The computation of the wave length from the recorded data is accomplish- 

 ed in a similar manner « It differs merely in the determination of T 

 which is obtained by dividing the time elapsed by the numlier of waves 

 (k) of relatively even height (T - t/K)o 



From previous investigations (3j ppo 24-27) it may be assumed 

 that the given relation between the mean values of T and L is sufficiently 

 accurate o The wave length thus determined from measurements can than be 

 regarded as the actual mean value for the vrave length to which estimated 

 wave length should correspond „ Figure 2 indicates that the observed wave 

 length is considerably smaller than those obtained by measuremento On 

 the other hand, the wave lengths estimated try counting the number of 

 passing waves are in better agreement with respect to order of magnitude c 

 In two-thirds of all cases 5 deviations are less than 5 meters or below 

 4.0 per cento The directly estimated wave lengths in figure 2 show the 

 same tendency for increase and decrease as the values obtained from 

 measurement 5 only the latter are four to five times as greato 



The estimates of observer II of lightship FEMARNBELT made in 

 the period 5 3-17 October 1936s were checked in the same manner as those 

 observations described above. The comparison between observed, measured, 

 and calculated data is presented in figures 3 and 4„ The picture is 

 essentially the same and occasionally this observer, too, has over 

 estimated the viave height o In the extreme case here a wave measured to 

 be O0S5 meters was estimated at lo50 meterSo Bsth observers were in- 

 formed of the results of comparison of measured and observed v/ave heights 

 and lengths, whereupon subsequently reported wave lengths increased o 

 Figure 6 indicates better agreement between measured and observed wave 

 lengths during the period 24 October to 13 November 1936, and according 

 to figure 5 vrave heights for this period are in good agreement; only at 

 very high seas were estimated heights greater than those measured „ The 

 observer noted that the buoy of the recorder undercut due to the high 

 sea and it is not to be assumed that the measured values are the more 

 accurate ones in this casco 



The estimates of observer III of lightship BOHKMRII'T'' were also 

 checked for a short period by means of measurements,, The result of 

 the comparisons v/as about the same as in the case of the estimates of 

 observers I and II on the lightship FEH'TARKBliLTo Figures 7 and 8 

 show the comparison between observed and measured wave dimensions for 

 the period, 4-10 November 1936 at the lightship R^RKUMRIFFo The wave 

 heights are generally in good agreement and deviations are only 

 appreciable in the case of verj^ high estimated values o The estimateu 

 wave lengths vrere too short but the values calculated from the counted 

 number of waves passing are in fairly good agreement with the measured 

 wave lent^tho 



The lack of measured data prevented comparison of the values 

 estimated by observer IV of the same lightship » It appears, neverthe- 

 less that the estimates of this observer followed the same trend as the 

 rest. 



