11 



Prom Havelock's investigation, the fact, mentioned before, follows 

 that k and k are not bounds of k . 



There exist two pertinent publications on forced heaving oscilla- 

 tions in the moored condition. They deal with the dependence of the hydrody- 

 namic masses upon the frequency of the oscillation. Figure 6 shows results for 

 a circular cylinder immersed up to a plane of symmetry following P. Ursell 9 

 and for a ship model following M. Haskind. 10 Remarkable Is the hump in the 

 k curve in the range of small frequency parameters F^ = u> yB/gl it indi- 

 cates that the condition d<f>/dz is approximately satisfied. We emphasize 

 further that in a certain region k becomes smaller than k = 1 , which again 

 means that the condition </> = does not furnish a bound for k . Unfortunately, 

 nothing is known about the method used by Haskind; we stress, however, the 

 importance of the curve of experimental values due to the same distinguished 

 scientist. We face now the question as to how far experiments in general have / 

 contributed to our knowledge in this field. 



2.6 When the derivation of theoretical solutions is difficult, the ex- 

 perimental approach suggests itself. In the field of wave resistance of 

 ships, an overwhelming number of mostly unsystematic tests have been performed 

 which nevertheless have yielded important results. Since the problem of 

 added masses is less important from a practical viewpoint, the amount of ex- 

 perimental work is small and correspondingly the success too is modest. The 

 last fact becomes especially evident because in older publications even the 

 decisive parameters of the problem remained unknown. 



Experiments dealing with added masses of ships in accelerated and 

 retarded translations (inertia tests) are especially precarious. The longi- 

 tudinal coefficient k is normally small; <* 

 additionally we must consider that the 

 wave resistance in its proper sense, and 

 even the frictional resistance, depend 

 upon the acceleration. In the light of 

 our present knowledge, the foundation 

 of inertia tests which in principle are 

 very suggestive, must be reconsidered. * 

 The determination of k by such experi- 

 ments is not promising, at least for 

 the time being; on the contrary, one 

 would prefer to estimate k and derive 

 thus conclusions about the magnitude 

 of the resistance. 



