72 PROCEEDINGS OP THE AMERICAN ACADEMY. 



away in pieces. It is certain, then, that RJiopalogaster cannot be placed 

 in the Secotiaceae. 



The two remaining groups, the Hysterangiaceae and the Hymeno- 

 gastraceae, are described by Fischer as possessing a " Gleba ohue oder 

 nur mit kurzer, nicht bis zum JScheitel reichender Columella." This 

 characterization, if adhered to strictly, would exclude RJiopalogaster from 

 both groups, but since it seems more than probable that it is most nearly 

 related to forms included in them, it will be necessary to examine into 

 the structure of the different genera in order to determine, if possible, 

 which are most nearly related to it. 



In the Hymenogastraceae the resemblances to RJiopalogaster are to be 

 found in the hymenial structure. In this group, especially Hymenogastev 

 and Octaviana, the basidia are clavate, septate, and are arranged sur- 

 rounding irregular chambers. The direction of the tramal plates is away 

 from the peridium, however, and the columella is but rudimentary or 

 entirely lacking. The genera Rhizopogon, Sclerogaster, and Lycoga- 

 lopsis seem rather to exhibit affinities with the next great group, the 

 Lycoperdineae. 



In the Hysterangiaceae the absence of a complete columella is the 

 most noticeable point of difference. Gautiera resembles RJiopalogaster 

 in its irregularly chambered gleba, and in its thin peridium, which is at 

 maturity lacking to such an extent as to expose the glebal chambers just 

 as is shown in fig. 11 of the accompanying plate. 



Gymnoglossum, a form reported by G. Massee (20) from New South 

 Wales, is distinguished from RJiopalogaster mainly by the absence of a 

 peridium at every stage of development, and by the fact that the colu- 

 mella does not reach more than half-way to the apex of the peridium. 

 The points of similarity are the " irregular cavities lined with large 

 clavate tetrasporous basidia, sterigmata elongated, slender, spores colored; 

 stem distinct, elongated, solid, passing up into the gleba." Protoglossum, 

 Mass., exhibits similar affinities to RJiopalogaster. 



The genus Hysterangium shows a columella reaching almost to the 

 apex of a well-developed peridium. A longitudinal section of an imma- 

 ture specimen of Hysterangium clatJiroides represented in a drawing by 

 Rehsteiner (18) is strikingly similar to some of my sections of RJiopalo- 

 gaster. It shows the thin flocculent peridium enclosing the stipe and 

 gleba, and from the sterile part which is a continuation of the stipe there 

 arise folds to form the gleba. The only essential difference in gross 

 structure which is here apparent is due to the fact that in H. clatJiroides 

 the columella falls short of the apex of the peridium. The basidia, how • 



