The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

April, 1909 should naturally be kept as low as pos- sible. Below a certain velocity, the per- 'formance of a skimmer model is very bad, owing to the formation of large waves, which allow the stern of the vessel to fall and greatiy increase the angle of the planes, thus rendering it more difficult for the vessel to mount to the surface and skim. This difficulty may be lessened by (I) extending the amount of supporting surface, or (2) reducing the weight of the vessel, the surface remaining the same. The great strength necessary to withstand the shocks to which this kind of boat is subjected, renders reduction of weight dificult. A boat very wide and short in shape leads to excessive air resistance, and this resistance becomes quite an im- portant factor at speeds of about thirty miles an hour. For. this reason too great a width must be avoided. The wave formation, when a skimmer is moving below the skimming phase, "TAE Marine REviEw still higher speeds the motion of the water becomes unsteady. Fig. 4 in- cidentally shows the house from which the author's experiments were conduct- ed. It contains a winding-drum for the tow-line fitted with rewwraing apparatus. Skimmers are near relations to flying- machines, and although smooth water would appear to form a definite plane on which to travel, a boat of this kind, when moving at high speed, is not content to be limited to motion in two dimensions, but tends to oscillate vertically and jump from the water surface, and under some conditions to dive below. The skipping action has been already referred to above, the author's model having been frequently made to leave the water temporarily in this way, at relatively high speed. The position of the center of gravity] of 'the relative merits of the vessels ig very important to get the best re- sult. In the models he had tested, im- Fic. resembles that of an ordinary vessel, but the waves are larger in proportion to the size of the vessel, and diverge at a wide angle. Some photograhps taken during recent experiments by. . the author, on a Ramus model, illustrate this. Fig. 2 shows the model used, and the lines by which it was drawn along. It will be seen that the boat is fitted | with a cross-piece,. or bar, to which the lines are attached. The lines join \some little distance ahead .and astern x the 0. to forma single. tow-line. ig ¢- an illustration' of. the, Ramus me at low speed, ar waves and echoes. . The contrast is great when the same model is made to travel fast. This is shown in Fig. 4, and here it will be seen that 'the volume of the waves is much less and the angle of divergence is small. At showing "the pH- 5; . provement appeared tu take place as this was moved aft up to the stage when skipping or- flapping commences. This dancing motion, if allowed to de-| velop, may become dangerous, but the best results seem: to coincide with its commencement. It has been suggested that the motor boat Gyrinus liffed to 'some extent, and so- reduced" its displacement, Tike a skimmer, ah 'the © photographs of "the boat at full speed clearly show that 'stich is not the case; for while the bow' is nearly at the same level at rest or speed, when moving fast, the stern is much lower. The fore part of this boat ap- pears to plough a channel, into which the stern falls, and with increase of "speed the resistance rises very rapidly, although the form of stern would ap- pear well adapted to avoid this result. The lines ot the Gyrinus produced. by vertical longitudinal sections rise very gently, and terminate in a sharp angle; therefore there is no surface which can suck the water up, and, by so doing, re- ° duce the pressure below that of -the atmosphere. The growth of resistance in a - skim- mer differs from that in an ordinary boat in a very marked manner; it rises very rapidly at first with increase of speed, but when the condition of skimming is established, it may even fall temporarily, and afterwards only Fic. 6. "rise very oes so. that ae. power re- -quired only inicreases a-Jittle- , faster than:>the velocity. This. is welh illus- trated by the diagram, Fig. 7, below in which the;iresistance,of the. Gyrinus and a skimmer: of thessame. weight is plot- ted, in which case the curves ,cross,jat about 17 knots. The boat- form is- much the better for lower speeds, while above the point where the resistances are equal the advantage is greatly in favor of the skimming model. . Experiments were made with the