TRANSVERSE SECTIONS UPON RESISTANCE. 309 



For changes of length within the limits that this data would be used, the second part 

 of his correction factor can be neglected, as clearly shown in the appendix. 



Mr. a. J. C. Robertson, Member: — In this very briefly worded paper, the authors 

 have presented this Society with a mass of most valuable tank-test data which it would take 

 many weeks to thoroughly assimilate. They have thrown a great deal of new light on the 

 resistance of the fuller types of ships, suitable for slow to moderate speed, and the more we 

 are able to go into these curves the more thoroughly do we appreciate the enormous work 

 and the great value of this contribution. 



A good model freight ship has been used as a starting point, and that model has been 

 varied radically in the shape of the transverse sections in the fore and after bodies, and in addi- 

 tion the lengths of these bodies are variously proportioned to the whole ship length ; and all 

 the models have been tested at four different draughts and the results furnished (except in 

 case of Plate 55). 



Plate 55 gives the first set of results, a full form of long run, and a fine form with 

 equal entrance and run, and shows each with two kinds of stem and one bow, and also 

 with two kinds of bow and one stem. The variation of the stem lines shows very small 

 change in the resistance values, but in general the U-shaped stern with the consequent nar- 

 rower water lines shows up somewhat better. This permits shipbuilders to indulge in a 

 reasonable amount of club foot of the stem lines, permitting more rigid support to the pro- 

 peller shaft and somewhat easier lines above water, preventing the slamming which occurs 

 with a full counter when the vessel is making its way against a head sea. 



The variation in form is much more interesting where the bow is changed, and in gen- 

 eral tlie results also help the naval architect because with the fuller model the V-shaped bow 

 shows up considerably better than the U-shaped bow for the useful range of speeds, whereas 

 in the fine model the U-shaped bow is the best. This apparently indicates that a certain 

 amount of fullness is permissible near the base line irrespective of the fullness of the water 

 line, but that excessive fullness, which is fatal to the ship's structure in head seas, is also 

 bad for the ship's propulsion. 



In Plates 56 and 57, a single model is shown at four different draughts, and this throws 

 a great deal of light upon the problem of U and V-shaped lines. In plates 56 and 57 the M- 

 bow and M-stem resistance curves are indentical, Plate 56 being devoted to the effect of vary- 

 ing the bow sections and Plate 57 the varying of the stern sections. 



Quite apparently the variation in the bow sections is the more important one, and con- 

 siderable savings in power can be gained by the adoption of the correct transverse form. 

 You will notice that, at the very large draughts shown at the bottom of the diagram, the 

 dift'erence between the V-bow and the M-bow is very much less than for the lesser draughts ; 

 this curious feature is in accordance with the observation I have made in analyzing other 

 models. A fullness of transverse section in the underbody affects the resistance in propor- 

 tion to its nearness to the load-water line. It is customary to think of the residuary resist- 

 ance of ships as wave-and-eddy-making, and though we have no means of measuring the rela- 

 tive amounts of these resistances this practice is a good one. That part of the ship which 

 is sufficiently far under the surface has relatively little to do with the development of the 

 diverging waves which are always lost, and the transverse waves which may be lost or may 

 be, in a large measure, recovered according to the length of the parallel bodies adopted. If 

 the lines low down are shaped for eddy-making only, the result will be good providing the 



