TYPICAL FORMS OF PLANT KINGDOM 107 



hermaphroditism of animals, these will become reduced to the 

 state of male flowers, that is to say, stamens ; only the fertile 

 leaves at the top of the catkin, those that represent the final 

 growth, preserve their female sex, and become carpels with 

 their ovules. Thus we have the actual flower of the 

 dicotyledons, as it is schematically represented and as it is 

 described in all the classical works on botany, with its carpels 

 in the centre forming the gyncecium or pistil ; a ring of stamens 

 forming the andrcecium, and a ring of sterile leaves constituting 

 the perianth. The latter is usually double, and comprises the 

 coloured leaves of the corolla and the green leaves of the calyx. 

 In order to explain this fact, we must turn to another order of 

 considerations. The elaboration of the germ-cells does not take 

 place without the resultant formation of special compounds, 

 waste products evacuated by these cells and thus coming into 

 contact with others contiguous to them, either by absorp- 

 tion or by way of a circulatory system. Zoologists and 

 physicians have long been aware of the influence exerted upon 

 the organism by these internal secretions, as they are called 

 to-day, and we have already indicated to what degree they are 

 capable of modifying the form, size, and the colour of the 

 organs they penetrate, especially those which form part of the 

 secondary sexual characters. So far as colour is concerned, 

 this action may be spent entirely in modifying the corolla, 

 but sometimes it is extended to the calyx, which then becomes 

 petaloid, notably in many Monocotyledons (Colchiaceae, 

 Liliaceae, Asparagoideae, Orchidaceae, etc.), and can even 

 succeed in modifying the bracts (various Sages) or the leaves 

 (Poinsetia). 1 



Naturally, if flowers have originated thus, the earliest of them 

 ought to preserve some traces of the primitive elongation of the 

 catkins and the indeterminate number of their constituent 

 elements : sepals, petals, stamens, and carpels ought to have been 

 numerous at the beginning, because of their common origin, and 

 to have approximated to one another by gradual transitions. 



1 It may happen, of course, that the cause for these phenomena is just the 

 reverse, and that the coloured bracts, the petaloid calices, and the petals owe 

 their particular development to the fact that they have arrested in their passage 

 the foods which the fertile leaves have attracted toward themselves, and thus 

 profited by additional nourishment ; but if this were so we might well ask why 

 these parts do not themselves become fertile. Experiment or chemical 

 analvsis must be left to decide this. 



