LYCOPERDAC] \r 121 



The abundant type collection from Jupiter, Fla. (Culbcrtson, coll.), shows this to 

 be a remarkable plant intermediate between G. coronatus and G. Jloriformis. In some 

 the rays are distinctly revolute with the center elevated, in others the rays are closed 

 in over or entirely under the sac. Though some of the rays are pliable they are dis- 

 tinctly hygroscopic, expanding quickly in water. Spore sac very faintly felted to 

 glabrous, not at all granular-dotted as in G. coronatus, about 1 cm. thick, practically 

 sessile; peristome varying from sharply outlined by a groove (about as in G. coronatus) 

 to a less emphasized boundary as in G. mammosus; mouth apparently brown in the fresh 

 condition, washing out pale and becoming lacerate. The outer layer of sand is thin 

 and the flocculence holding it is very delicate and obscure. Spores minutely and ir- 

 regularly warted, 3.2-3.8/z thick. Mixed with the types is a good specimen of G. 

 coronatus, and this is almost certainly the plant referred to by Lloyd in his footnote 

 as being an exception. 



From the appearance of the rays of G. arenarius, which are acute, but not acumi- 

 nate, the button is probably little if at all pointed. At the New York Botanical Garden 

 there are two plants from Florida labelled G. fimbriatus (Underwood, coll., No. 1801a) 

 that are almost certainly this species, though the mouth is only vaguely if at all deter- 

 minate. The plants are small, hygroscopic, covered with sand on the outside ; peristome 

 vague, spore sac nearly glabrous; spores minutely warted, 3.7-4.5yu thick, just like those 

 of Lloyd's plant and smaller than in G. ftoriformis. 



The species is very near G. floriformis (see Lloyd, Myc. Notes, p. 143) and this 

 approach is further emphasized by the vague peristome of the Underwood plants, which 

 have the same spores as Lloyd's plants. The rays of G. arenarius are distinctly less 

 rigid than in G. Jloriformis, but the smaller spores of the former are now the principal 

 character separating the two species. 



The extreme variation of Geasters is well shown by what appears to be a giant form 

 of G. arenarius that is represented in the New York Botanical Garden by two very large 

 plants from Greencastle, Indiana (as G. rufescens). They are 3.5-4 cm. thick with the 

 rays closed over the spore sac. The peristome is vague, but the spores are like those of 

 G. arenarius, minutely warted, 3.5-4^ thick, much smaller than in any other related 

 species. Another interesting case is that of plants from Lake Titicaca, noted under 

 G. jloriformis. 



Illustrations: Cunningham. Proc. Linn. Soc. N. S. W. 51: pi. 5, figs. 24, 25. 

 Hollos. 1. c, pi. 29, figs. 19, 20. 

 Lloyd. The Geastrae, figs. 53 and 54. 



Florida. Lloyd. (U. N. C. Herb.) Another collection by Underwood. (N. Y. Bot. Card. Herb., 

 as C. fimbriatus.) 

 Lake Helen. Mrs. Xoble, coll. (Lloyd Herb.) Mouth distinct. 

 Grasmere. Baker, coll. (Lloyd Herb.). 

 Texas. Denton. Long, coll. (Lloyd Herb.). Mouth distinct. This is variable in its hygroscopic 

 character. 



Geaster rufescens Pers. 



Plates 69 and 115 



Plants large, the button subspherical, not pointed, entirely buried until dehiscence, 

 when open 5 to 8 cm. wide; the thin outer mycelial membrane covered with earth or 



