255 
Groom.—Remarks on the Oecology of Coniferae. 
In different species of one genus at least some desert types show 
a decrease in the calibre of the tracheae when compared with mesophytic 
species. For instance, Volkens (’86) thus describes the wood of the small 
shrub Convolvulus lanatus'. ‘The wood-cylinder consists almost entirely of 
thick-walled libriform [fibres] and uniseriate medullary rays; and only 
after more close examination does one discover the few and narrow vessels.’ 
Apparently, in a number of other desert species, the limitation of the water¬ 
conducting channel is accomplished rather by a diminution in the number 
of the vessels ; for Volkens, describing the structure of the long, slender, 
poorly foliaged branches of Cocculus Leaeba , states that the ‘ water¬ 
conducting system consists mostly of tracheides with bordered pits ’, and 
scattered at intervals among these are ‘tracheae of unusual width’. It is 
evident that the tracheae in desert plants are not always very small, for 
Volkens describes those of Zizyphus Spma Christie which has small caducous 
leaves, as being ‘unusually wide’. But the descriptions of the wood of 
desert plants are generally too vague as regards number and size of vessels ; 
moreover, comparisons in these respects with more mesophytic species are 
required. 
Yet it is clear that, in many cases, in xerophytic desert types there is 
a tendency towards reduction in size or number of tracheae. But this 
conclusion, drawn from plants growing in the Egypto-Arabian deserts, 
seems to be at variance with results obtained by Cannon. 
Cannon (’ 05 ) has given an account of the comparative abundance and 
size of the tracheae in irrigated and non-irrigated specimens of identical 
species growing in the desert near Tucson (North America). His surprising 
results, which are illustrated by micro-photographs, must be regarded as 
absolutely reliable as regards facts. He found that on equal measured 
areas of the cross section the non-irrigated plants had more numerous and 
often wider tracheae. But in interpreting these results care must be taken, 
as Cannon used for observation stems of approximately equal thickness. 
But the irrigated stems thickened more rapidly than the non-irrigated. 
Hence the relatively smaller number of tracheae in the irrigated specimens 
may have been due to an increase in the width of the zone of growth, which 
was not balanced by a corresponding increase in the number of wide 
tracheae. For instance, a specimen of Quercus Robur growing in a poor 
situation (say on dry soil), when compared with one growing in a good 
situation, shows a larger number of wide tracheae in any measured area, 
because its annual rings are narrow and yet have a very heavy percentage 
or approximately the full number of wide tracheae. Again, as regards the 
width of the tracheae, comparison was made by Cannon between younger 
irrigated and older non-irrigated plants; but we know that often at least, 
if not always, the width of the corresponding tracheae increases gradually 
outwards up to a certain limit in the successive annual rings. Hence the 
