capacity of 1,650,000 tons of water. the arrowrock dam has a height of 348-5 ft. and the reservoir has a capacity of 340,000,000 tons of water. the pathfinder dam is 218 ft. high and the reservoir can contain about 1,300,- 000,000 tons of water. in all, the bureau of reclamation has 25 schemes either in opcration or in process of construction. canada irrigation in canada has been so far very partially developed and only in those provinces where extensive farming operations are in progress, such as alberta, saskatchewan and manitoba. developments are based upon the administration of the federal irrigation act of 1894, under which the ownership of all surface water supply is vested in the crown and the latter grants the necessary licences for its use. occasional droughts occur all over the wheat-growing belts; provision, however, is steadily and systematically being made to supply as much as possible irrigation water to counteract their effects. the government authorities are systematically obtaining data with a view to providing the maximum area financially possible with water and are yearly carrying out expensive systems of irrigation. on the eastern section of the canadian pacific railway co., a census of the yields obtained and the water used by a group of ro farmers during the past season, 1924, shows that the average vield of wheat on these 10 farms was 19} bu. per ac. with one irrigation of 4 in. deep plus rainfall; with two 4-in. irrigations the yields received were from 30 to 35 bu.; and with three 4-in. irri- _ gations were as high as 43 bu. per acre. in this district the total precipitation for 12 months was 11-24 in. of which 9-68 in. fell during the growing season. the water available for irrigation varies greatly in different provinces, some are much more favour- ably situated than others to receive it and certain provinces on the other hand hardly require irrigation. that the total water resources of canada, however, are enormous can be gauged from the fact that it is estimated that there could be developed with it 41,400,000 hi.p. of which only 3,227,414 h.p. was being used up to feb. 1 1924. some of the more important irrigation areas are mentioned below. alberta. —about 1,000,000 ac. were irrigated in 1923, of which 130,- 000 ac. were brought in about 20 years ago in the lethbridge section, canadian pacific railway, western section.—a scheme has been in operation in this district for about 16 years, the area being 218,980 acres. the 1923-4 season was, however, so ideal for crop production, owing to favourable moisture conditions, that little irrigation was used, and only 3,074 ac. were irrigated, chiefly for alfalfa and meadow grasses. canadian pacific railway, eastern section.—irrigation has been in operation in this section for about io years, the total irrigable area is 400,000 ac. which are gradually being developed, and of which 124,000 have been taken up. the conditions here render irrigation more necessary than in the western section owing to lower precipitation and slightly higher temperature. taber district—lirrigation was established in may 1924, in the taber district. of 17,244 irrigable acres, 3,625 ac. have already been taken up and irrigated. lethbridge northern irrigation district.—this project was com- pleted in may 1923. the irrigable area is about 105,000 acres. united irrigation disirtct—the irrigable area of this scheme ts about 36,000 acres. new west irrigation district—the irrigable area is 4,500 acres. bellaw lomond irrigation district-——the irrigable areca is about 55,000 acres. there are other smaller installations now working, and many more are proposed. : irrigation in australia owing to the lack of an adequate rainfall the advantages of irrigation appealed many years ago to australians. at first the object aimed at was to develop in unoccupied territory. while these efforts generally proved successful, in recent years the policy has been to extend irrigation to existing pastoral settle- ments, and some very large conservation of flood water schemes have been carried out under which considerable areas of land are how in process of intenser settlement. irrigation engineering new south wales,—-the principal works in this state include:— (1) murrumbidgee gravitation irrigation scheme.—the murrum- bidgee river is controlled by a dam 240 ft. high at burrinjuck which can conserve about 960,000,000 tons of water. the scheme is to irrigate about 200,000 ac. of which 120,000 ac. were settled by june 1923. they are mainly devoted to vegetable and fruit-growing and dairying purposes. (2) pump irrigation areas.—(a) at curlwaa on the murray river about 2,000 ac., out of an area of about 10,600 ac., are receiving irrigation. (6) at hay about 1,000 ac., out of an area of 4,500 ac. are receiving water from pumps on the murrumbidgee river. (3) projected schemes.— new south wales is investigating schemes which will utilise its share of the murray river waters and which it expects may amount to 150,000,000 tons. this should bring under irrigation an area of at least 100,000 ac. of land. it has also under consideration a number of smaller schemes affecting the lachlan, macquarie, hunter, namoe and peel rivers, but it is not possible to give any estimate as yet of the areas which may be irrigated by them. victorta.—in victoria the principal irrigation works are on the goulbourne, murray, loddon, werribee and macallister rivers. while the works for some of these schemes were com- pleted before 1910, the areas to be irrigated are still only in proc- ess of settlement and extension of the works are from time to time taking place. in 1923, 350,727 ac. were irrigated. under the goulbourne scheme, which comprises an area of about 850,000 ac. and where 800,000,000 tons of storage has already been provided for, a dam is being built at sugarloaf in two stages, the first one of which was so far completed that stor- ing of water took place in 1922. it may ultimately be 190 ft. high and may eventually contain about 1,100,000,000 tons of water. the murray river scheme already serves an area’ of 340,000 ac. and is capable of expansion when the hume storage reservoir 1s completed. this will form the largest sheet of fresh water in australia, as it will cover 47 sq. m. and may contain about 2,500,000,000 tons of water. the loddon river gravitation scheme serves 74,000 acres. the werribee river gravitation schemes at bacchus marsh and werribee serve 3,350 and 10,000 ac. respectively. the mildura pump irrigation scheme lifts water from the murray river and serves 45,000 ac., of which 13,000 ac. are under intense cultivation. the macdallister river scheme consists of a storage reservoir from which a gravitation supply will command about 80,000 ac. and is far advanced. these watering schemes are usually conjoined with domestic water supply schemes for towns and villages, and do not in the ordinary sense irrigate land, although they do, in fact, make great areas of land avail- able for pasturing purposes. there are also some flood protec- tion schemes, one of which at kooweerup and cardinia enabled 100,000 ac. of land to be brought into cultivation. quecnsland.—the dawson valley scheme now under construc- tion comprises a dam 140 ft. high at nathans gorge to impound about 3,100,000,000 tons of water. when completed it will be the second largest reservoir in the world. the area to be served is about 250,000 acres. the inkerman irrigation area of 4,500 ac. is served by 230 shallow well pumps. provision is being made to increase it to 10,000 acres. there are a number of smaller pump schemes at townsville, rockhampton, bingera and fairymead, which collectively serve about 4,000 acres. south australta.—the rennearth scheme serves 7,850 4ac., mainly fruit producing. the murray river pumping plants serve 17,800 ac. and are being extended to serve a further 11,000 acres. the cadett scheme serves about 1,200 ac. and is supplied with water pumped through go ft. of height. the waikcrie scheme serves about 9,800 ac. and has the water lifted through 150 ft. of head. the kingston scheme serves 500 acres. the moorook scheme serves 1,000 acres. the cobdoyla scheme is ready to serve about 3,600 ac. which can be increased to 30,000 ac. of irrigable land. the berni scheme serves 7,700 acres. the chaffey scheme will serve 14,000 acres. the murray swamp land scheme will even- | tually make available for irrigation 13,700 ac. of which 5,800 ac. are now cultivated. smaller schemes serve about 10,000 ac. in all. in western australia the harvey irrigation scheme serves 4,000 acres. : : 545 the union of south africa irrigation was at first confined to small schemes whose entire works usually lay within the boundary of one farm. works of greater magnitude were made easier of accomplishment when the govt. of the cape of good ihlope, to encourage irrigation, passed the act of 1876 known as the “ right of passage of water act ” in terms of which a proprietor was permitted to conduct water over the property of any other owner. ? a further act in 1877 provided for:— (a) the establishment of irrigation districts and boards, (6) the granting of government loans to private individuals, and (c) the repayment thereof at the rate of 8°, per annum over a period of 24 years. ie in the case of private schemes, both in the cape and in the other provinces also, no complete record of the irrigable area has been made and even in the case of board schemes such data as are available are frequently not based on actual surveys but are only estimates of the irrigable areas made during the earlier stages. since 1916 great progress has been made with irrigation based upon conservation of water schemes. the completed schemes, although numerous, are individually small in area, none exceeding 10,c00 acres. among those under construction or development are some of considerable magni- tude. these latter include the great fish river scheme, where 75,000 ac. are to be irrigated, the sundays river scheme of 36,000 ac. and the kamanassie river scheme of 28,000 acres. the area of land under irrigation in south africa as at march 31 1923 under board schemes was about 130,000 ac. and the area to be served by works under construction is about 225,000 acres. up to 1925 the state expended over {4,000,000 on completed schemes. neither interest nor redemption can be paid on this sum until the land is wholly settled. | _ work in china china with its huge population of about 300,000,000 has no doubt a very large area of irrigated land but no statistics are available as to its extent except in a few small special districts where europeans reside or have -commercial interests. hitherto engineering has been largely devoted to preventing the rivers in more than average floods overflowing their banks and inundating the land, an effect which has been many times accompanied by great loss of life. commissions have studied river control and the conservancy boards in recent years have reported on and carried out important works with this end in view. quite obviously there is a large field in china for this form of development, and no doubt in time it will be followed by the more usual irrigation works. the following opinion of the engi- neer reporting on the yangtse in 1923 gives a remarkably clear indication of the conditions there—it applies with equal force to the other great rivers of china. 