[ Narcissales.J 



TACCACE^. 



149 



Order XLIII. TACCACE^.— Taccads. 



Taccese, Pm^. Ikliq. Hcenk. 1. 149. (1830) ; Bartl. Ord. Nat. 82. (1830).— Taccaceje, Key &c 70 

 (l^Zb) ; Ed. prior, ccxxxix. ; Endl. Gen. l\in.— 3Ieistier, p. i02. 



Diagnosis.— NarcisscdEndoge7is ivitli tuhular half -tripetaloideous flowers andfleshij albumen. 



Large perennial herbs, \vith a tuberous root. Leaves all radical, stalked, undivided 

 or pedatifid, the segments pinuatifid and entire, with curved parallel veins. Stipules 0. 



Fig. XCVII. 



Flowers placed on the top of a simple 

 taper or angular fiu-rowed scape, m umbels, 

 , regular, siuTounded by undi^'ided bracts 

 forming an involucre. Perianth adherent, 

 ^^ ith a cyhndincal ribbed tube ; Umb petaloid, 

 the petals rather the longest, persistent. 

 Stamens 6, inserted into the base of the 

 ^-?si<^'AC;;^;^^p^^^^7'^ *^<, '^aLJ^ segments of the perianth, distinct ; filaments 



\i ^T~S^SSr-/ -*^'-x "x dilated, petaloid, hooded at the apex ; an- 



thers inserted Ijelow the pouits of their 

 filaments in their concavity, 2-celled, the 

 cells distinct. Ovary composed of 3 con- 

 nate carpels, 1 -celled, or half 3-celled, mth 

 3 parietal polyspermous placentae ; ovules 

 ascending and anatropal, or horizontal and amphitropal ; styles 3, connate ; stigmas 

 connate at the base, radiating, 2-lobed. Pei-icarp berried, indehiscent, 1 -celled, or half 

 3-celled, many-seeded. Seeds Imiate or somewhat ovate, striated. Albumen fleshy. 

 Embryo placed inside the albumen in the region of the hilum, or remote from it. 



Personally I have had no opportunity of examining critically the plants which com- 

 pose this small Order. They are in some respects hke Arads, in others hke Gmger- 

 worts (Tacca Isevis) ; but certainly have nothmg to do with Dicotyledons. Blume has 

 the followmg remarks upon Tacca. Enum. 1. 82. " The genus Tacca offers the 

 type of a new family between Araceee and Aristolochiacese. To the former it approaches 

 closest in habit, especially in the leaves, but it is very different from them in the 

 structure of the parts of fructification. For in no species of time Aracere is a corolline 

 perianth, properly so called, to be foimd ; what we have the custom of calling so in 

 Dracontium and others, is nothing but scales, and not even a calycine uitegument ; the 

 perianth is, moreover, superior in Tacca. By this superior perianth the affinity with 

 Aristolochiacete is evident ; but from those too Tacca diff'ers in the situation of the 

 stamens, which are not as in that Order adherent to the pistil with the anthers opening 

 outwards, but are placed on the perianth itself with the anthers tiu-ned inwardly." In 

 Tacca it is probable that there are several germmating points upon the embryo, analo- 

 gous to the double or triple plumule of Dracontium ; hence embryos of such a kind may 

 be said to be tubers formed in the fruit itself Brown long since stated {Prodromus^ 

 1810) that a relation is estabUshed between Arads and Bu'thworts by means of 

 Tacca. See also Agardlt's Apliorisms, 245. For my own part, however, this resemblance 

 to Birthworts seems so very slight as to be unworthy of notice. The true relation 

 is with the Arads, or at least with those plants which are now separated under the 

 name of Orontiacese, of which these seem to be an ei^igpious form. Endlicher compares 

 them with Yams, to which they appear to have even less resemblance than to the Birth- 

 worts. 



Fig. XCVII.— 1. Tacca integrifolia ; 2. fniit of T. pinnatifida : 

 removed ; 4. section of its albumen and embryo. — Gcertmr. 



3, seed of do. with half the testa 



