﻿114 BOTANICAL GAZETTE ^ [February 



in addition to its bringing out the germ pores it swells the spores 

 to their normal shape and size; drives out the air and makes 

 them more transparent ; by scattering the spores makes visible 

 any uredospores or paraphjses that would otherwise be over- 

 looked. By this means the uredospores of several species were 

 found, of which hitherto only the teleutospore stage had been 

 known. 



Sometimes it is necessary to make a fourth preparation to see 

 whether the teleutospores are all one-celled, or some two-celled ; 

 this will often be shown in the lactic acid mount where some of 

 the teleutospore heads will be broken up; but in case this does 

 not happen, it becomes necessary to macerate the material for 

 twelve to twenty-four hours in a weak solution of caustic potash 

 or even to boil it in this solution for a few minutes, when by 

 pressure on the cover glass the heads will fall apart and the inte- 

 rior structure can be seen. 



If permanent mounts are not desired, the cell feature can be 

 omitted and the mount made on the slide direct. Usually free- 

 hand sections of the softened specimen will show the position 

 of the sori in the leaf, and this can be done where a sufficient 

 quantity of material is available; but where only a bit, and that 

 of a type or rare species is at hand, it is best to use the micro- 

 tome. 



A study of the life history of the various species has not been 

 attempted in this article. One species (J?. Holwayi) was found 

 whose aecidial stage differs from the type in that it is without a 

 pseudoperidium. This species is placed in a new genus (Neo- 

 ravenelia), distinguished from Ravenelia by its caeoma type of 

 aecidia. The deep-seated character of the aecidiosori of this 

 species is in marked contrast to the very shallow sori of the 

 uredospores and teleutospores, which are usually between the 

 cuticle and the epidermal cells or just beneath the epidermis. 

 As the life history of the various species becomes better known, 

 others in all probability will be found to have a caeoma type of 

 aecidia, and will have to be transferred to Neoravenelia. 



The genus naturally falls into two sections, according to the 

 number of cells in the teleutospores, The first section (Rave- 



L 