7 to anyone who has inspected the yangtse from the goyes to the sea, and who has given the faintest consideration to the question of restraining a river of this magnitude in its course through hundreds of miles of low-lying land, thousands of square miles of its area being below high water level, and therefore subject to inundation, it must be obvious that draining works, to be effective, would have to be constructed on a gigantic scale, and although the river is, in itself, and its tributaries and creeks, practically the one means of transport and communication serving a population of nearly 200,000,000 people with ever-growing transport requirements, it must be obvious that, unless improvement works can be made self supporting, it is futile seriously to embark upon them. such are the conditions which prevail to-day and so they must remain for many years; in fact, until the growth of trade and growth of means which woukl follow thercon bring the necessity for works within the bounds of com- mercial! possibility. fh ly * developments in other countries ‘trag.—the construction of the hindia barrage was one of the first steps undertaken. in 1925 the diala cotton co. inaug- urated a great scheme whereby 108,000 ac. will be fertilised by water from the diala river, a tributary of the tigris. the com- pany, it is understood, intends to extend its operations as fast as possible under its concession; and a time can be envisaged when adequate control of the euphrates and tigris, under conditions 546 of peace and good government, will allow of great areas of the arid plains of “iraq to be once again cultivated. spain and portugal.—irrigation has been developed in a number of places in the peninsula since the beginning of the century and schemes for further works are being considered. none of those so far completed are of any great magnitude, but among the proposals there is one for the irrigation of 120,000 ac. on the guadalquivir. the possibilities of this river are being studied for other areas, and are great if by regulation of its excess flow in flood time its waters can be conserved for use in the drier periods of the year. on the tagus a scheme near villa franca is now being surveyed which, if carried out, will enable 30,000 ac. to be irrigated. arabia.—an interesting irrigation development is the possi- bility of reviving agriculture by its aid in the yemen. a syndi- cate is now (1926) studying a project to examine the many reser- voirs which in ancient times controlled flood waters for the benefit of agriculture in that region. mexico.—in mexico, and particularly in northern mexico where the rainfall is negligible in the lower or plain country, strips of land are cultivated near the rivers by irrigation. the total area is considerable, but there are no records of its extent. the works are somewhat primitive but have been constructed entirely by private enterprise and good results in crop are ob- tained from them. south america.—an irrigation scheme on the rio negro was carried out in 1914 whereby 230,000 ac. are being developed. there are other minor irrigation works in the country. there are irrigation works, carried out almost entirely by private enter- prise, in the other south and central american republics. (m. mac.) irwin, edward frederick lindley wood, 1st baron (1881- ), british politician, was born april 16 1881. the only surviving son of the 2nd viscount halifax, he was educated at eton and at christ church, oxford, being elected fellow of all souls college in 1903. in jan. 1910 he was returned to parliament as conservative member for the ripon division of yorkshire, and retained his seat until 1925. during the world war he served in france 1915-7 with the yorkshire dragoons, in which he held the rank of major, being mentioned in dis- patches, and was an assistant secretary to the minister of national service 1916-8. in april 1921 he was appointed under-secretary of state for the colonies. in this capacity he made a tour of the west indian colonies, and published his conclusions on the administrative and economic problems which he encountered in a report issued in june 1922. from oct. 1922 to jan. 1924 he was president of the board of education. in nov. 1924 he became minister of agriculture and fisheries in mr. baldwin’s second government. in oct. 1925 he was appointed governor-general of india in succession to the marquess (then earl) of reading, being raised to the peerage as baron irwin of kirkby underdale two months later. like his father, lord irwin, who had been sworn of the privy council in 1922, took a keen interest in ecclesiastical affairs, and published a study of john keble in the ‘‘ leaders of the church series.” he married in 1909 lady dorothy onslow, younger daughter of the 4th earl of onslow by whom he had three sons and two daughters. ishii, kikujiro, viscount (1866- ), japanese diplo- matist, was born at chiba, japan. like other members of his family, he was destined to enter official life. his intention was to become a lawyer, but his studies in international law gave him a deep interest in foreign affairs. his early days were full of incident: the meiji reconstructions took place before his eyes, and the constant stream of foreign influences soon aroused in him a desire to see and understand other countries. leaving the law faculty of the imperial university in tokyo in 1890, he was made attache to the japanese legation in paris. during his stay in paris he applied himself to the study of european languages and became proficient in french and english. he was made third secretary in 1893, and about this time he devoted much attention to the problems of economics and international irwin—ismet pasha trade. he was distressed to observe that japan did not hold such a position in the world’s mart as in the opinion of most japanese she deserved, and he set himself to find out the causes of the trouble. he endeavoured to form a franco-japanese mutual trade association, and although his efforts did not at first meet with the full success he desired, a certain stimulus was at once felt. some part of the fruits of ishii’s labours were seen in the commercial mission which was sent from france to japan in 1925. in 1896 he was appointed consul to chemulpho (jinsen) in korea. after this he was made second secretary and later first secretary at the japanese legation in peking. he was, there- fore, in peking in an official capacity during the boxer rising. in 1900 he was appointed secretary of the head office and chief of the telegraph section of the home office department, which post he held until he became director of the commerce bureau in rg04. in 1907 he was sent to san francisco and vancouver in connection with the anti-japanese riots there. his success in this matter was the direct cause of his appointment in the follow- ing year as vice-minister for foreign affairs. his love of france and understanding of french problems made him an easy favour- ite for the ambassadorship to paris in 1912, and on his return he took over the portfolio for foreign affairs in the new govern- ment. he was chief of the foreign office for the period 1915-6, and in the latter year he was created viscount, having been made baron in 1912. he played a great part in american-jap- anese relations and was a special envoy to the united states in 1917. he was nominated a member of the house of peers in 1916. in 1920 viscount ishii was for the third time officially delegated to paris. perhaps ishii’s greatest claim on his country’s gratitude was his work in connection with the so-called ‘‘ gentlemen’s agree- ment ” made between washington and tokyo in 1907. lansing was the united states representative in the negotiations, for which reason the arrangement is sometimes called the “ lansing- ishii pact.”” the western coast of the united states made rep- resentations to congress deploring the increasing immigration of asiatics, notably chinese and japanese. it was claimed that owing to the asiatic standard of living being lower than that of americans, there was a serious possibility of california becoming a japanese colony, and the aid of legislation was invoked. the lansing-ishii pourparlers resulted in the postponement of sug- gested legislative measures, and a yearly maximum (tentatively fixed at 150) was proposed. japan, through ishii, agreed not to issue passports for more than this number annually to enter the united states as residents, and it was through japan’s alleged breach of faith in connection with this agreement that the immi- gration restriction act was passed by congress in 1924. in substance, this act was one of the proposals originally shelved by the labours of lansing and ishii. in dec. 1920 viscount ishii became japanese delegate to the r1th session of the council of the league of nations. in aug. 1923 he became president of the council of the league and in the following month was one of the vice-presidents of the assembly. in sept. 1925 he was again a vice-president of the assembly. in march 1926 he was president of the council at its 39th ses- sion, and as such acted as president of the assembly (march 1926) until the election of the new president. (a. n. j. w.) islamism: see pan-islamism. ismet pasha (1884- ), turkish statesman, was born in smyrna of a turkish family of malatia. he received his educa- tion in the military schools of turkey, and in 1903 graduated as an artillery lieutenant. three years later he became a captain on the general staff. after the young turk revolution of 1908, he was sent several times on military expeditions to yemen. during the world war he served with distinction on the syrian front, and at the close of the war held the rank of colonel, and commanded an army corps. during the armistice he occupied important positions in the war office. at the time of the occu- pation of constantinople by the allies he escaped to angora dis- guised as a common soldier and was appointed by the great national assembly to be chief-of-staff during the military opera- isonzo, battles of the—isotopes tions of the nationalists against the caliphate army and ar- menia. later he became commander-in-chief of the turkish western front, and in this capacity insisted on the formation of a regular army in place of the irregular nationalist forces. in the meantime, the irregular forces were unwilling to submit, and the greek army had taken the offensive. at ineunu, ismet defeated the greeks, checked their offensive, subdued the irregular forces, and succeeded in completing the organisation of a regular army. he remained in command of the western front until after the battle on the sakharia and the fall of smyrna, in which events he was the right-hand man of mustafa kemal pasha. in 1922 he signed the armistice concluded at mudania with the allies, and succeeded in securing the evacuation of eastern thrace. in 1922-3, as minister of foreign affairs and senior turkish dele- gate to the lausanne conference, he most ably defended the interests of his country, and signed the treaty of lausanne. during the autumn of 1923, at the time of the declaration of the republic, he became prime minister. after a short rest, due to failing health, he became prime minister a second time, to- wards the end of 1924, and shared in the suppression of the shiekh said revolt in kurdistan. the abolition of the caliphate, the closing of the medressehs (religious schools) and of the tekkes (monasteries), the abolition of the tithe, the adoption of the swiss civil code, and the wearing of the hat are some of the mo- mentous changes which were realised during ismet’s tenure of office. isonzo, battles of the: see italian campaigns. isostasy.