158 I'sfilagfl. 



Spores olive-brown, variable in size and shape, often arranged in 

 chain-like rows and accompanied by long hyaline filaments, sub- 

 spherical to ellipsoid or oblong to linear, densely warted, 6-7 /' diam. 

 when subspherical, and 7-13 x 4-(U /' when elongated. 

 In spikes of Carex fseudo-cyperus L. 



Tasmania — Huon, Dec, 1894 (Kodway). 

 South Australia — Mount Lofty Eange, 1893 (Tepper). 

 A specimen was sent to Ludwig from South Australia hy Tepper on 

 a host-plant supposed to be Cyperus lucidus R. Br. He determined the smut 

 as a new species {U. catenata) but recognised it as being closely allied to U. 

 olivacea. Ludwig kindly forwarded me some of the original material, and it 

 was evidently the same as U. olivacea, but I presume Ludwig was mainly 

 influenced in making it a new species from being found apparently on a dif- 

 ferent genus of host-plant. On submitting the plant to L. Rodway, F.L.S., 

 he determined it as unmistakably Carex pseudo-cijperus, so that I have made 

 Ludwig's species a synonym. 



Spsgazzoni had already recorded it on this species from the Argentine 

 Republic, but Saccardo made it a variety — pseudocyperi — on account of the 

 spores being 6-12 x 6-10 //. We found no spores as broad as this, and have 

 retained the original species. This smut is confined to species of Carex and 

 chiefly those which grow under damp conditions. Only a few of the ovaries 

 of a spike are affected, but the rain readily spreads the spores over the parts 

 beneath. It is easily recognised, not only from the olivaceous tint over- 

 spreading the spike, but from the tufts of filaments which project from the 

 ovaries which are attacked, and it is this peculiarity chiefly Avhich has attracted 

 a considerable amount of attention to this species. Among the irregularly 

 twisted web of filaments there are two kinds to be distinguished — those 

 which produce the spores and those which remain barren. The spore-pro- 

 ducing filaments show the formation of spores at various stages. At first 

 the filaments exhibit swellings quite close to one another, like a string of 

 beads or a series of knots, then cross-partitions are formed separating each 

 swelling, the contents of which becomes a spore. 



There are other filaments, however, which are sterile and do not form 

 spores, but are mixed up with the spores like the filaments composing the 

 capillitium in the Puff-balls. How these filaments originate and Avhat is their 

 function has not been explained, but probably they serve in some way to 

 scatter the spores over a wide area. 



The characteristic elongated spores are usually found in chains from the 

 mode of their formation. 



Germination. — ^This has l^een determined bv BrefekU both in water and 

 in a nutritive solution. The spores germinate readily in water after a few 

 hours, putting forth an exceedingly fine germinal tube, which elongates and 

 then becomes detached as a conidium. In the case of small spores this com- 

 pleted the germination, but with large spores a second conidium is formed. 

 The detached germinal tube gave rise to a secondary conidium and ended 

 there. In a nutrient solution the spore germinated as before, only giving 

 rise to a succession of conidia, each of which when detached produced a 

 secondary conidium, or several might remain attached to the spore and bud, 

 but large colonies were never formed. It will be seen from this description 

 that there is, strictly speaking, no germinal tube, it is unicellvdar and does not 

 differ from the conidium directly proceeding from it. Hence the only means 

 of multiplication is by sprouting conidia. 



(Plate XXIX.) 



