242 M. B. WELCH. 
value for diagnostic purposes. The ‘closeness’ or ‘ open- 
ness’ of the grain, i.e., the relative size of the vessels or 
‘pores’ is probably one of the most valuable aids to classi- 
fication, but here again variation occurs, within limits, in 
the same timber, evidently due to different rates of growth 
of the tree, slow growth usually producing a closer grained 
timber and vice versa. 
On account gf the difficulties in exact determination 
as pointed out above, it would be of great value if some 
more definite method could be obtained, which while simple 
in application, would nevertheless, give a clear indication 
as to the nature of the timber. 
References are often met with in literature dealing with 
this subject, to the fact that extracts prepared from the 
timber may possess some value for diagnosis, and the object 
of this investigation was to determine whether’ such 
methods would be of value in the case of certain of our 
hardwoods. Fortunately, on account of the very large 
collection of timber specimens in the Technological Museum 
it was possible to obtain samples for comparison from 
different localities, and moreover, what is most important, 
the botanical names are based on herbarium material 
obtained from the same tree. There is thus practically no 
possibility of error in the scientific nomenclature, a point 
which cannot be overstressed in any work dealing with 
timbers. 
The method adopted was to boil 0°5 g. of shavings in 25 
c.c. of water for one minute and filter off the extract, which 
was diluted with a further 25 c.c. of water. This liquid 
was further diluted, if necessary, to give a colouration best 
suited for. comparison when additions were made of solu- 
tions of ferric chloride, caustic potash, or lime water. 
Stannous chloride, potassium bichromate, copper sulphate, 
ferrous sulphate, etc., were also tried, but the first-men- 
