202 LILIACE^, [Endogens. 



Yucca, which has the hard leaves of Phormium, with wliich however its distmct sepals 

 and petals forbid its being associated. 



With the SciLLE^ or Squills, we reach a division of the Order, aboundmg in beauti- 

 ful species, all of which are bulbous, with annual stems. Theu- peculiarity resides in 

 the anthers not being so Ughtly attached to the filaments as in Tuhpeae, and in the 

 leaves from whose axils the flowers proceed, acquiring a membranous condition. 



CoNANTHERE.^ are Squills with the ovary partially adhermg to the calyx and corolla, 

 and springing from tubers, not bulbs. They offer a direct transition to AmarylUds. 



Antuerice^ or Asphodels, agree with the last in having tubers or fleshy fascicled 

 roots and not bulbs, but theii- ovary is free ; they are therefore tuberous or fibrous 

 rooted Squills. Chrysobactron, a genus gathered by Dr. Joseph Hooker, in Auck- 

 and and Campbell's islands, is described as dioecious, but apparently is polygamous. 

 The fruit in these three last Orders is a capsule. 



APHYLLANTHEiE are plants with the habit of Rushes, and the bracts so membranous 

 and closely imbricated, as to give the appearance of Xyi'ids when the flowers are past. 

 They seem to form a connection between LiUes and some plants of the Juncal or X;yTidal 

 Alliances. The genera have been very insufficiently examined. Xanthorrhseas, called 

 Grass Trees in New Holland, are very different m habit from the remainder ; their 

 shrubby stems, which emulate small Palm trees m appearance, bear tufts of long 

 why foUage at their extremities, from the midst of which rise very long cylindrical 

 spikes of densely compacted flowers, resemblmg Bullrushes (Typha). By this genus 

 the Aphyllanths completely join Rushes, for the genus Kingia, included ui Rushes, 

 because of its minute embryo, has entu"ely the aspect of a Xanthorrhrea. 



The reason for refen-ing Wachendorfe^ hither, have been given in speaking of the 

 Bloodroots (p. 152). They are plants with ensiform or plaited leaves of a hard textm*e, 

 fibrous roots, with flowers usually in panicles and by no means remarkable for size or 

 bright colom'ing. If it is really true that their carpels are opposite the petals, as is 

 said, they will undoubtedly have to be removed from their present station. 



AsPARAGE^ are Lilies with a succulent fruit. They consist of plants extremely dissi- 

 milar in appearance, the common Asparagus and the Lily of the Valley being associated 

 under this title. In general their leaves are broad ; in the genus Cordyline they even 

 acqvure the expanded form and diverging veuis of the Amomal Alliance. Theu* stems, 

 although among the dwarfest that the Lihes comprehend, are in the common Asparagus 

 branched and of considerable size, and in the Dragon-trees they acquire the dimensions 

 and age of large trees. A tendency to the separation of sexes occurs here on the part of 

 the genus Ruscus ; but it is not carried so far as to constitute a diclinous structure. 

 According to Von Martins {Choix. p. 21.), the position of the sepals in Lily worts (in 

 which he includes Asphodelese) is V with respect to the axis ; while in Asparagese it 

 is A. He also finds thi'oughout the Lihaceous Order that the petaUne stamens are 

 larger and more perfect than the sepahne, it being the latter moreover which disappear 

 when there is any deficiency in the usual number of stamens. 



With respect to Aspidistre^, concerning whose structure we have very insufficient 

 information, they are principally kno\vn by a large muslu'oom-shaped stigma. Their 

 foliage is that of Giugerworts ; their flowers are dingy purple or gi*een, with a campa- 

 nulate perianth, on whose sides the stamens are inserted. In many respects they are 

 very hke Orontiads, to which, perhaps, they ought to be referred. 



In like manner the Ophiopogone^, or Teatworts, have a foliage hardly belonging to 

 Lilies, Pehosanthes Teta resemblmg a Ginger more than a plant of this Order. They 

 are remarkable for their seeds bursting through the sides of the ovai'y at a very early 

 period, growmg fi'eely though exposed to air, and finally acquiring the succulent ap- 

 pearance of a tuber. It is very uncertain whether they have any real claim to the rank 

 of Lilies. 



If we suppose that the doubtful members of this great Order are removed, we shall 

 find that its most immediate relations are as follows. From the Melanths it chiefly 

 differs in its anthers being turned inwards, and its carpels quite consolidated. To the 

 Amaryllids it approaches so nearly that there is perhaps nothing to separate them except 

 its free ovary ; and the group Conantherese exhibits a structure intermediate in this 

 respect. With Rushes Lilyworts are brought mto close contact by means of the Aphyl- 

 lanths as has been already stated. Towards Arads they extend in the dii'ection of Oron- 

 tiads, through the intermediate group of Aspidistrese. Finally, it is here that Dictyo- 

 gens are reached by means of the Asparagese, which, by most botanists, are actually 

 made to comprehend the genus Smilax and the Parids. For the affinity of Lihes and 

 Palms, the reader is referred to the observations under the latter Order.' 



The geographical limits of the Order are as wide as its diff"erences of structure. Upon 

 the whole, however, tlie species are much more abundant m temperate climates than 

 in the tropics, where they chiefly exist in an arborescent state. Aloes are mostly found 



