The burning of wood may be conveniently tested by taking 

 a splinter cleft from a block, and lighting it like a match. If 

 it burns while held upright, it burns well : if it goes out, and 

 requires repeatedly relighting, it burns badly. The residue may 

 be, as is usual, a glowing cinder which consumes away to the 

 ash, or may remain, as in the case of the Sweet Chestnut and 

 the Jarrah, a black carbonized stick. Crackling, spitting, or 

 even slight explosions (Spruce) may occur, caused by the heated 

 secretions in the wood forcing the fibres apart. The aroma 

 while burning, as already mentioned, should be noticed, because 

 in cases like the Dalbergias (True Rosewoods) the tarry smell 

 is practically a generic character. Certain juices, melted gums 

 or resins, come bubbling and spitting out of the pores, and their 

 colour furnishes evidence of some value. 



The hardness of timber is unfortunately just as much dependent 

 upon our impressions as are taste and smell, but its commercial 

 importance is much greater, hence many attempts have 'been 

 made to express a scale of hardness in words. 



Nordlinger expressed it in figures corresponding with the 

 weight of sawdust removed by a given number of strokes of a 

 saw ; another observer employed a rasp ; a third turned balls 

 of the wood to be tested, and measured the distance they re- 

 bounded when dropped from a given height ; another (Hough, 

 I believe) dropped a pointed weight upon the wood and measured 

 the depth of indentation. The methods are all useless because 

 Nordlinger's saw (to say nothing of his biceps) is needed to 

 produce the same result, and the saw must always be equally 

 keen ; and the like with the rasp. Again, the turned ball 

 could never be reproduced exactly, even in the same wood, and 

 in most woods a ball would soon become distorted by warping. 

 By Hough's method the bottom of the depression made by 

 the impact of the point would rise up in some woods from 

 their elasticity, while in others it would remain deeply im- 

 pressed. My own method, though more complicated, pre- 

 supposes nothing that cannot be reproduced by others, but un- 

 fortunately it requires a machine of considerable complexity, 

 which, as I have had no leisure so far to make exhaustive re- 

 searches in this direction, is merely described in the Appendix. 

 After all, it is not hardness alone that is measured, but more 

 accurately the resistance to impact, spoken of by Hough, or, 

 in other words, the amount of force which wood will absorb 

 when struck. We are, therefore, thrown back upon vague 

 terms, such as " hard," " very hard," " moderately hard," etc., 

 etc., coupled with the names of a few well-known woods for 

 comparison, so that they are not quite empty words. Gam- 

 ble's scale of hardness, expressed in this fashion, is good, but 



xxxvii 



