00-kilowatt power transformers and the necessary equipment for stepping the 44,000 volt transmission line pressure down to 2,200

volts.

High-power transmission line.-Under a contract dated March 31, 313, 794 double-track span bridges and 20 single-track span bridges Fere to be delivered on the Isthmus. These have all been erected in ace with the exception of five special towers which will be required at the Cristobal and Balboa terminals, and one bridge at Cristobal. Cader contract there have been purchased and received to date for be transmission line 1,562,208 feet of 2/0 Brown & Sharpe gauge tranded copper and 512,065 feet of five-sixteenths inch copper-clad vire. At the close of the fiscal year 1,408,443 feet of 2/0 conductor able were erected. The total amount expended on the transmission Je was $1,014,383.29, of which $701,222.62 were covered by conacts and the balance expended on inspection and installation. Cables-The total amount of cable on order, received and installed to the end of the fiscal year was 2,659,403 feet, of which 1,531,528 feet is lead sheathed and 1,127,875 feet is rubber covered, doublemid wire and cable. At the close of the fiscal year 1,462,684 feet flead-covered cable had been pulled into the ducts and 911,816 et of rubber covered had been used for the conductor slot feeds, trol connections, etc.

Telephone system. An elaborate system of telephone communicahas been designed for the operation of the locks and a contract Tarded for the complete equipment. It is to consist of three subsions: First, that required for the control of vessels passing ugh the locks; second, that required in the upkeep and mainCance work in the lock tunnels; and third, that required for local c service. The total amount of the contract is $6,949.35. Emergency dams. The emergency dams at Gatun were completed fre the close of the last fiscal year, but the final acceptance tests had been finished. During the year the two dams at Gatun were acted and the dams at Pedro Miguel and Miraflores were finished and arted, the first at Pedro Miguel on September 16, 1913, and the ed on October 17, 1913; at Miraflores the first was completed accepted January 14, 1914, and the second on February 7, 1914. ast was made at Gatun in May, 1914; the dam was swung, girders gates lowered, and the pipes driven to close the spaces between Leads of the gates. The upper lock was then filled with water ake level, the upper guard gates and upper operating gates were ned, and the intermediate and lower gates of the lock closed. upper lock was emptied through the culverts until the water was below the guard-gate sill. This brought the full head of feet on the emergency dam, which was found to have a leakage

63399-14-2

of 350 cubic feet per second. This leakage produced no dang current in the lock, and it would have been easy to close any lower gates in the face of the stream. Another purpose of th was to determine whether the emergency dam could be used i of a caisson for unwatering the locks so as to permit access gates for painting, but the leakage was too great to use the as it at present stands, in lieu of the caisson. Experiments are made to devise a means of stopping the flow. The total amou pended for the emergency dams was $2.206.984.67, of which $1 829.90 were covered by the contract for delivery of the materia its erection, and $248.654.73 for inspection and other exp assumed by The Panama Canal under the contract.

Floating caissons.-A description of the caissons for closin entrance to the locks, including the pumping plant for unwat the lock chambers, was given in the last annual report. Bids invited on May 21, 1913, for furnishing the material, constru and delivering the caisson on the Isthmus, and there were two ders, the lowest bidder offering to construct the two caissons an liver them at Balboa for the sum of $648,300, and the price for was given at $330,760. A contract was entered into for one ca under date of August 22, 1913.

Pontoon bridge. In order to maintain communication acros canal with the west side, it was finally decided to construct a pon bridge at Paraiso for carrying the Panama Railroad. This br www.conducted at the expense of the Panama Railroad Co., but design and construction were undertaken by Commission forces. approaches and abutments were built under the direction of Mr. Zinn, readent engineer; the pontoon and superstructure by dredging diynton under Mr. W. G. Comber; the track work by Pymma Railroad Co., and the operating machinery by the first to of the chief engineer's office. The bridge revolves about a p of com pot, amilar to pontoon bridges successfully operated meny years on the upper Mississippi, plans for which were furnis Praph the courtesy of Mr. C. F. Loweth, chief engineer of floor ago, Milwaukee & St. Paul Railway Co. The pontoon is 378 wall, 10 feet wide, and 6 feet 3 inches deep at the center Hedley end the al 38 feet above the bottom of the barge, or 30 Poop the wara level. The apron girders are 64 feet long, rest kad uppers at both ends, and consist of spare lock gate pa mide at each end of the girder for automatica Jana tha & Valecon of 6 feet in the water level of the ca Py hoy named the girders are lifted clear of the conc Na Adven mechanism and temporarily suppor ụ tập vàck of the barge. The bridge is revolved og å gi hot oðan fastened at each bank, which pas

་་

round an electrically driven wildcat on the deck of the pontoon, ear the west end. The mechanism for lifting the apron girders and turning the bridge, and also for operating the rail lift, the rail tches, and the main latch at the west pier, is operated from a cenal panel. The total cost of the bridge to June 30, 1914, was

218.331.78.

