﻿GRASSES FROM JOHORE. 103 



branches they are 15-19 in. long, and reach 7 lines in breadth; on 

 the branches and the culm below the inflorescence shorter, and 

 without the narrow base. The joints of the fragile rachis are 2£ lines 

 long, and, like the pedicels of the stalked spikelets, thick, sharply 

 triangular, and glabrous, except on the outer hairy angle. The 

 sessile spikelet, with its short callus (f line), is nearly 4 lines long, 

 the callus has a beard less than half its own length ; the stalked 

 spikelet is a little shorter (3 lines), and its pedicel is 1 line long. In 

 the sessile spikelet the lower half of the first glume is straw- 

 coloured and bears several strong transverse ridges, the upper 

 portion becomes thin and membranous, and is marked indistinctly 

 with several fine anastomosing green veins. The third glume is 

 nearly three lines long and includes a pale and male flower ; the 

 fourth, 2i lines long, is entire, with a weak and shortly setose apex 

 but no trace of an awn ; it probably includes originally a £ flower, 

 but only the fertilised ovary was present in the flowers examined. 

 The outer glume of the stalked spikelet is somewhat oblique with 

 a broad curled wing on the longer side turned away from the sessile 

 spikelet ; the incurved margins form strong keels on the inner 

 surface. Blakan Mate, Singapore. J. B. Feilding, 1892. 



Is closely allied to I. rugosum, resembling a large coarsely- 

 growing form of that species with unawned spikelets. The differ- 

 ence in habit is so great that the two species would be distinguished 

 at once. Thus while Hackel gives one metre as the extreme limit 

 for the culms in /. rugosum, in I. magnum they are twice as tall, and 

 unbranched for three- fifths of their length, not branching from the 

 base as in the former species. The hairy sheaths and glabrous 

 nodes also afford another distinction, while the racemes of the new 

 species are more than twice the length reached in I. rugosum; 

 similar differences are noticed in the details of the spikelets. 



We have in the British Museum Herbarium a plant labelled 

 "Malaya, G. T. Lay," which must evidently be included in the 

 same species as the above, though differing slightly in its somewhat 

 larger spikelets, with the outermost glume of the stalked spikelet 

 slightly larger than that of the sessile, and also more broadly 

 winged than in the type. The leaf-sheaths are glabrous, and the 

 lanceolate blades start from a narrow base, the midrib forming a 

 keel on the back for a short distance ; in the upper leaves the base 

 is rounded ; the greatest width was 8 lines. 



Note.— Ischasmum rugosum Salisb. is split up by Hackel (Mon. 

 Andropog. pp. 206-8) into three varieties,— var. *. genuinum, var. 

 (3. segetum, and a third. Salisbury's type, a plant collected by 

 Koemg and preserved in the British Museum, should, we presume, 

 under this arrangement, be included under var. *. Hackel makes 

 no reference to it, and probably never saw it. Comparison of the 

 plant with the varietal diagnoses shows that it falls under var. /3., 

 having the rachis-joints and pedicels ciliate, the callus of the 

 sessile spikelet bearded and the awn perfect and about 20mm. long, 

 with a glabrous twisted brown column and a whitish tortuous 

 subula of about equal lengths ; var. «. genuinum, on the contrary, 

 is characterised by glabrous rachis-joints and pedicels, a shortly 

 bearded or glabrous callus, and an imperfect awn 6-8 mm. long. 



