To Prevent Drawbridge Accidents. To the Editor of the SCIENTIFIC AMERICAN: If the approach to the drawbridge in Norfolk had been protected by a derail, that sad accident would not have occurred which resulted in the death of eighty people. The derail could be so connected that when the draw swung off, it would open the derail and, at the same time, throw a semaphore signal to danger (also mechanically connected with the drawbridge), which, located at a proper distance away, would notify approaching trains the draw was open. In case a train disregarded this signal, the derail would let it off on to the ground, thus preventing what might be a bad acci dent. F. H. SIDNEY. Boston, Mass., August 30, 1905. The Submarine Signal. To the Editor of the SCIENTIFIC AMERICAN: In your issue of September 9, on page 196, the last column, there was an article concerning submarine signaling, headed "An Improved Submarine Signal." We would like very much to correct the false impres sion which would be circulated by this article. The article as it stands contains facts which are perfectly true, except that this "fish" transmitter, as it is called, is an obsolete piece of apparatus which has not been in use by our company for a couple of years. Mr. Mundy at that time was connected with the com pany, and in conjunction with the other engineers in this company at that time devised this "fish," the point being that the noise in the boat was so great that if the receiver were put in the boat itself it would drown the sound of the bell for which the hearers were listening. Improvements in the receiving apparatus have entirely overcome this and the fish has been dis carded. It was an apparatus which possibly was a temporary help in overcoming a slight difficulty in the development of the system. We hope that you can by this explanation clear up the impression which was created by your article. SUBMARINE SIGNAL COMPANY. Boston, September 16, 1905. Chains Versus Cables on the Manhattan Bridge. To the Editor of the SCIENTIFIC AMERICAN: An editorial of September 9 makes a comparison between the new Elizabeth bridge at Buda-Pesth and the Manhattan bridge over the East River. If the Buda-Pesth bridge represents the "advanced ideas of the leading bridge builders of the world," these ideas must necessarily refer only to the rocking towers as a novelty for large bridges, because, outside of this feature, the Buda-Pesth bridge is built on old and reliable principles, like many other suspension bridges, and represents no special new feature. That the floor is suspended from eye-bar chains, in place of wire-cables, is a revival of an old construc tion and, in an economic sense, decidedly a retrograde step. The stiffening construction of the Buda-Pesth bridge is precisely the same kind and of the same proportions as in the Manhattan bridge in its present design. The claim that this design is inferior to the chain-cable bridge, designed under the former administra tion, may be a matter of opinion, but that it will cost $2,000,000 more than the latter is a decided error. The abolishment of the tower hinges is mentioned as one point of inferiority, but many engineers believe that this fact will make the bridge superior, though they all agree that it will make the towers more ex pensive. The present engineers of the bridge con sider the advantages of a tower without hinge, in gaining a precise knowledge of the stresses and in giv ing greater stability to the bridge, fully worth the money it costs more. In regard to a hinge in the floor beam, the matter is reversed, viz., floor beams with hinges will be more expensive, but they admit of a more accurate determi nation of the stresses. This matter is, however, a de tail of secondary importance and cannot be taken as a criterion for the quality of the bridge as a whole. The statement that the Manhattan bridge, on its present design, is costing $2,000,000 more than a chain-cable bridge would have cost, can easily be refuted as an impossibility. The breaking strength of steel wire is known to be three times greater than that of nickel-steel eye-bars, hence a chain made of the latter will, under existing conditions, weigh 4.84 times more than a wire cable of equal strength. The total weight of the wire cables, as calculated by the lowest bidder for the Manhattan bridge, is 6,328 tons, hence the nickel-steel chain would weigh 30,827 tons and would cost $5,654,000 at 9.23 cents per pound, given by the same bidder as unit price for nickel-steel, including the temporary working bridges. The bid for wire cables, however, is only $1,848,000, showing that the eye-bar chain would cost $3,806,000 more. To this sum must be added other sums for the increased cost of anchorages and anchor chains, found ations, etc., which would make the total cost of the chain bridge about $4,750,000 more. Granting all the minor economic advantages claimed for the eye-bar design and making liberal allowances for them, it still follows from the above given figures that the Manhattan bridge, on its present design, will, instead of $2,000,000 more, at the very least, cost $4,000,000 less than the eye-bar bridge would have cost. WlLHELM HlLDENBRAND. New York, September 13, 1905. [That the chain cable is "old" is not disputed. The "advanced ideas" are those which are shown in the use of nickel-steel, the rigid attachment of the chains to the towers, the hinged, footing of the towers, the hinged floor beams, and, in the Manhattan chain de sign, the trussing of the cables. We note that our cor respondent admits that throwing out the hinged con struction in the towers has incurred greater expense. In this connection it may be mentioned that the re jection of the hinged towers--a feature which the ex pert board had specially commended--has added 4,100 tons to the weight of the towers. This is 299 tons more than the total weight, 3,801 tons, of the whole suspended structure of the Buda-Pesth bridge. The question of the relative cost of the eye-bar and wire cable designs is discussed in our editorial columns.-- EDITOR.]