352 FOURTH GROUP. SEED-PLANTS. 



not coherent but free, the fact is expressed by the terms chorisepalous and choripetalous^ , 

 which, though here used in a figurative sense only, are at all events better than the 

 terms polypetalous and dialypetalous which have also been used. If there is only one 

 whorl of perianth-leaves present, and it is desired to express that it is composed of 

 coherent or of free leaves, the best terms to use are gamophyllous and choriphyllous ; 

 in some cases there are two perianth-whorls, but they cohere into a single whorl, as 

 in Hyacinthus and Muscari, where two alternating trimerous whorls unite to form a 

 single tube with six teeth. 



If the leaves of the outer and inner perianth are free and form a distinct calyx 

 and corolla, certain differences of form may usually be observed in them in addition 

 to the differences of structure before mentioned ; the leaves of the calyx are usually 

 broader at the base and sessile, of very simple outline and pointed at the apex ; the 

 leaves of the corolla are usually narrower at the base and often very broad above, and 

 there is often a distinction apparent between stalk (claiv) and lamina (limb], and the 

 limb is sometimes cleft or otherwise divided; ligules often appear on the inner 

 (upper) side where the limb bends away from the claw, and these when regarded as a 

 whole in a flower are termed a corona, as in Lychnis, Saponaria, Nerium, the Hydro- 

 phylleae, etc. ; if the corolla is gamopetalous, the parts of the corona also cohere, as in 

 Narcissus where it is very large. 



The general form of the perianth, especially when it is distinctly petaloid in 

 character and of some size, always stands in a definite relation to pollination by 

 means of insects, and large gaily-coloured delicate strongly-scented flowers only 

 occur where fertilisation is effected by them ; these characters are intended to induce 

 insects to visit the flowers ; the infinite variety and often strangeness of form in the 

 perianth are specially calculated to compel insects of a definite size and species to 

 adopt definite positions and movements of their bodies in their search for the nectar, 

 and thus the pollen is conveyed without intention on their part from flower to flower. 

 The multilateral or bilateral symmetry of the perianth is usually connected with that 

 of the other parts of the flower, and will be discussed therefore further on in con- 

 nection with them. 



Besides the perianth in the narrower sense which we have hitherto been 

 describing, other forms of envelope also occur not unfrequently in the single flower. 

 In the Malvaceae and in some other cases the calyx proper appears to be surrounded 

 by a second calyx, the epicalyx or calyculus, which, however, is not a calyx morpho- 

 logically ; in Malope trifida, for example, the three parts of the epicalyx represent a 

 sub- floral bract with its two stipules, in Kitaibelia vitifolia an epicalyx of six parts is 

 formed out of two similar sub-floral leaves with their four stipules (Payer) 2 . But the 

 epicalyx may also be only apparent, as in the Roseae and Potentilleae, where the 

 true calyx-leaves produce stipular structures which unite in each case in pairs, and 

 form a small and apparently simple leaf. In Dianthus, Caryophyllus, and other 

 genera, a kind of epicalyx is formed by two decussate pairs of small bracts imme- 

 diately beneath the calyx ; in the terminal flowers of the Anemoneae there is a whorl 



1 From xupis, separate. 



2 Eichler doubts the correctness of Payer's view of the epicalyx of the Malvaceae, and the 

 point will have to be again investigated. 



