ROOT TUBERCLES OF LEGUMINOS^E 259 



bacteria — by Beijerinck (1888), and the details of infection 

 studied by Prazmowski (1890). Since then much work 

 has been done on the nature of the symbiosis, the conditions 

 of nitrogen fixation, and the practical importance of the 

 association. Accounts will be found in Jost and Russell, 

 where the literature is cited. 



The bacterium, now generally referred to as Bacillus 

 radicicola, is found in soil in which leguminous plants have 

 grown, finding its way there from disintegrating roots. 

 It infects the young root through the root hairs. In the 

 hair it multiplies enormously and travels inwards in its 

 millions embedded in a slimy cord. The intervening cell 

 walls are dissolved and the thread of slime passes into the 

 cortical cells and stimulates these to abnormal growth. 

 The resulting hypertrophy of the root cortex forms the 

 familiar root tubercle or nodule, which may be found on the 

 roots of all leguminous plants. The nodules vary in size 

 and form with the species and the size of the root ; they 

 may be the size of a pin-head, or they may be as large as a 

 hazel nut, as on the stout tap-roots of the lupins (Fig. 34). 

 They consist of large-celled parenchymatous tissue, and are 

 traversed by forked vascular strands, arising from the central 

 cylinder (cp. Spratt, 1919). In the cortical cells the bacteria 

 leave the infection thread, and it is after this that the stimu- 

 lation to division of the root cells occurs. The parenchyma 

 cells in the mature tubercle are so densely packed with the 

 bacteria that they have a granular appearance under low 

 powers of the microscope. In the central cells of the tubercle 

 the bacteria lose their characteristic form ; they swell up 

 or form short forked bodies. These are degeneration or 

 involution forms, and have been named bacteroids. 



That the leguminous plant in conjunction with the 

 bacteria could fix free nitrogen was shown in the cultures 

 of Hellriegel and Willfarth, by comparison of the nitrogen 

 balance of plants grown in sterilised and infected sand in 

 which no nitrogenous compounds were available. In a 

 typical case a lupin grown in sterile sand attained a dry 

 weight of o'933 grm. and showed a nitrogen loss (as 



