STRENGTH AND TESTING OF TIMBEE 315 
timber, Lanza's tests on thirty-seven beams gave an 
average modulus of rupture of 4,451 lbs. per square inch; 
the minimum was 2,456, and he recommends 3,000 lbs. 
per square inch as the figure to be used. 
The results of tests made on superimposed beams by 
Major Moritz Bock on the Continent and by Mr. H. D. 
Smith in Liverpool are somewhat different to those which 
would be anticipated and generally assumed by designers. 
Li both experiments it was found that the bolting together 
of the beams gave no appreciable increase of strength over 
beams simply laid one upon another ; in one of those tested 
by Mr. Smith "the bolts bent at each joint of the baulks and 
pressed sideways into the timber in some cases to the extent 
of f inch," showing that the beams slid upon one another. 
The strongest of Major Bock's combined beams had only 
50 per cent, the strength of a solid beam of the same size. 
In the case of superimposed beams secured with oak keys 
and wedges at the joinings, as well as being bolted with 
inclined bolts, the strength was found to be 60 per cent, of 
that of the solid beam.^ 
In cases, therefore, where it is found necessary, because 
of heavy loads, to secure a greater depth of girder by having 
two or three beams superimposed, although it is advisable 
to have them secured with a few bolts, to ensure their not 
being shifted or knocked out of position sideways, the 
additional strength of the cumulative beams should only be 
assumed as about 50 per cent, extra per beam. Beams 
composed of three baulks are proportionally stronger than 
those composed of four, and those composed of two baulks 
are proportionally stronger than those composed of three ; 
but under the same unit stresses three beams are much 
stifle r than two, and, though in a less degree, two are 
' Mill, of Pioc. lust. C.E., Vols. CVI. and CXXVIII. 
