428 



THE BELL SYSTEM TECHNICAL JOURNAL, MARCH 1953 



pretation of the behavior of some of the ^^-inch stakes in the Gulfport 

 test plot the recorded per cent condition of the stakes at any one inspec- 

 tion period has been translated into terms of the proposed 5-number log 

 base system. 



The stakes were carefully sawed and planed units, %-inch square in 

 cross-section/^' ^^ Before treatment the stakes were selected so that they 

 would represent the normal distribution of density in the material avail- 

 able. Table IX shows the analyses of the four different creosotes used to 

 treat the stakes in respective 4- and 8-pound groups. Both empty-cell 

 and full-cell treatments were employed. The full-cell treatments were 

 made with toluene as a diluent in order to provide more uniform and 

 lower controlled retentions in the treated specimens. The empty cell 

 specimens were sorted after treatment to retain the middle group of 

 retentions, "svith a view to eliminating as far as practicable some of the 

 factors in the empty-cell treatment variation. All of the stakes were 

 treated between March 11 and March 26, 1941, and they were all placed 

 in the plot in the approximate period from April 8 to April 22, 1941. The 

 distribution of retentions in the 8-pound stakes set out in the Gulfport 

 plot are shown in Table X, along with data on average retention, stand- 

 ard deviation, and coefficient of variability. 



Table TX — Analyses, Water-Free Basis, of Four Creosotes 



Used in Treating %-inch Southern Pine 



Sapwood Stakes 



1941 series; Gulfport test plot 



