point maximum load. en-.. determined from the curve, the constants 

 given for each timber being reduced to the square inch basis of 

 comparison. 



3. The weight per unit volume of timber was calculated from 

 measurements and weighings of the whole test piece (some 80 cubic 

 inches in volume) made at the time of test, and reducing the weight 

 per unit volume so obtained, in the ratio of the weight of a small 

 piece (2J" long cut at a standard distance of 6" to 8f" from one end 

 of each specimen) weighed immediately after being cut to the weight 

 of same piece after having been oven dried. Thus the weight 

 per cubic foot tabulated is the weight in Ibs. of a specimen of the 

 wood oven dry which was one cubic foot in volume at the time of test. 

 Owing to the shrinkage which takes place in wood when it is dried 

 the figure so obtained is not the true weight per unit volume 

 of a piece of oven dry wood. This is, however, the method generally 

 adopted of determining the weight of timber. 



The specific gravity can be obtained by dividing each result by 

 <>2j (approx.). The air dry weight of an}' of the timbers can be 

 obtained by adding 12 per cent, to 15 per cent, to the oven dry weight. 



4. The results of the tests have been grouped together in 

 batches, each batch bpin' composed of several specimens obtained 

 from one tree and the average, maximum and minimum figures 

 (excluding results obtained from damaged specimens) for oven-dry 

 weight elastic limit, modulus of elasticity and modulus of rupture 

 have been tabulated. By comparing the average with the maximum 

 and minimum figures given an idea of the reliability or otherwise of 

 the limber can be obtained. The strength of an}' wood (the modulus 

 of rupture) appeal's to bear a distinct relation to its weight. This 

 relationship is fairly well expressed by the linear equation: 



R = .1.5-1- w 



" R " being equal to the modulus of rupture in tons per square inch 

 and " w" being the oven dry weight of one cubic foot of the timber. 

 An equation which is more cumbersome to use, but which appears to 

 express the relationship rather more accurately is of the form 

 n 



R=a w 



" a " and " n " hei'iiii 1 constants. Too few tests have as yet been made to 

 definitely determine the best values for "a" and "n" but it 

 appears that the equation 

 1.25 



" 1U.S 

 salaried the plotted values fairly well. 



5. Several notes are appended regarding the general behaviour 

 of the timbers tested and it is hoped that these may be of value. 

 The note.- arc not intended to In- exhaustive but are results of 

 observations marie on the specimens when under test. It should be 

 noted how in general the air-dried specimens are stronger than those 

 which contain more moisture. 



