Dr Dickson on DiplosfemoTions Flowers, d-c. 245 



pairs represented the five single stamens of the outer whorl 

 in Geranium congenitally deduplicated.* At first sight, 

 this view seems unexceptionable, since there is no doubt as 

 to the parts being homologous. It appears to me, however, 

 that if we were to invert Payer s statement, and say that in 

 Geranium there is a congenital connation, in pairs, of the 

 parts of the outer circle, which are distinct in Ilonsonia, we 

 should thereby be enabled at once to explain the apparent 

 anomaly of a younger whorl being external to an older one. 



If, in fact, we adopt a line of argument analogous to that 

 which Payer himself has employed in determining the sig- 

 nification of the epicalyx in the Potentillidee, the whole 

 difficulty, it seems to me, disappears. " In Fragaria, 

 colUna" he says, " we always observe a calicule, composed 

 sometimes of five leaflets alternate with the sepals, some- 

 times of ten leaflets grouped in pairs which alternate with 

 the sepals ; and, as this calicule appears always after the 

 calyx, it cannot be doubted that it is formed by the stipules 

 of the sepals." (Op. cit p. 503.) Now, in the Gera- 

 niacese we have an outer staminal whorl, which consists 

 sometimes (as in Gera^tium ) of five stamens alternate with 

 the parts of the inner whorl, and sometimes (as in Mon- 

 sonia J of ten stamens grouped in pairs which alternate with 

 the parts of the inner whorl. Moreover, these outer stamens 

 appear after the inner ones. The parallelism of these two 

 cases, as regards the number, position, and order of succes- 

 sion, of the parts, is quite complete ; for we have in the 

 Geraniacese an outer whorl whose relation to the inner one 

 in these respects is exactly the same as that which the ap- 

 parent outer calycine whorl of Fragaria bears to the inner 

 one, or calyx proper. 



The presumption, raised by this comparison, that the 

 outer stamens in Geraniacese represent the lateral lobes of 

 the inner or primary ones, distinct, or congenitally connate 

 in pairs like interpetiolar stipules, amounts almost to a cer- 

 tainty, when we consider the mode in which the penta- 

 delphous condition of Monsonia occurs, where each stamen 

 of the inner whorl is connate with two of those of the outer 



* Op. cit. p. 60. 



NEW SFRIES.— VOL. XfX. NO. II. APRIL 1864. 2 I 



