592 
THE TKOPICAL x\GRICULTUKlST. 
[March 1, 1905. 
"Manual of Inclian TimbRts," the wood specimens 
collected by Dr. Wallich were all examined and men- 
tioned, as wero the more reopnt collections o£ Sir D. 
Brandis, Mr. S. Kurz, Col. Ford, and others. In my 
new edition about to be issued 1 hive included also 
the collections of Sir J. D. Hooker, Mr. H. N. Eidley 
(Malay) and other*, deposited at Kew ; and one of my 
chief regrets is that the collection of the Biothers 
Schlagintweit could not also be included, because the 
names have been lost. .What has been don» in India, 
of course, I know more intimately, but I alio know 
that Colonial British explorers have not deserved to 
be accused of neglect, and that the work done by the 
French and the iJutoh has been as good as, indeed if 
anything, better than our own. 
I think'that in his remarks about Kew Dr. A. Henry 
was not quite just. I have been recently working in 
the Kew Museum and have been struck with the 
care taken to incorporate in it the woods sent by 
various correspondents. As I was not specially in- 
terestvid in Chinese woods, I can say nothing about 
Dr. Henry's collections, but they are doubtless there, 
ready for him or another to work up. The authori- 
ties at Kew were most kind in their help to me and 
were delighted that I should work in the museum ; 
but it is abundantly clear to anybody who inquires 
into the matter that the present miserably inadequate 
trnd ill-piiid Kew staff conld not be expected to do 
wood-deacriptions for explorers in addition to their 
own work. Tha Kew collections are national, and the 
herbarium and museum are maintained by the State, 
not only for study by the staff of the establishment, 
but to be conveniently at the disposal of workers in 
general. There are, unfortunately, not ma y specialists 
in the study of woods, and it is very good news to 
those interested in the subject that Mr. Stone is going 
to devote his scientific life to it. His 'paper was very 
interesting, and I, for one, shall eagerly look forward 
to the results of his further work. 
December 9th, 1901. J. S, Gamble. 
—Journal of the Socittij of Arts, 
THE EFFECT OF FORESTS ON THE 
CIRCULATION OF WATER AT THE 
SURFACE OF CONTINENT. 
(Continued from page 513. ) 
On level spots in the mountains, whether forest' 
grass or bare, there is a decidedly greater amount 
of moisture in the soil than in the plains under 
similar circumstances. There is more rainfall and 
the snow lies longer. The amounts evaporated 
directly, or utilised by plants, are also less. It is 
well-known that high plateaux, are often swampy 
or peaty. But level spots are the exception, moun- 
tains consisting principally of more or less steep 
slopes where surface flow is an important factor. 
In fact, the surface AjW is the great characteristic 
of unlevel countries and constitutes the great difference 
between these and the plains. There are numerous 
examples continually coming to light, showing the 
excellent effect produced by forests on the volume, 
the regalarity, and the maintenance of springs. 
This is perhaps the place to mention a certain nala 
in Lachiwala forest, coming down from Nag.sidh 
hill. Up to 1896 this nala carried running water in 
November-December. In later years it has always 
been dry by the end of October. The dryness is 
perhaps not premanent for the future, but it may 
be due to fellings that have been made on the 
slopes of Nagsidh. Nobody now denies the benefi- 
cent action of forests from a quadruple point of view, 
viz. the increaeo of rainfall, the protection of the 
Boii' from erosion, the more regular flow and diminu- 
tion floods, aud the maintenance and steady flow 
of flpringa. 
H c i.? it possible to explain the apparent con 
tradiction between this last point and the resnlt of 
the previously mentioned experiments in plains forc-sts, 
where it was shown that the level of subsoil waters 
was somewhat lowered ? Is it possible that the forest 
acts in opposite ways in different localities ? Before 
answering this question certain well known facts may 
be considered. Sappose a slope at 45,° wooded on 
its left half and bare or grassy on the right half. In 
winter both halves are covered with snow, more 
thickly and evenly so in the forests since there are 
no avalanches and the wind cannot sweep up or 
evaporate so much. The springs bring a rapid thaw. 
On the bare slope where nothing hinders the access 
of warm air the melting will proceed quickly and 
the greater part of the resulting water will disappear 
at once by the streams. It is well-known that in bare 
or grass mouuiains the melting of the winter snows 
causes great and sudden floods like the disas- 
ters of 1856. The quantity of water soaking into 
the soil depends, among other thing?, on the length 
of time during which the water remains in contact 
with the soil. This time itself depends partly on 
the steepness of the slope. A very steep slope re- 
moves the water quickly, so that there is little time 
left for absorption. Hence steep slopes di y np very 
quickly after rain, more especially if they arc not 
wooded. On the wooded half of the above supposed 
slope, the mel'ing will be quite slow, taking a fort- 
night or a mouth longer. Thus, even without any 
low vegetation, the surface flow will be much slower 
and more prolonged, with a corresponding decrease 
in flooded streams. But the forest soil always poss- 
esses a covering of dead leaves and humus acting 
like a sponge, able to absorb and hold as much as 
two or three times its own weight of water which 
it only parts with drop by drop. The soil below is 
protected from evaporation and sucks it all into the 
great benefit of subsoil supplies. The effect is still 
farther increased by the then dormant condition of 
vegetation, which has no need to absorb any water 
at that time. The same thin^ happens during tha 
heavy summer rains, except that the trees this time 
appropriate a certain share for their own use. If 
there is actually any water flowing along the sur- 
face it is much impeded by the network of roots, 
stalks, and obstacles of all kinds, so that a given 
flowing drop has every chance of finding a spot able 
to absorb it, or a small crevice by which it may 
get underground. The final result is a great diminu- 
tion or complete stoppage of surface flow. It is true 
that during excessive rains or melting of snows, the 
forests cannot always prevent floods, but in such 
cases the rise is much less sudden and severe than 
it might have been, and is due more to enlarged springs 
and less to erosive surface action. 
According to M, Imbeaux, State Engineer to tha 
important Municipality of Nancy, the fraction re- 
presenting the surface flow due to very heavy raina 
may be calculated from the rise in the streams effected. 
He found that in the exceptional floods of the 
durance on the 27th and 28th October, 1882, the 
26th, 27th October 1886, and the 8th, 11th Novem- 
ber 1886, the fraction at Mirabean was -33, "39 and 
•42, or over a third of the total fall. For less 
violent rains the fraction were -27, '22 and '18, in 
accordance with the general law according to 
which the surface flow follows the intensity of 
the rain. At the Junction with the Ehone, the 
figures were very similar but smaller. For the Dan 
ube at Vienna, M. Landa found 421 per cent, from 
28th July to 14th .August 1897. It is clear that for 
a given rainfall the surface flow will be greater as 
the soil itself becomes steeper and more impermeable 
or rocky, but the covering of the soil has a powerful in- 
fluence in all cases. 
On this point M. Imbeaux says : — " The re-forest- 
ing and returfing of bare mountains at high altitudes 
are, and have in France been long known to be 
among the most effectual methods possible for 
checking excessive surface flow and its consequen 
