188 TBINIDAD AND TOBAGO BULLETIN. [XVIII. 4. 



The 'second type of failure common on lime trees in the districts 

 indicated, and occurring sometimes on cacao, is produced when the 

 stem is girdled by the fungus while most of the root sj'stem remains yet 

 uninfected. In lime trees the earliest outward sign of the disease in 

 such cases, so far as I have seen, is the production of an abnormally 

 large crop of fruit. Presumably the production of flowers would be 

 equally striking, but this I cannot say from observation. Before this 

 crop of fruit has had time to ripen the foliage drops, often with such 

 suddenness that the ground is carpeted with leaves still green. The 

 appearance which the tree then presents is a familiar one in the 

 affected districts : its branches nearly bare of leaves and hung with 

 shrivelling and prematurely yellow limes. In an observed instance a 

 tree of perfectly healthy appearance, with abundant dark foliage, was 

 found on October 20 to have its bark all round and for some distance 

 up the stem infected through and through with Eosellinia ; it remained 

 green until November 17, and then the change from healthy foliage to 

 naked twigs was completed in fi-om twenty-four to thirty-six hours. 



When relieved from less of water by the fall of the leaves a tree 

 may put out a few small shoots and linger for some time before it 

 completely dies. It often happens that as the bark near the soil level is 

 killed, tufts of adventitious roots are pushed out from a callus formed 

 at the edge of the sound bark above. Quite a dense mat of small roots 

 is formed, some of which may get a good hold of the soil and thicken 

 up. This sometimes enables the tree to struggle on a little longer, and 

 even raises hopes of its recovery. Such an event is very improbable, 

 as the new roots soon become infected after they reach damp soil. 



Incidence of the Disease. 



The disease as seen on the Dominica lime estates is by no means 

 regular in its incidence. It is a matter of general and rueful experience 

 that it develops its attacks most widely in the best clearings, and that 

 the finest trees, which is to say those growing in the deepest and richest 

 soil, are the most liable to be infected. In clearings on thin soil and on 

 the considerable area of upland flats where the soil is underlain by a 

 stony ' pan ' which prevents anything like adequate drainage, diseased 

 trees are rare, and do not as a rule lead to further extensions unless 

 they happen to be situated in a deeper pocket in the one case, or on a 

 drier slope in the other. 



These experiences are exactly paralleled in Porto Rico, where 'the 

 disease often does most harm amongst the best trees, the sun-exposed 

 slopes of poor coffee plantations remaining quite free from trouble.' 

 ' The only things which retard or stop its progress seem to be excessively 

 dry or excessively wet soils, natural barriers, such as brooks, and the 

 scarcity of food material (decaying vegetation) in the soil.' 

 (Fawcett 1915.) 



This is not to say that slow-growing stunted trees are at all resistant 

 to the fungus. Where the roots of such trees happen to come in contact 

 with infected material they are killed as readily as any others. 



The explanation of this feature of the disease is to be found in the 

 favourable conditions of shade, moisture, and abundance of organic 

 matter, produced by the heavy canopy of well-grown trees. It is in 

 such situations that soil infections, without the intervention of stumps, 



