4 
siological importance, and its once supposed systematic value is daily diminish- 
ing more and more , the latter is apparently of as much systematic as physio- 
logical interest. 
If we consider what the plants are which are brought together by this 
circumstance, we shall find that they are obviously in many cases very clearly 
akin, and that in others they leave the series of orders from which they are 
extracted far more natural than while they were among them. Moreover, 
plants are thus brought together which although they have the closest possible 
relationship cannot be grouped together upon any other known principle. For 
instance, Umbelliferas are so nearly the same as Ranunculacese, as wiU be shewn 
under the former order, that they are in reality little more than epigynous or 
hypogynous forms of each other, and yet they are most widely divided by the 
French school. So again with Vitacese and Araliaceae, and with Pittosporacese 
and Dilleniacese. These orders, which are mutually connected by so many 
diflferent characters, are not, so far as I know, capable of being even approxi- 
mated, except by taking the peculiarity of the albumen into account. It may 
appear paradoxical enough to place the Vine in the same category with the 
Crowfoot, and the Gooseberry with Celery ; but it is to be remembered that 
resemblances or differences, affinities or discrepancies, are not to be measured 
in science by the standard of popular prejudice, but by the cool investigation 
of actual structure. It appears most strange to an ordinary observer that the 
Nettle and the Fig should be associated in the same natural order ; but the Bo- 
tanist is well aware how slender the structural differences between these plants 
really are, and consequently how extremely close their relationship undoubtedly 
is. What discrepancies or anomalies are to be found in this albuminous group 
are certainly not appreciable by any but a Botanist. They chiefly attach 
to the genus Dionsea and to Francoaceae, whose relationship to the orders 
with which they are associated is less evident than could be wished. But in 
addition to what is stated under those plants in the proper place, it is to be 
remembered that the group is by no means complete, and that there is nothing 
very violent in the supposition that the links which may be wanted to connect 
the plants in question with their neighbours stiU remain to be discovered. 
It may be m-ged that if the presence of a very large quantity of albumen is 
reaUy so important a physiological circumstance as it is here represented to be, 
I ought, to be consistent, to have added to the albuminous group those orders of 
Incompletse and Monopetalse in which albumen is equally abundant. I am by 
no means disposed to deny the justice of such an observation : on the contrary, it 
is unquestionable that Cinchonaceae, Stellatae, Caprifoliaceae, Menispermaceae, 
and Piperales, to which the observation would principally apply, would be 
better associated with Albuminosae than with the groups in which they at 
present stand. But I confess myself for the present unprepared to work out 
the principle to its full extent, and I hope that it is no objection to the arrange- 
ment I am proposing, that whatever merit it may possess it is capable of being 
rendered better. 
Alliance I. RANALES, 
Essential Character. — Herbaceous, rarely woody, plants, either with the carpels 
more or less distinct ; or if that is not the case, with parietal placentae. 
This alliance may be considered to be in some respects little more than an 
herbaceous form of Anonales, from which it is almost impossible to separate 
it by any very positive character. 
