48 



OBSERVATIONS AND RESULTS IN PHYSICAL OCEANOGRAPHY 



weight at the end of the wire, which was not kept constant, 

 and by the speed of lowering. The depth factor corre- 

 sponding to any given wire angle can be read off from 

 the curve in figure 1 and the wire depth obtained by mul- 

 tiplying the wire, length with this factor. 



In order to estimate the probable errors of the wire 

 depths which have been determined by this method, such 

 wire depths have been computed in the cases in which 

 the depth was determined independently by thermometer 

 and entered in table 1 together with the differences be- 

 tween the wire depths and the thermometer depths. 

 These differences, which are represented graphically in 

 figure 2, increase with increasing depth, which means 

 that the error in the wire depth increases with depth. 

 All points except three fall inside the two straight lines 

 which have been drawn in the figure, representing a dif- 



ference of 2.5 per cent of the depth. Of this difference 

 0.5 per cent can be regarded as owing to uncertainty in 

 the thermometer depth and the maximum error of the 

 wire depth is thus about 2 per cent of the depth. It is 

 evident from the graph and from the values in the table 

 that the error of the wire depth as a rule is considerably 

 smaller, especially if the depth is small. The result 

 must be regarded as very satisfactory, considering that 

 wire angles greater than 40° frequently occurred. 



Summarizing the preceding discussion it can be 

 stated that when sounding with piano wire has been under- 

 taken and the wire length and wire angle recorded, the 

 wire depth can be found by multiplying the wire length by 

 a factor which is read off from figure 1. The wire depth 

 which has been computed by this method has a maximum 

 error of 2 per cent. 