—when the great trigonometrical survey of india was initiated, it was found that the deflection of the plumb-line by the himalayas was much less than the calculated amount due to the theoretical attraction of the visible mass of the moun- tains. sir g. b. airy suggested that this might be caused by the presence of a mass of matter, of less than the average density, under the mountains; this explanation was further investigated by archdeacon j. h. pratt, who applied the term “ compensa- tion ” to the negative effect of the underlying defect of density, in compensating the direct effect of the attraction of the visible mass of the mountains. in 1892 maj. c. e. dutton, discussing the greater problems of physical geology, deduced a general principle that the weight of matter under any unit area of the earth’s surface tended to be- come uniform, and suggested that this was brought about by an underground transfer of material to balance the visible surface transport from regions of erosion to those of deposition. to this principle he gave the name jsostasy (ticos equal, and oraats position), not as a synonym for pratt’s compensation, but as a name for the principle and process, by which it was brought about. in 1909 there appeared a very complete and elaborate investigation of the subject by j. f. hayford, in which the word “isostasy ’’ is used as synonymous with what pratt called “com- pensation,”’ and this use of the term has since become general among geodesists. some inconvenience results from this two- fold use of the word, to express either a measurable effect, or one theory of the cause by which this effect is produced; but the usage is too firmly established to be overcome, and the incon- venience is lessened as it has become recognised that the deep- seated variations in density, and consequently in bulk, to which archdeacon pratt gave the name “‘compensation,’’ may be the primary phenomenon and cause of the major differences of sur- face level. (see also geodesy; geology.) see g. b. airy, phil. trans., 1855, vol. 145, p. 101; j. h. pratt, phil. trans., 1859, vol. 144, p. 745; c. e. dutton, bull. phu. soc. washing- ton, 1892, vol. 11, p. 51; j. f. hayford, the figure of the earth and isostasy, from measurements in the united states (washington, 1909); w. bowie, “ abnormal densitie#in the earth’s crust disclosed by analysis of geodetic data,’’ geog. jour., 1924, vol. 63, p. 26. {r. d. o.) isotopes (gr. tcos equal-+-rozos place), is the term first ap- plied by f. soddy in 1913 to substances which, though they had different atomic weights, yet had identical chemical properties and occupied the same place in the periodic table of the elements. over a century earlier dalton had postulated that atoms of the same element are similar to one another and equal in weight. 547 a little later prout suggested that the atoms of all elements were composed of atoms of a primordial substance which he en- deavoured to identify with hydrogen. if both these theories were right, the atomic weights of all ele- ments would be comparable with each other as whole numbers. this the chemists soon showed was quite incompatible with experimental evidence. it is true that many were very nearly whole numbers, far too many for the effect to be pure chance, but others, like chlorine, were hopelessly fractional. of the two alternatives dalton’s is much the simpler from the chemical point of view and was therefore quite rightly chosen as a working hypothesis. from this in course of time it developed into an article of scientific faith, and, despite the complete absence of positive evidence in its support, no serious questions as to its validity were raised until late in the 19th century. of such speculations those of crookes were founded on unsound evidence and were soon discredited. the question could not be settled by ordinary chemical methods, which employ countless myriads of atoms and could therefore only give a mean result, and it was only by the discovery of radioactivity and the development of accurate methods of weighing individual atoms that the exist- ence of isotopes was disclosed. the two advances were nearly con- temporaneous but the first definite convincing proof of isotopy was found among the radioactive elements and their products. the radioactive isotopes.—in 1906 boltwood discovered ioni- um and found that it had similar chemical properties to thorium. further research using the most delicate radioactive methods failed to indicate the slightest chemical separation of these two elements once they had been mixed, and even more surprising their spectra appeared to be identical. other pairs of elements in the radioactive group showed corresponding identities and later investigations on the chemistry of the products of radioac- tive disintegrations enabled the chemical law of radioactive change to be formulated (see radioactivity). this stated that a radioactive element when it loses an alpha particle goes back two places in the periodic table; when it loses a beta particle it goes forward one place. an alpha particle is a helium nucleus of weight 4, whereas a beta particle is an electron of negligible weight. it follows that if a body loses one alpha and two beta particles it will be back again in the same place in the periodic table although it will have lost a mass equal to four units of atomic weight. = isotopic bodies predicted.—supported by this law, which he was the first to state in its most general form, soddy boldly claimed that these ‘‘ isotopic ” bodies would be both chemically and spectroscopically indistinguishable. he also predicted that the lead produced by the disintegration of uranium would have an atomic weight 206, while that of the lead produced from tho- rium would be 208, and that consequently the atomic weight of the lead found in uranium minerals should be jess than that of ordinary lead (207-2) while that of lead from thorium minerals should be greater. these predictions were amply vindicated during the war by the work of experts on atomic weights (rich- ards, henigschmid and others), and it was shown beyond all dispute that the isotopic leads, though they differed by the pre- dicted amount in properties such as atomic weight, density, and solubility which depend directly on the weight of their atoms, in all others, which do not—atomic volume, boiling-point, melting- point, refractive indexand spectrum— were quite indistinguishable. practical utilisation of indicators—the impossibility of separating isotopes has been utilised in ingenious manner by hevesy and paneth. by the addition of a small quantity of a radioactive isotope to an ordinary inactive element, the latter is, so to speak, indelibly labelled and can be followed by the methods of radioactivity, which are incomparably more delicate than those of chemistry. in this way the solubility of very in- soluble salts can be readily determined. by the addition of a little thorium b, an isotope of lead, valuable information has been obtained on the assimilation of the salts of the latter ele- ment by living plants. the use of such radioactive indicators affords a direct proof of the ionic dissociation theory, it has led to the discovery of certain metallic hydrides and to the deter- 548 mination of the velocity of diffusion of molecules among them- selves, an otherwise insoluble problem. | postiive ray analysis.—lt is clear that the reasoning which led to the discovery of isotopes among the radioactive elements is inapplicable to those much more abundant ones which do not exhibit this property. among the latter the presence of isotopes can only be demonstrated by weighing their individual atoms. this can be done by the analysis of positive rays, also called kanalsirahien or mass rays, which are electrically charged atoms of matter moving with such high velocities that they can be de- tected by means of their impact on a fluorescent screen or photo- graphic plate. the usual method of producing these is by the discharge in gases at low pressure. in the region of the crookes dark space and negative glow in front of the cathode the gas is ionised; that is to say, its atoms are split up into positively and negatively charged parts. the latter are the same whatever the nature of the gas, they are electrons and fly away from the cath- odes constituting the well-known cathode rays. the remaining positively charged particles will fly toward the cathode. their masses may vary from that of the lightest atom to that of the heaviest molecule in the gas, and their energies from an indefi- nitely small value to a maximum expressed by the product of the charge they carry, multiplied by the total potential applied to the electrodes. if the cathode be pierced, the rays pass through the aperture and form a stream of particles heterogeneous both in mass and velocity, which can be subjected to analysis. the “ parabola”’ method.—in sir j. j. thomson’s “ parab- ola ’’ method of analysis, the particles, after reaching the sur- face of the cathode, pass through a long and very fine metal tube. by this means a narrow beam of rays is produced, which is passed through electric and magnetic fields causing deflections at right angles to each other, and finally falls upon a photographic plate. it can be shown that, if the mass of any particle is m and its charge e, when both fields are on together, the locus of impact of all particles of the same e/m, but varying velocity, will be a parabola. since e must be the electronic charge, or a simple mul- tiple of it, measurements of the relative positions of the parab- olas on the plate enable us to calculate the relative masses of the particles producing them—that is, the masses of the individ- ual atoms or molecules. the fact that the streaks were definite, sharp parabolas, and not mere blurs, constituted the first direct proof that atoms of the same element were, even approximately, of equal mass. for some time the results of the application of this method of analysis appeared to support the hypothesis of dalton, as the elements introduced into the discharge tube gave single, or apparently single, parabolas in the positions expected from their chemical atomic weights. (see gases, electrical properties of.) = | . examination of neon.