[Vol. 3 

 332 ANNALS OF THE MISSOURI BOTANICAL GARDEN 



7. S. Patouillardii Burt, n. sp. 



S. (very near) Lcprieurii (Mont.) Patouillard, Soc. Myc. 

 Fr. Bui. 16:55. 1900. 

 Type : in ]3urt Herb. 



Fructification resupinate, effused, coriaceous, dry, velu- 

 tinous, aniline-black at first, becoming fuscous in the her- 

 ))arium, the margin rather thick and determinate; in struc- 

 ture 200-400^ thick, with (1) next to the 

 substratum a thin layer of loosely inter- 

 woven hyphae 3/t in diameter, buffy brown 

 under the microscope, which form (2) a 

 layer of hyphal pillars each about 30-50ai 

 in diameter, 100-200ai long, about 3-4 to a 

 millimeter, whose hyphae spread apart 

 above and form (3) the interwoven hyme- 

 ^ig- 5 nial layer containing some probasidia and 



a^twopTobSa;b, ^^'i^^i the surfacc composed of numerous 

 spore-bearing organ; ercct, nearly straight, fuscous hyphal 



p, four paraphyses or , , , n • t j 



hyphal ends, x 640. branches or paraphyses 2/x m diameter; 



probasidia hyaline, subglobose, 15-20/i. in 

 diameter, erect on short branches of the colored hyphae; 

 no spores found; the only possible spore-bearing organ seen 

 is 46 X ^i^/M, acuminate at the apex. 



Fructifications 2-31^ cm. long, 1-2 cm. broad, 200-400 mm. 

 thick. 



On living branches of ash, Liquidanibar, and Nyssa. Flor- 

 ida to Louisiana. Xovember to March ; a January collection 

 has a few probasidia. 



This species may be recognized by its thin fructification 

 resembling a piece of black velvet, slightly raised from the 

 substratum on such short and slender pillars as to be barely 

 visible without the aid of a lens. Patouillard determined 

 this species for Mr. Langlois as very near to S. Leprieurii. 

 Since Corticimn Leprieurii was originally described as gla- 

 brous, shining, and chocolate-colored, and since no specimens 

 like ours have yet been collected in the re,gion between Guiana 

 and the United States, our specimens are probably a distinct 

 species which should have a definite name. 



