74 DISEASES OF CROP-PLANTS 



Grenada and probably of St. Vincent and Montserrat, the control 

 approaches to that in Dominica. In the drier districts of these 

 islands the fungi, while present, are not so effective, or are more 

 restricted to a seasonal activity. With reasonable accuracy it 

 might be said that excepting the shield-scale fungus and the 

 Aspergillus on sugar-cane mealy-bugs, the limits of possible 

 cacao cultivation are the limits of the effectiveness of scale- 

 destroying fungi. 



Consideration must, however, be given to the fact that the 

 matter is not quite so simple as has been assumed for the purposes 

 of general statement above. The rapidity of effective repro- 

 duction by the insects has to be taken into account as well as 

 the powers of attack of the fungus parasite, and in this the 

 condition of the plant is usually, perhaps always, the determining 

 factor. 



Even in the favoured districts of Dominica and St. Lucia, 

 plants transferred from the nursery to the field may become 

 badly infested until they get estabhshed, and require, or would 

 be the better for spraying. In connection with older trees, 

 the influence of other factors stands out where fungus control is 

 less complete, as in cases observed in Grenada, where particular 

 mango trees were heavily infested, though the shield-scale fungus 

 was present in some abundance, while others near by were quite 

 clean. 



Resuming the consideration of the effects of humidity on the 

 fungi in question, their distribution in Barbados is instructive. 

 The rainfall is low, and the island in general is open and wind- 

 swept. With the two exceptions noted, the fungi in question 

 are absent save in deep gullies, and in closely sheltered spots 

 among the hills where the rainfall is greatest. A similar, though 

 perhaps less extreme, situation seems to prevail in Antigua, 

 St. Kitts, and Nevis — islands of relatively low rainfall. 



Considering next the fungi mentioned above as exceptions 

 whose distribution is not so strictly limited, the case of one of 

 them, Aspergillus fiavus, the sugar-cane mealy-bug parasite, 

 is readily accounted for. It is not less dependent on moisture, 

 but finds it accompanying its hosts in the enclosure formed by 

 the base of the leaf-sheath around the nodes of the cane, a 

 situation in which water collects and lingers. 



The shield-scale fungus is more remarkable. Even in Bar- 

 bados it persists through the driest season with little shelter, 

 and becomes very active with somewhat small encouragement 

 in the wetter months. The only relevant difference known to 

 the writer between its spores and those of the other scale fungi 

 is that they are enclosed in a drop of mucilage. Whether this 

 or some constitutional factor accounts for its hardiness would 

 be difficult to determine. 



Having considered the influence of weather conditions on the 

 natural occurrence of scale-destroying fungi, there remains to be 



