Astelia.| LILIACES. 709 
sent. Female flowers: Staminodia present. Ovary sessile, broadly 
ovoid or oblong, 1-celled with 3 parietal placentas, or 3-celled with 
the placentas in the axis; ovules numerous on each placenta ; style 
very short; stigma 3-lobed. Fruit an indehiscent more or less 
fleshy oblong or ovoid or subglobose berry. Seeds several, ovoid or 
oblong, straight or curved, terete or angular; testa black, crus- 
taceous ; embryo small, cylindric ; albumen fleshy. 
A small genus of 12 or 13 species, with its headquarters in New Zealand, 
but with 1 species in south-eastern Australia and Tasmania, 1 in Fiji, 2 or 3 in 
the Sandwich Islands, and 1 in antarctic America. In New Zealand it forms a 
prominent part of the vegetation, especially in the northern forests, to which 
one or two of the species often give a peculiar aspect. The species are by no 
means easy of discrimination, partly from a certain amount of similarity in the 
foliage, and partly from the flowers being dicecious, thus making it difficult 
to match the sexes. When dealing with fresh specimens these difficulties in 
great measure disappear, particularly if due attention is paid to the structure of 
the ovary and the size and shape of the ripe fruit, both of which afford excellent 
characters. The student will find some valuable remarks on this point in 
Mr. Kirk’s notes on the genus, published in Trans. N.Z. Inst., Vol. iv., pp. 241— 
247. With respect to the remarkable diversity existing in the genus in the 
placentation of the ovary, reference should be made to the ‘‘ Flora Antarctica,” 
Vol. ii., p. 357. 
In the following arrangement I have adopted Hooker's identification of the 
two species originally described by Cunningham. But Cunningham’s diagnoses, 
such as they are, do not satisfactorily match Hooker’s plants, and it has been 
suggested that he had other species in his mind. The question is one which 
cannot be settled, if settled at all, without reference to Cunningham’s collections, 
all of which are outside the colony, and beyond my reach. But taking into 
account Hooker’s remarks in the ‘“ Flora of New Zealand’’ (Vol. i., p. 251), 
where he says, ‘‘ Cunningham confused all the species and sexes, examined 
none, and referred at random to Banks and Solander’s drawings and notes, 
substituting names of his own for theirs,” it appears highly doubtful whether 
such an examination would be at all conclusive. An alteration of the present 
nomenclature, which has received universal acceptance, would be a matter much 
to be deplored. 
I have had much trouble with the species, 10 in all, described by Colenso in 
the Trans. N.Z. Inst. Unfortunately, few of them are repr-sented by named 
specimens in his herbarium; and his descriptions are so vague, and so much 
overloaded with trivial details, that it is difficult to come to an opinion respect- 
ing them. But I can see no grounds for supp sing that they are really distinct, 
or represent anything more than individual differences. 
A. Berry 1-celled. Ovules attached to 3 parietal placentas. 
Small, subalpine. Leaves 1-8 in., glabrous or scaly. 
Scape few-flowered. Berry oblong, 4-4 in. long sells Aig LULEATUS. 
Leaves 2-5ft. x 4-lin., glabrous or silky. Scape 
panicled, many-flowered; female prostrate in fruit. 
Berry globose, tin. diam. Seeds terete 46 .. 2A. Cunning- 
ham. 
B. Berry 3-celled. Ovules attached to the inner angles of the cells. 
* Perianth not enlarged or coloured in fruit. 
Leaves 2-6ft. x 4-l}in., not conspicuously 3-nerved. 
Female scape stout, erect in fruit. Flowers }in. long. 
Berry 4in. long, ovoid, purplish-black. Seeds angled .. 3. A. Banksii. 
