usually found on bare ground having no need to raise up the tortile 

 portion, either develop a sterile base very slightly or not at all. The 

 sterile base has no value in classification, generically or even spe- 



The genus Bovistella belongs to the tribe of Lycoperdae and is dis- 

 tinguished from all the Bovistae, not by the presence or absence of a 

 sterile base, but by its method of spore dispersion. Hence, little globose 

 species that do not have a sterile base but have subflaccid peridia and 

 a strong rooting system that holds them to their place of growth when 

 ripe, we do not call Bovista, but Bovistella. 



In examining the gleba of the Lycoperdae under the microscope 

 two very strong characters are found. First we find certain species 

 that have short, separate, capillitium threads (fig. i2)4 Second, we 



. 





Fig. 125. 



Fig. 124. 



find certain species that have permanent pedicels to the spores! 



(fig. I2\)lT 



Frequently these two characters are associated in the same plant, 

 but not always, and we believe it will simplify classification to embrace 

 in the genus Bovistella all plants of the tribe Lycoperdae that have 

 either or both of these characters. $ Taking the genus Bovistella in 

 this broad sense we can readily divide it into a number of sections, 

 which will no doubt in time be raised to generic rank by those who de- 

 light in multiplying the genera. 



tThe only idea in connection with Quelefs genus Globaria apparently is "puff ball- that 

 are round." As now classified it embraces in his Bncbiridion three species ol Bovist.. <>nt 

 Catastoma, one Calvatia, one Lycoperdon, one Bovistella and two species, genus unknown to 

 me. It illustrates the value of a genus based on the absence of a sterile base. 



I Statement has been made that pedicellate spores in the Lycoperdx- are not of pennnnent 

 value and that all spores lose the pedicels with age. I think that this is incorrect and that 

 there is no better character among these plants than the character of permanent, pediot llaie 

 spores. It is true that all I.ycoperdons perhaps have the spores pedicellate when young: in 

 some these pedicels are absorbed in deliquescence, in others remain, but are sfparatf from the 

 spores in the ripe gleba. In these the articulation is at the tof> of the pedicel. Truly prdirel- 

 late spores have the articulation at the base of the pedicel which remains f>ci manfully attache^ 



> the spores. We have examined specimens ol this type of spores in Kay's herbarium, 

 collected two hundred years ago, and have found every pedicel attached. 



'i All Bovistellas to our knowledge have pedicellate spores, but we would not exclude from 

 the genus a species having short, separate, capillitium threads and spores not pedicellate, if wf. 

 u a p.ant. In the genus Bovista there is such a species, Bovista pila. 



278 



