models, as in human beings, the selection of good parents does not necessarily lead to 

 better - or even as good - offspring! 



Although the original conception of the project was to derive a series of related parent 

 forms which would serve as a point of departure for future model programs, and which there- 

 fore should have reasonably good but not necessarily optimum resistance qualities (the quest 

 for which might indeed last forever), it was evident from the very lengthy and valuable dis- 

 cussion on the paper that the members of the profession desired something better in quality 

 than Series 57 as a basis for any such systematic program. 



The panel thereupon reviewed the original series and agreed that the real merits of 

 the Series 57 models could best be established by comparison with the performance of actual 

 successful ship designs. In this way, differences in proportions and in LCB position could 

 be eliminated and the effects of differences in shape of area curves, waterlines and section 

 shapes evaluated. 



Five designs were chosen as being typical of good, modern, single-screw ships, which, 

 of necessity, had to meet many requirements in addition to those of good resistance qualities. 



Three of these were Maritime Administration vessels of the MARINER, SCHUYLER 

 OTIS BLAND, and C.2 classes. The other two were Bethlehem Steel Company designs. 

 One was the tani<er PENNSYLVANIA. The other did not represent any built ship but was a 

 design for a 0.70 block coefficient ship given by Mr. H. de Luce in his contribution to the 

 discussion on the Series 57 paper. 



Models of the first four were available at the Model Basin, and a model of the fifth 

 design was made and tested. 



For comparison with each of these, an equivalent Series 57 model was made to lines 

 drawn out from the contours. Each pair of models represented a ship of given length, beam, 

 draft, displacement, and position of LCB so that the differences in each case were restricted 

 to the shapes of area, waterline, and section curves. 



The results of these model tests are given in full in Reference 45. Briefly, at speeds 

 appropriate to the different fullness coefficients, the Series 57 models were in general some- 

 what wOTse than those of the actual ships by amounts up to a maximum of 6 percent. 



The area and load waterline (LWL) curves of any pair of these models were not very 

 different in shape or character, and the chief differences lay in the shape of the cross 

 sections. An analysis of the bow and stern lines indicated that the actual ships had, in 

 every case, more U-shaped sections than the Series 57 models, and the Panel decided that 

 new contours should be drawn using the sectional area and waterline curves for these actual 

 designs as guides, thus giving a more U-shaped character to the transverse sections while 

 paying due attention to stability considerations. This change was also expected to lead to 

 improved propulsive efficiencies. 



These new contours formed the basis for Series 60. 



II1-2 



