438 



THE BELL SYSTEM TECHNICAL JOURNAL, MARCH 1953 



The straight lines in Figs. 23, 24, 25 and 26 are drawn through the 

 average points for per cent condition and retention of the respective 

 4- and 8-pound groups, and projected to intercept the 100 per cent con- 

 dition hne. The empty-cell data are represented by the solid line and the 

 full-cell data by dashes. The threshold concentrations necessary to pre- 

 vent all decay are estimated from the intersection of the gradient lines 

 and the 100 per cent condition line. This method assumes that the rela- 

 tion of condition to treatment retention is linear at or near the threshold, 

 providing the average condition points through which the lines are drawn 

 are established by the logarithmic based rating system described. The 

 method also resembles in a way the procedure of the Madison investi- 

 gators who have used the intersection of straight lines drawn through 

 operational losses and decay weight losses in estimating the threshold 

 retention in creosoted blocks.^^ As in the case of the latter the method 

 would probably be more precise if one had more points for average 

 condition at average retention nearer the thresholds. At any event the 

 system represented by Figs. 23-26 seems to be about the only one that 

 indicates probable thresholds for these particular creosotes and these 

 particular sets of data. 



Table XIII — Estimated Threshold Retention and 

 Average Life 



^-inch southern pine sap wood stakes; (see Tables X-XII); 1941 series; 

 Gulfport test plot. 



* See text for additional data on method used in estimating the threshold re- 

 tentions at treatment and the average life figures. 



