Deyeuxia.| GRAMINE. 871 
4. D. avenoides, Buch. Man. N.Z. Grasses, 6.—Culms tufted, 
erect, slender, rigid, smooth, 9-24in. high. Leaves much shorter 
than the culms, narrow, in slender forms almost filiform, smooth, 
strongly involute; sheaths smooth, deeply grooved, the uppermost 
long; ligules short, broad. Panicle erect, 2-5in. long, ++in. 
broad, narrow, contracted, usually dense; branches short, erect, 
sparingly divided. Spikelets pale-green, 4-}in. long; pedicels 
shorter than the spikelets. Two outer glumes subequal, lanceolate, 
acuminate, sharply keeled, rigid, 1- or rarely 3-nerved, keel scabrid, 
sides smooth or minutely rough; 3rd or flowering glume slightly 
shorter, hard, convolute, scabrid, shghtly silky at the base, 
minutely 2—4-cuspidate; awn from below the middle, stout, re- 
curved, twisted below the bend, longer than the spikelet. Palea 
almost as long as the flowering glume, linear, hyaline, 2-nerved. 
Rhachilla produced into a silky bristle nearly 4 as long as the 
palea.—Agrostis avenoides, Hook. f. Handb. N.Z. Fl. 330; Buch. 
N.Z. Grasses, t. 24a. 
Var. brachyantha, Hack. MS.—Culms taller and more slender, 1-3 ft. 
high. Leaves narrower, filiform, often strict and wiry. Spikelets smaller, 
about 4in. long; rhachilla shorter and more delicate. 
Norte anp Souru Isntanps, Stewart Isnanp : Var. brachyantha common 
throughout, ranging from sea-level to 3500ft., the typical state apparently 
confined to mountain districts in the South Island. Nelson—Wairau Gorge, 
T. F.C. Canterbury— Craigieburn Mountains, Petrie! Rangitata Valley, Sun- 
clair and Haast; Tasman Valley, 7. #. C. Otago—Not uncommon in upland 
districts, Hector and Buchanan! Petrie! Cockayne ! 
The typical state is well characterized by its large spikelets, which no doubt 
induced Hooker to give it the name of avenoides ; but the var. brachyantha is 
an exceedingly puzzling form. Professor Hackel has no hesitation in placing it 
under D. avenoides, but most New Zealand botanists, including myself, have 
been accustomed to regard it as a state of D. quadriseta, to which it seems to 
show avery near approach, principally differing in the rather larger spikelets, 
with the rhachilla always produced at the back of the palea. It appears to me 
to be very much a matter of taste whether it should be placed under D. avenoides 
or D. quadriseta. 
5. D. Youngii, Buch. Man. N.Z. Grasses, 6.—‘‘ Similar in habit 
to A. avenoides, but larger, 2-4 ft. high, more robust. Leaves flat, 
4-tin. diam. Panicle 4-6 in. long, very slender, flexuous ; branches 
very short. Spikelets 4-tin. long. Empty glumes oblong-lanceo- 
late, acuminate, rigid, smooth, glabrous, nerveless ; flowering glume 
as long, pedicelled, hard, scabrid, 2—4-cuspidate; awn very short, 
almost terminal. Palea as long as the glume; pedicel stout, with 
long silky hairs.’’—Agrostis Youngii, Hook. f. Handb. N.Z. Fi. 
380; Buch. N.Z. Grasses, t. 29. 
Sourn IsnandD: Canterbury—Dry hillsides, sources of the Waitaki River, 
Haast. 
