130 MORrilOLOflY AND PARASITISM 01" lUlIZOC'TON'I A. 



iiltlunitjli its vinilciKo is groatcst on its own particular host. Thus 

 tlic HhizoclDiiid from cotton if infected upon jjrouiul nut. jute oi- cow 

 pea has a mortality of ahout 40 — oO" ^„ but upon cotton it has a mor- 

 tality of SO",,. 



The behaviour of the jute plant and its Ehlzoctonia is. however, 

 lather different. Jute is attacked very vigorously by its owmi Hhiz- 

 orlonia. the mortality rarely falling l)elow 70 of infected plants. 

 RhizortnxKt of cotton and ground nut is considei'abK- less \-irulent 

 on jute, the infect i\-e powci- of the latter being niuch higher if tlic 

 jute be first woundeii. This ap])lies in an even greater degree to 

 the Rhizoctonid of cow pea. which does not attack uninjured jute at 

 all l)ut has a mortality of JO",, on wounded jute. 



The Rhizoctr,nia of jute shows a high degree of sy)ecialisation as 

 it does not. under normal circumstances, attack the other three 

 hosts at all. ISotli on dt-s'i and "khaki" cotton and wounded and 

 nnw'onnded cow pea infections failed lo take. In the lattei' plant 

 tlie host successfullv reacted to the attack by cork formation round 

 the infected area (IM. VI. Fig. 2); in cotton the hypha' seemed unable 

 to penetrate the epidermis and the plant was not injure<l in any way. 

 In the early stages of attack on the cow pea the hyphse are more or 

 less intercellular in growth. In those cases in which the plant 

 reacts successfullv against the fungus {p-fj.. attack of the Rhizoi-tu- 

 iilu of jute) the hyphaj appear to remain in this state (IM. \'l. Fig. 

 2) : should, however, the Rhizocionia prove fatal, intracellular hy- 

 phic are speedily formed (PI. VI, Fig. 1). 



Infections which were made on ground nut early in the year 

 failed to give anv result, but on repeating the (>x])eriment foui' 

 months later death I'csulted in the case of one infection. Thegrounil 

 nut seed in this case, however, was not above sus])icion ; at the same 

 time the fact must not be overlooked that after nearly a year in 

 culture till' infective powers of a fungus may change. Onthewli(de. 

 considering all the facts with regard to the Rlnzovloiiia of jute, it 

 .seems (dear that we have to do here with a physiological s])ecialisa- 

 tion unaccompanied liy any oulwaid morphological distinction. 

 Details uf the experimental work are shown opposite on page ; 



