22 
disk; styles 2, distinct ; stigmas simple. Fruit consisting of 2 carpels, separable from a 
common axis, to which they adhere by their face {the commissure) ; each carpel traversed 
by elevated ridges, of which 5 are primary, and 4, alternating with them, secondary ; the 
ridges are separated by channels, below which are often placed, in the substance of the 
pericarp, certain linear receptacles of coloured oily matter called vittce. Seed pendulous, 
usually adhering inseparably to the pericarp, rarely loose ; embryo minute, at the base of 
abundant horny albumen; radicle pointing to the hilum. — Herbaceous plants, with fistular 
furrowed stems. Leaves usually divided, sometimes simple, sheathing at the base. Flowers 
in umbels, white, pink, yellow, or blue, generally surrounded by an involucre. 
Anomalies. Sometimes there are three carpels. The leaves, or rather the dilated 
leafless petioles of some, as Er^nigium, have the appearance of those of Endogens. Leaves 
opposite in Spananthe. 
Affinities. I trust I have sufficiently proved already the near relation 
of this order to Ranunculacese ; its affinity to- Araliacese is such that the two 
orders hardly differ, except in the number of the parts of the flower. With 
Saxifragacese, Umbelliferse agree in habit, if Hydrocotyle is compared with 
Chr}^sosplenium, and if the sheathing and divided leaves of the two orders are 
considered. To Geraniaceee, De Candolle remarks that Umbelliferae are allied, 
in consequence of the cohesion of the carpels around a woody axis, and of the 
umbellate flowers which grow opposite the leaves, and also because the affinity 
of Geraniacese to Vitacese, and of the latter to Araliacese, is not to be doubted. 
The resemblance of Umbelliferae to Geraniaceae is however veiy feeble compared 
with the likeness they bear to Vitaceae. The arrangement of this order has 
only within a few years arrived at any very definite state ; the characters upon 
which genera and tribes could be formed having been for a long while unset- 
tled ; it is, however, now generally admitted that the number and develop- 
ment of the ribs of the fruit, the presence or absence of reseiwoirs of oil called 
vittae, and the fomi of the albumen, are the leading peculiarities which require 
to be attended to. Upon this subject see Koch’s Dissertation, Lagasca in the 
Otiosas Espanolas, and De Candolle’s Mt'moire, especially the last. The clas- 
sification of De Candolle has, however, been recently attacked by Tausch, in 
the places above quoted, who asserts that the albumen is a fallacious guide. 
He says that some species of Bupleurum are campylospennous, and others or- 
thospermous, and the same is true of many other genera. He adds, that in 
Hasselquistia the fruit of the ray is orthospermous, while that of the disk is 
coelospennous. The arrangement which this author proposes to substitute has 
not yet been examined critically. 
Geography. Natives chiefly of the northern parts of the northern hemis- 
phere, inhabiting groves, thickets, plains, marshes, and waste places. They 
appear to be extremely rare in all tropical countries, except at considerable 
elevations ; where they gradually increase in number as the other parts of the 
vegetation acquire an extra-tropical, or mountain character. Hence, although 
they are hardly knowui in the plains of India, they abound on the mountains 
of the Himalaya. 
Properties. The properties of this order require to be considered under 
two points of view : firstly, those of the vegetation ; and, secondly, those of 
the fmctification. The character of the fonner is, generally speaking, suspicious, 
and often poisonous in a high degree ; as in the case of Hemlock, Fool’s Pars- 
ley, and others, which are deadly poisons. Nevertheless, the stems of the Ce- 
lery, the leaves of Parsley and Samphire, the roots of the Skin-et, the Carrot, 
the Parsnep, the Arracacha, and the tubers of GEnanthe pimpineUoides and 
Bunium bulbocastanum, are wholesome articles of food. Tlie leaves of Prangos 
pabularis, afford a nourishing and abundant fodder for cattle ; it is thought 
to have been the Silphium of the ancients. The fruit, vulgarly called 
the seed, is in no case dangerous, and is usually a warm-and agreeable aroma- 
tic, as Caraway, Coriander, DiU, Anise, &.c. From the stem, when wounded. 
