37 



On January 7, 189G, 15 pots of the cowpeas, wliicli were now 6 to 8 

 inches liiiili, were re-inoculated with the melon fungus. There were 89 

 plants in these 15 pots. The fungus was derived from 5 pnre test tube 

 cultures on bruised, steam-sterilized seeds of the cowpea, and was put 

 into the top layers of the soil very copiously. The cultures were young 

 (1 and 2, December 21; 10 and 11, December 16; and 7, December 31) 

 and growing vigorously — i. e., each tube contained 10 to 15 cubic centi- 

 meters of the moist peas entirely overgrown and interwoven with the 

 fungus. In these 50 pots at least 175 plants of cowpea and as many 

 more of cotton were exposed to infection under conditions which 

 appeared to be admirable. 



This experiment was continued 14 months, during which time no 

 positive results were obtained, i^one of the cotton plants contracted 

 the disease and none of the cowpeas. In one only of the cowpea plants 

 a very little of the mycelium of a Fusarium was found in a few of the 

 vessels near the earth, but the symptoms were not tyiucal for the cow- 

 pea disease, and the fungus may have been that of the non-parasitic 

 Nectria mentioned below. On February 4, 1897, both varieties of cot- 

 ton were still alive. A few of the cowpeas were still alive, but the 

 majority ripened their seeds and died in the fall of 1890. At this time 

 the stems of some of the latter bore sparingly and in a saprophytic 

 manner (in no case high on the stems, but always near the moist earth), 

 the pinkish eonidia beds of a Fusarium. These first began to appear 

 in ]Srovember— i. e., 11 months after the soil was inoculated. Some of 

 these compact sporodochia may have been the eonidia beds of the 

 melon fungus, but they did not appear to be parasitic— i. e., they were 

 not preceded by an extensive occupation of the vessels of the plant 

 and were not arranged in rows up and down the stem,' but rather were 

 clustered on the moist bases of some of the decaying stems. Moreover, 

 on one stem they appeared in connection with the red perithecia of a 

 Nectria. These perithecia in shape and color much resembled those 

 described in this bulletin, but the ascospores were colorless, thin- walled, 

 elliptical-pointed, smooth, 1-septate, and 8 to the ascus — i. e., typical 

 Nectria sporidia. 



To determine whether the melon fungus was alive in the soil, the 

 earth was knocked out of the pots and the cotton and cowpea Stems 

 buried in it to form the substratum of a bed on one of the hot-house 

 benches. This earth was then covered with three pailfuls of clean 

 sand, on top of which an inch of good potting soil was spread. On 

 February 10 this bed was planted with the seeds of two varieties of 

 watermelons. Seeds enough for 500 plants were put into the earth, but 



1 This fungus did not iu any case fill the vessels and brown their walls the whole 

 length of thejilaut, as the cowpea fungus does, nor even for a few inches. Neither 

 did the external eonidia beds appear in the proper way. In the cowpea disease these 

 external eonidia beds occur by thousands on the dry stems at all heights from the 

 surface of the earth to the top of the plant (3 feet or more), and are arranged very 

 regularly in parallel rows lengthwise of the stem. (PI. 1, 11, 12.) 



