128 in cyn:es. Or-O.-T^. produced above its union ^i'dth the ovary into e 



short, distinct tube and 4-5-lobed above, or lobed neevlj or quite 

 do'"n to its union vrlt^. the ovrrv; tvro of the lobes often large and 

 le!?f^. Petals numerous, slender, t ineer or linear-filiform, united 

 pt, the brise into a short, distinct tube and arising around the top 

 o-^ the ovar-"". Stpmens numerous, erect, arising: frora the corolla- 

 tube; filercents severe"^ times as lon?^ es the anthers. Stigmas 

 usually 5, rarel;"- 4, erect, filiform. Ovary partly or more than 

 he If- superior, usuallj?" 5, rarely 4-celled; rlscentas axile. Capsule 

 with 5, rarel-"- 4, valves and cells; valves vith the expanding- 

 kee'is contiguous, forming a central and often acute keel, v/ith a 

 broad infleTed or erect sub -membra nous flap or iving on each side; 

 c^Ils OT^en, without cell--'"inps or tubercles. Seeds numerous in 

 eac^ cell, small, compressed, somewhat D-sbaned, very minutely 

 tube-^cijia te or a-nooth. — N.E.Bp. ±^1 l^e hardeners* Chronicle, 1925, 

 Vol.. LICTJIII, n. 412, and in ^'hilliDS, ^en. of 3. Afr. Fl. Pi., 

 p. 245. 



ST>ecies seveal, natives of South,>frica, the i-^diterranean 

 region, Arabia, Persia, Bo^nchn stan , Hirdisten and -ti-tlsndtic is- 

 lands, pnd naturalised in Opiifornis. 



The generic name is derived from the_GTeek, kyyos, ice, and 

 nhyton, a ■nlcnt , because the ■'."■ell kno'-.-'n "-'•ce-plant ' (-1. crj'^stallin- 

 um, I^. "S. 3r.) is the t^^-pe of the "genus. 



The illustration of the generic structure (^'ig. 54) is dravm 

 from a f lover of 0. grandiflorum, K.S.;:!p., ^'^hich is nearljr allied 

 to the type. 



It would appear that the broad, i"ing-like margins of the valves 

 of the C'^'^sule in this genus (as in others of similar structure) 

 develop ge-^ias so as to enfold some of the seeds betv;een themselves 

 and the expanding keel of the v^lve, and when the rain soaks in 

 so as to ■'•'et the expandina--keels and cause the valves to open, 

 these vnn^p clasn some of the seeds so that they are torn from their 

 funicles (stalks) and raay then be easily v/ashed out of the capsule 

 by the rain. A very ingenious and effective arrangement, ensuring, 

 the eas;"- dispersal of the seeds, and v/idely different from, the stru- 

 cture of the cansules and mode of dispersal of the seeds in such 

 o-ener" as C^rTanthea and ^lotti'ohyllum, xvhere ever^r r^recaution 

 see^s to exist to i^revent the seeds from getting out of the cells 

 of the cansule, althou.gh they certainly do become dispersed, but 

 I have failed to discover ho-"^. 



Ti'ei," species ere in cultivation, as m.ost are not shoT^^y plants, 

 and those that are, being annuals or biennials (although C, Bark- 

 lyi appears to be of lonpier duration), reruire to be renev/ed from 

 seeds, ""'hich doe s not alva^/s ri^^en in the British ^sles. '-'n the 

 Continent they might do better. Therefore, as J^ am unable to m.ake 

 a ke3'' to al^ the srecies from living material, -'■ merelj^ give a list 

 of the species des-r-ribed •'•'ith descriptions of som.e that are incom- 

 pletel-"^ described, and the new ones that I place under this genus, 

 grouped accord^' rg to the shape of the leaves and colour of the floxv- 

 ers. As in previous gener'^, thp letter ^^ in the synonymy stands 

 for --esembryanthemum. 



2,— FLOV-TSRS \7HITE. 



1.— C. Aitonis, IT. E. Bj.... .... .^itonis, "rcc., ^^ort. Vind. 



Vol. Til, p._3, t. 7;_HaTv. I'isc. Nat.^, p. 48, "^ynon. 247,-, end Rev. 



158; Berger, '"es. und Port., p. 38. ^-'^. angulatum, -^hunb. ""^rod. 



91, rnr"- Fl. Cr^, ed Schult. 425; Merger, ^>-es. und Port., 36; i^'.E.Br. 



