I'sfilago. 157 



and fertile cells. The developiuent of this species, as described by l)e l^>arv, 

 will show so clearly the nature and origin of the different parts that it will 

 be given briefly before discussing the necessity for a new genus. 



When the ovule is fully formed, the hyphae pass through the flower-stalk 

 into the ovary and reach the ovule through its funicle or stalk. There the 

 parasite penetrates the tissue to such an extent that the ovule is replaced 

 by a compact mass of densely interwoven hyphae with gelatinous walls. 

 The fungus is confined entirely to the ovule, and becomes differentiated into 

 a thick colourless outer wall, a central axile cylinder or columella, and be- 

 tween the two a dense mass of violet-coloured spores. The base of the ovule 

 remains undifferentiated, and there new hyphae are constantly being formed 

 which add to the height of the parts already formed and consequently give 

 them a cylindrical form. 



The fully-formed wall consists of several layers of minute round cells, 

 similar in size to the spores and produced in the same way fi'om the hyjjliae 

 of the primary tissue, but with a colourless membrane. The columella has 

 generally the structure of the wall, and the spores are developed around the 

 columella in a dense ring about equidistaiit from the centre and circum- 

 ference. The wall of the ovary is not attacked by the fungus, and does not 

 follow the growth of the spore-receptacle so that it is soon ruptured. The 

 wall of the spore-receptacle itself is very fragile and readily bursts at the top 

 to allow the escape of the spores. The development and germination of the 

 spores are the same as in Ustilago, the compact mass of much-branched spore- 

 bearing filaments having gelatinous Avails as in that genus. 



It will be seen from this description that the species differs considerably 

 from that of any other Ustilago, particularly in the distinct spore-receptacle 

 made up of fungus cells, but there are various gradations in the formation of 

 a fungus membrane enclosing the spores in different species, and the difficulty 

 is where to draw the line. The presence of a central columella is also not 

 uncommon, although it is often composed chiefly of the plant-tissues, and 

 the differentiation of similarly produced cells into colourless sterile cells, and 

 coloured fertile cells is foui\d in various other species of Ustilago, such as 

 U. cruenta, Cintractia as ('. leucoderma, and Sorosporium as S. panici-nnhacei, 

 so that the presence of a fungus membrane enclosing the spores cannot be 

 regarded as of generic value. While there is a general resemblance in this 

 variety to the type species which has already been recorded on Poh/c/oDum 

 minus in other parts of the world, yet there is sufficient distinction, perhaps, 

 to make it worthy of being classed as a variety. The cylindrical projecting 

 columella is very characteristic, and, on comparing the spores with those 

 from U. hi/dro pi peris in Rab. Fung. Eur. No. 2<)0l, they are slightly sniidler. 



(Plate XXXII.) 



( '(irc.r. 

 1:3. Ustilago olivacea (DC.) Tul. 



Tulasne, Ann. Sci. Nat. Bot. III.. Vol. VIT.. p. 88 (1847). 

 Brefeld, Untcrs. Gesammt. Mvk. \'.. p. 1l'9 (1883). 

 Sacc. Syll. VII., p. iM (1888). 



Ustilago catenata LudwiiX. Zeitschr. Pflanzenkr. III., j). 139 

 (1893). 



8oi'i only j)roduced in some of the ovaries of the inflorescence, often 

 more or less concealed at first by the perigynium ; olive-brown 

 spore-masses, firmly agglutinated at first ; finalh' more or less 

 powdery and intermixed with conspicuous filaments. 



