SECTION i.-BOVISTELLA TRUE. Capillitium short, separate threads. 

 Spores pedicellate. Sterile base well developed. 



SECTION 2. BOVISTELLA LYCOPERDON. Capillitium long, inter- 

 twined threads. Spores pedicellate. Sterile base well developed. 



SECTION 3- BOVISTELLA BOVISTA. Capillitium short, separate 

 threads. Spores pedicellate. Sterile base scanty or none. 



SECTION 4. BOVISTELLA-GLOBAR1A. Capillitium long, intertwined 

 threads. Spores pedicellate. Sterile base scanty or none. 



The Species of Bovistella. 

 Section i. Bovistella True. 



Capillitium short, separate threads. Spores pedicellate. Sterile base well 

 developed. 



BOVISTELLA OHIENSIS (Plate 86). Peridium usually glo- 

 bose or depressed-globose, with a strong tap root ; dehiscing by a 

 large but definite opening. Cortex of soft, well developed, cruciate 

 spines somewhat detersive in its nature. In old specimens it falls away, 

 leaving the peridium smooth. Sterile base of large cells, usually oc- 

 cupying about half the interior. Gleba olive or brown, never purple. 

 Capillitium of short, separate, much branched threads. Spores smooth, 

 subglobose or slightly oval in shape, 4-5 mic., with slender permanent 

 pedicels, 10-12 mic. long. 



This is a very common species in the United States and of a south- 

 ern range. It occurs from Washington, D. C., west to Missouri and all 

 over the Southern States. It is not found in the extreme east, north 

 or west. At Cincinnati it is the most frequent puff ball that we have. 

 It takes very kindly to cultivated ground and I gathered it abundantlv 

 one season in Professor Morgan's garden. 



SYNONYMS. It was always a mystery why Schweinitz never found this 

 plant, but it is not to be found in his herbarium. Considerable correspondence 

 [passed between Ellis and Morgan before the plant was named, and at one time 

 it \vas proposed to call it Lycoperdon velutinum.t Finally Cooke wrote Ellis 

 that it had the capillitium of a Mycenastrumt and Ellis published it (Jour, of 

 Myc. 85-89) as "Mycenastrum Ohiense, Ellis and Morgan." De Toni has com- 

 I piled it as Scleroderma Ohiense. When Morgan wrote his work he based on 

 ;it a new genus, calling the plant Bovistella Ohiensis, and the genus Bovistella 

 I based on the best of structural characters will always stand. No one has yet 

 'attempted to juggle the generic name. 



FORMS. We have received from Dr. Glatfelter, Missouri, a form with 

 inotably larger spores (6-7 mic.) but we hardly feel it worthy a name. Also in 

 iEllis' herbarium is to be found an unusually robust specimen (See Plate 86, fig. 

 |7) collected by Gentry in Indiana. It rarely occurs so large. 



SPECIMENS IN OUR COLLECTION. 



Alabama, A. S. Bertolet ; Florida, C. E. Pleas; Indiana, R. V. Con- 

 ifrse; Iowa, J. F. Clarke; l\cn'.ucky, John Nelson; Louisiana, Rev. A. B. 



t Specimens bearing- this name are still to be found in Ellis' collection, and in Cooke's 

 iierbarium I found a specimen sent by Ellis which Cooke has labeled " Mycenastrum [Ster- 

 j>eeckia velutinum, Ell. and Morg.' 1 



t Cooke's ideas of Mycenastrum capillitium were evidently -very vague at that time. 



I Which is a fair sample of considerable of the compilation made by De Toni in Vol. 7 of 

 nccardo. It has not one single character of the genus Scleroderma. 



279 



