THE TROPICAL AGRICULTURIST. [September * 1889. 
is aoraped ; there comes out a skein of fibres ad- 
hering still to the " heel," which has not been 
scraped. This heel is a thiok and fleshy portion. 
In order to scrape it, the following machine, 
called by the makers rattletrap), is employed, 
(fig. 2). It is the very same pulley turning in the 
same direction. Only instead of the preceding 
oatch, there is a savate S ; this is a plank 
having the form AS, movable round an axle A. 
A roller r rests against the exterior surface ; this 
roller is supported at the extremity of a lever I 
movable round the axle a which is fixed ; at the 
other extremity a rod b fastened to a pedal p 
movable round the axle d which is fixed. On 
pressing with the foot upon p, S is pressed against 
the pulley, and on withdrawing the foot iS is 
allowed to fall back, and it goes away from the 
pulley. At c is a hook fixed to a piece of the frame- 
work of the machine. And now for the working : 
the skein of fibres whioh comes out of the preceding 
machine is taken and fixed to the hook by rolling 
it round several times and letting the unscraped 
heel be taken ; at this moment the savate 
is down, the pedal is pressed upon and this raises 
the savate S. This in its progress drags the 
heel and presses it firmly against the pulley, and 
the soraping begins ; the reserve motion is put into 
aotion and the bundle of fibres is taken out 
completely scraped. These fibres, still moist and 
gummy, are put for 12 to 15 hours in hot water, then 
dried and packed. This proceeding is rough and coarse; 
the loss also is very great in this rattletrap ; 
these are the machines which were used at Eouge 
Terre, and whioh are still used in two or three places. 
The second machine which is about to be described 
is an improvement of the first in its mechanism. 
This is the machine invented by M. Marabal and 
constructed by Messrs. Rosnay & Francois, which 
has worked at Rouge Terre pretty regularly for 
nearly a year. I will not enter into any detail of 
construction, for it is rather complicated. This is 
the principle of it : T is a drum (fig. 3) of cast-iron 
backed by a good thickness of hard wood. At m is 
a chop, set into the wood ; t a table doubled at p 
by a leaf of thin copper ; P is a copper comb ; 2" 
a soraper in cast-iron with grooves absolutely 
similar to those already desoribed ; only it has a 
less diameter ; it is about two feet. The movement 
of the drums is in the direction of the arrows/; 
the drum T turns with a speed of 30 revolutions per 
minute, 2" of 200 or 300 revolutions. The following 
is the working : — The leaf is presented on t by the 
heel; the automatic comb P buries itself in the 
heel, and pushes it on at the moment when the 
chop nt presents itself on a level with the table ; 
the ohop, which is automatic, is open ; it closes 
rapidly and grips the heel, dragging with it the leaf, 
and forcing it against the drum ; the rotation con- 
tinuing, the leaf arrives at a where it is cleaned by 
the scraper which turns in a reverse direction. The 
scraping continues till the chop arrives at the 
position m indioated in the sketch. This latter 
then opens, and rejects the leaf by a rapid move- 
ment on to a leaf of sheet-iron F, where it is 
gathered up. Then the chop takes up another leaf, 
and so ti goes on. 
Here, again, it is the point which is scraped, and 
therG remains a heel. There is indeed a 
Marabal, rattletrap Dut it has been put aside, and it 
never acted satisfactorily. At Rouge Terre they 
made use of the rattletrap, described above, in 
order to crush the heel. 
It must be remarked here that in these two 
machines the leaf has a reverse movement from 
that of the scraping drum ; this is a main point 
whioh has oausod the Vigor maohine, which is now 
about to be desoribed, to be rejeoted. 
It is essentially composed (fig. 4) of a large 
drum T, of polished cast-iron, 2J feet in diameter, 
turning round an axis A ; three small cylinders 
c, c', c", of cast-iron, mounted on axts turning in 
bushes; scrapers Grand G' in cast-iron, with T irons 
are supported by a triangular balance a b d mov- 
able round an axis a. To this balance is fixed a 
rod (which is not represented in the sketch) ; 
at the end of the rod a small wheel which revolves 
on a wedge-shaped cam upon the axis A ; the balance 
thus receives an automatical and periodical oscil- 
lating motion. Lastly at S is a metal plate actiDg 
as a guide to the leaf, c, c' and A are furnished 
with a tightening screw for regulating ; c touches c' 
but does not touch T ; c' touches T; and so does 
c". The directions of the rotations are indicated 
by the dotted arrows, the passage of the leaf by 
black ones. The drum, the cylinders, and the 
scrapers have an equal length of 6 feet, and are 
contained between two vertical frames of cast-iron. 
The working is easily explained : the leaf is intro- 
duced between the two cylinders c and a' the point 
foremost ; it is squeezed, comes up between c' and 
T where it is again squeezed, but more strongly. 
Directly the point arrives at h the balance oscillates, 
the scraper G' comes in contact with T and the 
scraping begins, the leaf being retained by c\ 
When the heel is released by c' the balance oscil- 
lates in the contrary direction, and it is G which 
then comes in contact with T and which scrapes 
the heel, the leaf being then retained by c". When 
this first leaf is done with, the balance oscillates 
again, and G encounters at h the point of a second 
leaf. And so on. So the leaf comes out completely 
scraped. The drum T and the scraper G in effect 
only form a Marabal machine, the cylinder c" acting 
as chop ; G scrapes in a reverse direction to the 
passage of the leaf; G" scrapes in the same direc- 
tion as this passage. At the trial of the machine, 
G gave an excellent result,- whilst G' did not scrape 
at all ; or rather the leaf submitted to the action 
of G' bore a series of transverse scratches pro- 
duced by the grooves, the skin and the pulp were 
not raised. This caused the machine to be rejected. 
Many othermachines, among which those of Messrs. 
Digard, Carcenag, d Unienville, &c, may be cited, are 
the result of patient and laborious researches, but 
do not yet appear to realize the dream pursued : 
In all these machines the end which has generally 
been aimed at has been to scrape the leaf com- 
pletely, without having to sorape first one end and 
then to turn it round to scrape the other end. It 
is perfectly evident that in that case there would 
be a less expenditure of workmanship, but would 
it be worth the trouble ? This cannot be done 
without complicating the mechanism, which ought 
to be as simple as possible, for it needs great pre- 
cision, and as was seen above the juice of the aloe 
attacks iron. The wear and tear produced by these di- 
fferent instruments would without doubt destroy their 
good working powers. We will consider further on the 
present machine, whose value consists principally in 
its simplicity, its easy management, the nicety 
with which it is constructed, and its moderate price. 
The machine sent by M. Paul d'Unienville to the 
Exhibition of 2Sth October last is similar to the 
first one described with this modification, that a 
" rattletrap " has been adapted to it ; the catch 
and the scraper are identical, exoept the diameter 
of the latter which is less being about 1£ foot (fig. 5). 
A scraper A and a catch S, that's the old 
apparatus. At c and c' two cylinders of wood on 
smooth cast-iron, turning in the direction of the 
arrows. At a a scraper identical with A ; <S" is a 
joist or catch movable in two slides and which is 
worked by the hand by means of a lever. The leaf 
is introduoed by the point between c and c', the 
