years past in ttie Botanic garden at Montpellier, where it 

 had beeji obtained from a collection at Bourdeaux. It 

 has been referred to Scilla by M. De Candoile, and specifi- 

 cally named by that celebrated botanist, from the circum- 

 stance of the blossom expanding after mid-day. 



We suspect the plant to be of the same species with a 

 dilapidated sample from the Cape of Good Hope, preserved 

 in the Banksian Herbarium under the title Anthericum 

 scabrum ; but from the imperfect state of the specimen we 

 offer our opinion as a mere guess. 



Rootstook bulbicipitous (terminated by a bulblike bud), 

 with broWnii^ fibrous coarsely reticulated integuments. 

 Leaves radical, buUrush-shaped, several, ambient, somewhat 

 flaccid, broadly subulate, 3-sided, channelled with a sharp 

 keel beneath, glaucoiDs at the upper surface and streaked 

 with shagreenly roughened nerves, outer ones about a foot 

 and a half long, revolute along the edge, smooth and green 

 underneath. Fhwers fugitive, tender, twisting their seg- 

 ments together as they decay, white, scentless. Peduncles 

 one-flowered, green, round, smooth : cotolla about an inch 

 and a half across, sixparted, recurvedly stellate, converging 

 downwards; segments linearly ligulate, obtuse, upwards 

 wide apart, marked down the middle of the back by a pur- 

 plish green line, inner ones rather broader and blunter. 

 Stamens even with the corolla, uprightly divergent : JUaments 

 very shallowly fixed to the base of the segments, filiformly 

 subulate, smooth, somewhat widened downwards and in- 

 wardly channelled. Germen green, oblongly tapered, se- 

 veral times shorter than the style, roundedly 3-comered, 

 and marked with six paler lines ; cells 2-seeded : style 

 setaceous, inclining slightly upwards with a gentle curve, 

 even with the stamens ; terminated by an inconspicuous 

 notch for the stigma. 



