of gauge, level, or joints; showing faults and excess power in consequence of faults, side by side, thus placing a value on faults, and then instead of representing faults and excess power in inches or foot pounds, make the instrument show them in dollars and cents per ton of load when capitalized, which would show, multiplied by the ton-miles on any road tested, the amount that could, with good management, be expended on track reconstruction or rebuilding. The apparatus devised for this purpose is what we call an indicator car. A description of the construction and method of working might be of interest to not a few, for I do not know of another in use by a street railway company, and found only one other, although very dissimilar, in existence on a steam road, after I had the plans of mine finished.

The results shown by it are high and low rails, low joints, gauge, drawbar pull and the variation of the track level. Each one of these results is automatically platted on paper eighteen inches wide. The car consists of a platform eight feet by ten feet, mounted on a single truck, no springs being used. Midway between the two end axles is one which is fitted with wheels which record defects of joints or gauge. After use and calibration of instrument it will be more valuable and the dynamometer will not be required, as any man using this car constantly will become so accustomed to the value of defects that a glance at the profile will tell him the money he may, with judgment, spend for rebuilding a track or repairing it.

The dynamometer consists of two draw bars, one at each end of the car, and extending beneath the platform to within a distance of about one foot of each other. Between the two adjacent ends of the draw bars a spring is placed, and the amount of pull required to draw the car along the track in either direction is recorded by a recording arm, which is connected to the spring with a wire. The apparatus for showing the variation in the level of the tracks is mounted on the platform and consists of two cups of mercury (having a connection between them by means of a pipe) and into which dip two plungers connected to a recording arm. The paper on which the record is taken unwinds at a uniform speed of one foot per thousand feet of track. The car indicator weighs 3,865 pounds. The car is fastened to an ordinary car, which is drawn over the tracks by horses (or any motor car) made to go at as even a speed as possible. The results shown by a recent test trip are very interesting to compare.

Fig. 11 is from an old track-the joints are uncast, spread far apart, and the rails are low at each joint. This is distinctly brought out by the record. The gauge is also uneven. It would appear from the record that the rails spread at each joint. The dynamometer pull is very unsteady, showing the effect of low joints and uneven gauge.

Fig. 12 is a record from a new track. Here we can observe hardly any joints or variations in gauge, and the dynamometer pull is more steady. Occasionally high points are observed, but they are due more

to unsteadiness of speed. A person riding on this track will not be able to detect any joints. This track has the cast welded joint. We find from the data obtained by use of this car that we could afford to expend $7,383 per mile to repair the old track. The estimated cost of repairing this track, leveling and casting new joints, is $1,740 per mile.

The track maintenance during the year 1895 for 184 miles of track cost $158,217 and represented 17.75 per cent. of the total operating ex

penses. This excessive cost is largely due to a partial or complete rebuilding of many miles of track.

In conclusion, I wish to acknowledge receipt of facts and data bearing upon the subject of this paper from Mr. A. J. Moxham, Mr. Augustine W. Wright and Mr. G. A. Mead; also to thank Mr. W. G. Price for valuable aid rendered me in designing and perfecting the details of the indicator car, which can be inspected in the Exhibition Hall during this Convention.

Respectfully submitted,

М. К. BOWEN.

The President: What disposition will you make of Mr. Bowen's paper?

Mr. Seely, Lock Haven: I move that a vote of thanks be extended to Mr. Bowen for his valuable paper, and that it be printed and spread upon the minutes. Carried.

The President: We would like to hear from other gentlemen present in regard to track construction, track joints,

etc.

Mr. Robert McCulloch, St. Louis: There is one important matter in connection with Mr. Bowen's paper. I do not wish to criticise it, but merely wish to give my own experience with regard to the carrying of the current over a welded track without bonds. We have one piece of track which is thirteen miles long and all welded together, the most of it being sixty-foot rail. We have not an ounce of copper nor a bond of any kind on the entire track. We have made tests of all sorts, with delicate instruments, graded to the thousandth part of a volt; tested as long and as short lengths as possible, sometimes a mile and sometimes a single joint, and have found in most instances that the carrying capacity of the joint was greater than that of the rail at any other place. We have found no necessity for bonding; our contact is perfect at the extreme ends of the rail, and I do not believe it is necessary to bond a welded track.

Mr. Harry Scullin, St. Louis: Has the gentleman found any difference in the carrying capacity of the rail, after the weld has been in use some time?

Mr. McCulloch: All the tests we have made have been on track that was laid about ten months since. The tests were made about two months ago, so that the oldest joints tested were about six or eight months old.

Mr. T. J. Minary, Louisville: Were the tracks welded by electricity or cast welding?

Mr. McCulloch: Cast welding.

Mr. Seely: I ask Mr. Bowen whether it is necessary to own a plant for repairing purposes in case the joints are broken? How do you repair joints after the welding plant is taken down? I know that to weld a mile, or four or five

miles, it is not expensive to run a cupola; but it would be in the case of repairing a few joints now and then.

Mr. Bowen, Chicago: For a small road it would be better to let the contractors repair the broken joints. For a large system it would probably be better to keep a cupola. You are always building more or less track, and doing a considerable amount of repair work. A large system is generally building new track, and you could use a machine and make it pay for itself.

Mr. Dodge, New Haven: You spoke of something like 200 joints out of 1,700 which parted. How soon after the casting of these joints did they part; was it very soon after they were welded?

Mr. Bowen: We lost the joints on account of contraction, due to very cold weather in mid-winter. With the first cold. snap, the contraction of the rails pulled 154 joints apart. The joints were then repaired, and since then they have gone through the summer the same as the other joints.

Mr. Dodge: Were they cast in the winter; were they made in the winter?

Mr. Bowen: No, sir; in mid-summer, during the night hours.

Mr. Harry Scullin: I ask Mr. Bowen if most of the joints were examined, or simply those found broken? I want to ascertain if the joints were not loosened as well as brokensome of them loosened in some way or other, and which was not perceptible until they were closely examined?

Mr. Bowen: There were 154 re-made. When they do part, it is for two or three inches, so that you can see them readily. If there is any oxidation between the joint and the rail, I hardly imagine it would do any harm unless it was sufficient to loosen the cast iron from the rail proper. In that case it would come apart two or three inches, due to the contraction of four or five blocks of steel.

Mr. H. H. Littell, Buffalo: What kind of a job do you make of repairing a rail with a three inch space?

Mr. Bowen: We saw a piece of rail and cut it in, and recast around the entire two joints in that case.