324 STUDIES m symbiosis, ii., 



In Macrozamia deeply seated secondary roots are generally not infected, 

 and develop in the normal way. From this fact, it would appear that the bac- 

 teria which cause infection are most abundant in the upper strata of the soil, 

 where they have access to greater supplies of oxygen. If the surface of the main 

 tap-root and of a tubercle-root be gently scraped on to a slide, and the scrap- 

 ings mounted in water, examination will reveal the presence of numerous bac- 

 teria. Many of these bacteria gain access to the tissues of the main root and 

 of the upper secondary roots and develop pathogenically. There seems no doubt 

 that the peculiar form of these tubercles is a pathological effect of bacterial 

 stimulation. The root-tissues are not injured in any way, but there is a greater 

 development of the root-sheath (root-cap) which is permanent, since the tubercles 

 grow into the air, and a more extensive formation of cortical tissue. In both 

 sheath and cortex the cells are considerably enlarged as compared with the cor- 

 responding tissues of a normal secondary root, while the intercellular spaces are 

 also more extensive. This structure obviously lends itself to efiicient aeration, 

 and suggests the probability that the bacteria are aerobic. This assumption is 

 supported by experimental evidence for, when transferred to a water-drop on a 

 slide, the bacteria generally arrange themselves round the edge of the cover-slip. 

 When placed in a water-drop containing a Spirogyra thread, which is illuminated 

 strongly from the mirror of the microscope through a small diaphragm, they 

 mass in the vicinity of the illuminated spot. It would seem, therefore, that one 

 function of the tubercle is aeration; but the structure is developed for the benefit, 

 principally, of the bacteria in the cortical cells, although no doubt, the tap-root 

 derives some beneiit in this respect; that the aeration of the primary tap-root 

 of Macrozamia by the tubercles is a subsidiary function; the aeration of the 

 pathogenic bacteria, the primary function, is supported by the fact that many 

 tap-roots do not develop tubercles. 



The presence of the bacteria, therefore, produces certain definite effects upon 

 the development of the tubercle, namelj', (1) the production of a more extensive 

 sheath and cortex, (2) the enlargement of the sheath and cortical cells, (3) the 

 production of a more extensive system of intercellular spaces for aeration, (4) 

 the disappearance of starch from the infected cells, and (5) the production of a 

 more active meristem which however is free from bacterial infection — Text-fig. 14 

 shows these meristem cells with starch and no bacteria. The absence of bacteria 

 from actively dividing cells of the tubercles of Macrozamia recalls the parallel 

 case of Dipodium punctatiim roots in which the mycorhizic fungus is not present 

 in the meristematie zone. 



Cultures of bacteria were made by placing smaU fragments of the cortex of 

 a tubercle on sterilised serum-agar, on sterilised turnip jelly, and on a sterilised 

 jelly made from the mashed tap-root of Macrozamia, in test tubes. The cultures 

 were developed at ordinary laboratory temperature and at 26° C. in an in- 

 cubator. In the course of a few days a fairly vigorous growth of bacteria had 

 taken place. The colonies were ovoid to circular, raised, shining, and from 1 mm. 

 to 2 mm. in diameter. From these cultures, others were made with the platinum 

 needle, on similar sterilised media, and in a liquid nutritive medium composed of 

 Cane sugar 1 gram, K. phosphate 1 gram, Am. sulphate 1 gi-am, CaCOa -5 

 gram, Distilled water 1000 c.c. 



The bacteria were examined, and two forms were always present, namely a 

 rod-like bacillus form, and a small spherical coccus form (Text-fig. 7). Both 

 forms were motile; this motility was apparent even in the cells of the tubercles. 

 Transverse and longitudinal sections of the tubercles were cut by microtome, and 



