In order to keep the original program at the Taylor Model Basin within reasonable 

 bounds, a graphical method of delineating the models was adopted, and except for the fact 

 that comparisons were made with existing "good" ships, no attempt was made at that time 

 to explore the vast field of possible changes in the shapes of area curves, sections or 

 waterlincs. At the time of publication of the earlier papers, this adoption of certain area 

 curve and section shapes was subject to some criticism as having been done too arbitrarily 

 and accepted too easily. But the fact that it has taken 15 years to reach the present position 

 is sufficient indication of how long it might have taken had we been led astray in the early 

 days by the temptation to explore all the delectable byways, opening up vistas of attractive 

 changes in area curves, waterline and section shapes. Now that the main framework has 

 been finished, such exploration is undoubtedly necessary; it could well form the subject of 

 a number of research projects in different tanks. 



There are a number of possible approaches to this type of research. One would be the 

 trial and error method of trying different shapes of area curves, waterlines and sections, be- 

 ing guided in successive choices by the results of each step in turn. A second would be to 

 apply statistical methods to the results of previous model tests— both Series 60 and others— 

 to determine the influence of the different design parameters, and so approach closer to an 

 optimum combination to suit any given design conditions. Considerable success has been 

 achieved in this way in the particular field of trawler design. ^^ 



Thirdly, one may seek guidance from the mathematical work being carried out in the 

 field of wavemaking resistance. As a matter of history, it is perhaps worth recording that 

 much thought was given to this aspect of ship resistance research when the original series 

 was being planned. At that time. Dr. Weinblum was a consultant at the Taylor Model Basin, 

 and he took an active part in the planning and in the early phases of the project. The ques- 

 tion of using mathematically defined lines was seriously considered, and it is perhaps of 

 interest to record some of Dr. Weinblum's views as Set out in his discussion on the first 

 series paper (discussion on Reference 44, pp 722-4). 



"For a considerable time attempts have been made to establish a rational theory of 

 ship resistance as the function of its form by using analytical methods and pertinent basic 

 experiments. Although this approach is developing successfully, if slowly, the choice of 

 proper ship forms for practice has still to rely widely upon experimental data, obtained by 

 testing methodical series or single models. Clearly, the latter procedure is the most waste- 

 ful way of getting results which are capable of appropriate generalization. Therefore, from 

 a practical point of view, the need for Series work cannot be denied at present. On the 

 contrary, the substitution of methodical experimenting for single testing promises within 

 plausible limits decisive advantages in various respects . . . when the present series was 

 being planned the authors received proposals to base the work on algebraically defined lines 

 . . . There is no magic in mathematical lines. Their use in research work is desirable 

 essentially 



XIII-3 



