DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagođeno pretraživanje po punom tekstu



ŠUMARSKI LIST 1-2/1966 str. 11     <-- 11 -->        PDF

Such investigations are becoming more and more topical, and their success
naturally depends on the extent of development of forest tree improvement
in individual countries. Thus, for instance, in the report of the School of
Forestry, North Carolina State College, for the year 1963, it is mentioned that
in a few years it will be possible to determine a selection index by means of
which it will be possible to determine much more accurately the value of a
selected superior tree to serve as a parent for the seed orchard. More has been
done in this direction with coniferous than with breadleaved species. The
tropical species have had the least attention.


At present we do not have at our disposal selection indices based on progeny
tests of plus trees. Therefore it is necessary to develop selection methods based
on mathematical calculations and indices, on the basis of which the assessment
of trees will be srfer and the subjective factor eliminated. Cec h (8) came
to the same conclusion on the basis of a comparison of 5 systems for the
selection of superior trees used in the Southern States of the U.S.A. On the
basis of his results he concludes that the system is not so important as the
estimator, and therefore he considers the numerical system to be better than
the less objective descriptive systems.


The attempt by Douglas s (10) to work out a method of selection of
Japanese larch trees which are superior in wood volume production is most
interesting. This selection method is based on the ratio of tree volume to crown
volume. In previous methods for the selection of superior trees there is a lack
of connection with the stand mean, which is necessary, a Douglas s says,
to predict the degree of improvement to be expected. It is clear that a final
evaluation of the superiority of trees would have to be based on testing the
heritability of this property. Such a selection method, when established in
detail, will surely mean a pace ahead compared with these based on the volume
of the selected tree in relation to neighbouring trees, or when taken as the
largest tree of a number of trees or the tree displaying the largest volume
increment or some other good characteristics. Growth rate estimation in
the selection of southern Pines in the U.S.A. as described by Brow n
and Goddar d (4) also merits discussian. According to them, assessment
of growth rate in even-aged stands should be based on the ratio
basal-area increase to crown size (length x mean radius). Rudolp h (35)
recommends use of the ratio of d-h (square of diameter b.h. x tree height)
to D2L (square of crown diameter x crown length) for tree evaluation. By
plotting these data on a coordinate system it is possible to construct curves for
average and plus trees. The most reliable results from the use of this system
are obtained from estimates of trees in even-aged stands.


We should continue such work toward early development of reliable indices
for selection of plus trees, both in eve-aged and selection forests. Clearly
success will be greatest in an even-aged stand where spacing between trees is
equal and the site is of equal quality, and if the selection is made among trees
belonging to the same diameter class.


3. SELECTION FOR ONE, TWO OR MORE CHARACTERS
Selection of plus trees will be much more successful if it is made for one
or two characters (6, 36, 40). The difficulty and the possibility of errors increase
in geometric ratio to the increase of the number of characters, except if these