358 BELL SYSTEM TECHNICAL JOURNAL 



and cable terminals. The need for these dry spots had become apparent 

 from analyses of line tests which showed that after the braid on the wire 

 had weathered and had begun to disintegrate, a considerable period of 

 time elapsed after rain ceased before the line insulation returned to its usual 

 dry weather excellence. The insulated wires passed through the insulator 

 and at the point of exit the conductors were soldered to a metal insert 

 moulded into the insulator. The bridle wires themselves supported the 

 insulators at a point close to the connections to the line wires. It is of 

 interest to note, in passing, that a patent was granted to Jewett on some 

 design features of this insulator. 



Returning to the discussion of the New York-Chicago loaded line experi- 

 ment, it was found that the insulation improvements described above were 

 insufficient to provide satisfactory transmission performance during periods 

 of continuous bad weather. During fair weather periods, however, the 

 transmission was as good as had been expected. The experiment thus 

 proved beyond question the need for a new type of line insulator having 

 substantially better insulating properties than those of the standard toll 

 line insulators, which were made of glass and had a single petticoat. 



Renewed studies of this particular question led to the rush development 

 of a moulded double-petticoat porcelain insulator. The possibilities of 

 porcelain insulators had been under consideration for several years, not- 

 withstanding adverse cost factors. The accumulated test data on porcelain 

 insulators generally similar in design to the standard glass insulators in- 

 dicated that after a long period of exposure on roof racks the wet weather 

 insulation was about twice as good as that with the glass insulators. More- 

 over, theoretical studies indicated that a properly designed double-petticoat 

 porcelain insulator should be about twice as good as the single-petticoat 

 porcelain insulator. The possibility that the opacity of the porcelain might 

 unduly encourage insects to build their nests under the petticoats and 

 thereby impair the wet weather insulating properties, however, could not be 

 allowed for quantitatively in the preliminary estimates of the potential 

 over-all improvement, due to the limited and conflicting evidence on this 

 question. 



Consideration of all of the factors involved, including favorable price 

 estimates, led to a decision in October 1909 to substitute the new double- 

 petticoat porcelain insulator on the experimental loaded line. An ac- 

 cumulation of manufacturing difi&culties delayed the completion of the 

 installation, so that the transmission observations and over-all line insula- 

 tion tests did not get under way until the spring of 1910. 



After a suitable test period it was found that although the wet weather 

 line insulation was not so high as had been expected it was sufiiciently good 

 to warrant the general commercial use of loading on 165 mil circuits. Ac- 



