74 



seasons may be long. They are often unnoticed because of their 

 small size and inconspicuous colours and consequently supposed 

 not to be present. The size of the flower is no guide as to the 

 size of the fruit or seed, small flowers often producing large fruits, 

 and vice versa. The term fruit is here used in its botanical sense; 

 one fruit may contain any number of seeds and may be of any size 

 from that of a pin's point to that of a durian or jack fruit. As a 

 rule, plants raised from seeds are of a superior constitution and 

 less liable to disease than those propagated by any other method, 

 but as a matter of convenience, and a means of saving time, other 

 methods are sometimes adopted when seeds are just as easily ob- 

 tainable. Seeds should always be selected from clean, healthy, 

 vigorous plants, and in the case of plants that vary in any impor- 

 tant matter, and there are few that do not, from the very best 

 types obtainable. That they should be thoroughly ripe at the time 

 of gathering is also important, for although imperfectly ripened seeds 

 may germinate; they will not give such satisfactory results ns those 

 that have properly matured. Recent observation on this point in 

 connection with gutta percha (Palaquium gutta) seeds, confirms 

 this opinion. In this case great numbers of seeds were knocked 

 off the trees by bats before they were quite ripe, but being anxi- 

 ous to raise a large number of plants, every seed was sown. A 

 good number of these immature seeds germinated, but in many 

 cases dwindled away and died in the course of a few weeks, while 

 fully ripened seeds sown under exactly the same conditions, but 

 separately, have grown well. The demand for para rubber seeds 

 up to the present has been so great that seeds from any tree has 

 readily found a buyer, but now that it has been proved beyond a 

 doubt that there is a great difference in the yield of trees of the 

 same age, growing side by side, under exactly the same conditions, 

 preference should be given to seeds from trees that are known to 

 yield freely. Personally, if I were forming a plantation, I should 

 not hesitate to pay twice as much for seeds from proved trees as 

 for those gathered indiscriminately. 



To lay down a rule applicable to sowing all kinds of seeds is 

 impossible, but the principal points to be observed, and which will 

 prove successful in most cases, may be summed up in a few words. 

 Drain well, cover lightly; water carefully; shade moderately. Use 

 light sandy soil with a large proportion of leaf-mould for sowing 

 seeds of most kinds no matter what the ultimate requirement of 

 the plant may be. A smooth level surface, so that all the seeds 

 may be of the same depth, and a covering of very fine soil not ex- 

 ceeding in thickness the diameter of the seed is generally sufficient. 

 Under natural conditions seeds dropping from the trees get no 

 covering at all but under such conditions only one in tens of 

 thousands germinate except under the most favourable weather 

 conditions, therefore it is not a system to be imitated. From 

 observations made in regard to the natural reproduction of the 

 more important local timber trees I have come to the conclusion 

 that these bear seeds freely only at intervals of several years and 

 thai their chances of germination depends mainly on -the state of 



