286 BOTANICAL GAZETTE [APRIL 
the morphology of the genus. In general the mycelium, aecidiospores, and 
uredospores offer no striking characters different from other rusts; a few points 
of interest, however, are note mycelium is usually intercellular and 
no ed by haustoria; but in some species it penetrates the cells, and in R. 
atrocrustacea on Swartzia it penetrates the vessels, developing abundantly within 
them. In several species the mycelium is perennial. Aecidia occur in only a 
few species and present no unusual peculiarities. The uredospores correspond 
to those of other rusts; but in the species on Cassia the uredospores are formed 
between the epidermal wall and the cuticle. 
The teleutospores, which constitute the most interesting spore-type, are 
treated with some detail. A great many modifications exist as to the number of 
cysts, spores, and hyphae making up the spore-heads;_ but the author has arranged 
these into types within which the mode of building up the spore-head is remark- 
ably uniform. These types are briefly as follows: (1) all spore-heads have a 
definite number of stalk-hyphae, which is uniform for each species, and each 
hypha gives a definite number of outer and inner spores; (2) the number of 
hyphae bearing the spore-heads is variable, but each hypha bears a definite 
number of spores, which is uniform within each species; (3) there is no uniformity 
in the number of hyphae and spores; the simplest forms of this last type have a 
single stalk-hypha which bears several spore-cells. 
Heretofore the genus Ravenelia has been regarded as related to Puccinia, 
but on account of the occurrence of longitudinal divisions in the heads, DIETEL 
thinks the genus bears a closer relationship to the genera Diorchidium, Antho- 
myces, and Sphaerophragmium. 
n the purely taxonomic part of the work 81 species are described, all of 
which are placed in the genus Ravenelia. The genera Pleoravenelia and Neo- 
ravenelia, recently separated from Ravenelia by LonG, are not retained, being 
considered insufficiently distinct. The author divides the genus into two sections: 
Haploravenelia and Pleoravenelia, the former comprising Ravenelia of LONG. 
ELBRING. 
Bacteria and mineral salts.—BENECKE,; in an investigation of the pigment- 
producing powers of B. pyocyaneus and B. fluorescens, lays emphasis upon the 
need of careful purification of chemicals and selection of glass receptacles in bac- 
teriological experiments with synthetic media. He used several kinds of glass 
control, including quartz, which is too expensive for general use, Jena glass 
which is K-free but contains Mg, resistance glass, Vienna glass which is probably 
Mg-free, Bohemian glass, and ordinary glass. With the various flasks he could 
obtain practically all the contradictory results of previous investigators, such as 
NoesskE, THuM, JoRDAN, GEsSARD, SULLIVAN, and Loew. By means of the 
controls he could trace most of the discrepancies in their results to impurities of 
chemicals or glassware. BENECKE’s chemicals were washed and recrystallized 
5 ,» W., Untersuchungen iiber den Bedarf der Bakterien an Mineral- 
stoffen. Botanische Zeitung 65:1-23. 1907. 
