1914] Wilson: Anthracnose of Grasses 109 



size, the conidia from Arundinaria are in the main somewhat less curved 

 than is usual in C. cereale. The difference, however, does not appear to 

 be sufficient to warrant the separation of this form as a distinct species. 



11. Colletotrichum lineola Corda 



This species was first described from dead Umbelliferous stems, but 

 later almost any species of Colletotrichum with elongate acervulae was 

 referred to it. As a result C. lineola ' ' Aut. " is a species of astonishing vari- 

 ability in conidial and other characters, to say nothing of its extreme di- 

 versity of hosts. As Corda's figure is such as to give no warrant for sup- 

 posing that his fungus and the grass-inhabiting forms which have been so 

 named are the same species it is impossible to include this name among 

 those of the grass anthracnose. In the specimens on grasses which have 

 been referred to this species the conidia average 20 to 22 x 2 to 4^t, and the 

 setae about 100 to 150/^ long. This places the material among that which 

 must be considered as belonging to Colletotrichum cereale. The material 

 examined is listed below: 



On Andropogon nutans (Kansas), Festuca heterophylla (France), Festuca 

 sp. (Italy), Panicum virgatum (Louisiana), P. Crus-galli (Louisiana), 

 Sorghum saccharum (Connecticut, New Jersey, France) Sorghum sp. 

 (Alabama), and Zea Mays (New Jersey). 



12. Colletotrichum Bromi Jennings 



This species is known only from the type collection on Bromus unioliodes 

 in Texas. It is merely a form of Colletotrichum cereale with small sporo- 

 dochia, but otherwise typical. 



13. Colletotrichum lineola pachysperma Ellis & Kellerm. 



The fungus so named on Chrysopogon nutans from Kansas has conidia 

 which measures 20 to 30 x 4 to 7/jl. While its other characters would indi- 

 cate that it is identical with the species described by Manns, the extreme 

 size of the conidia will appeal to some as entitling it to the dignity of at 

 least a trinomial. However as the forms considered as identical in this 

 paper have a range of conidial measurements of 18 to 26 x 3 to 6/x this one 

 does not materially change these measurements. In case a separate name 

 is desired for the large-spored form, Roumeguere's Vermicularia Lolii 

 appears to be the earliest published name for such a fungus. As the 

 present name merely appeared on exsiccati labels, it cannot be considered 

 as properly published. 



