Any one sending a sketch and description may a quickly ascertain our opinion free whether an invention is probably patentable. Communications strictly confidential. HANDBOOK on Patents sent free. Oldest agency for securing patents. Patents taken through Munn & Co. receive special notice, without charge, in the Scientific American. A handsomely illustrated weekly. late circulation MUNN&Co.361 B™"way,N ew York Branch once. 035 I' St.. Washington, D. C. Specialties We wish to caU attention to our exceptional facilities for producing molded specialties of all kinds for new and old inventions. We have specialized on this line of work for years, and have developed superior equipment and experience which we place at your service: Pump Valves and Valve Discs for high and low pressure; Diaphragms for Heaters and Pumps; Printing Contact Mats for Blue Print work. We solicit an opportunity to figure on these or any other class of molded specialties. MORGAN 6, WRIGHT Manufacturers of Grad Rubber Goods 14 Bellevue Avenue, Detroit, Mich. Manufacturers! investigate r.Portsmouth,Virginia Portsmouth offers unequalled advantages for the profitable operation of almost any manufacturing indMtry. Itis the natural gatewaybetwn toe resources and markets of the North and South, draw- ingfrom and catering alike to both. Cheap fuel at the very door, an abundance of reliable labor and an inexnamitible 8upply ofrawmaterial. Nine lines of railway and a score of lines of water transportation assure exceptionally low freight rates. A delightfully even climate, pure water, good schools, good health and beautiful homes. Splendid factory sites and busine.. opportunities are now available. Valuable assistance will be given investigators. Full particulars on request. Address l. T. VAN PATTEN. SECRCTARY. PDRTSMOUTH. VIRGINIA. the purpose of estimating, and, in the usual run of cases, it should not fall below that figure. This per cent, moreover, is usually based upon the effective horse-power of the bare hull of the ship. In practice, for vessels with unusually efficient machinery and few appendages, the efficiency of propulsion thus figured may rise as high as 60 per cent. But it is hardly safe when king to rely upon such a high figure, it being obviously desirable to provide a reasonable margin. It will be evident from the nature of; the case that the accumulated records of experiments of the model basin increase in value as they increase in number, and after several years of systematic experimentation it is possible from the records of the United States model basin without making and trying a model to make a fair approxition to results to be expected of almost any normal type of vessel. Fig. 3 illustrates such a case. Suppose we wish to determine for an Atlantic liner of 40,000 tons displacement and 800 feet in length the approximate curves of effective horse-power according as the boat is made with a fine cylindrical coefficient or a full cylindrical coefficient. The cylindrical coefficient is the ratio between the actual submerged volume of the vessel and the cylindrical volume obtained by multiplying the area of the mid-ship section by the length of the vessel. After displacement and length the cylindrical coefficient is the most important variable affecting resistance. The area of and hence the power absorbed by the wetted surface depend almost entirely upon the displacement and the length, and we are enabled to draw one curve of frictional effective horse-power, it being the same in the first approximation for any cylindrical coefficient. From accumulated data of systematic experiments we are able to further calculate curves of residuary horse-power for the various coefficients, which, a dded to the curve of frictional horse-power, give the curves of Fig. 3. In developing for an actual vessel of the size and length the most I desirable lines we could improve some- I what upon the results of Fig. 3, but not very much. It is seen that up to any speed which might be regarded at present within the realm of possibility, say, up to 30 knots, the friction is the main fac- 1 tor. This is generally true ; that is to say, there are very few vessels where the residuary resistance is greater than the frictional resistance. A few highspeed navy cruisers, torpedo boats, and torpedo-boat destroyers, and fast passenger vessels for short runs may have residuary resistance greater than the frictional resistance. Even for fast vessels, as a rule, the residuary resistance is not more than a third of the total, or one-half the frictional resistance. Excessive residuary resistance means that the vessel is too short for her speed, and it can be reduced by increasing length without much increase in frictional resistance. ! It will be observed in Fig. 3 that at practicable speeds there is much more difference in total effective horse-power between cylindrical coefficients of 0.60 I and 0.65 than between 0.55 and 0.60. At 0.55 we are close to the minimum. It will also be observed that at the top speed the curve for the 0.55 coefficient is beginning to rise beyond that for 0.60 co- I efficient. This is typical. Numerous have shown that for speeds which are moderate in proportion to the length of the vessel the cylindrical co' efficient should be low; while for speeds , which are high in proportion to the length, say, speeds in knots twice or ! more the square root of the length in feet, the cylindrical coefficient may be I given with advantage surprisingly large values, as high as 0.65 or 0.66. For excessive speeds, at which the vessel begins to rise bodily from the water, the cylindrical coefficient may be made even greater, or, in other words, the ends may be made very full