288 



Mycologia 



This fungus is clearly different from Epichloe ? nigricans Speg., 

 which has stromata 5-10 mm. long. Atkinson worked with one 

 of Stevens's collections of this form and labeled it in his herbarium 



Fig. 1. Dothichloe subnodosa sp. nov. Cross section of a stroma showing 

 the sharp definition between the host and fungous tissue. (Outlined with a cam- 

 era lucida ; X 50) . 



as " Balansia subnodosa sp. nov." He apparently, however, never 

 published it. The writer feels that the species should be trans- 

 ferred to Dothichloe on account of the absence of the intimate 

 fusion of fungous and host tissue, pseudosclerotium, characteristic 

 of Balansia. 



Material examined: 



On Ichnanthus pollens Munro. Cornell University Explora- 

 tions of Porto Rico (Whetzel & Olive), No. 690, Mayaguez, Mar. 

 6, 1916 (type); id. id. (Whetzel & Olive), No. 689, Mayaguez, 

 Mar. 2, 1916; id. id. (Whetzel & Olive), No. 692, Maricao, Mar. 

 22, 1916; id. id. (Whetzel & Olive), No. 691, El Yunque, Apr. 22, 

 1916. 



Dothichloe Aristidae Atk., Bui. Torr. Bot. Club 21 : 224. 1894. 



Characteristic in that the stroma completely surrounds the culms 

 of the host, as in Epichloe. However, the stromata are black and 

 carbonaceous (fig. 13). 



Material examined: 



On Aristida portoricensis Pilger. Cornell University Explora- 

 tions of Porto Rico (Whetzel & Olive), No. 695, Mayaguez, 

 Mar. 7, 1916. 



