Tas, 8545. 
KNIPHOFIA carrata. : 
South Africa, : 
LILIAcEAE. Tribe HEMEROCALLEAE. 
‘Kyrrnorta, Moench; Benth. et Hook.f. Gen. Plant. vol. iii. p. 775; A. Berger in 
Lingl. Pflanzenreich, Lil.--Asphod.-Aloin. p. 81. 
+ ileal 
Kniphofia carinata, C. H. Wright; species K. comosae, Hochst., affinis, fila- 
mentis luteis, perianthio vix duplo longioribus differt. 
Herba perennis. Folia 7+5 dm. longa, e basi 2°5 em. lata ad apicem acuminatum 
gradatim attenuata, supra profunde canaliculata, subtus acute carinata, 
tenuia, glabra, marginibus levibus anguste albo-cartilagineis instructa. 
Scapus cylindricus; bracteae lanceolatae, longe acuminatae, scariosae, 
8 mm. longae, 2 mm. latae; racemus 15 cm. longus, densiflorus; pedicelli 
breves ; flores nutantes. Perianthiwm claro-luteum ; tubus anguste urceo- 
latus, supra ovarium leviter constrictus, 18 mm. longus, prope apicem 
5 mm. diametro ; lobi erecto-patentes, obtusi, 8 mm. longi, 3-5 mm. lati, 
Filamenta circiter 30 mm. longa, perianthio concolora; antherae oblongae, 
3mm. longae. Ovariwm ovoideum, 4 mm. longum, trilobum, leve ; stylus 
subulatus, staminibus paullo longior. Ovula plura.—C. H. Wriaat. 
The Kniphofia which we depict is a South African species 
for the introduction of which horticulture is indebted to Miss 
Ayliff of Rose Cottage, Grahamstown, South Africa, by 
whom seeds were presented to Kew in 1892. The plants 
raised from these seeds have been grown in ~ in a cool 
frame, where, however, they throve indifferent y until 1912, 
when they suddenly began to make vigorous growth and 
flowered for the first time in September of that year. 
They flowered again in September, 1913, when the oppor- 
tunity was taken of preparing the present illustration. 
During the twenty years that this species has been in 
cultivation without flowering at Kew it has also been tried 
in the open border with other species of Kniphojia, but has 
never in that situation proved a success owing to its 
being manifestly less hardy than the majority of the species 
under cultivation, The species when in flower was seen by 
Mr. Berger, who at once recognised it as one not enumerated 
in his recent scholarly monograph of the genus. It is 
apparently most closely allied to the Abyssinian A. comosa, 
Marcu, 1914, 
