>49 
August I, 1890.] Supplcmmt to the '•'■Tvopical Agvicultuvisi 
from a depth of 1 1 feet. This kind of lift is not 
luiknowii in Ceylon. It is used tor raising -water 
from -wells in tlie public bathing|places of Colombo, 
and to some extent in the irrigation of tobacco and 
other crops in Jaffna, and a fe-sv other places by the 
Tamil inhabitants. 
The follo-ndng calculations are from Professional 
Papers, Vol. I. 
Water raised 16 ft. Content of bucket='45 
c. ft. IMo. of discharges per minute=3. Discharge 
per hour=81 c. ft. 
Actual discharge per houi—72'9 c. ft.=45d’4 
gallons. 
(To be continued. ) 
4 - 
ABOUT THE GROWTH OP TREES. 
Dr. William Somerville -who made a special 
study of Sylviculture in Germany, — where so 
great attention is paid to this subject, — and vdio 
was last year apfiointed lecturer on Fore.stry in 
the University of Edinburgh, has published a 
most useful paper on the growth of trees ; in 
which the results of foreign workers are noted. 
■ That the crude sap of trees ascends by the 
wood, and that the elaborated sap descends by 
the bast, he says, can be well proved experi- 
mentally. If a ring of bark (which contains 
the bast) be removed from a tree at the height 
of a few feet from the ground, and if the stem 
below be destitute of branches, no growth will 
take i)hice below tlie ring after the first year — 
the little that does take place being due to the 
consumption of reserve materials ; above the rings 
ordinary growth takes place. The upward flow 
does not support cainbiuni activity, and is 
thus practically not plant food till it is elaborated 
in the leaves. 
Death of a “ringed” tree however will occur 
before very long owing ( 1 ) to the death, of 
the exposed alburnum under the action of 
minute fungi, or (2) what usually puts an end 
to it, the death of the root hairs which are no 
longer nourished (jwing to the check given to 
the downward flow. This can be proved by ring- 
ing one of the limbs of a forked stem. The 
behaviour of this limb will be similar to that 
oi the stem, only that the tree will continue to 
grow for a much longer period, for the linil) or 
limb.s which were not ringed will send down 
nourishment to the root to keep them healthy 
so as to send up water and mineral matter. 
Under the most favoural)le condition.s of growth 
a tree produces more nutriment than is actually 
r(!(|uired to supply the want of the cambium. 
I f a cross-section of a healthy tree be examined 
with the microscope, it will 'be found that the 
cells of the medullary rays as well as certain 
cell.s lying parallel to the axis of the stem hold 
large (piantities of re.serve food, especially .starch, 
these being scarcely found near the centre of 
hirg(' trees, but abundantly so towards the outer 
))arts e.sj>ecially in the al1)urnum. Experiment 
tends to prove that the capacity to bear seed 
is regulated by tlie amount of reserve materials 
collected by the tn-e. 
The almost total ah.scuce of starch in dura- 
men, and its presence in alburnum, is a circum- 
stance connected in a very direct manner -v\ ith 
the difference in durability of these two parts 
of tree. It is well known that the outside 
■«’ood has not the same po-wer of reshsting the 
attacks of fungi as tlie wood lying nearer the 
centre. The fungi which cause decay cannot live 
without food, and the food best suited to the 
growth and development of many fungi is just 
the kind found in the storehouse of the tree. 
When the spores germinate on heartfood the 
young fungi do not find the means to sustain 
life and quickly die ; -whereas in the sap-wood 
all is present that they require, and consequently 
if moisture and heat be forthcoming, rapid in- 
crease takes jilace and the -wood is soon destroyed. 
The absence of starch from the duramen is not 
the only cause of its greater durability — the 
depositions of gums, resins, &c., being even more 
influential. 
In an annual ring of -wood the spring-wood 
is kiio-wn to be not so firm as the autumn--wood. 
The preponderance of the latter it is that re- 
gulates the specific gravity of woods, and specific 
gravity is proliably the best test which can be 
conveniently ajiplied to discover their quality. 
Two theories have been advanced to account 
for this peculiarity of this wood-ring. The older 
and more generally accepted one is known as 
De Vries and Sachs’ “ Bark-pressure Theory.” 
Rut shortly it is as follows; — In winter owing 
to the mollifying action of -vuater, to its freezing 
and thawing, and to changes of temperature 
generally, the bark is considerably loosened, so 
that in spring when activity of the cambium 
recommences, much less i)ressure is offered to 
the growth of new cells than is the case in the 
hitter parts of the season, when the bark presses 
on the cambium very tightly, owing to the wood 
having increased in volume. In the early part 
of a growing season the cambium rapidly ]u'o- 
duces wood under reduced pressure, the Av.ood 
being consequently very porous, whereas in 
autumn the wood produced is veiy dense, because 
the cambium is then working under greater 
pressure. The other and newer theory may be 
called Hartig’s “ Nutrition Theory,” according 
to which the spring-wood is not so firm as the 
autumn-wood, because, in the early part of a 
growing season, the conditions are not jiresent 
which enable a tree to assimilate rapidly, and 
therefore there is not so much plant-food avail- 
able for the formation of cell-wall materials. 
An argument in support of the “ Bark pressure 
theory ” is, that if the pressure on the cambium 
be relieved, naturally or artificially, the part of 
ring formed at the point of relief is broader 
than any other’. For example, if the outside 
rings of a section of an old thick-barked stem 
bo examined, it will be seen that they clo not 
follow a regular cour.se, but jnirsue a more 
or less undulating one. A careful examination 
will show that the points of swelling corre- 
spond (o places where deep fissuues occur in the 
bark, that is to say, places where pressure is 
reduced — while the jroints where the rings bend 
in towards the centre correspond to places where 
the bark is firm and entire. The same thing 
may be seen on a larger scale, where by any 
means a tree has been longitudinally cracked. 
If the rupture is of long standing, the circuhu- 
