246 Dr Dickson on Diplostemonous Floivers, dtc. 



-—one on either side, — offering the closest analogy to a leaf 

 with lateral lobes, or with adnate stipules. That the pre- 

 sumed interstaminal lobes — if I may so call them — in 

 Geranium should so closely resemble, in all essentials, the 

 primary stamens, need not surprise any one who bears in 

 mind instances like Galium cruciatum, w^here the interpetio- 

 lar stipules differ in no respects from the leaves between 

 which they are placed.* 



It is evident that if the foregoing reasoning holds good 

 as to the Geraniacese, we must extend its application to all 

 the numerous cases where a similar diplostemonous ar- 

 rangement occurs. All such plants, if my view be adopted, 

 must be considered, strictly speaking, as isostemonous, the 

 members of the outer staminal circle consisting merely of 

 the lateral lobes of the primary stamens which form the 

 inner circlcf 



We may now proceed to examine what bearing the posi- 

 tion of the carpels may have on this question. I have 

 already stated that where the younger whorl of stamens is 



^ It may perhaps be thought that I am begging a question a little, in this 

 allusion to the stipules of the Galiacese. Although the stipulary nature of 

 these organs has been admitted by many very eminent botanists, yet I would 

 not thus have assumed their opinion to be correct, had I not satisfied myself 

 on the subject by examination of the course of development in Galium aparine 

 where I can positively testify to the appearance, in the first place, of opposite 

 leaves, followed afterwards by the development of intervening lobes, two or 

 three on either side of the axis. I can hardly doubt that the leaf develop- 

 ment in the Galiaceee has been traced by others, but I have not succeeded in 

 finding any references to it. 



t It is perhaps worthy of remark that Payer has shown that the " stami- 

 iiodes" of Linum are not developed until after the fertile stamens are so far 

 advanced as to indicate the distinction between anther and filament, and after 

 the carpels have made their appearance. (See Organogenic, pi. 13, figs. 6 and 

 7 ; with description, p. 67.) Now, if these staminodes in Linum represent, as 

 Payer suggests {loc. cit. p. 66), the staminodes in Erodium, and these last con- 

 stitute a true whorl of sterile stamens, it is very difficult to understand 

 their very late appearance. If, on the other hand, we adopt the view stated 

 above, as to the younger and outer stamens being merely the lateral lobes 

 of the primary ones, and analogous to leaf lobes or leaf stipules, this diffi- 

 culty disappears ; since there is nothing surprising in such structures appear- 

 ing at a comparatively late period, and it is quite in accordance with what 

 one observes in the case of staminal groups, where frequently the greater 

 number of the stamens (lobes of the compound stamensj are developed after 

 the appearance of the carpels. 



