104 Doa , 5478, and a cultiveted plpnt. 



Dried specimens of this plant are so like those of C. communis 

 that until I noticed the dark fuscous membranes of the calyx-lobes 



1 had considered it to be that species, but these indicate a dif- 

 ference that lA'Ould -probably be more apparent if living plants of 

 them tv7o species v.^here compared side by side in flov/er and fruit. 

 In its fuscous calyx -membranes it agrees with C. cspensis, but 

 differs from that species in its smooth and terete, not rough and 

 Rnp"ular nedicels. I'r. T. K, Leslie sent me a photograph of a flov/er- 

 ing branch, produced at ?ig. 42, together with seeds of this spe- 

 cies, from v/hich I raised a plant that dies after flov/ering. Its 

 origin v^^as unknovmnto I-r. Leslie. 



11. G. Iluirii, K. S. B^. ^ in Gard. ^hron. , 1928, L:ca:iV, 

 25? (?-ig. 43). — Rootstock a fleshy cylindric or Hadish-like tuber, 

 8-17 inches long and 4-] inch thick, deeply descending and emitting 

 from the top a sessile tuft of leaves end 5-17 radiating, pros- 

 trate or decumbent flov;ering branches 6-15 inches long and In- 

 line thick; the whole plant (except petals and stamens) glabrous. 

 Leaves of the central tuft 4-7 inches long, 2-3 lines broad and 1-^- 



2 lines t^ ick, those on the branches smaller, somev.hat sharply 

 triangular in section, with slightly concave faces, acute, entirely 

 green or with the basal part dark purple passing into light viola- 

 ceous, glaubous, Fiovrers three to several to a branch. Pedicels 

 Z-7> inches long, 1 line thick, terete, tapering upv^^ard, smooth, 

 prreen, Gpiyx with the ovary-part shallov/ly hemispherical, 5 lines 

 broad and 2 lines deep, with 5 slight ribs, each ending between 

 the bases of the lobes in a small, blunt tooth; lobes i-lj^ inch 

 lonfT, much longer than the petals, linear-trigonous from a broad, 

 ovate base, the inner with broad fuscous, or flackish membranous 

 m-ars-ins. Corolla I2-2 inches in diameter, opening in daytime; 

 TDetals in 5-6 series, snreading in different planes, the outer 5-9 

 lines lonc', 1/3 - 2/3 line broad, linear, very acute or acuminate, 

 ciiiate at the basal part, with short, fine hairs; the inner grad- 

 U'^ily smaller with hair-like, curly points; all bright lemon- 

 yeiTow, the outer dull red or ^urpiish-red on the back, ■^taminoder 

 filiform, bearded at the base, with tortuous hair-like tips inter- 

 minerling in a v.'eb-like manner, lemon-yell ov/. 2-tamens about 2§-3| 

 lines Ion;?, erect, v/ith recurved tips, filiform, pale yellow. 0- 

 vary flattish at the top vdth the centre slightly raised, light 

 green. Cepsule 11-14 lines in diameter, with the cone on the top 

 about 5 lines high and v/ith 10-20 valves and cells, the rim around 

 the base of the cone about l-l| line broad. Seeds rather more 

 than ^-line in (?iameter, compressed-globose, smooth, mottled with 

 lip:ht and dark brovn, not shining. 



Hiversdale Division : In sandy places to 100-500 feet altitude, 

 -'uir 999 and 4128. Flowering from August to December. 



To this species I feel almost certain the plant figured_^5s 

 Chrvsanthemum aizoides af ricanum triangulari folio, ^reyne, Sxot. 

 Pi., p. 163, t. 181 (1678), should be referred, as it has the seme 

 kind o-^ foliage and flowers, and the root is stated to be like that 

 of G. fusiformis but larger. As it was in cultivation in 1568, this 

 plant is one of the oldest known members of the i'^sembryanthemum 

 e-roup, and the second species of Conicosia brought into cultivation. 

 Bregma's figure is copied in '.Prison's Plant. Kist., Ill 12, 507, 

 t. 5, -^ . 11 (iS'^o) and has also been wrongly identified as being 

 ''. molle, 3preng. (^ M. molle. Ait.) by Klinsmann. 



Dr. l^iir informs me that the Dutch name for this plant is 