—but when, in 1912, neon was examined, the trace obtained was definitely double. the brighter curve cor- responded roughly to an atomic weight of 20, the fainter com- panion to one of 22, the atomic weight of neon being 20-20. the line 22 could only be explained as due to a hitherto unknown elc- mentary constituent of neon. this agreed well with the new idea of isotopic elements which was just then emerging from the in- vestigations on radioactivity, so that it was of importance to investigate the point as fully as possible. the first line of attack was an attempt at separation by fractional distillation, but the result was entirely negative. the second method employed was that of fractional diffusion through pipe-clay which gave a small, but definite, positive indication of separation. it therefore seemed probable that neon was a mixture of isotopes. _ the mass-spectrograph.—by the time that research on the subject was resumed at the cavendish laboratory in 1919, the existence of isotopes among the products of radioactivity had been proved beyond all reasonable doubt by the work on the atomic weight of lead. this fact automatically increased the value of the evidence of the complex nature of neon and the ur- gency of its definite confirmation. it was realised that separation could only be very partial at the best, and that the most satis- factory proof would be afforded by measurements of atomic weight by the mcthods of positive ray analysis. these would isotopes have to be so accurate as to prove beyond dispute that the ac- cepted atomic weight lay between the real atomic weights of the two constituents, but corresponded with neither of them. the parabola method was not equal to this, but the required accuracy was achieved by means of an instrument shown in diagram in fig. 1. in ot ; i i 3) eo a la ee ee ee ee ——_— = = b fic, 1—diagram of mass spectrograph. positive rays are sorted out into a thin ribbon by two parallel slits sis, and are then spread into an electric spectrum by means of the charged plates pi:ps. a portion of this spectrum de- flected through an angle @ is selected by the diaphragm d and passed between the poles of a powerful magnet o, the field of which is such as to bend the rays back again through an angle @ more than twice as great as @. the result of this is that rays having a constant mass (or, more correctly, constant em) will converge to a focus f, and if a photographic plate is placed at gf as indicated, a spectrum dependent on mass alone will be obtained. on account of its analogy to optical apparatus, the instrument has been called a mass-spec- trograph and the spectrum produced a mass-spectrum.! mass-specira.—fig. 2 shows a number of typical mass- spectra that are obtained by this means. the numbers above the lines indicate the masses they correspond to on the scale o=16. it will be noticed that the displacement to the right with increasing mass is roughly linear. the measurements of mass are not absolute, but relative to lines which correspond to known masses. such lines due to hydrogen, carbon, oxygen and their compounds, are generally present as impurities or purposely added, for pure gases are not suitable for the smooth working of the discharge tube. the two principal groups of these reference lines are the c; group, due to c (12), ch (13), ch, (14), ch, (15), ch, or o (16), and the c, group (24-30) containing the very strong line c,h, or co (28). these groups will be seen in several of the spectra reproduced, and they give, with the co, (44) line, a very good scale of reference. alass measurements —mecasurements of mass can be made with an accuracy of 1 part in 1,000. it must be remembered that the ratio of mass to charge is the real quantity measured by the position of the lines. many of the particles are capable of carrying more than one charge. a particle carrying two charges will appear as having half its real mass; one carrying three charges as if its mass was one-third, and so on. lines due to these are called lines of the second and third order. lines of high order are particularly valuable in extending our scale of reference. when neon was introduced into this apparatus, four new lines made their appearance at 10, 11, 20 and 22. the first pair (not shown in diagram) are second order lines. all four are well placed for direct comparison with the standard lines, and a series of consistent measurements showed that to within about one part in a thousand, the atomic weights of the isotopes composing neon are 20 and 22 respectively. ten per cent of the latter would bring the mean atomic weight to the accepted value 20-20, and the relative intensity of the lines agrees well with this propor- tion. the isotopic nature of negn was therefore settled beyond doubt. spectrum i. on fig 2 (see next page) shows the first order lines of neon and some of the reference lines with which they were compared. | | analysts of chlorine.—the element chlorine was naturally the next to be analysed, and the explanation of its fractional atomic weight (35-46) was obvious at once. its mass-spectrum is char- 1 a full description of the mass-spectrograph is given in isotopes by f. w. aston, cambridge university press, pp. 44 e¢ seg. isotopes acterised by four strong lines 35, 36, 37, 38. the simplest expla- | nation of the group is to suppose that the lines 35 and 37 are due to the isotopic chlorines, and the lines 36 and 38 to their corre- sponding hydrochloric acids. the elementary nature of 35 and 37 is indicated by their second order lines at 17-5 and 18-5, and also when phosgene was used, by the occurrence of lines at 63 and 65 due to coc]® and coci*’. later it was found possible to obtain the spectrum of the nega- tively charged atoms of chlorine. this showed only two lines 35 and 37, so that the lines 36 and 38 cannot be due to isotopes of the element. these results show that chlorine is a complex element, and that its isotopes are of atomic weight 35 and 37. spectra ii., iil. and iv. show the results with chlorine taken with different field strengths (see chemistry). other elemenis—as the work progressed with other elements further interesting results were obtained. some elements, such as carbon and oxygen, were found to be “ simple’; that is, not mixtures of isotopes. this was to be expected from their whole- ys wv) :. wit 549 the original hypothesis of prout can now be restated with the modification that the primordial atoms are of two kinds: protons and electrons, the atoms of positive and negative electricity. according to the modern theory of the nucleus atom (see atom; matter) all the protons and about half of the electrons are packed very close together to form a central, positively charged nucleus, round which the remaining electrons circulate, some- what like the planets round the sun. all the spectroscopic and chemical properties of the atom depend on the nct positive charge on the nucleus, which is the excess of protons over nuclear electrons. this is also clearly the number of planetary electrons in the neutral atom; it is called the “ atomic number” and is actually the number of the element in the periodic classification: 1 for h, 2 for he, 3 for li, and so on. the whole-number w eight of the atom, on he other hand, will be the total number of neutral pairs of protons and electrons it contains. this is also the number of protons in its nucleus, and is called the ‘‘ mass-number ”’ of the atom: 1 for h, 4 for he, fig. 2. typical mass spectra. 93 number atomic weights. even more proved “ complex,’’ some consisting of 7, and in the case of xenon possibly 9 isotopes. the complexity of mercury is indicated by the blur of its unresolved third and fourth order groups seen in spectrum iv. its lines have been resolved by means of a more powerful instrument. mass rays of the metallic elements, which are in the majority, cannot in general be produced in the ordinary vacuum discharge. they can be investigated by means of anode rays. the constitution of the alkali metals was first discovered by the use of an anode consisting of a platinum strip coated with salts of the metals, and heated electrically. dempster, at chi- cago, produced mass rays of metals by heating the element in a furnace and ionising the vapour produced by electron impact. to analyse the rays so formed he used low velocities and an electrical method of detection. by this means he made the first analyses of magnesium, calcium and zinc, and also confirmed the results already obtained for the lighter alkali metals. more recently a large number of elements, including some of the rare earths, have been successfully attacked by means of the special method of ‘‘accelerated anode rays,” and in all 56 out of the 80 known non-radioactive elements have been analysed into their constituent isotopes or shown to be simple. the whole-number rule. by far the most important general result of these investigations is that, with the exception of hy- drogen, the weights of the atoms of all the elements measured, and therefore almost certainly of all clements, are whole num- bers to the accuracy of experiment. with the mass-spectro- graph, this accuracy is generally one part in a thousand. of course, the error expressed in fractions of a unit increases with the mass measured, but with the lighter elements the divergence from the whole-number rule is extremely small. this enables sweeping simplifications to be made in our ideas of mass. 6 and 7 for the isotopes of li, and so ou. for the purpose of dis- tinguishing isotopes it is customary at present to use the chemi- cal symbol of the complex element with an index corresponding to the mass-number of the particular isotope, e¢.g., ne”, rb®7, packing ieffect—the whole-number rule is not mathemat- ically exact, for owing to the extremely close packing of the charges of opposite sign in the nucleus, this will have a weight slightly less than the sum of the weights of its constituents. this loss of weight is called the “ packing effect.’?’ hydrogen, which has for its nucleus a single proton, will have no packing effect, and may therefore be expected to have an abnormally large weight. comparison of this clement with others by means of the mass-spectrograph, using special methods, shows that, within experimental error, the weight of its atoms on the chemi- cal scale (o= 16) is indistinguishable from that obtained by chem- ical methods, namely, 1-0077. the divergences of other elements are barely measurable at present, but the isotopes of lithium appear heavier than whole numbers by about 1 part in 1,000. most of the heavier elements show a slight defect, which is to be expected if we take oxygen as having normal packing. the defect in the case of the isotopes of tin appears to be at least 2 parts in 1,000, which 3s 0:2 of a unit of atontic weight. these considerations are of the greatest interest theoretically, for upon them rests the hope of the future liberation of the so- called “‘ atomic energy.” (see atomic energy.) the theoreti- cal importance of chemical atomic weight has been somewhat reduced by the discovery that for so large a number of elements it merely represents a statistical mean. its position as a natural numerical constant associated with an element has now been taken by the atomic number, which indeed defines the element, though from the point of view of chemical analysis, the mean 990 atomic weight is as important as ever. the anomalies shown by those elements which, by their atomic weights, appear out of their right order in the periodic table, are now open to the sim- plest explanation. thus argon, in which the heavier of two isotopes predominates, has a greater mean weight than potas- sium, in which the reverse is the case. the same explanation applies to cobalt, nickel, tellurium and iodine. since the atomic number only depends on the wef positive charge on the nucleus, arithmetically any element can possess an indefinite number of isotopes. the table herewith shows that those present in detectable quantity are restricted both in num- ber and range of weight, though the causes of these restrictions are at present unknown. no element of odd atomic number has more than two isotopes and, above atomic number 9, the mass- numbers of the isotopes always differ by 2, and the lighter is the more abundant constituent. the number of nuclear electrons tends to be even. that is, in the great majority of cases even atomic number is associated with even mass-number, and odd with odd. beryllium and nitrogen are the only elements con- sisting entirely of atoms whose nuclei contain an odd number of electrons. if the mass numbers of the various species of atoms are plotted against their relative abundance in the earth’s crust, a strong preponderance of those of type 8n may be seen. there is an extreme difference of range between the abundance of isotopes in an element and elements in nature. in the case of elements of an odd atomic number this cannot be ascribed merely to lack of delicacy in the means of detection of their isotopes. thus while there are only about three cl® atoms to one cl? and about two ga® atoms to one ga”, yet there are a thousand million more atoms of chlorine than of gallium. this suggests that isotopes have some relation in common more fundamental than that of identity of nuclear charge, an idea which is supported by other independent lines of reasoning. spectra of isotopes.—as regards their series spectra, in which, on bohr’s theory, the two bodies concerned are an electron and an enormously more massive atomic nucleus, the prediction that isotopes should be indistinguishable is satisfied to a high degree of precision. so far the only effect detected is a minute difference of wave-length between the lines of carnotite lead (206) and ordinary lead (207:2). the most accurate measurements by merton indicate that this has a maximum value of o-o1r a for the line x =4058. smaller shifts are detectable in a few other lines, the jwave-length for the lighter atom being the greater in all cases. in band spectra, where two nuclei are concerned, the isotope effects are much larger and in excellent agreement with theory. in the case of hcl bands in the infrared region the duplicate peaks are as much as 14 a apart, and in position and relative intensity, correspond exactly with the results expected from the presence of hcl and hce”. more recently by investigation of band spectra produced in the visible region by boron oxide and silicon nitride, millikan has shown that separate band heads appear corresponding to the two isotopes of boron, in the one case, and to three isotopes of silicon in the other. these results constitute valuable inde- pendent confirmation of the results of the mass-spectrograph. separation of isotopes.—it is perhaps a fortunate thing for the simplicity of chemical arithmetic that the artificial separa- tion of isotopes is excessively difficult, while at the same time no process tending to that end in nature appears to exist at all. of the artificial methods the only one giving complete separation is the actual analysis of the mass-rays, during which the isotopic atoms strike the plate at different points and therefore, if col- lected, would yield pure specimens. the quantities so produced would, with the means at present available, be far too minute to be of any practical value. _ a large number of methods for partial separation have been suggested and tried. the first successfully used, which is only applicable to gases, is that of free diffusion through pipe-clay or other suitable porous material. the diffusion rates are inversely proportional to the square roots of the masses concerned. it follows that if a large volume v of a mixture of isotopes is al- isotopes lowed to diffuse, leaving a small residue », the latter will be richer in the heavier constituent than was the original gas. the actual numerical value of this enrichment, under ideal conditions, with isotopes, such as those of neon, which differ by 10% is only 20 (v/»)4 so that only by the use of very large volumes, or laborious repe- titions, can any measurable change be achieved. the original experiments with neon gave a shift of atomic weight of rather more than o-1 of a unit. harkins, at chicago, by the use of 19,000 litres of hcl, was able to obtain considerable samples in which the atomic weight of chlorine differed by 0-055 unit. table of elements and isotopes mini- ‘ . mum mass numbers of element att eke a as isotopes in order of 2 ae intensity isotopes 1 i 1-008 i i e 2 4-00 i , li 3 6-94 2 ; 6 be 4 9-02 i 9 b 5 10-82 2 it, 10 c 6 12-00 i 12 n a ae 14-01 i 14 o 8 16-00 i 16 f 9 19-00 i 19 ne 10 20:20 2 20, 22 na ii 23-00 i 23 mg 12 24°32 3 24, 25, 26 al 13 26:96 i a7 si 14 28-06 3 28, 29, 30 p 15 31-02 i 31 » 16 32-06 i 32, 33, 34 cl 17 35°46 2 35, 37 a 18 39°88 2 40, 36 kk 19 39°10 2 39, 41 ca 20 40°07 2 40, 44 5c 21 45:1 i 45 ti 22 48-1 1 48 v 23 51-0 i 51 cr 24 52-0 i 52 mn 25 54°93 i 55 fe 26 55:84 2 56, 54 co 27 58-97 i 59 ni 28 58-68 2 58, 60 cu 29 | 63-57 2 | 63, 65 zn 30 65:38 4 64, 66, 68, 70 ga 31 69°72 2 69, 71 ge 32 72°38 3 74, 72, 70 as 33 74:96 i 75 se 34 79:2 6 80, 78, 76, 82, 77, 74 br 35 79°92 2 79, 81 kr 36 82-92 6 84, 86, 82, 83, 80, 78 rb 37 85°44 2 85, 87 sr 38 87-63 2 88, 86 y 39 88:9 i 89 zr 40 (91) 3(4) | 90, 94, 92, (96) ag 47 107 88 2 107, i cd 48 112-41 6 113 312, 116; 314, iii, f16 in 49 114°8 i 115 sn 50 118-70 7(8) 120, 118, 116, 124, 119, 117, 122. (121) sb 51 12177 2 i2i, 123 te 52 127°5 3 128, 130, 126 i 53 126-92 i 127 xx 54+ 130-2 7 (9) 129,132,131, 134, 136, 128,130,(126), (124) cs 55 132-81 i 133 ba 56 137-37 (1) 138 la 57 138-91 i 139 ce 58 140-25 2 ijo, 142 pr 59 140-92 i i4! nd 60 | 144-27 3(4) | 142, 144, 146, (145) hg 80 200-6 6 202, 200, 199, 198, 201, 204 bi : ; 83 209-00 i 209 another method, following much the same numerical laws, is that of bronsted and hevesey, which consists of free evapora- tion from a liquid surface at very low pressure. they obtained two samples of about 0-2 cc. of mercury differing in density by 5 parts in 10,000, or o-1 of a unit. the atomic weights showed a corresponding difference, but the electrical conductivity of the two samples was indistinguishable to one part in a million. israels—italian campaigns other methods of separation such as chemical action, centri- fuging, ionic migration and thermal diffusion have only yielded meagre or entirely negative results. a very large number of attempts have been made in recent years to discover any varia- tion in the chemical atomic weight of elements known to be complex, which would indicate a change in the proportions of the isotopes present. boron, silicon, chlorine, iron and nickel have all received attention, but in no case with any certain posi- tive result. from their experiments on silicon from no less than 12 different terrestrial and meteoric sources, jaeger and dijkstra concluded that these gave products not differing in density by more than 0-00004 %. the accumulation of negative evidence of this kind is very impressive, and supports the idea that the evolution of the ele- ments, apart from those produced by radioactive disintegration, must have been such as to lead to a proportionality of isotopes which was constant from the start, and, since we know of no natural process of separation, has remained constant ever since. see f. w. aston, isotopes, 2nd ed., 1924. (f. w. a.) israels, josef (1824-1911), dutch painter (see 14.885b), died at the hague aug. 12 1g11. itagaki, taisuke, count (1837-1919), japanese states- man, died in rg1g. true to his liberal principles, he forbade his son to apply for the succession to his title and it lapsed. italian campaigns.—italy entered the world war as a combatant on may 23 1915, when she declared war on austria. the following article constitutes an outline of the struggle be- tween italy and austria from that date until the signing of the austro-italian armistice on nov. 3 1918. the article has been divided into five sections, namely: i. italy’s strategical position; ii. the first year’s campaign; iii. battles of the isonzo and the carso;iv. the austro-german effort; v. the collapse of austria. more detailed accounts of the outstanding battles in the campaign are given under separate headings, namely: asiago, battle of; caporetto, battle of; and vittorio veneto, battle of. i. italy’s strategical position at the outbreak of the world war the italian general staff had no worked-out plan for an offensive campaign against aus- tria-hungary. the great military superiority of the habsburg empire and the unfavourable frontier drawn in 1866 seemed to deny the possibility of italian offensive action. a glance at the map makes the position clear. original italian plan.