116 THE GASTEROMYCETES OF THE UNITED STATES AND CANADA 



Illustration: Lloyd. As cited above. 



Patterson and Charles. U. S. Dept. Agric. Bull. 175: pl.36, fig. 1. 1915. 



South Carolina. Ravenel, No. 953; Curtis, No. 3041. (Curtis Herb., as G. Curtisii Rav., co-types; 



also in Farlow Herb, from Ravenel.) 

 Florida. Sarasota. On a cedar stump. (Path, and Myc. Herb.) 

 Bermuda. Millspaugh. (N. Y. Bot. Gard. Herb.) 



Geaster mirabilis Mont. 



G. caespitosus Lloyd 

 G. lignicola Berk. 



Plates 65, 67, 68, 115 and 116 



Densely cespitose to single in considerable numbers, and arising from an obvious 

 white mycelium which binds together the leaves and twigs. Buttons spherical, not 

 pointed, but a few with a small umbo when dry, 0.8-1.2 cm. thick, strigose-tomentose, 

 reddish brown to ochraceous buff, a few blackish; mycelium basal, the buttons quite 

 exposed or partly covered with loose trash (completely covered in No. 7173). Ex- 

 panded plants 1.4-2.2 cm. wide; lobes usually 6 or 7, expanded (tending to become 

 elevated when dry), the basal half bowl-shaped and holding the completely sessile inner 

 peridium which is 6-9 mm. thick, subspherical, gray, with a delicately felted surface; 

 peristome distinct, silky, more or less elevated, conical, immediately after exposure pale 

 from a granular-looking, minute pubescence, with a darker line around the base, soon 

 darker than the peridium; mouth fimbriated, not sulcate. Fleshy layer when first 

 exposed about 0.5 mm. thick, smooth, light flesh color, soon darker and drying to a 

 thin, adnate, continuous, gray-brown membrane. 



Spores (of No. 7077) very dark, minutely warted, 3.2-3.8/*, most about 3.6/u. 

 Capillitium threads sinuous, thick- walled (practically no lumen), much paler than the 

 spores, about 3.4/* thick, the tapering ends smaller. 



From G. saccatus the present species is distinguished by cespitose habit, much 

 smaller size, hairy surface, less pointed buttons when dry and larger spores. Both have 

 the habit of growing exposed (epigeal). 



The species was originally described from French Guiana and is known also from 

 Brazil, Africa, and the Orient. 



Our plants are the same as plants received through the kindness of M. Patouillard 

 and collected in Brazil by Rick. The young buttons have the same strigose-tomentose 

 surface ; the older ones more felted-tomentose ; spores exactly as in our plants, faintly 

 warted, 3.4-4/i thick, not like the smooth and smaller spores of what we are calling 

 G. subiculosus. Rick considers G. trichifer the same (see Myc. Notes, p. 804), but 

 Lloyd thinks it different (see that species). In our No. 7077, some of the plants are 

 apparently as strigose as G. trichifer (which see), while others show only a matted tomen- 

 tum, exposure to weather having a considerable effect. 



The species is represented in the Kew Herbarium by five cards of material: three 

 from Cuba (Wright), two from Ceylon (G. H. K. T.), and one from Uganda, Africa, 

 (Diimmer, No. 1462). All these have the spores of G. mirabilis, and in appearance are 

 the same as the Chapel Hill and Brazilian specimens except that none shows quite so 

 strigose a surface. The spores of the plant from Ceylon, No. 184, are 3.2-3.8/* thick, 

 minutely rough. The African plants connect up directly with G. papyraceus, with 

 outer peridium more papery-looking than in the others. The spores of this collection 

 are 3.5-4.2/* thick, with surface as in the others. 



