fective mill practice in failing to crop the blooms to sound steel. A rail of this kind does not generally fracture in the track, but mashes down gradually and is detected by the track men and removed before it becomes unsafe. Another type shown in Fig. 2 is the split head. In the illustration the flowing over upon both sides of the head is shown. This section has been polished and lightly etched with iodin, and it will be seen that the metal is thoroughly unsound, again due to failure at the mill to crop the blooms to sound metal. As a result, the steel has too little tenacity to support an ordinary load, and crushes. In Fig. 3 we look down upon the top of a 100 lb. rail from which a layer about three-eighths inch thick has broken out in service. A seam runs lengthwise through the rail, and by carefully observing the fracture it will be noted that the failure began at the seam line and extended gradually in service until the crack reached one side of the head and nearly extended to the other. A failure of this kind is particularly dangerous since it is gradually developing while the rail is in service and no outward indication may be given of the presence of the defect. The rail in question was made by the open-hearth process and had the following composition: This composition was normal and gave no indication as to the cause of failure. Fig. 4 shows an etched section from the same rail taken at the same location. It is seen that the steel is thoroughly unsound and porous and shows clearly the spot at the seam line. The exact method of formation of this type is beyond the scope of our present subject, but it is an evident defect in mill practice. Fig. 5 represents another longitudinal crack of the same type. In each of these cases the composition was normal, but the same condition of extreme unsoundness, resulting from defective mill practice, was present. Fig. 7 gives another example of a radically unsound rail with a split lengthwise in two planes, and Fig. 6 shows the upper portion of the head with the slag-line at which the fracture began. This condition obviously would have been avoided by proper mill practice. Fig. 8 shows the type of failure known as an "internal transverse crack." This is also particularly dangerous since it is not present in the rail when rolled, but begins to develop in service at some point of weakness within the steel, and gradually extends across the head until fracture of the rail occurs. The rail in this case was made by the Bessemer process and had the following composition: This rail was removed from track and broken under the drop, and a similar crack, though smaller, was found in the lower part of the head near the web, about three feet from the first, while about a foot further along the rail a small crack was found in the process of development as shown in Fig. 9. A characteristic form is shown in Fig. 10 and the nucleus of weakness at which the fracture began is clearly seen near the center of the crack. Fig. 11 illustrates again the gradual growth of what later on would have been a failure of this nature and the spot of weakness at the center is evident. Owing to the dangerous character of this type of failure and the need of definite knowledge as to the prime cause of the growth, we made a critical study of the subject, gathering data throughout the country. This defect has been found in low carbon steel and in high carbon steel-in rails as heavy as 100 lbs. per yard, and as light as 80 lbs., while in some cases, where rails as light as 67 lbs. were used and subjected to the same wheel pressures, none of the defects had developed. This defect has been found in relatively small numbers in both Bessemer and in open-hearth steel, made at different mills and in service in various parts of this continent. Thorough study proved beyond question that the composition of the steel in its ordinary elements was not the cause of the failure, and also that the conditions of service were not the prime cause. Analysis of the small central core in these spots was not conclusive, and microscopic investigation proved that the spots were mere cracks and did not extend longitudinally along the rail. In other words, they had no depth, but on examination of the steel in the immediate vicinity of the central core we always found decided porosity or other source of weakness, this condition being caused by various defects in the mill practice. In one of these unsound rails shown in Fig. 12 two internal transverse cracks had developed within about one inch of each other. Another characteristic form of internal transverse crack is shown in Fig. 13, that of a badly segregated open hearth rail in which the manganese content was 1.62 per cent., although the heat average approximated .80 per cent. Fig. 14 shows a threefold enlargement of this spot. By noting the formation of the ridges it is apparent that the fracture began inside the head and gradually extended until it reached the surface. The spot was about three-quarters of an inch