388 Dr Fulton on the Infiorescence, Floral Structure, and 



These floral features, vrhen considered in relation to the 

 injurious action of bees, are, I think, strong arguments for 

 concluding that they are modifications produced by the 

 action of general laws, in order to protect the flowers from 

 the bees' visits. Darwin* tells us that the freely-exposed 

 nectar of Epipactis latifoUa, which is habitually visited by 

 wasps, is never sucked by bees, which apparently dislike 

 it. Whether bees would prove injurious to this plant, 

 which is apparently protected from them by a change in 

 the character of the secretion, or be less effective in per- 

 forming cross-fertilisation, I do not know ; but it may 

 occasionally happen, that by reversion, or by retrograde 

 variation, a plant will escape from the keen competition 

 for the visits of bees, by limiting itself to the attraction of 

 a less engaged species ; and in that case it would be 

 advantageous to keep the nectar for it alone. Darwin says 

 that the continuance of this Orchid is dependent on the 

 existence of w^asps, and the same may probably be said of 

 the Figwort. 



The effect of the habit or mode of working of the 

 insect has thus a marked influence on the inflorescence, as 

 may be seen by comparing that of the Figwort with that 

 of almost any of the Labiatce fertilised by bees. Both are 

 constructed on the same type, but in the latter the 

 dicbasial cymes are modified by vertical and horizontal 

 aggregation, by partial suppression, and by more rapid 

 development of the flowers, so as to bring the dichogamy 

 into relation with the development of the inflorescence as 

 a whole. Many of the Scrophulariaceoe probably attain 

 the same end by suppression of all the flowers of the 

 cymes except those of the first generation, retaining their 

 lengthened raceme — such, for example, as in Digitalis. 



Before concluding, I may say a word about the marked 

 p»roterogynous dichogamy of Scrophularia. Dichogamy, 

 or the separation of the sexes in hermaphrodite flowers by 

 successive development of the reproductive organs, when 

 it exists in anemophilous, or wind-fertilised, flowers, is 

 almost invariably proterogynous. In entomophilous dicho- 

 gamic flowers, on the other hand, the great majority are 

 proterandrous, especially, and most markedly, in crowded or 



* Cross and Sdf-FertUisaAion of Plants, pp. 375, 423. 



