TIMBER: ITS STRENGTH, AND HOW TO TEST IT. 13 
So far I have been considering merely compression applied 
endwise and with the grain, that is, compression upon the column 
of wood as it would ordinarily be used when built into a structure, 
but it frequently enough happens in actual practice that a piece 
of vertical timber stands or rests upon a horizontal piece, and 
transmits a crushing load to this horizontal piece, and the 
horizontal piece is, therefore, subjected to a crushing across the 
grain. Only comparatively few experiments have been made to 
determine what is the strength of wood under such conditions. 
Tredgold many years ago found in his experiments that rooo 
Ibs. per square inch was sufficient, when applied in this way, to 
distinctly indent Memel fir, while about 1400 lbs. per square 
inch was required in the case of good English oak. 
Of course it is very much a matter of judgment as to what 
constitutes the limiting crushing load in such a case, that is, what 
amount of indentation shall be considered to be the limit which 
can be allowed. Some authorities give ,,th inch in depth of 
indentation. Professor Johnson took 3 per cent. compression as 
a working limit, which would amount to an indentation of 2 inch 
in a beam 1 foot thick, and 15 per cent. as a destructive indenta- 
tion, which would amount to about 1? inches in a 12-inch bar. 
He experimented with specimens 2 inches and 4 inches thick, 
and he found in the case of pines or soft woods that the com- 
pressive strength, when dealing with this cross grain compression, 
was only about one-fifth of that which could be developed when 
the compression was applied along the grain, and in the case 
of American oak it was about one-third. 
I made a couple of tests for this lecture; for the white (Wey- 
mouth) pine I found a compressive strength across the grain of 
only 0°79 ton, as compared with 2°31 tons when tested in direct 
compression along the grain, that is, the strength was about 
one-third, and this, too, when the direct compression strut was 
a fairly long one. 
I have said enough to indicate the importance of considering 
this matter of strength across the grain when designing structures 
in which timber is used in this way. 
Cross-BENDING TESTS. 
Tests of timber beams are a very favourite method of determin- 
ing the mechanical properties of any specimen of timber, partly 
because large pieces can be used, and the loads are not excessive, 