long. Fig. 15 represents a similar crack in a 110 lb. open-hearth rail in which the manganese was also excessive, 2.65 per cent. being present, whereas the heat average was approximately 90 per cent. In both of the latter cases the abnormal composition was caused by failure in manufacture properly to mix the mangagese added in the ladle. Our experience has been that a defective condition is very seldom characteristic of an entire rolling, but that it is generally present only in individual ingots. Consequently it is clear that in order to guard most effectively against the various defects and at the same time utilize the largest possible proportion of the rolling, it is desirable to test each ingot in the rolling, stamping each rail with its ingot number, in addition to the heat number and the letter indicating its position in the ingot. Also, in order to forestall as far as possible the various defects in the rails, we found it advantageous to place a considerable number of inspectors throughout the mills to watch the progress of the manufacture in its various details and to report all variations which would tend to lessen the value of the product. The plan has now been in operation for about one year with very satisfactory results. These extraordinary, though now needful, precautions on the part of a railway company should be unnecessary since they are intimately connected with the manufacture of the steel and should properly be observed by the mill itself to safeguard the quality of its own product. Of the routine tests we have found it advisable to make a drop test upon a butt from each ingot in the heat in order to avoid brittleness, rejecting all rails from each ingot which fail under the test. The ductility test of Dr. P. H. Dudley has also been found extremely useful in helping to avoid qualities which result in some of the types of failures shown above. The test to destruction under the drop to detect piping is also of great value, as it serves as a check upon the work of the men at the shears and tends to reduce to a minimum the number of piped rails in any given shipment, particularly when the tests are made upon the ingot plan mentioned above. The deflection measurement under the drop test enables one to avoid undue softness, while base seams may be detected if present by making the drop test with the head of the rail downward, as is our regular practice. but with the base slightly inclined so that the drop will strike the extreme side of the base and thus open up as a result of the blow any unwelded seams which may be present. If seams are found the top rail of the ingot might properly be rejected. This modification does not interfere with the effectiveness of the drop test in other respects. I wish to emphasize with all the force of which I am capable, the absolute unfairness of forcing upon the railroads the detecting of defects which if the rails were properly made would seldom be present in the steel. TRACK SCALES.* By H. T. PORTER, Chief Engineer, Bessemer & Lake Erie. In looking over competitive bids for track scales from time to time, the writer never was able to find that there was sufficient information in the plans and proposals of the scale companies to enable one to make a comparison of the relative merits and capacities of the scales. The writer submits herewith a proposed set of specifications for a 46-ft., 150-ton, four-section track scale. These specifications are not submitted as a finished product, but as a first effort. This is a subject on which there is very little information readily available, and the writer thinks that a move should be made to standardize track scales. We once asked for bids on track scales without any attempt to give particular specifications, and were advised by one of the scale companies that by reason of the variations in details and methods of figuring, they could furnish 49 different kinds of scales under those requirements. This is merely mentioned to show the necessity of standardizing, as a great many of the variations in the construction have no special merit, but became necessary to satisfy the whims of the draftsman in the office of the purchaser. The following are the general specifications for 150-ton 46-ft. track scales on the Bessemer & Lake Erie : Plans. Each bidder shall submit a strain and stress diagram of the scale he intends to furnish, showing thereon the proposed sections of the various parts under strain, and all plans showing details similar to those intended to be used in the construction of this special scale; these plans to be submitted at the time of making his bid. Length. The length of this scale shall be known as 46 ft.; this length to be the distance between the ends of the live rail. The ends of the live rail should be vertically above points located not less than one-fifth of the distance between the first and second sections distant from the center of the end section of the scale. Number of Sections.