118 I. H. Burkill — Flower of Ranunculus arvensis. [No, 2, 



Summary. 



I liave shown first of all (Tables I-IV) how the flowers of Banun- 

 cuius arvensis in the races studied, vary; and how each set of organs 

 varies in a different way ; so that the curves which may be plotted 

 for sepals, for petals, for stamens, and for carpels are unlike, most of 

 them neither perfect Quetelet-Galton nor perfect half Galton curves. 



I have shown secondly (Tables V-XIV) that a correlative in- 

 crease and decrease occurs between the different sets of organs; so 

 that when the stamens or any other set of organs depart from normal, 

 it is probable that all other sets of organs will depart from normal, 

 but chiefly those which follow. This is important as it indicates a 

 division of vigour among the various sets, to be distinguished from 

 an increase of the one at the expense of another. 



In Tables XV-XVI and in the graphic representation of them 

 on page 106 I have followed this up by showing how if we take the total 

 number of organs in the flower as a measure of the vigour in the bud, 

 we find that the ring of sepals, being the first-formed of the sets of 

 organs, has the. first pull on the vigour and is most likely to get a full 

 complement, the ring of the petals being the next in order, is the next 

 to be satisfied, and that stamens and carpels obtain the surplus the 

 stamens chiefly so. T consider that the curves might with some little 

 trouble be translated into formulae by a mathematician. 



In Tables XVII-XX, I show that the power to produce organs 

 diminishes as the plant grows weaker towards its death. Sometimes 

 a slight 1 ecovery occurred at the very end : I do not feel justified in 

 suggesting a cause for it. In Tables XXI and XXII, I show that 

 sterility of the stamens increases towards the death of the plant. 



In Tables XXIII-XXV, I show that the stamens — the organs 

 which profit chiefly as we have seen by the extreme of vigour — lose by 

 its loss ; and consequently the flowers are most male when blossoming 

 begins. 



In Table XXVI, I have represented Table XV in a different way, 

 so as to bring out sharply the division of vigour (i.e., number of organs) 

 between the different sets (moods), I can show by it that the flowers 

 with more than 20 organs, there apparently is a setting aside ah initio 

 of so much vigour for the carpellary mood, the staminal mood becoming 

 residuary legatee ; while in flowers with fewer than 20 organs the 

 carpellary mood has to jostle with the preceding ones for its place. 

 I show also by it and by the Table which follows it (XXVII), 

 that there are certain irregularities which seem to be due to a 

 borrowing of organs by the staminal set from the petals, which 



