233 
laboratories, may be directed. There is almost a virgin field 
for such work in the tropics, and great results may be hoped 
for. Already there are indications that such work will be of 
the very greatest importance in a modern Department of 
Agriculture. 
As the introduction of new products has become of less and 
less importance in the agriculture of a country, the need for 
detailed study of the old has become more and more clear, But 
to pursue the study of the improvement of crops, tools, methods, 
and the like is more directly agricultural than botanical, and 
requires more space than is usually available in a botanic 
garden. This has led in the more progressive countries, such 
as Java, to the opening of definite Experiment Stations as well 
as botanic gardens. The duty of the latter, so far as this great 
line of work is concerned, is to have a fine collection of a 
few specimens of every kind of useful plants that can be ac- 
climatised, to provide the necessary seed or cuttings with 
which the Experiment station can then try detailed experiments 
upon cultivation or other matters. Laboratories for the scien- 
tific study of the agricultural problems that occur are usually, 
as in Java, opened on the stations. 
To understand how well such lines of work have been fol- 
lowed in the Java-gardens, one need only look at the numerous 
scientific publications that have emanated from the institution 
in the last 25 years, more especially the well known “Annales”. 
Other gardens, such as those of Ceylon, Manilla, and elsewhere, 
have also turned out valuable work. But a study of these 
publications will soon reveal one feature of some importance. 
Many foreign students have been attracted by the reputation 
of the gardens as centres for scientific work, and especially to 
the Buitenzorg Institution. At first the work of such men was 
of great importance for the general progress of botanical science, 
but it has steadily decreased in value, while the value of that 
tured out by Dr. Trevs and by other regular residents in 
the tropics has not done so. In other words it would seem as 
if the problems that could be solved by a residence of a few 
Ann, Jard. bot. Buitenz. 2e Sér. Suppl. Ill. 
