186 TBINIDAD AND TOBAGO BULLETIN. \XVIIL 4. 



protection provide conditions especially suitable for the fungus. It 

 creeps along the line like a smouldering fire, killing oil the trees and 

 their seedlings, and most of the shrubby and herbaceous vegetation, as 

 it comes to them. In a gap so produced, from which the stumps had 

 been removed, a line of living fence posts subsequently put in consist- 

 ing mostly of white cedar [Tecoma leiicoxijlon) were three months later 

 infested to soil level. Wind-breaks of pois doux and galba {Galoj]hyllum 

 Calaba) are very susceptible to the disease. - 



The evidence thus shows clearly that the spread of the fungus may 

 take place in two ways : (a) underground along the roots of diseased 

 trees or infested stumps, in which case one or more outlying roots of 

 the new contacts are usually first infected and the fungus travels along 

 them to the collar, infecting other roots as it crosses them; (b) by the 

 growth of the fungus through rich and damp (which usually means 

 shaded) surface mould and vegetable debris, in which case the fungus 

 first attacks surface roots or directly infects the collar. 



How the infection originates is not known. Tradition, the whole 

 course of planting experience, and many definite observations by 

 agricultural ofl&cers, combine in associating the first outbreaks which 

 occur with the presence of stumps in new clearings, and of dead or 

 sickly shade trees in older plantations. The theory as to the general 

 course of infection which seems to the writer to best fit the facts at 

 present known is this : (1) that the fungus is able, by means of its spores 

 (of which the conidia are by far the more abundant and the more likely 

 to be distributed by the wmd) to infect any accumulation of decaying 

 vegetable matter in damp soil ; (2) that the required conditions are most 

 often presented in the immediate neighbourhood of mouldering logs and 

 stumps ; (3) that a surface infestation thus begun may or may not com- 

 municate the disease directly to the cultivated trees : in the earlier stages 

 of a plantation the chances are considerable that it will not, owing to 

 (a) disturbance of the humus-bearing surface soil in planting the trees, 

 and (6) the lack of shade conducing to rapivl destruction of organic 

 matter in the soil around ; (4) that infestations round about the stumps 

 of certain trees, on the other hand, are communicated to the roots at 

 and about soil level, the fungus finding in the buried roots a food 

 supply situated in permanently congenial conditions, so that it is able 

 to follow them out to their extremities however far and deeply they 

 may run, thus establishing a long-enduring and wide-spreading source 

 of infection for the roots of cultivated trees which extend into contact 

 with them. 



As the trees grow large and produce a deeper shade, so protecting 

 and keeping moist the organic matter which accumulates beneath them, 

 the conditions for surface infection are greatly improved, and by this 

 time, owing to the production of spores on the remains of the trees 

 previously killed, the chances of any suitable patch of soil becoming 

 infected are much greater. At this stage, therefore, the proportion of 

 cases originating by direct infection from the soil will have risen, and 

 such cases may in time far outnumber those associated with stumps. 



