Jan. 1, 1901.] THE TROPICAL AGRICULTURIST. 
457 
safer to consult a foreman of long experience with the 
particular class of goods concerned, than it is to refer 
to the soientifio man unless the latter, which is not 
always the case, has studied the manufacture closely in 
the works. And this seems to hold just as strongly 
with the raw rubber; at the present time this is 
bought by the practical man who employes his senses 
alone and extreme caution should be evinced by those 
who say that this position of affairs is all wrong, 
and that rubber should be bought and sold on chemical 
analysis. 
Now, as regards chemical analysis, everything 
depends upon the sampling. Unscrupulous mining 
agents can get any result they wish from a gold reef 
by omitting to work upon a really average sample for 
their assays. This sampling is a matter really of 
great complexity and occupies a considerable amount of 
time, but its importance cannot be exaggerated. At 
the Rio Tinto copper and sulphur mines in Spain, the 
sampling, in order to get a representative assay, ia 
carried out in a very comprehensive manner, the 
company priding themselves on the effectiveness of the 
system which they have worked out. With rubber, 
the case is very similar unless a really average sample 
were taken from bulk, the results of the chemicil 
analysis, say the deteraiination of dirt, water and 
resin, might easily be very misleading. No doubt, 
with proper methods, an average sample could be 
obtained, but at what cost of time and labour ? Rubber 
is much more difficult to work than is the case with 
the general run of minerals, and where time enters 
largely into the question, the difficulties, in my opinion 
at least, in the way of getting an accurate valuation 
of a few tons of African ball rubber would be found 
practically insurmountable. In some cases where the 
quality is pretty uniform, the chemical method might 
prove of utility, but we have to remember that unifor- 
mity is the exception rather than the rule, so we must 
base our calculations accordingly. It certainly seems 
desirable that some agreement could be come to bet- 
ween dealers and buyers as to the amount of water in 
raw rubber ; at present, where the amount of water 
Taries to such an extent, it is rather like buying a 
pig in a poke. The subject has before uow come up 
for discussion in the correspondence columns of this 
journal, but I do not remember that anything definite 
was come to. Not so long ago I remember reading 
of a case in London, in which a rubber firm made a 
claim for loss of weight, the difference between the 
weight invoiced and that actually delivered being 
accounted for by tha water which had percolated the 
cask. 1 forget how the claim resulted, but the point 
is an isaportant one. If some arrangement could be 
come to with the exporters to send the rubber on board 
ship with a maximum of, say 10 per cent, of water, 
there would be an advantage in freight charges, and 
it would certainly prove an easier matter for the 
purchaser in England to arrive at the value of the 
ihipment. As the case stands at present, the com- 
moner qualities of rubber are bought in a rather 
speculative way, and it becomes of importance for 
the management of the works to estimate their value 
for the several purposes to which it is intended to put 
them. Now, here con«i4erable progress has been made 
compared with former years, tbou-h it must bo con- 
fessed that the very elaborate methods of testing 
which have been worked out at soma of the German 
works have not commended themselves to the trade 
as a whole, the suspicion having gained ground that 
unless you have a scientific staff employed at a low 
rate of remuneration the time and trouble involved 
in carrying out the tests are not justified on the score 
of expediency. It may not be without interest to give 
a few details of the testing system, as carried ont, 
or said to be carried out regularly at one of tha 
largest German rubber works. The apparatus em- 
ployed is very similar to that used in testing tex- 
tiles or metals ; that is, a weighing machine, to 
which is attached an iron rod with a grip, and at 
9ih9i; ejid fv« irva rpd with a ,grip, worked by 
a cog-wheel, as in the ordinary lifting jacks. This 
explanation may not be particularly clear, but it is 
difficult to describe machinery in a few words or 
without the advantatious aid of diagrams. Fine 
Para is taken as the standard, and its stretching 
power and resistance to rupture ascertained, and a 
series of comparative tests are made with other 
rubbers, in which way it can be ascertained whether 
a particular rubber tested is not too dear, compared 
with Para reduced to the same standard. For instance, 
when the rubber tested is found to be 35 per cent, 
under the quality of Fine Para, while its price is 
only 25 per cent, under, an indication of moment 
is obtained. For mill purposes, a table is made 
showing the stretching and resistance to rupture of 
the varisus compounds and is referred to on occa- 
sions when it is proposed to alter the composition 
of the mixings. In this way it is found possible to 
avoid any variation in the quality of the goods, 
moreover, these tests, it is said, l,h .ve proved useful 
in the case of claims brought against the firm for 
supposed defective work. This brief account of 
what is done in some quarters in Germany will 
suffice for the present occasion ; it is a good illus- 
tration of the scientific bent of the German mind, 
and in this case, there is every reason to suppose 
that advantage accrues from its adoption. The 
German passion for detail, however, is not always 
justified by rrsults, and tho Englishman is apt to 
treat its excesses with but scant respect. I am open 
to correction in the matter, but I do not think that 
the German method of testing rubber finds its coun- 
terpart in England, though I am not at all inclined 
to say that it could not be introduced into our works 
with advantage. Some years ago I adopted a com- 
parative method of testing the value of washed raw 
rubbers, depending on the melting points. Naturally 
the rubber which contains the most resinous consti- 
tuents melts the soonest, and the relation between 
tensile strength and melting point is a very close 
one. The table which I made naturally commenced 
with Para and ended with African Flake, the other 
brands of rubber taking up their positions at inter- 
mediate stages. In this way, by strict attention to 
details of time, best results of value can be 
obtained, showing not only the difference which exists 
between the various brands of rubber, but also to 
what all extent variations occur in the tensile strength 
of rubber which is always sold under one name, 
and which is commonly supposed to be of uniform 
quality. There is no doubt that failure to recognise 
the differences existing in the quality of rubber is a 
fruitful source of trouble, because it leads to incorrect 
vulcanisation. Especially is this likely to be th« case 
where the eoods are submitted to mechanical tests of 
great stringency, as, for instance, in the case of buffers 
for certain of our railway companies, and it is a fain 
presumption that trouble which has occured in tha 
past might have been obviated if the rubber, previous 
to mixing, had been put through some sort of test, 
such as indicated above, to enable an opinion to be 
arrived at with regard to its axact quality. In the 
absence of such testing, there must always be an 
element of chance about the results of vulcanised 
mechanical rubbers as regards the important parti* 
cular of tensile strength, and when trouble arises it 
is rather too much the tendency to put the blama 
on some workman or other, whereas it is more pro- 
bable that the real cause lies in the failure of the 
workman's superior to recognise that the time of 
vulcanisation, or the percentage of sulphur might; 
have been altered with advantage. However it may 
be said that here I am somewhat transcending tha 
limits of my subject, though it is always permissible 
when advocating a change of policy or procedure to 
state clearly in what direction benefits may be ex« 
peoted to accrue from the innovation. The subject 
as referred to above, is clearly seen to come under 
two heads, which have little in common, viz ; tha 
testing of xaw rubjjer Ijefgre purchase, au4 th^ lestj 
