9 BULLETIN 145, U. S. DEPARTMENT OF AGRICULTURE. 
the deficiencies of present practice in wood preservation and pave 
the way for increasing its efficiency by suggesting lines for ge 
research. 
From 40 to 90 per cent, or an average of over 60 per cent of the 
total cost of treating wood, is chargeable to the preservative alone... 
Jn ordinary treatments with coal-tar creosote it is common practice 
to inject 5 to 10 pounds of the oil per cubic foot of wood, although 
toxicity tests indicate that about one-half of a pound will prevent fun-_ 
gous growth; in other words, exclusive of subsequent changes, from 10 
to 20 times as much creosote is used per volume of wood as is theoretic- 
ally required. The possibility of safely reducing this amount, and | 
consequently the cost of treatment, is one of the problems referred | 
to. The tests described herem should therefore be considered simply | 
as preliminary to the more difficult problems involved. 
PROPERTIES INVESTIGATED. 
The practical value of a preservative depends very largely upon 
the conditions under which it is used, and investigations to deter- 
mine its value must necessarily be broad. The following points 
were considered in the tests: | 
(1) The important chemical and physical properties of the pre- 
servative. 
(2) The effect of the preservative on the strength of the wood 
treated with it. 
(3) The ability of the preservative to penetrate and diffuse through 
wood. 
| (4) The permanency of the preservative after its injection into 
| wood. This involves a study of its volatility and leachability. 
| (5) The combustibility of wood treated with the preservatives. 
(6) The toxic efficiency of the preservative in preventing the 
growth of wood-destroying fungi. 
(7) The corrosive action of the preservative on steel. 
(8) The effect of the preservative on paint applied to the wood 
subsequent to treatment. 
No systematic tests were made on the effect of the preservative 
as an electrolyte or in contaminating drinking water, nor any tests 
which relate to a special or limited use. 
METHODS OF TEST. 
The various tests were conducted as follows: 
Thoroughly air-seasoned eastern hemlock (Tsuga canadensis L.) | 
was selected as the wood best suited for the experiment because of — 
its low inherent resistance to attack by fungi, the comparative uni- 
formity with which it can be treated, and its availability. Only 
perfectly clear, straight-grained, and uniform material, free from all 
