312 SCIENCE PROGRESS. 



abundant.^ In Java, lastly, repeated inoculations have 

 failed to produce any effect, and this fungus is therefore to 

 be regarded as purely saprophytic.-^ There seems to be 

 evidence that this form, belonging, as it does, to a sapro- 

 phytic alliance, is acquiring parasitism. 



It has been advanced that the altered behaviour of the 

 fungus in various parts of the world is due to the different 

 varieties of cane o-rown there. The determination and 

 formation of resistant varieties is being prosecuted with 

 vigour in many parts of the tropics. The rind-fungus 

 appears to be only parasitic upon the soft, juicy canes of 

 the Bourbon type. The "hard" Caledonian Qtieen and 

 White Transparent varieties, even in Antigua, where the 

 Trichosphaeria is rampant, are usually successful in resisting 

 it.^ But the parasite appears to be gaining power. In a 

 certain area in St. Kitts, in which island the Caledonian 

 Queen has been largely grown for nearly twenty years, 

 this hardy cane is already severely attacked ; * while one 

 of the " hardy " varieties is reported to have been cleared 

 out of parts of Martinique by the same disease.^ 



15. Of a similar nature to this incipient parasitism is 

 the occasional change of habit in insects which may cause 

 much destruction in the crops. This may be the case with 

 respect to a small beetle Xyleborus perforans which has 

 appeared in incredible numbers in West Indian canes 

 during recent years. The case is interesting because this 

 form was previously described as destroying wine casks, 

 and the change from dry wood to juicy rotting cane is 

 an extreme one.'' 



There appears to be no doubt that such a change of 

 food has been observed in a minute Tomicine beetle found 

 in Nevis." This species has been described as Hypo- 

 thenemus ei^uditzis, from its first discovery in the binding 

 of a printed volume, and it has since been found in various 

 dry substances. In Nevis, however, for a short time, its 



1 Boname (i). - Went (5). 



^This has been frequently noted; see especially Watts (i). 



* Barber (2), p. 150. ^ 



^ Blandford (i). " Barber (3) p. 122. 



