( xlvi ) 



their hosts, they would be less i:)revalent than they are, but 

 the variety of resting spores and chlamydospores which they 

 produce, gets over the difficulty in which the seasonal preva- 

 lence of their hosts would otherwise involve them. 



" In connection with Dr. Perkins' observations on the 

 condition of the leaves of trees in Brisbane and elsewhere, 

 with their black coating of mycelium, it may interest him to 

 learn that Citrus trees generally, which are notoriously beset 

 with these fungi, are here extraordinarily free, and it is quite 

 exceptional to see them. 



" But to return to the Cremastogaster fungus, we were and 

 are quite alive to the possibility that the fungus may be 

 exclusively saprophytic on the Coccid secretions, and that 

 proof is required that the ants actually eat the fungus or rear 

 it. But pending proof, we laid stress on one or two points. 

 The fungus is not a sooty mould, but belongs to rather a 

 remarkable group not generally found in such situations. 

 Though too much stress cannot be laid on the point, the 

 curious analogy of fungi of the same family occurring on 

 Termitaria was pointed out. 



" From the outside standpoint of the mycologist I would 

 put forward this suggestion. Is it not probable, if this 

 fungus is growing at the expense of the Coccids, and so at the 

 expense of the ants, that the latter would take measures to 

 get rid of it? Even a comparatively small stroma such as 

 the fungus forms would extract a great deal of moisture alone, 

 apart from the other nutritious matter which the ants are 

 likely to be after. Such stromata during their formation 

 can frequently be seen covered with quite large drops of 

 water— that is in confined spaces. In the open this of course 

 evaporates, but the drain on the substratum would be even 

 heavier. 



" Further, as I pointed out, the effect of chemotropism 

 would be to make the vegetative mycelium grow downwards 

 inside the shelter, and, while interfering with the ants, might 

 even more seriously impede the Coccids. (This is a point 

 on which as a mycologist I am not quite clear, that is in what 

 way the pathogenic fungi on Coccids kill their hosts.) 



" Mr. Lamborn, who has had a much wider experience of 



