the ' Crown-Gall ' Fungus of Lucerne 365 



More than 100 nuclei are formed in the spore as it matures; they 

 are on the whole larger, and show a reticulum better than those in 

 the collecting cells. 



Figs. 3 and 4 show this method of prohferation, spores of four 

 different ages developed in succession on one portion of a thallus 

 being shown in fig. 4. 



The branched (haustorial) processes are not found to persist 

 on the ripe spores, but as a rule the small depressions from which 

 they arise can be made out. 



From this description of the fungus it would appear that it can 

 no longer be regarded as forming its resting spores as a result of 

 the conjugation of two hyphae, in the manner described by Magnus* 

 and Schroeterf; this is the conclusion also reached by Jones and 

 Drechsler % ; their description of the spore formation agrees exactly 

 with that in this paper. It should be noted that the fungus even 

 at the earhest stage in the host plant is bounded by a thin wall, 

 forming a definite mycelium. No trace of a plasmodial stage as 

 described by Wilson§ has been observed. 



Host Plants 



Lucerne {Medicago saliva) has been the only host plant observed 

 associated with the fungus in this country; M. falcata. is reported 

 to be about equally attacked in the United States with M. saliva, 

 under the same conditions, and M. denticulata to be immune. 

 Several attempts to infect M.falcata and M. lupidina have failed, 

 but are being repeated. 



Unsuccessful attempts to bring about infection have also been 

 made with all commonly cultivated leguminous crops, and a 

 number of common leguminous pasture plants and weeds. 



Infection has been induced at all seasons of the year and with 

 Lucerne plants of all ages from six months old upwards. It is found 

 however that under normal conditions actual infection of the host 

 tissue does not take place during the summer months, the most 

 favourable time being from September to February. From observa- 

 tions made in the field it seems probable that a very wet condition 

 of the soil is favourable for infection, though actual flooding is not 

 necessary. All attempts to cause infection of the youngest seedhng 

 stages have so far failed. 



I should like to take this opportunity of expressing my thanks 

 to Professor Biffen for suggesting the work, and for supplying some 

 material for investigation, to Professor Seward for laboratory 

 accommodation and to Mr F. T. Brooks for directing the work 

 and for much stimulating criticism, 



* Magnus, P. Ber. Deut. Bot. Gesdl. Bd 20, Heft 5, pp. 291-296. 

 t Schroeter, J. (1882). Bot. Centbl. Bd 11, Nos. 31, 32. p. 219. 

 J Jones, F. R. and Drechsler, C. (1920). Jouni. Agric. Res. (U.S.A.), vol. 20, 

 No. 4, p. 295. § Wilson, O. T. (1920). Bot. Gaz. v. 70, No. 1, pp. 51-08. 



VOL. XX. PART III. 24 



