January i, 1890.] THE TROPICAL AGRICULTURIST. 
475' 
1 
marks on the sterlity of the Blood Oranges bearing 
the Navel marks that he considers this sterility due 
in some way to the direct influence of pollen having 
occurred ? From what he says further on in his article 
on the character of the seeds of the Navel Bahia 
Orange when influenced by the pollen of the St. 
Michael's variety, one would infer that seedlessness 
does not always occur in fruits which show the direct 
influence of pollen. It would be very interesting to 
know how far it is usual to find sterility accompany- 
ing the direct influence of pollen on the fruit. In 
the account given by Maximowicz and others of this 
phenomenon, I do not remember any mention of 
sterility appearing in the fruits described by them. 
It is possible that those fruits whioh, through culti- 
vation or other causes, have become more or less 
sterile, such as Gourds, Cucumbers, &c, may be those 
which are most easily influenced directly by pollen; 
for, according to Mr. Phelps, the Imperial Blood or 
Maltese Blood usually only produce one or two seeds. 
It would also be very valuable to know whether Mr. 
Phelps or other observers have ever noticed any after- 
effect on the mother after the direct influence of pollen 
on the fruit has occured; i.e. whether a tree which 
bas once had its fruits directly influenced by pollen 
has ever been known to produce more anomalous fruits 
without again receiving the pollen of the variety or 
species which affected it before. Mr. B'lrbidge, in 
his book on Cultivated Plants, quotes mi instance of 
a white Calceolaria plant producing re I flowers on a 
branch on which some flowers had been crossed with 
pollen from a red variety ; and these red flowers 
were not merely those which had been crossed with 
the foreign pollen. It is p ssible that Mr. Phelps or 
others who have studied this interesting subject, 
might be able to add some more instances to the few 
already known of this further influence of pollen on 
the mother plant.— A. B.— Gardeners' Chronicle. 
♦ 
THE SO-CALLED MTJSSiENDA COFFEE OF 
BEUNION. * 
[Communication jrom the. Research Laboratory of the 
Pharmaceutical Society.] 
BY PEOFESSO W. E. DUNSTAN. 
A short time ago, Mr. Thiselton Dyer, the Director 
of Kew Gardens, suggested to me the desirability 
of making a chemical examination of a supposed sub- 
stitute for coffee, concerning which a good deal has 
of late beeu heard in France. It has been stated, 
on the authority of M. Lapeyrere, a pharmacist re- 
siding in the island of Keuniou, that the seeds of a 
plant abundant in the island which he calls Mus- 
samda boronica, known to the natives as " wild orange," 
closely resemble coffee berries, not only in appearance, 
but also in their chemical constituents, so that they 
may be regarded as a formidable rival to coffee. 
This announcement appears to have caused some 
scare in the coffee trade, and it is said that many 
inquiries for the seeds shave been made. M. La- 
pcyn re's paper on the subject is printed in the Bul- 
letin Bimensuel de la Societe Nationale d' Acclimatation 
(}, France for 1888, pp. 285-300. It contains an ac- 
count of the botanical characters of the plant, which 
are stated to differ from those of the known re- 
preseutatives of Mussmnda, and the plant is therefore 
adjudged by M. Lapeyrere to be a new species. The 
results of a chemical analysis of the unroasted seeds 
recorded, from which it appears that they contain 
from 0 3 to 0 -5 per cent, of caffeine. 
An examination of the seeds at Kew revealed the 
fart ilia I the plant in question is really a member 
of thi! natural order Loyamiacea, named Gcertnera 
Vaginata, and not a new species of Mussosnda, which 
belongs in tl": Rubiacece. In a despatch to the Foreign 
Office, Mr. vSt. John, the British Oorsul in Reunion, 
corrects some misstatements regarding the plant which 
have been inadi: i.. connection with its proposed em- 
~*~Read before the Pharmaceutical Society of Great 
Britain, at an Bveniug Meeting in London, Wednes- 
day, November 13. 
ployment as a substitute for coffee. Mr. St. John 
points out that the shrub is not plentiful in the islandt 
that it yields fewer berries than the coffee shrub 
the fruit growing only at the end of the branches 
He further asserts that the Gmrtnera seeds are less 
fragrant than coffee, and would probably be more 
expensive. 
These facts do not, however, affect the interest and 
importance attaching to the discovery of a new 
source of caffeine, and a possible substitute for coffee. 
M. Lapeyrere's statement that the seeds contain caf- 
feine is not so remarkable when made in reference 
to a plant belonging to the natural order Rubiacea. 
In view, however, of the fact brought to light at Kew 
that the plant belongs to an entirely different natural 
order, Loyaniaeece, the statement becomes not only 
remarkable but improbable. A special search was 
therefore made, not only for caffeine but also for 
theobromine and other compounds belonging to the 
same family. In the first instance 150 grains of the 
very hard seeds were powdered, mixed with half 
their weight of magnesia, and made into a paste 
with water. This paste was dried and reduced to 
powder. Having been packed in a continuous extrac- 
tion apparatus the powder, which would not contain 
caffeine in the free state, was successively exhausted 
with boiling chloroform, boiling alcohol, and finally, 
with boiling water. The residues left by the evapor- 
ation of these solvents were examined for caffeine. 
The residue left by the evaporation of the chloro- 
form, consisting largely of fat and resin, was extracted 
with dilute hydrochloric acid, and the acid liquid 
was tested with a solution of iodine and with Mayer's 
solution, in both cases with a negative result. Another 
part of the same liquid was oxidized with nitrio 
acid, and the residue mixed first with a little am- 
monia and then with potash, but neither a red nor 
a violet coloration was observed, so that caffeine and 
theobromine were certainly absent. The alcoholic and 
aqueous residues were similarly dealt with, but here, 
again, no trace of caffeine or of any similar sub- 
stance could be detected. Since a plant belonging to 
the Loganiacea? might be expected to contain strych- 
nine or brucine, each residue was tested for these 
alkaloids, but neither was found. In fact, at no 
stage of the inquiry was there found any evidence 
of the existence of an alkaloid in these seeds, unless, 
indeed, there is included the doubtful indications, 
afforded by the alcoholic residue of the presence of 
choline or of a substance resembling it. The seeds 
contain, besides much proteid and a small quantity 
of sugar, a fat which seems to be similar in some 
respects to the fat contained in Strychnos Nuxvomica, 
but owing to our supply of the Gcetnera seeds having 
failed, it could not be examined minutely. It must 
be remarked, in conclusion, that the seeds bear 
only a slight superficial resemblance to coffee berries. 
They are flatter and are further distinguished by 
carrying on both surfaces a series of deeply marked 
radiating lines. 
It had been intended to publish the facts in re- 
ference to the botanical nature of " Musstfnda coffee," 
in the first instance in the November number of the 
Bulletin of information which is issued by the Kew 
Department. The notice was, however, crowded out 
of this publication at the last moment, and under 
these circumstances Mr. Thiselton Dyer generously 
placed at my disposal, for the purpose of this com- 
munication, the information which will now subse- 
quently appear in the December number of the Kew 
Bulletin. 
The paper gave rise to the following discussion : — 
The Pkfsident said this paper was rather unsatis- 
factory for the purpose of discussion, the results be- 
ins purely negative, but such results involved a con- 
siderable amount of labour and skill, and if anyone 
would like to ask for further 'information Prof. Dun- 
stan would be happy to afford it. D was perhaps 
one of the advantaaes of a Research Laboratory that 
substances of this kind, which were stated to have 
properties of great commercial value, could be quickly 
examined and the results promptly published- 
