﻿1903] RAVENELIAS OE 7 HE UNITED STATES 1 19 



tical-ovate, 27-33 t>y 14-17 fi; germ pores 8 in two rows, one 

 row of 4 germ pores in the equator, the other row near the base 

 of the spore ; paraphyses numerous, light yellow to semi- 

 hyaline, clavate to capitate; pedicel hyaline, 50-60 /i long; 

 heads 13-20/^ in diameter, very variable in shape. Teleutosori 

 dark brown, on both sides of the leaves, small, i "*'" or less in 

 diameter, oval to orbicular; teleutospore heads brown, usually 

 smooth, but margin of some of the fully mature spores some- 

 times faintly verrucose, orbicular to oval, 83-lOOft in diameter; 

 6-8 spores in cross section, marginal spores IG-12, spores 40-45 

 by i6-i8 /x, with apex brown for 4-5/^; cysts hyaline, periphe- 

 ral, oblong-ovate, few, 24-30 /a, swelling and finally bursting 

 in water; pedicel compound, short, slightly fulvous, deciduous. 



On Acacia Greggii: Gillespie co., Texas, Jermy 557 (U. S. Dept. Agr.) ; 

 Tucson, Arizona, Aug. 1892, /. W, Tourney \ San Bernardino, Cal., June 4, 

 1894, Parish 3377; Mescal mts., Arizona, May 24, 1890, M, E. Jones, 



This is one of the very widely distributed Ravenelias as the above cita- 

 tions show. It is also a very marked and easily recognized species, as its 

 uredospores are very characteristic, having two rows of germ pores, one in 

 the equator and the other near the base of the spore, the lower half usually 

 being hyaline or nearly so, and when boiled in lactic acid the lower half 

 swells up and the outer part of the epispore appears gelatinous, givin^sf the 

 entire spore somewhat the appearance of an acorn in its cup, the upper half 

 representing the acorn and the lower the cup. 



In all the material examined except the Texan specimen, only uredosori 

 were present, with an occasional teleutospore ; the Texan specimen on the 

 other hatid was teleutospore material with only a few uredospores intermixed. 

 A sufficient number were present, however, to fully identify the plant and to 

 show all the peculiar characters of the "typical uredospores of R. versatilis. 

 The Texan plant was reported on Acacia filicina Willd., which proved on 

 comparison with authentic specimens to be not A. filicina hyxl A, Greggii, 

 This comparison was made by the writer at the Missouri Botanic Garden, 

 St. Louis, with Jermy's original collection from near San Antonio, Texas. 

 Even the plant in the herbarium was infected with the rust. The species 

 heretofore called R, decidiia (Peck) Holw. proved on careful comparison Xo 

 be in every respect identical with 7?. versatilis {^\.:) Diet. It is reported on 

 all sgecimens examined as being on Prosopis pttbescens, but a comparison of 

 authentic specimens of this species with the host proved the previous deter- 

 minations to be erroneous, and the host to be Acacia Greggii. Type mate- 

 rial of R. decidua was examined and compared with authentic specimens of 

 R. versatilis, with the above results both as to hosts and parasites. 



■ 



