I 

 NATURAL CLAJSfsll ICATIOX. 13.5 



bably a consequence of abortion or union, as we have al- 

 ready remarked of the Labiatae, and as in tlie Glt'dilschia 

 triacantha and Spartium scoparium wc often actually observe 

 two pistils more or less united. 



The absolute size of organs is a very insignificant cir- 

 cumstance in the theory of classification. A more innx)rtanl 

 consideration is the proportional size of similar or dissi- 

 milar parts. Wc may assume the following rule as a funda- 

 mental law of vegetation. The parts of one and the same 

 system are by nature equal in size, and only become unecjual 

 from changes, which are more or less intimately connected 

 with the general nature of plants. We have already noticed, 

 that the Labiatae and Cruciform plants shew an abortion and 

 alteration of parts ; hence proceeds the want of uniformity in 

 the length of the filaments. 



But, in general, the regularity or irregularity, and also the 

 similarityor dissimilarity, of the size of parts, depend very much 

 upon their position on the stem. If a flower stands alone at the 

 tip of a branch, where no other flower hinders its evolution, 

 it will necessai'ily be regular, even when it belongs to a fami- 

 ly with irregular flowers. Parnassia and Saiivagesia have 

 regular flowers, although they belong to the Resedeas with ir- 

 regular flowers. Asarum stands among the Aristolochiae, be- 

 cause it has always a stalk with but one flower. But let us 

 make the supposition, that around and near this blossom 

 others arise, it will then become a whorl, a head, an ear, or 

 an umbel. The uniformity is now overturned. The cen- 

 tral blossom continues regular, but those on the margin musl^ 

 be irregular, as we every day see in the Umbellatae, the Ag- 

 gregatae, and in the compound flowers. Hence the most irre- 

 gular flowers never stand single, and never on the tip of the 

 branch, if we except some of the Orchideae. When in the 

 Labiatae flowers appear on the tip of a branch, even they 

 sometimes are regular, as is the case in Tcucrium campanula- 

 turn and some species of Galeops'is. 