-There shall be four sections. Rating. The rating of this scale shall be known as 150-ton 46-ft. track scale, and this load shall be used in the design, each section to be considered as taking one-fourth of the rated tonnage, or 37.5 tons live load. Dead Load. The dead load of any part may be neglected when the strain produced by it is less than 21⁄2 per cent. of the strain produced by the rated load. Impact.-An impact equal to 100 per cent. of the rated load shall be everywhere added, in designing. Grade. The maximum grade to which the scale rails will ever be laid is 0.3 per cent., and such grade shall be obtainable by tapered shims to be placed between the bottom of platform stringers and supports, these shims to be omitted should the purchaser desire to have the scale rails level. Load Adjustments-Means should be provided by wedge blocks or other device for taking up any minor irregularities of the platform supports, on account of variations in manufacture or setting. Deck. The scale deck shall be of the rigid type, supported by the pit walls, and shall overhang the ends of the platform stringers to correspond with the design of the platform length, relatively to the lengths between sections. Stringers.-The stringers shall be I-beams of Carnegie structural steel, Manufacturer's Specifications for Steel for Railway Bridges. The span shall be considered as the c. to c. distance between any two sections, and the maximum bending stress calculated shall not exceed 16,000 lbs. per sq. in., assuming no continuous action over supports. The minimum depth of I-beams shall be 24 in. The stringers shall be thoroughly braced laterally by angles not less than 3 in. in width, and by transverse *From bulletin No. 148 of the American Railway Engineering Association. diaphragms. Connections to have 3/4-in. rivets or larger. No structural parts shall be less than 3g-in. thick. Special Clause, Structural Steel.-The manufacturer will submit a general plan of structural steel showing sizes and total estimated weight, at the time of making his bid, and all necessary scale dimensions in the final plan, after contract is signed, which will enable a structural manufacturer to fabricate the steel to fit the scale details. Suspension Bearings.-The load of the platform shall be transmitted to the main levers by means of suspension bearings. No forged links shall vary more than 1/16-in. from the dimensions shown on the plans. Levers-The main levers shall be of cast steel. Other levers may be of cast iron or of cast steel. Level tabs shall be cast on the tops of levers about 12 in. c. to c. and faced accurately. The unit stresses in levers shall not exceed the following values: Cast-iron, 3,000 lbs. maximum tensile stress; cast-iron, 18,000 lbs. maximum compressive stress; cast-steel, 8,000 lbs. maximum tensile stress. Knife Edges.-The knife edges shall be separate from the bearing in main and extension levers and of silico-manganese steel; the load per linear inch of the knife edge shall not exceed 5,000 lbs. The line of knife edges in any one lever should coincide as nearly as practicable with the neutral axis of the lever. Bearing Blocks.-Cast-steel bearing blocks, self-centering the bearing on supports, shall support the knife edges. Supports.-The scale levers shall be supported by cast-steel standards resting on concrete pedestals, and no parts subject to tensile strain shall have a unit stress exceeding 8,000 lbs. The bearing on the concrete shall not exceed 300 lbs. per sq. in. Check Rods-Check rods shall be provided transversely and longitudinally. Multiple. The multiple of the beam pull shall be 800, but an additional rod whose pull is 400 shall be provided for attaching to the 2 to 1 lever of a Streeter-Amet recorder. Beam. The main beam shall be graduated in thousands of pounds, up to 200,000 lbs. An auxiliary sliding poise attached to the main latched roller poise shall move over graduations of 50 lbs. for 1,000 lbs. All weights in excess of 200,000 lbs. to be hung from the end of the beam. The graduated scales shall be of brass. Painting. All parts, except wearing points and brass, shall be painted before shipment. Foundation Plan.-The scale manufacturer shall furnish a foundation plan showing all necessary dimensions to install the scale. The pit shall be amply wide to permit thorough inspection and make all parts accessible. Purchaser Furnishes.-The purchaser will construct the foundation to the dimensions shown by the manufacturer's plan and will place all bolts and structural steel that is to be built into the masonry. The purchaser will also furnish and construct the deck, weather guards around rail chairs, scale house, and provide foundation bolts, 100 lbs. scale rails and fastenings. Purchaser will furnish structural steel to fit dimensions shown on scale plans. Easer Rails.-Easer rails will be provided for both ends of the scale by the purchaser, but the scale manufacturer shall not consider the length of the platform as affected thereby. Superintendence.