CLASSIFICATIONS OF ORDERS AND GENERA. 
lvii 
Erect trees. Leaves stipulate. Ovary 3-celled . . . Fagus, p. 248. 
Trees or shrubs. Anthers solitary in the scales of a 
cone. Ovules solitary or few in the axils of scales, 
without ovary or perianth Conifer.®, p. 255. 
CLASS XXIV. CRYPTOGAMIA. 
(See p. lxvi.) 
III. CLASSIFICATION OF THE NEW ZEALAND NATURAL ORDERS ACCORD- 
ING TO THE NATURAL SYSTEM, AS ADOFTED IN THE PRESENT WORK. 
The number of New Zealand genera is so small, and many of them are so exceptional in 
characters, that they afford but an imperfect idea of the relations that subsist between the 
Orders they belong to ; or between these and the same and other Orders in the world at 
large. The natural sequence of the Orders of plauts and their relationships being determined 
by the characters of the majority of all the known genera which they severally include; 
this sequence and these relationships cannot be demonstrated by a small flora, like that 
of New Zealand, in which many large Orders and more small ones are unrepresented, or 
represented by one or two uncharacteristic genera only. 
The following system is not altogether natural, but it is so in the main, and as much so 
as any hitherto devised. The primary divisions Phaenogams and Cryptogams are perfectly 
natural and well-defined, and the Subclasses and Orders of Cryptogams follow, on the whole, 
a very natural sequence. The two Classes of Pham ogams, Dicotyledons and Monocotyledons, 
are also natural and well defined, as are, on the whole, the Subclasses and Orders of Mono- 
cotyledons. With the Dicotyledons the case is different : the subdivisions Angiosperms and 
Gymnosperms are distinct and natural; but no natural arrangement of the Orders of An- 
giosperms has yet been devised. The Five Subclasses ( Tlialamifiorce , etc.) are only in so 
far natural, that each consists of Orders more or less naturally related to oue another; for 
the cross affinities between certain Orders of all the Subclasses are very numerous, and 
in some cases so strong that the single technical character of the Subclass aloue keeps 
them where they are. In the case of Subclass V., Incomplete , matters are still worse, the 
Orders of this Subclass having no common relationship, but consisting of Thalamifloral, or 
Calycifloral, plants in which the perianth happens to be incomplete or absent. There- 
fore it must be borne in mind, that the Subclasses of Angiosperma are merely artificial de- 
vices, to enable us to find our way to the Order we seek ; and that these Orders are con- 
nected by so many cross affinities that no one has ever yet been able to arrange them in a 
linear series. 
For further information the reader is referred to Lindley’s ‘Vegetable Kingdom,’* which 
contains an excellent history of this subject, besides being the best Encyclopaedia of all that 
relates to the Natural Orders which we possess. 
A. Phcenogamic or Flowering Plants. 
Class I. DICOTYLEDONS. 
Stem, when perennial, furnished with pith, concentric layers of wood and 
bark. Leaves usually with netted venation. Organs of the flower generally 
4 or 5 each, or multiples of those numbers. Seeds having an embryo with 
2 cotyledons. In germination the radicle lengthens, and forks or branches. 
The exceptions to one or other of these characters are too numerous to be indicated. 
* A work that should be in every naturalist’s library (one thick volume, 8vo, with 500 
woodcuts), price 36s., 3rd edition. 
9 
