616 BELL SYSTEM TECHNICAL JOURNAL 



cable splicing were well founded and that it is scarcely practicable to 

 make a satisfactory wiped joint with a lead-tin solder containing less 

 than 38 per cent or more than 42 per cent of tin. New solders may 

 well grow out of our study of why and how the old-fashioned solder 

 works. 



The interests of the telephone system extend of course to many other 

 metallic materials, notably iron and steel, brasses and bronzes, die-cast- 

 ing alloys, etc. For the most part, however, progress in these fields 

 has been along the lines of that of other industries. Part of our metal- 

 lurgic shops are largely devoted to the melting, casting, and fabrication 

 of a great variety of alloys into wire rods or sheets of specified dimen- 

 sions for experimental trials in electrical apparatus design. 



Wood and Its Preservation 



It is a far cry from metals to wood, and particularly to wood preser- 

 vation, which is one of our important chemical interests. The tele- 

 phone pole is our most urgent concern. Large numbers of poles of 

 cedar, chestnut, and southern pine are in use and the greater part have 

 been subjected to a preservative treatment. Pine poles possess a layer 

 of 2 or 3 inches of sapwood which is subject to impregnation under 

 pressure, and such a treatment with creosote has long been the standard 

 practice in the system. 



The problems in this field are innumerable. We will mention only 

 a few. Given a train load of telephone poles, how does one determine 

 the average quantity of creosote they contain and the uniformity of 

 distribution from pole to pole and in various parts of a given pole? No 

 two trees grow alike. Soil, climate, sun exposure, accidents in past 

 histories such as fires in forests — all make for peculiarities of growth in 

 each individual tree. These peculiarities reflect themselves in the 

 absorption of creosote, so that it is entirely possible for two poles treated 

 simultaneously in the same cylinder to differ by a factor of 5 or even 10 

 in the over-all creosote content per unit volume. How can one obtain 

 a sample which will be representative of a large group of such poles? 



The sampling problem was approximately solved by taking a suffi- 

 cient number of cylindrical solid cores with an increment borer and 

 splitting these borings diagonally along a length which represents the 

 approximate radius of the pole. Mathematically such a tapered cylin- 

 der approaches a wedge such as would represent a true sample of the 

 pole's cross section. The borer holes are plugged with a creosoted peg 

 and the poles are still fit for use, so one can sample as many poles as he 

 likes. 



This method is gradually being applied to a study of the content and 



