VI. CRUOIFERiE. 58 



mences, but lengthen out as it advances. Sepals 4. Petals 4, 

 equal, or 2 (on the outer side) larger. Stamens 6, of which 

 2 are generally shorter or very rarely deficient. Ovary solitary, 

 2-celled. Style single, often very short or almost none, with 

 a capitate or 2-lobed stigma. Fruit a pod, divided into 2 cells 

 by a thin partition, from which the valves generally separate 

 at maturity ; or, in a few genera, the pod is 1 -celled or in- 

 dehiscent, or separates transversely into several joints. Seeds 

 without albumen, attached, in each cell, alternately, to the right 

 and left edges of the partition. 



x\n extensive and very natural family, widely spread over the globe, 

 but chiefly in the northern hemisphere ; scarce within the tropics, and 

 in some districts entirely unknown. The number of sepals, petals, and 

 stamens readily distinguish Grucifers from all other British plants, but 

 the discrimination of the numerous genera into which they are distri- 

 buted is a very difficult task. The characters are derived chiefly from 

 the pod and the seed, and are often very minute. It is therefore 

 absolutely necessary, in order to name a Crucifer, to have its fruit ; 

 and to examine the seed, it must be ripe ; it should then be soaked, 

 and the outer coating removed, in order to lay bare the embryo, and 

 observe the position of the radicle on the cotyledons, which affords 

 the most essential among the generic characters. 



A few terms specially made use of in describing plants of this family 

 may require some explanation. The calyx is said to be bisaccate when 

 two of the sepals, a little outside the two others, are broader at the 

 base, forming little protuberances or pouches. The pod is termed a 

 silique or siliquose when linear, at least three or four times as long as 

 broad ; a silicule or siliculose when short and broad — not twice as long 

 as broad ; and a lomentum or lomentose when it does not open its valves. 

 The nerves on the pod, often used as a generic character, can be best 

 seen on dried specimens ; they are sometimes quite imperceptible on 

 the fresh pod. The seeds are said to be in one roio when, from the 

 narrowness of the pod or the length of the seed- stalk, they occupy the 

 centre of the cell, the two rows being as it were blended into one ; or 

 in tivo rows, when the two rows are distinct without overlapping. In 

 the embryo the radicle is said to be accumbent when it is bent down on 

 the edges of the cotyledons, incumbent when bent over the back of one 

 of them ; in the latter case the cotyledons are either flat or conduplicate, 

 that is, folded longitudinally over the radicle. 



Notwithstanding all these nice distinctions, the genera of Crucifcrs 

 as at present defined are often as artificial as they are difficult, under 

 which circumstances I have selected those adopted in the best modern 

 Floras, or such as have appeared to me the most natural. The following 

 table is founded, as much as possible, on less minute characters, but 

 even in the few British species the examination of the seed cannot 

 always be dispensed with. 



( Pod with a longitudinal partition, generally opening in 2 valves ... 2 

 1 -j Pod not dehiscent, with 1 seed, or with several seeds separated by trans- 

 ( verse partitions (Lomentose) 29 



