(294 ) 
nearly so, about 1.6 mm. wide: grain flattened, oval, about 
Dry sandy soil, Florida to Louisiana. The following 
specimens are referred here: 
Fiorina: A. W. Chapman, two specimens, one num- 
bered 69. 
Mississippi: S. M. Tracy, Biloxi, July 12, 1891. 
H. Kearney, Jr., near Waynesboro, Oct. 2, 1896, 
no. I14. 
Lovuistana: T. Drummond, New Orleans, 1832. 
A. B. Langlois, Slidell, ‘‘in pine woods clearings,” Oc- 
tober 5, 1891 (the type, in my own herbarium). 
Related to Z. ambzgua, but quite distinct. In that species 
the panicle is more open, the spikelets smaller and broadly 
oval when mature, the flowering scales only about 3 mm. 
long, and the palet, in which lies the most marked difference, 
is only ciliolate on the nerves which are much more curved 
at the base than at the rounded and apiculate apex of the 
palet, the internerve therefore broadly ovate-oval. The 
nerves are decidedly gibbous at the base. 
In the herbarium of Columbia University there are two 
sheets from Dr. Chapman labeled 7ricuspis ambigua- 
The specimens on one of these belong to the above described 
new species; the other contains two specimens, one the 
herein described species and the other the Poa ambizgua of 
Elliott. To this latter sheet is fastened a small pocket en- 
closing a few spikelets from Elliott’s herbarium. This con- 
clusively settles the identity of Elliott’s species which Chap- 
man cites in his Southern Flora. It will thus be seen 
that Chapman’s idea of TJrrcuspis ambigua was a com- 
posite one, including both the Poa amb/gua of Elliott and the 
species described above. 