-The manufacturer will be notified when the foundation is completed, and he shall furnish a competent erecting foreman to superintend the installation of the scales; any assistance this foreman may need will be supplied by the purchaser. Erecting foreman will be paid for at a rate per day as specified in the bid. Inspection of Material, etc.-All material or workmanship shall be subject to the inspection of the purchaser's representatives. Guarantee. All parts failing as a result of defective material or workmanship of any parts furnished by the manufacturer shall be replaced free of extra cost to the purchaser. Assistant Engineer, Chicago, Burlington & Quincy, Chicago. In many parts of the Mississippi valley embankments are often subject to damage from waves during high water, and this action is particularly severe when the water is about at the level of the shoulder of the bank. It is usually not economical to protect banks with riprap where they are not exposed to the current or liable to be overtopped by high water. For such places a cheap and efficient protection is afforded by the use of brush or bushy saplings, willows being usually the most available and being peculiarly suitable for the purpose. The larger end is secured to the embankment either by a stake, or better by a rope, at the water's edge, the brushy part floating on the water at the side of the embankment. A fringe of this brush 2 or 3 ft. wide will almost completely prevent the cutting action of the St. Louis & San Francisco, Pittsburg, Kan. It is the practice in most places to use a roller bender for curving rails when relaying curves. While a gang was curving rails for a 15-deg. curve recently the roller bender was broken and the foreman had to resort to the Jim Crow bender to finish the job. He found that by setting the plunger so that it barely touched the web of the rail he could pull the lever down and put a slight bend in the rail. Then, lifting the lever he would slide the bender forward about 14 in. and push the lever down, repeating the performance, from one end of the rail to the other. One adjustment of the plunger was all that was necessary. To the foreman's surprise he found that he could curve a rail with the Jim Crow bender in one-third the time that it took with a roller ender, and it did not take as many men to do the work. *Entered in the Track Kink contest, which closed December 25, 1911. General News. The United States Civil Service Commission announces examinations November 25 for mechanical draftsmen. Candidates are wanted for temporary service at $5 to $6 a day at Ellis Island, New York harbor. The Vandalia Railroad has established a department of safety, with general and division committees like those on many other prominent roads. J. W. Coneys, superintendent of the Pennsylvania, at Indianapolis, is chairman. Professor Edward C. Schmidt announces a gift to the University of Illinois from the Midvale Steel Company of Philadelphia, of four axles and four pairs of supporting wheels for the new locomotive laboratory which is now under construction. The northbound Shasta Limited of the Southern Pacific was stopped by robbers at Delta, Cal., on the night of November 8, and the registered mail was rifled. The trainmen fought the robbers, and a brakeman killed one of them. The Glenfield & Western Railroad has been authorized by the New York State Public Service Commission to suspend operations from December 15 to May 1. This road, about 20 miles long does little business except in connection with lumber operations. It connects with the Ontario & St. Lawrence division of the New York Central at Glenfield, N. Y. The mail car of southbound passenger train No. 7 on the Louisville & Nashville was robbed on the morning of November 8, a few miles north of Birmingham, Ala. The robbers rode in the mail car 26 miles, and then succeeded in binding the mail clerks and escaping before their presence on the train was known by the trainmen. Captain A. C. Baker, U. S. N., who was connected with the Chicago Exposition in 1893, the Paris Exposition in 1900 and the St. Louis Exposition in 1904, has been appointed director of exhibits of the Panama Pacific International Exposition, to be held at San Francisco in 1915. Captain Baker is especially well-informed in the transportation field. Miss Annie Criley has retired from the position of station agent of the Pennsylvania, at Glen Loch, Pa., after 45 years' continuous service for that company. She began as a telegraph operator in 1868. Miss Criley will receive a pension and will be allowed to continue to make her home in the company's house which she has occupied for many years. An informal meeting of the Iron and Steel division of the American Institute of Mining Engineers was held in New York, on November 7, at which Benjamin Talbot of The Cargo Fleet Iron Works, Middleborough, Eng., described a new method of getting rid of piping and segregation in steel ingots by compressing the ingots after the outer portion has cooled and while the interior is still liquid. The Licking River Railroad in