124 THE GASTEROMYCETES OF THE UNITED STATES AND CANADA 



core projecting downward from the central point where the mycelium was attached. 

 (This core is lacking in No. 8702 from Porto Rico.) 



Spores (of No. 7436) sometimes faintly angular as well as minutely warted, 3.5-4.2^ 

 thick. Capillitium threads up to 7.4ju thick, strongly encrusted. 



We have not seen the types, but as our plants agree almost perfectly with the 

 description of G. Lloydii, we refer them to it with confidence (see Myc. Notes, p. 50. 

 1901). Lloyd, who collected the types in Samoa, later merged the species with velutinus 

 (Geastrae, p. 35). If our determination is correct, we think this a good species. 



The species is in the rufescens-fornicatus group and almost exactly intermediate 

 between them in habit. The outer layer separates, not as in velutinus, beginning at the 

 tips, but as in fornicatus, radicans, and limbatus, beginning at the center. It is dis- 

 tinguished from fornicatus by the imperfectly fornicate habit and by the strigose and 

 much less trashy surface. The spores are larger than in G. Hieronymi and the spore 

 sac is not asperate; G. fimbriatus differs in more saccate base, smaller spores, entire 

 absence of hispid hairs, and in usually lighter color. In an ample collection from 

 Florida, consisting of 8 buttons and 12 open plants, all but one open plant were very 

 dark; another collection of a single plant from the same section was pale brown. 



This species, as well as G. fimbriatus and Astraeus hygrometricus, shows that in 

 some, if not all, Geasters with so-called universal mycelium, the true mycelium is really 

 only basal and the felted "mycelial layer" only an outgrowth from the button. Ex- 

 panded plants of G. fornicatus often show a hole in the bottom of the sac-like base as if 

 it had been pulled away from a firmer basal attachment. 



Florida. Gainesville. On an old stump. Weber, coll. 1923. (U. N. C. Herb., No. 7436, and Univ. 

 of Florida Herb.) 



Porto Rico. Seaverand Chardon, colls. (N. Y. B. G. Herb, and U. N. C. Herb.). Outer tomentose 

 layer almost entirely stripped off but fragments present. Spores dark, minutely warted, 

 3.5-4/i. Capillitium threads paler than the spores, up to 7.5/j thick. 



Geaster fornicatus (Huds.) Fr. 

 G. marchicus Hennings 



Plates 70 and 115 



Plants of medium to large size, the outer peridium splitting beyond the middle 

 into four (usually) to five (rarely more) lobes which bend strongly backward and down- 

 ward, the tips remaining attached to an equal number of shorter lobes on the margin of a 

 membranous cup, sunken in the earth and trash, which, as the outer layer of the button, 

 was stripped off from the middle layer when evagination occurred. This cup is always 

 completely covered with adhering trash and earth due to the flocculent mycelium weft- 

 ing it. Rays firm, the fleshy layer dark brown and adherent or partly peeled off, in old 

 plants worn away; the exposed fibrous layer whitish tan. Inner peridium seated on a 

 distinct, whitish, short stalk about 2-3 mm. long (in our plants), subglobose or urn- 

 shaped to depressed, constricted near the base so as to form a ring; surface dark brown, 

 finely velvety; mouth large, slightly elevated, fibrous lacerated, wrinkled, without a 

 distinct peristome, but the paler brown fibers in older plants have a distinct limit 

 exactly as in G. limbatus (sense of Hollos), which is very near. Columella not large, 

 about 3 or 4 mm. high and 2.5 mm. broad (in the Smith's Island plant). 



Spores spherical, blackish brown, distinctly warted under high power with rather 

 regular, blunt warts, 3.7-4.5^j thick. Capillitium threads 3-7. 5/j. thick, about as dark 

 as the spores; walls roughened. 



