Mr. George Dickie on the r umariaceae. 441 



ullel, the two outer either one or both saccate at the base, 

 the two inner callous and coloured at the apex^ where they 

 cohere and inclose the anthers and stigma. Stamens six, 

 in two parcels, opposite the outer petals, very seldom all sepa- 

 rate. Anthers membranous, the two outer of each parcel one- 

 celled, the middle one two-celled,^^ &c. 



Lindley, in his ^ Introduction to the Natural System^, con- 

 siders the parts as divided upon a binary plan ; the bodies 

 commonly called deciduous sepals he considers as bracts, the 

 outer series of the supposed petals he considers as a calyx, 

 and the inner only as petals. He also considers the lateral 

 one-celled anther of each parcel as belonging to a common 

 stamen, the filament of which is split by the separation of the 

 two parcels. This opinion might appear at first sight forced, 

 in order that the parts may conform to the arrangement in a 

 normal flower ; if however the structure of the parts be exa- 

 mined it will be seen that this opinion is exceedingly plausible. 



Each common filament is traversed by three distinct fasci- 

 culi of vessels ; the central one, which passes to the two-celled 

 anther, is double the size of those which pass to the lateral 

 anthers; their relative position must also be attended to. 

 They are placed at equal distances from each other, but the 

 two lateral ones are very near the margin of the filament. In 

 all perfect stamens of different plants which I have examined 

 there has been seen only one bundle of vessels, which passed 

 straight through the centre of the filament to the anther. 



If the arrangement in Fumaria is normal, it is then a re- 

 markable exception to the usual method. It appears much 

 more probable that the filament is compound, and composed of 

 one entire filament and portions of two others. In some spe- 

 cimens of Fumaria the adhesion of the lateral portions is very 

 slight, and they may be readily separated from that in the 

 centre with very little force; and in very young flower-buds 

 there is no adhesion, the single-celled anthers are quite sepa- 

 rate, and the apices of the inner petals as they are commonly 

 called are also free. It has been already mentioned that the 

 fasciculi of vessels passing to the single-celled anthers are very 

 near the margin of the common filaments : now in some of 

 those cases where a partial adhesion only is found, these ves- 



