«'.\ot»\*^ Mr. B. Clarke on Relatme Position ; ■ v'^'^^A^vx' 



the following reasons : in Selaginese the calyx sometimes consists 

 of two lateral sepals which enclose the flower much like the 

 bracts of Callitriche ; the embryo has short blunt cotyledons re- 

 sembling those of Selaginese (but still more reduced), and ap- 

 proaching Stilbe as described by Bndlicher ; but the raphe is 

 next the placenta in Selaginese, and on that account the afl&nity 

 of Callitriche is rather with the latter Ordei', between which and 

 Podostemacese or Elatinese it may be regarded as forming a con- 

 necting link. 



Valerianace^. In Valeriana officinalis the mode of growth 

 is the same as in Scleranthus, &c. The fertile carpel in the 

 forking of each division of the inflorescence stands with its ven- 

 tral suture towards the inner branch of the forking and the dorsal 

 next the outer branch, the gibbosity of the coi'oUa being ante- 

 rior and coming forward between the two branches (PI. XIII. 

 fig. 1). The regularity of the position of the fertile cell is perhaps 

 connected with the irregularity of the corolla, as they always have 

 the same relation to each other, although in Valerianella, where 

 the corolla is quite regular, the position of the carpel is the 

 same. But in Valei'ianacese the fertile cai'pel has sometimes the 

 appearance of being posterior; this however arises from the 

 mode of growth being the same as that of Centranthus ruber 

 (PI. XIII. fig. 2), viz. dichotomous with the fiower sessile be- 

 tween the rami, one branch of the dichotomy being regularly 

 suppressed, except near the principal axis (PI. XIII. fig. 2d); 

 in such a case therefore the fertile cell is next the remaining 

 ramus, and constantly so, as it is always the external ramus that 

 remains*. In this diagram the successive branchings are repre- 

 sented as springing from each other at right angles (instead of 

 obliquely as in the plant itself), that the relative position of the 

 fertile cell may be more evident ; — the cross indicates the position 

 of the spur of the corolla. The same mode of growth occurs in 

 Fedia sagittifolia, where the barren cells are inflated. 



DiPSACEiE. In Marina and Scabiosa the placentation is lateral, 

 and the former agrees closely with Valerianacese in the carpel 

 being lateral and external ; and in Dipsacus sijlvestris it is pos- 

 terior with remarkable regularity, although, from the elongation 

 of the style, the position of the usually unilateral stigma becomes 

 variable. (PI. XV. fig. 6.) 



CALYCERACEiE. The ovule is here attached to the posterior 

 side of the ovary, but nearer its apex than in Dipsacus sylvestris, 

 which occasions its placentation to be less obvious ; and the 

 raphe is anterior, so that a partial adhesion of the anthers may 

 prove to be the only distinction between the two Orders, as the 

 involucel of Dipsacese occurs here also. 



* See also the description of the figure. 



