[Vol. 10 



134 ANNALS OF THE MISSOURI BOTANICAL GARDEN 



suggestive as to almost, if not entirely, cancel the force of the 

 above alternative. Furthermore, the appearance of seeds in 

 pods from the most highly diseased plants leaves little room to 

 doubt that, they were killed by the pod-blight organism before 

 reaching maturity. 



Attempts to isolate the pod-blight organism from lesions 

 found on cotyledons after seed germination have failed in every 

 case save one. In the spring of 1921, 100 cotyledonary lesions 

 taken from seedlings which had just come through the soil were 

 surface-sterilized and planted in agar plates. All these gave 

 rise to growths of bacteria or Fusarium or both; the pod-blight 

 organism was obtained from none of them. The one successful 

 attempt was that of strain 7 described above, in which case the 

 diseased tissue produced the pod-blight organism unaccompanied 

 by bacteria or other fungi. This isolation strongly supports 

 the belief that actual parasitism of the embryonic tissues does 



occur. 



Inoculations 



Inoculations have been attempted in the laboratory, the green- 

 house, and the field. Field inoculations have not been uniformly 

 successful, due entirely, the writer believes, to the unfavorable 

 influence of the dry weather which prevailed. Preliminary 

 field inoculations were made during the first week of September, 

 1 920. Spore suspension of material from diseased pods was atom- 

 ized on stems and half-grown pods. The rains which prevailed 

 during the greater part of July and August gave way to dry 

 weather a week or 10 days before these inoculations were made 

 and no infections resulted. The summer of 1921 was remarkable 

 for its deficiency in rainfall, the usual rainy season in July and 

 August failing to develop. A condition approaching serious 

 drought prevailed during the entire growing season for soybeans. 

 Field inoculations were made at 4 different times both by atom- 

 izing suspensions of pycnospores on wounded and unwounded 

 plants, and by inserting mycelium into stem and pod tissues. 

 Infection was obtained in one case only. In this instance, the 

 plants were growing on low ground and were large and very 

 bushy, probably maintaining a higher humidity by reason of 



