Sharma 



Exact theoretical curre 

 by direct coaputatlon 



With seBl-emplrlcaL 

 corrections after Inul 

 and author, aee Ref. 4 



Q Prom analysio of 

 five transverse 

 C cuts, see Bef. 



From analysis of a 

 single longitudinal 

 C cut (present work) 



KlJJ'LJuiD^' 



-?5 



7.S 



1Q0 



12,5 



^5.0 [rry-1] 



Fig. 7 - Amplitude spectrum of Inuid 

 S-201 for 7o = 6.1 or F„ = 0.287 



The overall result of this experimental verification can be tentatively sum- 

 marized as follows. The transverse- and longitudinal-cut methods yield fairly 

 consistent results and appear to be essentially equivalent, thus allowing the 

 great practical advantages of the longitudinal-cut methods to be fully exploited 

 in the future. If wave resistance alone is used as a criterion, then the differ- 

 ences manifested in Table 4 can be almost dismissed as due to experimental 

 error. However, it should be noted that the discrepancy is least in Case 2 

 (Fig. 6), where the amplitude A^ is relatively small, and highest in Case 3 

 (Fig. 7), where the amplitude A^ is relatively large. In view of the numerical 

 experience gained on theoretical wave systems (Figs. 2 and 3) this could be evi- 

 dence of a systematic error arising from inadequate truncation correction and 

 should therefore be further investigated. 



Another possibility of improving the accuracy of the longitudinal-cut analy- 

 sis would be to record two or more cuts simultaneously and use the redundant 

 information to obtain more reliable least-squares fits. It may be recalled that 



754 