—when italy was preparing to enter the war, the fact that austria-hungary was already heavily engaged elsewhere offered the chance of an italian attack, and gen. ca- dorna, who took command on the declaration of war, had worked out his scheme in expectation of simultaneous offensive action on the part of russia and serbia. the plan was based on the idea that italy must hold on the north and push towards the east. the eastern front, though difficult enough, seemed less impervi- ous to an offensive than the alpine masses on the north. an attack in this direction was calculated to occupy a much larger number of enemy troops, which was obviously a part of cador- na’s duty, and was further a direct threat against a vital part of the monarchy. these advantages were rightly held to compensate for the fact that an advance eastward meant the lengthening of a front already very long, and increased, moreover, the menace of the trentino salient. this threat was mitigated by the expectation of allied action on the northeast and southern fronts of austria-hungary, which would prevent the enemy taking advantage of the weakness indicated. but, on the eve of italy’s declaration of war, the situa- tion had changed. the russian armies north of the carpathians had given way before the attacks of mackensen and behm- ermolli. there was no word of movement, even of demonstra- tion, on the part of serbia, in spite of requests from the allies, and in may the austro-hungarian troops on the serbian front were reduced by five divisions, which were sent to the isonzo- carso front, their place being taken by three newly-formed ger- man divisions. : 551 disposition of the forces —on the eve of war cadorna’s dis- positions were as follows: gen. roberto brusati’s i. army, with five divisions and 1o alpine battalions, was to conduct a limited offensive in the trentino, with the object of shortening the line and securing strong defensive positions. gen. nava’s iv. army, five divisions and seven alpine battalions, was to attack the cnemy communications in the pusterthal and co-operate in an advance towards tarvis; this advance was to be carried out by the “‘ carnia force,” consisting of one infantry division and 16 alpine battalions under gen. lequio. the ii. and iii. armies, under gen. frugoni and the duke of aosta respectively, were to attack with all speed on the isonzo front. frugoni had eight divisions and the duke of aosta six, with three cavalry divisions. but on may 24 only three corps (seven divisions) and three cavalry divisions were ready to attack. on may 20 the austrian armies, which had been placed under the command of the archduke eugene, with gen. krauss as his chief-of-staff, formed littke more than a screen. gen. dank], lately in command of the austrian i. army, had about two divi- sions in tirol, with some landstiirm battalions; gen. von rohr had a similar force to oppose an advance on tarvis; and the isonzo front, from monte nero to the sea, was lightly held by three divisions under gen. ludwig von goiginger. but rein- forcements were close at hand. within a few days rohr had two more divisions at his disposal, and by may 27 gen. boroevie von bojna, to whom the isonzo front had been entrusted, had go battalions between monte nero and the sea. ij. the first year’s campaign the opening moves of the italian offensive, all-important as they were in relation to the future operations, failed to obtain the results hoped for in cadorna’s design. the i. army per- formed its limited task quickly and satisfactorily, but the iv. army was very slow. here, almost certainly, a good chance was missed. lequio’s carnia force was quick t® move and found that the enemy was equally quick. this sector was all- important to the austrians, from the point of view both of offence and defence. it was essential to prevent a break-through to tarvis and villach, and if they could hold the frontier line it preserved for them the chance of the attack down the valleys leading to the tagliamento. here alone the way was blocked against any but an overwhelming force. italy takes the offensive—meanwhile, the ii. and iii. armies were on the move. frugoni, with the five divisions of his army that were ready, attacked along the line of the isonzo from the saga to opposite gorizia. the duke of aosta, with a single corps and two cavalry divisions, was to force the passage of the lower isonzo and push on towards the carso. the austrians had withdrawn beyond the river, except at the two strong bridge-heads opposite tolmino and gorizia, which were held in force, and south of gorizia the line of defence chosen was the carso plateau. speed and initiative were essential if the opening moves of the italian offensive were to be successful, and at vari- ous points speed and initiative were lacking. the bridge-heads were invested, as were several other strong positions held by the enemy. | on the other hand, the country is extraordinarily difficult, and roads were few and mostly bad; and those which were suitable for the movement of troops and guns led only to the points which the enemy was holding in some force. fortune, too, was against the attacking armies. appalling weather made movement in the mountains almost impossible during the critical period, and when the chance of a surprise had gone the great barrier of the julian alps was an insuperable obstacle to such forces as the italians could bring against it. guns, machine-guns and trans- port were lacking. farther south, in the plain west of the carso, the isonzo came down in sudden and violent flood and held up the duke of aosta’s advance. the fords were impassable and bridges had been blown up by the retreating enemy. by the end of the third week in june the italians had gained several important positions, some of them after very hard fight- ing, but they had been brought to a stop before the austrian italian campaigns . * mitte wy z meat? 7” q z= en? s) ep rg i area ¥ $3 f° hay eat ist 7 oe ; : ete dees sie? avp fi & 14 ing, oat “2, boy <= < ad ;pieve di al ays vi ly os = 784 wp tote 7 teeev- 4 4 me = va sy, ms aad es “nile b- ef 4, yn aee » = ‘ 6 ra ae 5 ; y : _ ny igrap nin ne %, « ¥ brecten | my ; mbassand’” ‘brescia hg ya as italian campaigns i9i5 — 18 ieegeer se bs mites 10 40 5 “5 10 20 30 ocs kilometres ' j & 3 main raileau r ea aa ata ry | a , ("oe © otero oe main lines, and it became clear that the hopes of a war of move- ment must be given up, that only the slow processes of trench warfare coultl lead to success. the italian mobilisation was now completed, and cadorna had ready some 35 divisions. against these the archduke eugene had some 20 divisions, including the bavarian alpenkorps, which had been sent to tirol, although germany and italy were not yet at war. boroevie and dankl were much weaker in infantry strength than the armics opposed to them. but they were strong in artillery, were very much better equipped with machine-guns, and held positions that were naturally ideal for defence and had been well prepared. first and second battles of the isonzo——~on june 23 began what is known as the first battle of the isonzo. the ii. army, with a strength of 10 divisions and two alpine groups, attacked the enemy lincs in the tolmino sector, and from plava to podgora, while the iii. army, with six divisions, moved to the assault of the western rim of the carso plateau. the italians had not suffi- cient heavy artillery; boroevi¢e had now nine divisions, with their march battalions, and further reserves were on the way. after very heavy fighting at plava, where the earlier italian ad- vance had established a small bridge-head, at the gorizia bridge-head (m. sabotino and m. podgora) and on the edge of the carso the action was broken off on july 7. during the fort- night’s fighting the italians had gained a little ground at the cost of about 19,000 esha one austrian casualties to date were 22,000. after 10 days’ breathing space casini attacked again in the same sectors. boroevi¢ had now 13 divisions under his command, but cadorna had concentrated all his slender supply of heavy guns on the isonzo front. no headway was made by the ii. army but a fierce struggle took place on the carso. monte san michele was taken on july 20 but lost again, and six days later a fresh attack had the same result. the second battle of the isonzo ended on aug. 3 with the italians close under the crest of monte san michele and the village san martino del carso, and in possession of most of monte dei sci bus, farther south. the italians lost some 34,000 men, but the austrians also suf- fered severely and lost over 10,000 prisoners. : the autumn offensive, oct. ~dec.—various isolated actions were carried out on the long front during the summer, but the bainsizza pill te ; eu fe “aly: £ ey “oper si j chloe nv sae mag tuanny ‘ “a s. martino del hes ey al faj “ ‘ay, \ mt ust:e etal si wn, brestoviz za wontraicone ne, ls wea. ney. cor \ mn, by, quarcito maal hermada) pale | m wenn > time was devoted chiefly to preparation for a big offensive in the autumn. by the middle of oct. cadorna could dispose of 312 battalions on the julian front. the ii. army now consisted of 12 divisions and the iii. of seven, while a reserve of five divi- sions lay ready in the friuli plain. when the attack began boro- evi¢ had about half this number of troops, but within a fortnight he had the equivalent of 15 divisions at his disposal. it was on the iit. army front that the italian numcrical superiority was great; on the carso the duke of aosta had no marked advantage in numbers over the archduke joseph, who had assumed the command in this sector in july. but the terrain on boroevit’s right was such that he could expect to hold with greatly inferior forces, especially in view of the italian weakness in artillery. the offensive went badly. by his main attack cadorna had hoped to turn the gorizia positions from both north and south, and as a secondary operation, after crossing the middle isonzo, to threaten tolmino from the south, as well as from west and north. after a first phase lasting a week, the attack was renewed four days later on a narrower front, but the third battle of the isonzo closed on nov. 4 with little tangible result. six days later the fourth battle of the isonzo was begun, and the struggle went on for more than three weeks, the offensive being finally broken off on dec. 5. as men and munitions be- came exhausted the scope of the action had been gradually re- duced to an attack on the gorizia bridge-head and still another struggle for monte san michele. once more the artillery proved insufficient, and the main feature of the fighting was the heroic efforts of the infantry on both sides. during the two battles, which were fought in persistent bad weather, the italians lost over 113,000 men and the austrians about 90,000. the result of the first seven months’ campaigning was disappointing to those who had hoped for far greater effects from italy’s intervention. it was not generally recognised how poorly the italian army was provided with the material necessary to modern war. in- evitably, moreover, the austrians had a great advantage in their nine months’ war experience. they were clearly superior in skill to their opponents, and their supcrior skill was backed by a spirit which the armies of the dual monarchy sometimes failed to show on the eastern front. | preparations for 1916.