Operation of locks.-The lockages made during the fiscal year gave ■ opportunity to try out the locks and their machinery. The first te at Gatun was made on September 26, 1913, when the tug Gatun as put through the Gatun Locks, followed on October 14, 1913, when part of the dredging equipment was locked through the Pacific cks to the lake level. From these dates throughout the year various aft belonging to The Panama Canal were passed back and forth the necessities of the work required, in addition to the tows that nere instituted for handling freight from the Balboa terminals to lon and Cristobal for the Panama Railroad Co. To try out the wing locomotives, the Panama Railroad steamers Allianca and Iacon were locked through Gatun Locks and returned, and through e courtesy of the agent of W. R. Grace & Co. the Santa Clara was cked through the Pacific Locks and returned.

The operation of the locks has developed certain facts in regard the action of flowing water which had not been anticipated. As reviously noted, the gates of the upper locks of each flight and of e Pedro Miguel Locks are duplicated. The upper pair of gates is led the guard gates and the lower pair of the upper two the lock ates At the lower end of the locks the upper pair of gates is called e safety gates and the lower pair of the two the lock gates. The "ace between the guard gates and the lock gates is regulated by an xiliary culvert, while the space between the safety gates and the wer gates is regulated by a T culvert. When the water in the per lock is low and the valves are open there is a sudden drop of water level in the forebay. This is more noticeable at Pedro gel, where the canal above the forebay is relatively narrow, than aat Gatun and Miraflores, where the forebay opens immediately the lake. This drop is faster than can be followed by the water the space between the guard and upper gates, and the result is a se head on the guard gates, causing them to open at the miter. The reverse head lasts but a short time.

In the first lockages the T culvert regulating the space between lower and safety gates was in free communication with the side all culvert near its lower end, but above the lower valves. When upper valves were raised the water in the space between these pates rose faster than in the lock; consequently there was a reverse sure on the safety gates, causing them to open at the miter, the ere being sufficiently great to compress fully the springs in the

gate maneuvering struts. Due to the possible danger of cri the moving apparatus by such reverse stress, the valves in culvert were partially closed, choking the communication be the side wall culvert and the space between the gates. By e menting, a degree of closure was reached which caused the wa the space between the culverts to lag behind the water in the when filling, and at the same time to fall rather more rapidly emptying. In this way a positive pressure was always kept a the safety gates. With the use of both side and center wal verts, as the rapidity with which the water level in the lock ch is greatly increased, a different adjustment will be necessary.

When the valves in the side culvert are raised and the water the look the flow of water is greater through those openings i laterals which are nearest the middle wall than through those n the side eulvert. As the water rises in greatest volume next t middle wall, there results a slight slope of the surface towar side wall When both side and middle culverts are used no suc Firm fe nofievable. The first gush of water from the side cul appears to come from the highest laterals and then successively The urlers No far as can be ascertained, however, it can not be s that the drevhange a greatest from any one of the laterals.

În fling or empty ing a small canal lock the water levels appr why they with a espidity depending upon the square root of me hoed Tensed the end of the operation the chang bai dooma »lowy, and the flow of water is supposed to fo: wt the cousilization of the levels. Freque my arsonar, mit vegy slight head against them to a mabag in maing for the head to vanish enti by the lacks of The Panama Canal there ch, water, so that the water in the char Celeb Fighe that the level in the chan down any, the Cates separating the ch At Pedro Miguel

at the overtravel of the water from the emptying lock into the tail y may result in continuing the flow of the fresh water through the Ivert beyond the point at which in theory it should cease and, refore, in obviating the danger of the resultant pressure. Hower, no trouble is experienced from the resulting pressure against e lower gates. The difference in density of the water above and low the gates is, however, noticeable in another way. When the wer gates are opened the fresh water rushes out with considerable locity on the surface of the salt water below, and there is a correonding rush of salt water along the lower strata. This current ntinues for a considerable time and has a noticeable effect on ves's leaving the lock, sheering them away from the middle wall as on as they pass the angle of the side wall.

The slope given to the water in the lock by use of the side culverts s already been noted. The same effect was noted during the filling the lowest lock with fresh water when the lower gates have been it open long enough to insure salt water filling the lock. The slope from the middle wall so that the ship moves to the side wall and towing locomotives are unable to hold a large vessel central in lower lock during the entire process of filling when the side vert only is used.

As noted in previous annual reports, in the formula for filling and tying the locks the coefficient of flow C, used to determine the L. velocity, was assumed at 0.65, this being slightly more favorable in experience at other locks has shown to be readily obtainable, the nature of design and construction warranted the assumption. om experiments made at Pedro Miguel when the lake level was at erence 84.8 and the water in the lock was at reference 50.9, the e of C for the side culvert, with both valves open, was determined be 0.886, much more favorable than was anticipated. Similar obations, using only one valve, were made both at Pedro Miguel in the upper lock at Gatun as a check, and the value of C was and to be 1.177 and 1.272 at the two locations, respectively. The e of the coefficient for emptying the lock at Pedro Miguel is ewhat less favorable than that for filling, the coefficient C for ade culvert being found to be 0.804 when both valves were used. Gatun and Miraflores, where the culvert is turned up at the harge, the coefficient of flow for emptying the lower lock is et 8 per cent less than through the level discharge at Pedro el When the middle wall culvert is used in combination with side culvert the time of operation is greatly reduced. The Pedro zel Lock can be filled or emptied in about eight minutes without ing too great a disturbance in the chamber. Observations have yet been made with the middle culverts at Gatun and Miraflores.