1.W0PERDACEAE 117 



Caster papyraems B. & C. (Proc. Amer. Acad. Arts and Sciences 4: 124) in the 

 Curtis Herbarium from Japan and Benin Islands (authentic, but not type) is closely- 

 related if not the same. The Japanese plants are all open ; peridium extremely thin and 

 paperv, pliable, pale creamy straw color, outer surface with a concolorous coat of very 

 thin soft weft in places. Mycelium membranous; fleshy layer almost all gone; inner 

 peridium broken or worm eaten, peristome not shown. The Bonin Island plants are 

 strigose-hairy outside; two cespitose buttons, one grown button apparently torn open 

 at the top, and one open plant; even the last is strigose. These look like G. mirabilis. 

 Fleshy layer intact and rays much stiffer than in the Japanese plants. Peristome 

 distinctly silky. A microscopic examination, however, shows the spores to be identical 

 in the two collections. In the Japanese plants they are spherical, finely but distinctly 

 warted, 3.5-3.8/i thick, with a very distinct and shiny oil drop. In the Bonin Island 

 plants they are 3.6-4^ thick, surface and oil drop as in the others. We can confirm 

 Lloyd's opinion (Myc. Notes, p. 181) that Coilomyces Schweinitzii B. & C. is an un- 

 opened Geaster related to, if not the same as, G. mirabilis. The plants in the Curtis 

 Herbarium have spores exactly like our Chapel Hill G. mirabilis, spherical, minutely 

 warted, 3-An thick (pi. 116, fig. 15). 



Geaster caespitosus is apparently the same, with the usual slight variations which 

 are to be expected. The ample type collection in the Lloyd Herbarium has the exact 

 appearance of our plants except that the strigose character is not so obvious. The 

 surface is softly felted, becoming smoother; peristome silky, sharply defined by a ridge; 

 the plants attached by a pointed base to the abundant mycelium. Spores slightly 

 smaller than in our collections, very minutely warted, 2.3-3.4^. We find at the New 

 York Botanical Garden a box of miscellaneous material without names but labelled 

 Ohio, Lloyd, No. 255, in which, mixed with three other species, is a dense clump of small 

 plants that are G. caespitosus and just like the types. The type of G. lignicola Berk. 

 (Journ. Linn. Soc. Bot. 18: 386. 1881) at Kew from Rockingham, Australia (Thozet, 

 No. 908), is apparently not different from G. mirabilis. There are two crushed buttons 

 and one open plant on fragments of plant material. Mycelium not very obvious; 

 surface of both buttons and mature plants nearly smooth. Spore sac dark brown with 

 fimbriate mouth. Peristome characters not obvious on account of the way the plant is 

 pressed and glued down. Spores spherical, minutely warted, 3.2-4m thick. 



Illustrations: Lloyd. Myc. Works, pi. 100, figs. 2 and 3; also figs. 6-9 (as G. caespitosus). 

 Rick. Broteria 5: pi. 4, fig. 3. 1906. Freshly opened plants. 



7077. On trash and leaves of deciduous trees and a few pine needles, wooded hillside, August 10, 1923. 

 7208. In rich, rocky soil near cedars, November 11, 1923. 

 7502. In humus in mixed woods (mostly deciduous) , Aug. 14, 1924. 

 AlsoXos. 7331,7413, 7414. 



Virginia. Warrenton. Upland frondose woods, Sept. 6, 1926. Coker, coll. (U. N. C. Herb.) 



Ohio. Lloyd. (X. Y. Bot. Gard. Herb., unnamed.) 



Kansas. Cragin, coll. (X. Y. Bot. Gard. Herb., as G. saccatus). Outer surface clean, tomentose; 



spread when expanded 1.8 cm. Spores minutely warted, 3.5— 4.2^. 

 Cuba. Wright, Xo. 696. (Curtis Herb.) Button very small, strigose to tomentose. 

 Jamaica. Bancroft. Ellis, coll. (Farlow Herb.). 

 Brazil. Rick, coll. (U. X. C. Herb.). 



