DURING recent years there has been an increasing tendency to examine in closer detail the methods applied to the design of structures, and to question, to some extent at least, the simple methods used in the past. Much experimental work has been carried out
with the object of comparing the calculated stresses used in the design with the actual
stresses to which the structure is finally subjected. Such measurements have indicated
some considerable discrepancies, in spite of which, however, structures built under the old rules have withstood successfully the loads applied to them, the structures having been endowed with a sufficiency of strength. In the following paper an attempt is made to consider the extent to which refinements in the determination of design stresses can confer an economic benefit, in the use of the material, and to indicate methods by which refinements in calculation can be made.