BANUNCULACE2E. 63 



in number' like those of the Pa3onies. Each consists of a one-celled 

 ovary, tapering into a short style, which swells at the tip into an 

 oblique discoid stigmatiferous head. In the inner angle of the ovary 

 is a double placentary cord, bearing in two parallel rows the 

 indefinite, horizontal, anatropous ovules, each having a little ciliated 

 frill round its hilum. This becomes an aril with long filaments 

 around the reniform seed,- which contains within its thick coats a 

 curved, fleshy albumen surrounding the embryo. The fruit is dry 

 and dehiscent.^ C. caJifornica, the only known species, is a small 

 branching shrub,^ with alternate, simple, obovate-oblong leaves 

 tapering at the base, with a short petiole and a penniveined blade. 

 The flowers are solitary and terminal. 



Now alone, after all the genera of this family are known to us, 

 are we qualified to study its general characters. Some are constant : 

 in all the genera studied, we have observed that albumen is always 

 present surrounding the embryo ; that the ovule is entirely or 

 incompletely anatropous ; that the pieces of the perianth and of the 

 androceum are free from all adhesion ; and that the number of 

 stamens is never strictly defined.* 



Other characters, not absolutely constant, are very frequently 

 observed, and have therefore great value. Such are : the alternation 

 of the leaves f the absence of stipules -^ the spiral arrangement of 

 the parts of the flower f the independence of the carpels ; the convex 

 form of the floral receptacle ; and hence the hypogynous insertion of 

 the exterior whorls." 



Others, finally, are essentially variable, and hence can only serve 

 to distinguish altogether secondary groups. These are : the per- 

 sistence of the pieces of the perianth around the fruit,'" and the 



' There are said to be from three to five. by their hitcral wings, come very near to true 



- These seeds recal in their conformation stipules. Generally the Ranunc-nlaceee are with- 



those of several Menisipermaceai — among others out stipules, but it is difficult to refuse the name 



the Indian Berry (Pr. Coque dti Levant). to the lamclhe seen at the bases of the leaves of 



^ Separating, it is said, into two valves. ThaUctrum, Isopijrum, &c., and especially of the 



"• The bark of the branches is said to be very floral leaves of the latter. 



bitter ; and the dry leaves are slightly so, — a ■'* In the Columbines there ajijiear to be true 



character which, joined to some others, suggests verticils ; but the rays of stamens are perha])s 



affinities between it and the SimaritbacecE. only very conspicuous secondary spirals, vertical 



Multis notis Simarubaceis convenii, sed recedit or nearly so. 



staminibus ao , ovulo, et arillo (Bentii. & '■* The Pwonics and Crossonoma are alone jieri- 



HoOK., I. cit.). gynous. 



^ Characters which by themselves have no '" Considerable value has been attached to this 



taxonomic value, being found in many families. character, which it has been said distiniruislie."' 



^ The genus Clematis is the only exception. this order from Dil/eniaretr. Wc jjivc it hardly 



' There are petioles with dilated bases which, any, as wc have said several times. ^Ve knuw. 



