2 
ultimately bursts : Exogens have two or more cotyledons : Endogens have but 
one. In this country the trees and shrubs, and larger herbaceous plants, are 
nearly all Exogenous; while our native Endogens are chiefly confined to 
grasses, sedges, orchises, bulbs, and submerged water-plants. To this it may 
be added, that the flowers of Exogens are usually formed upon a quinary or 
quaternary type, while those of Endogens are most commonly ternary. 
Although the difference between Exogens and the other four classes is in 
general very precise, yet there are certain points at which the distinctive cha- 
racters become less obvious than usual ; and where, in fact. Exogens seem to 
be in a state of transition to other parts of the system. For instance, some 
species of Ranunculus are strikingly similar to the genus Alisma, both in gene- 
ral appearance and in structure, and thus establish a transition to Endogens ; 
Nymphseacese on the one hand, and Hydrocharacese on the other, Menisper- 
macese and Smilaceee, Aristolochiaceae and Araceae, are further examples of the 
same fact ; while Callitrichaceae or Ceratophyllese are instances of a reduced 
state of organization in Exogens, analogous to that of Pistiaceae among Endo- 
gens ; and all these three, especially the last, connect their several classes 
more or less obviously with those forms of Acrogens, which like Marchantiaceae 
have their stems and leaves all fused as it were into one homogeneous mass. 
It is a point of great difficulty to form any good natural divisions of Exo- 
gens which shall at the same time be tolerably weU defined, and yet not offer 
violence to strong natural affinities. The Monopetalous, Apetalous, or Poly- 
petalous structure of the flowers, is what is generally adopted ; but it must 
be confessed that the numerous exceptions which occur to the characters of 
these divisions, render them sometimes embarrassing to the student. For 
instance, Stackhousiacese, which are by common consent placed in the vicinity 
of the Polypetalous Rhamnaceae, and Euphorbiacese, are to all appearance 
Monopetalous ; Glaux, again, which is Apetalous, is unquestionably a genus of 
the Monopetalous Primulaceous order ; and the cases of Apetalous genera and 
species belonging to Polypetalous orders are extremely numerous. Yet I 
doubt whether any better method of division than that of Polypetalae, Apetalse 
or Incompletae, and Monopetalse, is likely to be devised ; and at all events 
nothing with the slightest claim to adoption has as yet been brought before 
the public. I therefore adhere to the customary division in the following 
arrangement, premising only that the various exceptions that have just been 
mentioned produce in a much greater degree theoretical than practical diffi- 
culties, and that the student will do well to dismiss them altogether from his 
mind until he has become in some degree familiar with his subject. 
Sub-Class I. POLYPETAL^. 
Essential Character. — Floral envelopes consisting of both calyx and corolla ; the 
latter composed of distinct petals. 
MTiatever reasons there may be for adhering to the characters of the 
French school for the sub-classes of Exogens, there does not appear to me to 
be any sufficient motive for following their system in regard to the groups 
subordinate to the sub-classes. In Polypetalse there are only two, or at the most 
three, distinguished by the origin of the stamens ; which are either hypogynous, 
perigynous, or epigynous. Now such groups have neither the merit of being 
well defined, nor of fomiing natural combinations, nor of breaking up the orders 
into assemblages of small extent ; on the contrary they abound in exceptions, 
and are practically liable to much uncertainty ; they place side by side plants 
having but slight affinity, as Papaveracese and Crucifera', which coiTespond in 
little excej)t the form of the fruit of certain species, they separate orders like 
