THE FLORA OF MOUNT PULOG. 329 
Aniselytron agrostoides Merrill sp. nov. 
Glabrum, laxe caespitosum, circiter 40 cm altum; foliis planis, linea- 
ri-lanceolatis, quam culmus brevioribus; paniculis terminalibus, laxis, 
ramulis gracillimis, patulis vel adscendentibus; spiculis viridibus, pedi- 
cellatis, circiter 3.5 mm longis; glumis vacuis 2, valde inaequalibus, I 
-minuta, subobsoleta, vix 0.10 mm longa, truncata, I] lanceolata, acu- 
minata, uninervia, 1.5 ad 2 mm longa, II] ovata, 3.5 mm longa, 5-nervia, 
acuminata, leviter carinata. 
A laxly tufted, glabrous, perennial grass 40 cm high or less. Culms 
slender, unbranched, striate, smooth and shining, slightly geniculate ; 
nodes two or three, black. Sheaths about equaling, or somewhat exceed- 
ing the internodes; ligule hyaline, ovate, about 3 mm long, acute or 
irregularly cleft at the apex; leaf-blades linear-lanceolate, flat, thin, 
smooth, 10 to 12 cm long, 3 to 4 mm wide, acuminate. Panicles 9 to 13 
cm long, erect or somewhat nodding, the branches few, distant, slender, 
obscurely scabrid, the lower ones 4 to 5 cm long, ascending or spreading, 
comparatively few-flowered. Spikelets ovate-lanceolate, green, about 3.5 
mm long, pedicelled. Empty glumes 2, the first subobsolete, sub- 
orbicular, truncate, hyaline, 0.10 mm long or less, the second lanceolate, 
acuminate, 1.5 to 2 mm long, 0.5 mm wide, 1-nerved.. Flowering glume 
ovate (when spread), 3.5 mm long, 1.5 to 2 mm wide, slightly keeled, 
minutely scabrid on the keel, with 5 rather obscure nerves, submem- 
branaceous, acuminate. Palea nearly as long as the flowering glume, 
similar to it in texture, acuminate, 2-keeled, minutely scabrid on the 
keels. Styles 2, plumose. Anthers 3, 1.2 mm long. Caryopsis brown, 
loosely enclosed by the palea, about 2 mm long. Callus obconic, with 
very few short hairs, the rachilla produced back of the palea into a 
straight, glabrous, slender, 0.8 mm long awn. 
In the mossy forest, altitude about 2,400 m, not common, Merrill 6483, May, 
1909. 
This new genus is a member of the Agrostideae, and is undoubtedly most closely 
allied to the recently described Aulacolepis Hackel, which genus at present has 
two species, A. japonica Hack., of Japan, and A. treutleri Hack., of the Himalayan 
region. Suspecting the allinnte of the Philippine plant to the above genus I 
sent a specimen to Doctor Hackel for comparison, regarding which he writes as 
follows: “The specimen of Aniselytron, which you have sent me, shows in the form 
of its spikelets great affinity with Aulacolepis (even in the furrowed palea), but 
it can scarcely be placed in that genus on account of the rudimentary first 
glume * * * I think it therefore better to consider it as a separate genus.” 
In habit Aniselytron is very similar to lax-panicled species of Agrostis, and is 
doubtless closely allied to that genus,in spite of its glume characters, unawned 
flowering glume, and produced rachilla. The genus is well characterized by its 
strongly unequal empty glumes, the first being reduced to a mere rudiment, and 
the second one-half as long as the flowering glume. 
During the ascent of Mount Pulog this plant was noticed in widely scattered 
loose tufts in the mossy forest, and was mistaken for Agrostis elmeri Merr., 
<i ee oe ee 
wooed 
Pee} te 