eastern Kentucky is to be torn up because it does not pay to operate it. This information is taken from Kentucky newspapers, which say that the people of the Licking river valley need the road and will miss it. This road is 32 miles long, 3 ft. gage, and connects the Chesapeake & Ohio at Salt Lick with the lumber works of the Yale Lumber Company. It was built as a lumber road and has had little traffic besides lumber. According to its last report it had four locomotives, three passenger cars and 71 freight cars. Disastrous Collisions in Louisiana and Georgia. In a rear collision on the Yazoo & Mississippi Valley, at Montz, La., on the morning of November 11, about 12:30 o'clock, 13 passengers were killed and 30 or more were injured. Nine of the killed and about half of the injured were negroes. Of the injured some were said to have been subsequently burned to death, the cars having taken fire immediately after the crash. A north-bound passenger train carrying a large number of excursionists, while stopped on a curve, was run into at the rear by a following fast freight train. It is said that the collision was due to the neglect of the rear trainman of the passenger, who went back only about 1,100 ft. The freight engine ran through the whole length of the rear passenger car, and this car and the two next to it took fire and were burnt up. The fourth and fifth cars from the rear also were badly damaged. There was a dense fog. The passenger train had been standing 35 minutes. In a butting collision between a freight train and a work train on the Louisville & Nashville at Emerson, Ga., on the 8th of November, seven employees were killed and 15 were injured, most of the victims being laborers on the work train. The foreman and the assistant foreman on the work train were killed, and two enginemen and two brakemen were among the injured. Disastrous Collision in Indiana. In a butting collision between a westbound passenger train and an eastbound freight on the Cincinnati, Hamilton & Dayton at Irvington, Ind., five miles east of Indianapolis, on the morning of the 13th at 3 o'clock, 15 persons were killed and 17 injured. The wrecked passenger cars took fire, but, according to the reports, the flames were extinguished before any persons were burned. The accounts indicate that the collision was occasioned by a misplaced switch, the responsibility for which is placed on the men in charge of the freight train. Telegraph and Telephone Pensions. Theodore N. Vail, president of the American Telephone & Telegraph Company, and of numerous subsidiary corporations, and also of the Western Union Telegraph Company, announces that on January 1 these allied companies will put in operation an elaborate pension and benefit scheme for their employees; and it is said that it is the most liberal and comprehensive scheme of the kind ever adopted. The total number of employees in all the companies is about 175,000. The plan contemplates the accumulation of a fund of $10,000,000 which is to provide for the employees of the companies named and also for those of the Western Electric Company. The total yearly pay rolls of these companies aggregate $115,000,000. The compulsory retirement age for male employees is 60 years (ten years younger than on most railroads); optional age 55 years; the pension age of females is five years younger in each case. As on railroads, the pension rates are to be based on the years of service and the amount of pay, and will be 1 per cent. of the average annual pay for ten years, multiplied by the years of service. No pension will be less than $20 a month. For accidents while at work the employee receives for total disability full pay for thirteen weeks and half pay until he returns to work, up to six years. Partial disability is compensated on the same terms, except that half pay is given until the employee can earn a livelihood, not exceeding six years. Full pay for thirteen weeks is provided for sick employees who have been ten years in the service, together with half pay for 39 weeks. Five to ten years' service brings the same full pay time, with thirteen weeks of half pay. Two to five years' service brings four weeks' full pay and nine weeks' half pay. Heads of departments may pay benefits to employees under two years' service. In case of death caused by accident in performance of work the employee gets insurance equal to three years' pay, the maximum payment being $5,000. Death from other causes will be met with one year's pay for employees ten years in service and six months' pay for five to ten years' service, the maximum being $2,000. State laws providing for more liberal compensation will prevail. Employees may have option as to legal rights in accident cases, as against the benefit plan. Trainmen Want More Pay. A recent convention of general committees of conductors and brakemen held at Rochester, N. Y., and representing the brotherhoods of these two classes in the eastern states, drew up a schedule of rates of wages on which, it is said, will