16 H. MARSHALL WARD. 



that the rapidly following septa permit no recognition of primitive 

 cells destined to form the outer walls, ascogonium, &c, as before. 



After a few radial, vertical, and horizontal walls have been formed, 

 tangential septa (fig. 42, d.) make their appearance, cutting out series 

 of cells which are to form the outer wall, and which become firmer 

 and more deeply coloured, from an inner cell mass which gives rise to 

 the ascogonium much as before. Only a few asci are formed, in each 

 of which arise eight small oval uniseptate spores, which acquire a pale 

 brown colour as they ripen (fig. 42,/., and fig. 43). 



The mature perithecium is shaped like a pear or top, the broad end 

 attached to the hypha by a short pedicel, the narrow free end, or apex, 

 becoming thin and diffluent in order to allow of the escape of the 

 spores (fig. 43). Very few or no setce are formed around the peri- 

 thecium, and these of the same simple type as those scattered on the 

 mycelium (fig. 41). The whole structure of the fruit-body is, there- 

 fore, much simpler than that of the above described forms, and, from 

 the semi-translucent characters of the thinner cell-walls, allows the 

 main details to be made out by optical sections only. In some of the 

 dark-coloured cells of freshly prepared specimens, a bluish tint is often 

 observable ; I have not seen this in any other similar form. 



In no case have I succeeded in tracing a distinct alterative or de- 

 structive action of the Meliolas on the cells of leaves to which they 

 are attached. In many instances, as, for example, thick leathery 

 leaves like those of Memecylon cajntellatum, &c, the haustoria seem to 

 have no function beyond that of holdfasts ; in others, such as Pavetta, 

 Triumfetta, &c, attacked leaves certainly suffer from the presence of 

 the fungus. Nevertheless, I cannot trace this to any direct action of 

 the mycelium ; the contents of the cells show no effects which can be 

 regarded as due to the fungus mycelium directly. We must conclude, 

 therefore, that where the life of the leaf is interfered with at all, it is 

 indirectly ; the dense crust of a well-developed Meliola no doubt ob- 

 structs the play of physiological functions in an obvious manner by 

 obscuring it from light, blocking up stomata, &c. 



It is now possible to consider the question of the systematic position 

 of these remarkable and interesting fungi. Bornet, 1 following Fries 

 and Leveill6, places Meliola near the old group of S})haerias with 

 especial reference to Erysiphe, I have already quoted the view of 

 Fries that the Meliolas may be considered tropical representatives of 



1 Op, ciU, p. 260, 



