84 
because unless the data is absolutely correct, in records of experi- 
ment, the whole is useless. I may here remark Jffiat as the 
seed-pods burst, the seeds are scattered broadcast, and there would 
appear to be some difficulty, in fact, an apparent impossibility, 
in selecting the particular seeds of any one tree from the host of 
others lying intermixed on the ground. With a little care and 
close observation as to shape, colour and size, the task becomes 
simple. Further, it may be pointed out that no two trees have 
exactly the same seed formation. There is always some little 
* peculiarity in the seeds of each tree, and the original pods which 
contained the seeds vary. Should doubt still exist pull a pod 
from the tree in question, open it and place a supposed seed in one 
of the sections. The observer will at once see whether it fits or 
otherwise, the outline of all seeds from the same tree being more or 
less alike. Again, one known seed of any one tree is always kept for 
comparison and from quite a large collection of mixed seeds we found 
no difficulty in separating those known from the unknown. 
Having now detailed the first course of procedure, the next 
operation is to plant out the selected seeds as quickly as possible 
after collection, otherwise they soon lose their vitality, in a specially 
prepared nursery, each lot of seeds being kept entirely by itself : 
thus providing a safe-guard against any chance of the lots being 
inadvertently mixed when the seedlings are pulled for planting out. 
While the seeds were germinating, an area of 15 acres of virgin 
jungle, slightly undulating and naturally drained, had been cleared, 
holed and made ready for planting out the young seedlings in the 
month of November. That mouth is selected because it comes at a 
time of yeai 1 when the wet-season is with us, and every other day is 
dull or cloudy : thus ensuring a much better chance of successful 
life to the young plants. 
Each respective lot of seedlings was carefully drawn from the 
nursery and planted out in the prepared holes in the formation of 
squares, fifty to a hundred seedlings in each square, 20' x 20' distance 
apart. Every square is numbered, aud on the original plan-sheet 
all particulars of the parent tree are recorded ; also where situate on 
the estate, so that when the young seedlings reach the age of 
maturity comparison can be made. 
Up to date, we can observe a great similarity to the parent tree 
in so far as growth and general appearance goes, and we have every 
reason to feel justified in thinking that the experiment well warrants 
the time and trouble taken. 
It is not our intention to tap those trees at an age when girth 
might warrant it, as is usually done in the ordinary course. They 
will be given every justice, with the sole idea of keeping them 
purely for seed-selection in the future, should the experiment bear 
out our anticipations. 
