386 Transactions of the Society. 



described as sessile this must ratlier be interpreted to intimate tliat 

 the stem is reduced to such a minimum as to be little more than a 

 mere point of attachment. It is not at all improbable that certain 

 conditions, such as an unusual lateral compression when growing 

 closely together, may serve to elongate the point of attachment into 

 a very distinct stem. We mention this tendency because we have 

 observed undoubted indications of a stem to some of the capitules 

 in those which are described as sessile species. In so far as North 

 American specimens are concerned, it seems difficult to define the 

 limits between R. glandulfeformis and B. sessilis. That, liowever, is 

 a technical point which need not be discussed here. 



The species called Bavenelia sessilis B. occurs in North America 

 on leaves of Te])hrosia, and in Ceylon on leaves of Acacia LehbeJc. 

 The capitules are convex, but rather flattened, about 0*12 mm. diam., 

 with from six to eight cells in each direction. Surrounding the capi- 

 tules is a dehcate hyaline frill of sterile cysts, the lateral margins of 

 which it is sometimes difficult to define, so that the frill appears to 

 be continuous all round the capitule. We prefer to call these bodies 

 " capitules " instead of " pseudospores " of Berkeley, as we shall here- 

 after show that the latter is a misnomer. (Fig. 6.) 



Closely allied to the last is Bavenelia glabra K. and Cke. on 

 leaves of Caljmrnia sylvatica at the Cape of Good Hope. The capi- 

 tules are large, as much as ■ 15 mm. diam., convex above, without 

 any visible stem or barren cysts, by which latter feature it may at once 

 be distinguished from B. sessilis. There are from six to eight cells 

 in each direction, so that the individual coloured cells are about the 

 same size as in the preceding species, " 02 mm. or rather more. It 

 is not improbable that if examined fresh and living, sterile cysts in 

 some form, perhaps diminutive, would be found concealed beneath 

 the lower margin of the capitules. (Fig. 9.) 



Following the two sessile species just named, which have the 

 capitules smooth, we may note two species in which the capitules 

 are surrounded at the base by a series of hyahne processes re- 

 sembling spines. Bavenelia aculeifera B. occurs on the leaves of 

 Megonemium in Ceylon, and some unknown tree in India. The 

 capitules are about • 1 mm. diam., with from six to eight cells, the 

 outer series of which are furnished with acute spines, fully as long 

 as the diameter of the cells, the residue of the cells being unarmed 

 and smooth. Neither stem nor barren cysts visible. (Fig. 11.) 



The other species, if really distinct, is Bavenelia Hobsoni Ckei 

 on leaves of an unknown tree in India {B. stictica B. and Br. in 

 ' Grevillea,' vol. v. p. 15), with rather small and flattened capitules 

 •05- '07 mm. diam. containing about four cells in each direc- 

 tion, the outer and lower series armed with thin hyaline pro- 

 jecting spines about half the length and thickness of those in B. 

 aculeifera B. This would alone be insuflicicnt ground on which to 



