58 P.O. KEEGAN [sULY 
which is destroyed by acids and increased by a trace of alkali. Fraxin 
is a colourless crystalline glucoside of a feebly bitter taste, and seems 
to be related to quinic acid or hydroquinone. ‘The tannin of the Ash 
is totally different from that of any of our native or denizen trees: it 
is distinctly iron-greening, is not a glucoside, does not yield anhydrides 
by the action of acids, but only by heating dry or by repeated evapora- 
tion of its solution, when brown substances (recalling the dun shade of 
the autumn leaves) are produced, and finally on potass-fusion it yields 
protocatechuic acid but no phloroglucin. In fact, it is doubtful if any 
constituent with a phloroglucin nucleus occurs in the entire organism ; 
for the quercetin found in the leaves from birth till late in August 
shows at all times reactions more like those of a tannin than of a 
mere tannoid compound. The leaves may be regarded as among the 
wonders of British botanical chemistry. Replete with chlorophyll and 
carotin, they contain much starch, fat, and resin, and from 6 to 9 per 
cent mineral matters (ash), but they are specially distinguished by the 
number and variety of decomposition products, which constitute an 
exceptionally high non-nitrogenous extract consisting of quercetin, 
tannin, inosite, mannite, glucose, gum, mucilage, malic acid and its calcium 
salt in astonishingly large quantity. On the whole, we see that the 
small and short-lived leaves of the Ash are extraordinarily active, and 
we are impressed by the apparent contradiction between the enormous 
percentage of mineral matters indicative of an intense transpiration and 
the small number (150) of stomata per square millimetre of epidermis ; 
the carbohydrates produced on assimilation are largely oxidised to 
acids, but the chlorophyllian protoplasm itself in its descent on 
exhaustion stands hesitating, so to speak, on the first round of the 
ladder, the not very oxidised tannoids. 
Much instruction and edification would doubtless be gained by a 
specific recital and description of the chemical constituents of the 
arborescent Rosaceae, e.g. the wild cherry, the rowan tree, etc., with 
their wondrous plethora of products of de-assimilation and of carbo- 
hydrate degradation; but as these are assuredly scattered and not 
forested, I now pass on to a tree which, although not a sterling native, 
has yet been frequently artificially planted in our parks and groves 
on such a plan and with such effect that the serried outskirts of a 
dense forest—vast columns upholding a dome of leaves and flecked 
with white clusters of blossoms, have at least been suggested. This 
is the beautiful Horse-Chestnut (Aesculus Hippocastanum), and truly 
there is something very satisfactory in the chemical distinguishment 
and examination of so many constituents that are comparatively simple 
and afford atomic groups more or less harmoniously proportionate. 
The well-known tannin, C,,H,,O5, for instance, has a number of atoms 
of hydrogen nearly equal to those of carbon, and exactly double those 
of oxygen ; hence its reactions come out very decisively, the deficiency 
in carbon being a great help towards the ready production of a series 
