252 Dr Dickson on Dtplostemonous Flowers, dtc. 



sary correspondence between the mode of succession of the 

 staminal and foliar lobes, wben both stamens and leaves, 

 in the same plant, happen to be lobed. Thus, in the 

 Hypericaceee, Myrtacese, &c., we have examples of families 

 characterised by their compound stamens, and yet remark- 

 able for the simplicity of their leaves ; and in CajopJiora 

 (Loasa) laterttta, where the stamens are developed in 

 succession from above downwards, or basipetally, upon 

 the staminal cushions,* I find the pinnae of the leaf to 

 be developed from below upwards, or basifugally. From 

 such considerations it may be inferred that we need not 

 expect any necessary association of lobed or stipulate leaves 

 with pseudo-diplostemonous flowers.! Should any one 

 be inclined to imagine that the facts I have just been 

 stating at all invalidate Payer's determination of staminal 

 groups as compound stamens, I would have it borne in mind 

 that it is no more surprising that there should be entire 

 leaves and compound stamens in the Myrtacese, &c., than 

 that in many other plants there should be lobed leaves and 

 simple stamens ; or, again, that in CajopJiora there should be 

 basipetal development of staminal lobes with basipetal 

 development of leaf-lobes, than that these two modes of 

 development should often occur together in the same leaf, 

 as they do in the so-called mixed leaf-formation.J 



* Payer, Organogenie, p. 391, pi. 84. 



t I i^^y observe, however, that in some plants with pseudo-diplostemonous 

 flowers, the leaves are not only stipulate, but exhibit a tendency to the forma- 

 tion of interpetiolar stipules. A. P. de Candolle has remarked that " several 

 Geraniacefe present this peculiarity [fusion of stipules] in a very evident man- 

 ner." [Organographie Vegetale ; Paris, 1827 ; tome i. p. 339.) 



In some Geraniacese I find a very remarkable condition, which, so far 

 as I know, is without parallel in other plants. In Erodium hymenodes, for 

 example, where all the leaves are opposite, there are invariably, between 

 each pair of leaves, on the one side of the axis a single, entire or slightly 

 bifid, interpetiolar stipule, and on the other side a pair of free stipules, the one 

 of which pair overlaps the other, from their bases passing each other. I find 

 a similar arrangement in E. cicutarium, Pelargonium zonale and allied forms, 

 P. scutatum, &c., whenever the leaves happen to be opposite. 



Again, in Spergula (which, like Malachium, has ten stamens, and five carpels 

 superposed to the petals) there are interpetiolar stipules ; and similar stipules 

 exist in the allied Lepigonum. 



X Cohcea scandens affords a very pretty example of the association of basipetal 