—the winter months were busily em- italian campaigns ployed, especially in the munition factories. it was beginning to be recognised in rome that estimates regarding numbers of men and shells had to be revised. during the winter the small total of heavy and medium guns was increased sevenfold, and. great if still inadequate efforts were made to increase the supply of shells. an important innovation of the winter was the bombarda, or big trench mortar, large numbers of which were constructed to make up for the deficiency in heavy artillery. the bombard was in fact much more than what is usually understood by the term trench mortar. its range was much longer, and the de- structive power of its big projectile was very great. the supply of men, no less than that of material, required to be replenished and augmented. in seven months the italian losses in the field were close upon 280,0c0c0—66,o90 killed, 190,400 wounded and 22,520 missing or prisoners. the austrian losses were 28,000 killed, 97,000 wounded and 31,000 missing or prison- ers, in all 156,000. the italian figure was in addition to casual- ties from sickness, which were heavy, including as they did the losses from an outbreak of cholera in the iit. army. men had to be found not only to fill up the gaps but to make new formations. during the winter the gaps were filled, and eight new divisions were ready in the spring, while others were in process of forma- tion, and cadorna had succeeded, after some difficulty, in having the classes required for drafts called up well ahead of his imme- diate needs. iii. battles of the isonzo and tiie carso during the early months of 1916 the only fighting of any im- portance took place in march, when cadorna opened a big dem- onstrative action on the julian front, with the object of preventing the dispatch of austrian forces to verdun. the action contin- ued from march 1i1 to march 29, and received the name of the fifth battle of the isonzo. although it was only a demonstration, some hard fighting took place, and both sides suffered consider- able loss. meanwhile, preparations for a real italian offensive on the julian front were well advanced, when news came that the austrians were preparing a big attack in the trentino. austrian failure in the trenttno—this attack, which is de- scribed elsewhere (see astago, battle of) had a big initial suc- cess, but it was already condemned to failure when brusilov, answering the appeal for co-operation made by cadorna on may 19, attacked the weakened austrian lines in front of him on june 4, and won the great victory that, if cramon may be believed, came within an ace of being decisive. the attack in the trentino, based on a miscalculation, nearly ended in the collapse of aus- tria’s eastern front, and brought only a slight territorial gain that was no compensation for the defeat elsewhere and for the losses suffered. nor did the penalty end with these. cadorna refrained from knocking his head against the lines upon which his retreat- ing enemy turned and stood. the positions he reached were ade- quate to his aims in the trentino, which were purely defensive, and instead of persisting in his counter-offensive he rapidly swung his reserves back to the julian front, smashed through the gorizia bridge-head and took gorizia, and drove the austrians from the western section of the carso plateau. the taking of gorisia.—during the winter of r915-6, in prep- aration for an attack upon gorizia and the carso, the right wing of the it. army had been transferred to the iii., so that the front from north of monte sabotino down to the sca, was under the duke of aosta. at the end of july, when the delayed attack was imminent, the duke had 16 divisions and a dis- mounted cavalry division. he had 1,250 guns, of which 520 were heavy or medium, and these were supplemented by nearly soo bontburde. on the carso and sen gorizia boroevie was badly daniel to meet the italian attack. he had only five divisions in line between monte sabotino and the sea, and one in immediate re- serve, when the duke launched his attack, and the austrians were taken by surprise. on aug. 4 the duke began with a feint against the low hills east of monfalcone, and two days later the real attack developed, when the vi. corps attacked the gorizia bridge-head and the xi. the summits of monte san michele. dod the vi. corps, commanded by gen. luigi capello, had outgrown the dimensions of an army corps, for it consisted of no fewer than six divisions. capello’s attack was brilliantly successful. monte sabotino, which had resisted so many attempts at cap- ture, was taken on the run in 4o min., while the greater part of the podgora ridge was torn from the austrians and some italian detachments reached the river at sunset. the austrians defended with the most obstinate valour, and gained precious time for their hard pressed commander. italian troops crossed the river on the night of aug. 8, and the town of gorizia was occupied next day without resistance, while a general attack on the carso was breaking down the stubborn defence which had survived the loss of monte san michele early in the first days’ fighting. on aug. 10 the austrians were driven back across the vallone, the deep cut that separates the san michele-doberdo section of the carso from the main plateau. both to the east of gorizia and on the far side of the vallone the advancing italians found themselves faced by new lines hidden among the woody slopes beyond the town and the stony undula- tions of the carso. attempts to continue the offensive were not successful. the sixth battle of the isonzo was an important italian success, but there was not sufficient weight of guns and ammunition to push the attack home. closing operations of 7916.—ia sept., oct. and again at the beginning of nov. the duke of aosta attacked on the main carso plateau, between the vippacco and the brestovizza valley, punching out a big salient on the northern half of the carso, driving the austrians back to their last jine of trenches and occupying the important position of dosso fajti. but the seventh, eighth and ninth battles of the isonzo were each broken off as soon as the attack slowed down. cadorna was attempting to gain position for a bigger attack later on, when men, guns and shells should be more plentiful. bad weather prevented another blow, and prevented also an attack in the asiago uplands, which had been planned for the middle of november. the year had seen much heavy fighting, and both sides had suffered severely. the italian casualties were nearly 120,000 dead, 285,000 wounded and 78,oc0 prisoners, the bulk of the latter taken in the first days of the austrian offensive in may. the austrian losses were also heavy, well over 200,000 killed and wounded and some 60,000 prisoners. if the territorial gains at the end of the year’s fighting were not great, cadorna’s continued attacks, following upon the costly failure of the austrian offen- sive in the trentino, had done their work in occupying an in- creasing number of theenemy’s troops and wearing down his resist- ance. cadorna’s rele was clearly marked out; so long as the plans of the allies were based upon the policy of attrition, he had to hammer when he could, with what means he could collect from month to month as the output of guns and munitions increased and fresh troops were trained, kecping always in view as an essen- tial aim that of attracting to his front, and wearing out, the maxi- mum number of enemy forces. juaged from this standpoint, the italian effort of 1916 was of the greatest value to the allied cause. although cadorna was strongly opposed to the dispersal of his forces in pelits paquets and had resisted the suggestion of an expedition to libya to quell the rising which had reduced the italian occupation to a few points on the coast, the importance of the balkan front had not been lost sight of by the italian government. fresh troops were sent to strengthen the italian position in albania, and in aug. a strong force arrived in salonika under the command of gen. petitti di roreto to take part in the allied advance on. monastir. early in oct. an italian column occupied argyrokastron, and before nov. the italians were in touch with the left mine of the allied forces based upon salonika. iv. the austro- german ef fort at the allied conference held in rome in jan. ror7 the formal proposal was made that an allied force should join the armies of italy in an attempt to smash the weaker of the two central powers.- in spite of mr. lloyd george’s advocacy the plan did not commend itself to the french and british military authori- 554 ties, who offered, however, to send 300 heavy guns to italy on condition that they were returned to the french front by the month of april. cadorna declined the offer, on the ground that the season was unsuitable for an offensive on his front, and that the guns would have to be returned at the moment when they would be most useful. the question of closer co-operation was dropped for the time being, but the conference organised a line of communications through italy to salonika, via the southern italian ports, a route which greatly lessened the dangers from submarine attack, and at the same time made a much smaller demand on the diminishing tonnage of the allies. cadorna’s plan.—cadorna embodied his proposals in a mem- orandum written after the rome conference; they were as fol- lows: if the allies would give him at least 300 heavy guns he would make two attacks on the trentino and julian fronts—his own artillery was insufficient for this double offensive—and so find the enemy’s weak point. he had the advantage of interior lines, and would move his reserves of guns and men from the venetian plain according to the devclopment of the two actions. if, on the other hand, the allies would send a minimum of eight divisions in addition to the heavy guns, he would concentrate upon the julian front and attack from tolmino to the sea, with the object of breaking through towards ljubljana (laibach). such an attack, in cadorna’s view, would have had decisive re- sults. he believed that austria could not recover from so severe a blow. the plan was tempting, but it did not commend itself to the allied commands. french and british opinion was against any further diversion of effort from the western front, for there was the chief enemy, upon whose defeat the result of the war de- pended. and there were obvious technical difficulties in the way of supplying large french and british forces on the julian front. nivelle and robertson, who visited the italian front in the spring, agreed to the principle of direct co-operation by the dis- patch of troops and guns in the event of necessity, but both were inclined to prefer co-operation by simultaneous attack, and while a scheme for the quick transport of troops from france to italy was prepared, no definite engagements were taken. it was agreed, on the other hand, between cadorna and nivelle, that the italian and french spring offensives should be timed to coin- cide as nearly as possible. the iialian spring offensive—cadorna’s attack was slightly delayed owing to a threat of an austrian offensive in the tren- tino, and later, by bad weather. but the italian guns, which had been reinforced by 11 batteries of british 6-in. howitzers and 35 french heavy guns, opened fire on may 12. cadorna feinted with the iii. army on the carso, making his real attack with the ii. army, now under gen. luigi capello, against the hills north and east of gorizia. north of the town the greater part of the long ridge (kuk-vodice) running southward from above plava was gallantly stormed and held against the most determined counter-attacks, but little progress was made east of the town. as soon as the occupation of the kuk-vodice ridge seemed as- sured cadorna moved the bulk of his heavy guns southward and attacked with the iii. army on the carso. useful progress was made here also, a number of positions being captured, and the vii. corps on the right carrying one line after another till they were half-way up monte hermada (querceto). but ammunition was running very low; the offensive was broken off at a moment when it seemed as though further success lay very near. on the evening of may 26, when the attack on the hermada was stopped, the defenders of the battered hill were reduced to under roo men, each of whom received the maria theresa medal. there was only a short breathing space. on june 4 the aus- trians on the carso counter-attacked in the most determined manner. finding a weak resistance on the part of the troops who has come into line as reliefs, they freed the lower slopes of monte hermada, and took alarge number of prisoners. during the four weeks’ fighting cadorna used 31 divisions and lost 132,000 killed and wounded and 25,000 prisoners. boroevie had held his ground, or nearly, with 17 divisions, and his losses, including 25,000 prisoners, were close upon 120,000. italian campaigns the rumour of battle had scarcely ceased on the julian front when the italians attacked in force north of asiago, on a front of nine miles. the attack failed. progress was made at one point only, on monte ortigara, and here too, after a long and bloody struggle, the attacking troops were thrown back. between june 10 and june 29 the italians lost 24,000 killed and wounded and 2,000 prisoners. the austrian casualties were over 0,000. cadorna’s second offensive, auyg.