29 
posed. In that case the alliance would possibly have to be combined with 
either Anonales or Ranunciilaceae. At present I know no more of its affinities 
than is stated under the order itself and Grossulacese. 
Order XVI. BERBERACE^. The Berberry Tribe. 
Berberide^, Fent. Tahl. 3. 83. (1799) ; DC. Syst. 2. 1. (1821) ; Prodr. 1. 105. (1824) ; 
Lindl. Synops. 14. (1829.) 
Essential Character. — Sepals 3-4-6, deciduous, in a double row, surrounded exter- 
nally by petaloid scales. Petals hypogynous, either equal to the sepals in number, and 
opposite to them, or twice as many, sometimes with an appendage at the base in the inside. 
5 'famen 5 equal in number to the petals, and opposite to them; anthers generally with two 
separated cells, opening elastically with a valve from the bottom to the top. Ovary solitary, 
1 -celled ; rather lateral ; orbicular, berried or capsular. .Seetis attached 
to the bottom of the cell on one side, 1, 2, or 3 ; albumen between fleshy and corneous; 
embryo minute, occasionally as long as the axis of the albumen. — Shrubs or herbaceous 
perennial plants, for the most part smooth. Leaves alternate, compound, usually without 
I stipules. 
I Affinities. Botanists appear of one opinion in considering Menisperma- 
I cese the nearest order to this, agreeing in having the stamens opposite the pe- 
tals, the floral envelopes regularly imbricated, 3 or 4 in each row, never 5, the 
j fruit usually baccate, and fleshy albumen. These, however, differ in their ha- 
' bit, the separation of the sexes in distinct flowers, and the presence of several 
distinct carpels, while in Berberacese there is never more than one, which is 
perfectly simple, as is demonstrated by the position of the placentae, the single 
, style, &c. I am, however, of opinion that under all circumstances Menisper- 
maceae must go to Incompletae, and that consequently the supposed relationship 
i of Menispermaceae is more slight than is believed. {See that order hereafter.) 
j; The true affinity of Berberaceae appears to be on the one hand with Grossu- 
1 laceae, with which the genus Berberis agrees even in habit, and on the other 
j hand with Umbelliferae, through the section Nandineae. This latter differs ex- 
ceedingly in habit from the section Berberideae, and shows of how little real 
value external form is in determining affinities. In its pouched petals, and 
general appearance, the section Nandineae may be compared to Fumarieae. 
1 With Podophylleae the order is connected through Leontice and Diphylleia, 
which have a near relation to Jeffersonia and Podophyllum itself. In the sin- 
gular structure of their anthers there is a striking analogy with Lauraceae, 
Atherospermaceae, and Hamamelaceae, orders not otherwise akin to Berbera- 
ceae. Leontice thalictroides offers one of the few instances of seeds being ab- 
solutely naked, that is to say, not covered by any integument originating in the 
I pericarp. In this plant the ovary is ruptured in an early state by the expan- 
sion of the ovule, which, having been impregnated, continues to grow, and 
ultimately arrives at maturity, although deprived of its pericarpial cover- 
: ing. The spines of the common Berberry are a curious state of leaf, in 
which the parenchyma is displaced, and the ribs have become indurated. They, 
as well as all the simple leaves of the other species, are articulated with the pe- 
tiole, and are therefore compound leaves reduced to a single foliole ; whence 
the supposed genus Mahonia does not differ essentially from Berberis in foliage 
any more than in fructification. Berberaceae are related to Anonacese through 
the genus Bocagea ; their ovary is described as being sometimes strikingly 
like that ofDavilla in Dilleniaceae. {FI. Bras. 1. 47.) Aug. de St. Hilaire re- 
marks, that the opposition of the stamens to the petals, and the erect ovules. 