be based a communication to the fifty-two eastern roads asking for an increase of pay. The conductors and brakemen had their pay advanced about 10 per cent. in 1910. In the scheme which has now been drafted they propose to ask for rates of pay graded according to the length of trains; also according to the grades of the road, where these are very steep (over two per cent.). It is also proposed to ask for overtime at a higher rate after the first hour with another increase after the second hour. On passenger runs monthly guarantees are to be demanded, as follows: Conductors, $135 a month; baggage masters, $87; rear brakeman, $84, and other brakemen, $81. Some of the demands for through or irregular freight trains are as follows: For trains of 60 cars or less, conductors, $4.20 a day; flagmen, $2.90, and brakemen, $2.80. For each additional 10 cars in the train, 3 per cent. of the foregoing rates are added until the maximum trainload of from 150 to 160 cars calls for a rate for conductors of $6.82 a day; flagmen, $4.63, and brakemen, $4.53. These rates are to be increased 12 per cent. upon all runs upon which the controlling grade exceeds 2 per cent. The demands will also stipulate pay at the rate of time and onehalf for crews handling double-head trains. If more than two engines are used, or if either one or both of the engines used in double-head trains are of the Mallet type, double time. Why Cars Are Scarce. J. F. Holden, vice-president of the Kansas City Southern, has issued a bulletin to the employees regarding the car shortage situation. He ascribes the situation to the fact that the railways have been so restricted as to be unable to finance the purchase of the needed amount of new equipment during the past three years. "While every other business, including agriculture, has been prospering and expanding," he says, "the railways have been prevented from doing their part in keeping up with the growth and expansion of the country, for the reason that the investing public has not been willing to put its money into an industry which has been hampered in so many ways and to such an extent as the railroads have been." Stoppage of Work on the Southern New England. On Saturday last the contractors at work on the construction of the Southern New England Railway from Palmer, Mass., southeastward to Providence, R. I., were notified to discontinue their work, and about 2,300 men were thrown out of employment. According to Providence newspapers about $1,500,000 has been spent on this road already and large sums are yet to be paid. A good deal of work has been done on cuts, embankments and bridges, and it was given out a few weeks ago that the officers of the road expected to have it finished so as to go into operation next summer. E. J. Chamberlain, president of the Grand Trunk, which controls the Southern New England, and which has been promising the people of Providence that they should have a competing railway to the west, when asked the reason for the suspension of operations on the Southern New England replied that it was because of the stringency of the European money market, following the disturbances caused by the war in Turkey. It is reported, however, in Boston and Providence that the reason back of this is that some agreement has been reached with the New York, New Haven & Hartford, which opposed the building of the Southern New England, and which would probably suffer by its completion. Officers of the New Haven road say that negotiations have been going on with the Grand Trunk looking to a traffic agreement under which the G. T. would be able to solicit freight at all stations on the New Haven and the Boston & Maine lines, on terms satisfactory to itself; but they declare that these negotiations have not yet been brought to a satisfactory termination. The Boston & Maine has withdrawn an application, which was pending before the authorities of the state of New Hampshire, for permission to build a road between Windsor, Vt., and White River Junction, on the east side of the Connecticut river, an agreement having been reached with the Central Vermont (Grand Trunk) under which the Boston & Maine will be satisfied to continue the present arrangement by which trains are run over the C. V. between those points. This agreement between the New Haven and the Grand Trunk interests is taken as a confirmation of the belief that what has been done in Southern New England is the result of an understanding between the two companies. On Tuesday of this week representatives of the Grand Trunk repeated their statement, made in the beginning, that the stoppage of work was temporary. President Chamberlain was quoted as denying that negotiations between the New Haven and the Grand Trunk promised a settlement of the controversy upon the basis of the admission of the Grand Trunk into Boston & Maine and New Haven territory on equal terms with other roads in competition for business. . . . So far as he knew, there was no prospect of a compromise with the New Haven, Con |