—sepi.—the general situa- tion at the end of june gave cause for disappointment and some anxiety. russia was going out of action. the prospect of an entente victory with which the year had opened was clearly removed to a distance, and war weariness was making itself in- creasingly felt in italy. the question of allied co-operation on the italian front was once more discussed but without result. it was decided that cadorna should attack alone in august. the army was strung to the highest point of tension, awaiting the order to attack, when pope benedict xv. launched his appeal for peace. parts of the army were shaken, for the pope, in his impartiality, placed the two contending groups of powers on the same level; he held out the hope that germany and austria were ready to consider certain territorial questions “in a conciliatory spirit,’ taking into account “ the aspirations of the peoples”’; and to the long and weary struggle he attached the label “ useless slaughter.”’? some of the commands were anxious about their men when the attack began, on the night of aug. 18. as a matter of fact, the troops put aside their questionings, and the blow dealt to the austrians was a very heavy one. the isonzo was crossed in many places between tolmino and plava and the greater part of the bainsizza plateau was occupied by troops of the ii. army, while the southern end of the chiapo- vano valley was passed and a footing obtained on the western corner of the ternova plateau. but a long sustained effort brought no further success. cadorna intended to renew his offen- sive at the end of sept. by an attack against the ternova plateau, in the hope of definitely turning the gorizia positions from the north and cutting the main line of communications be- tween the austrian right and left. but towards the middle of the month news came of increased enemy forces and a probable counter-offensive at an early date, and cadorna, after taking stock, decided he must stand on the defensive. the four weeks’ fighting in aug. and sept. had cost him over 166,000 men—4o0,000 killed, 108,000 wounded and over 18,000 prisoners. the toll taken by sickness had also been very heavy. the units were at low strength and the new drafts had not been satisfac- torily absorbed. a breathing space was urgently needed. british and french reinforcements -the austro-german suc- cess against the ii. army, the retreat of the italian forces to the piave, and the resistance in the new positions are described in a separate article (see caporetto, battle of). when the gravity of the situation became clear, england and france acted with all possible speed. the order was given for six french and five british divisions to entrain for italy, and foch and robert- son hastened to the spot. an allied conference at rapallo began on nov. 4, and from its discussions were born the supreme allied council, which was to meet, once a month if possible, at ver- sailles, and the versailles military council, which was to sit per- manently. it was agreed that the failure of the italian armies to resist the enemy attack called for a change in the italian com- mand, and cadorna was appointed italian military representa- tive at versailles. he was succeeded by gen. armando diaz, commander of the xxiii. army corps, and the functions of gen. porro, who was also relieved of his post, were divided be- tween gen. giardino, who had been minister of war during the summer, and gen. badoglio, commander of the xx vii. corps. v. the collapse of austria after the failure of krauss and conrad to break through to the venetian plain, the italian front saw no action of first class importance for nearly six months. the time was well occupied in reorganisation, but there were several minor combats. the german divisions left italy at the beginning of 1918, in anticipa- tion of the great offensive which was being prepared on the italian literature western front. gen. plumer, who commanded the british forces, also left italy to take up his old command. he had acquired a great popularity, and his departure was much regretted. fortu- nately he had a worthy successor in the earl of cavan. when the german offensive in march 1918 pierced the line of the brit- ish v. army four french and two british divisions were immedi- ately withdrawn from italy. they were followed by the italian ii. corps under gen. albricci. this left diaz with 55 divisions (so italian and five allied) as against 60 freshly organised aus- trian divisions, who were preparing for an early offensive. austrian altacks fail_—the original proposal of the austrian general staff, now under gen. arz von straussenburg, was to make a drive on both sides of the brenta. conrad, now in com- mand in the trentino, pressed for an attack in the asiago up- lands. krauss disapproved of both plans and urged an offensive on both sides of the lake of garda. conrad’s plan was chosen, but boroevie urged that this offensive should be accompanied by a straight drive by his armies across the piave. this was agreed to, and both army groups attacked on june 15. conrad attacked with scheuchenstuel’s xi. army from south of asiago to monte grappa, while the archduke joseph attacked the montello and werzel von wurm crossed the lower piave. conrad had 27 divisions at his disposal, and boroevie 23. conrad’s attack was a complete failure. it went well to begin with, but at the end of the day all hope of success had gone, and by the evening of june 16 he was finally beaten. boroevi¢, on the other hand, made good headway on the first two days, for the archduke joseph took half the montello, and werzel von wurm established an extensive bridge-head opposite san dona di piave. but at the end of a week’s fighting the austrians were closely held, and the order was given to retire across the piave. the failure was complete, and very costly. conrad lost 36,000 men and boroevie over 60,000. the defeat broke forever the offensive power of the austro-hungarian empire. the conscious- ness of impending disaster grew and spread through the mon- archy, and the troops were greatly disheartened by failure. final italian offensive —it was felt in many quarters that diaz should have followed up the victory by a strong counter- attack, but he was unwilling to attack in force without careful preparation. he had had heavy casualties, over 40,000 killed and wounded and a very large number of prisoners, and he pre- ferred caution. plans were drawn up and preparations made for an offensive between the vallarsa and the brenta in september. early in sept. diaz went to paris to discuss the situation. he was still pre-occupied in regard to his reserves, and asked that a strong american force should be sent to italy. this was refused, and in spite of criticism he delayed his offensive still further, while working out secretly a more ambitious scheme which was to be adopted if the chance should offer. in the middle of sept. the victorious advance from salo- nika began, and the chance seemed to have come.’ ‘on sept. 25 or- ders were issued for a rapid concentration of troops, artillery and technical services in the sector chosen for the attack, which was no longer the plateau, but the middle piave.” (gen. diaz’s report). the attack (see vittorio veneto, battle of) was launched exactly a year after the disaster of caporetto, and it shattered the armies of austria-hungary. seldom in history has so great a disaster been followed by so complete a triumph. yet the final overwhelming success of vittorio veneto was not italy’s greatest victory. the way to it was paved by greater deeds, the wonderful recovery on the new line after the great re- treat, and the successful resistance against the last austrian of- fensive that was the first ray of light to break upon those gloomy months when the fortune of the allies seemed at their lowest. nor can the sum of italian achievements be judged by the issue of those battles which were crowned with victory. conclusion.—italy’s contribution to the long effort that led to the triumph of the allies can only be gauged by a review of the campaign as a whole, by a realisation of the extent to which she drained the resources of austria-hungary, and of the price which she paid. her dead totalled 600,000; and 570,000 men were per- manently disabled for military service by wounds or disease. for doo two and a half years the italian armies were a constant threat to the habsburg empire and kept employed a number of divisions that increased from 20 to 40 (in the summer of 1917). during the last three months of 1917 the number of enemy divisions rose to 55, and in 1918 practically the whole effective strength of austria-hungary was arrayed against italy, the number of divi- sions at one time approaching 70. the figures speak plainly, and ludendorff, in an interview published in the spring of 1919, placed among the chief causes of the german defeat “the lack of support from austria, gripped ever more tightly at the throat by italy.” cadorna rightly claimed that italy’s ‘‘ grip on austria’s throat from 1915, com- pelling her to immobilise against us ever increasing forces, con- stituted the most notable result of our war, though it was little apparent to the eyes of civilians. it contributed largely to the victory of the allied arms and to our final triumph.” bibliography.—italian official papers, diario della guerra d'italia (1915-7); w. k. mcclure, jtaly’s part (1918); l. capello, note di guerra, 1915-20 (1920); l. cadorna, la guerra alla fronte italiana, 1915-7 (1921); t. n. page, italy and the world war (1921). (see also world war: bibliography.) (w. k. mcc.)