66 APPENDIX. 



i. e. laciniis 3 anticis et 3 posticis/^ These are hard words that 



mi 



ahnost 



M 



alludes to this feature. I have searched in vain for 

 its constancy in dried specimensj and it must be confessed that 

 it does not appear to be likely, from its very nature, to prove a 

 character available for practical purposes, because by the mere 

 torsion of the pedicel, in so small a quantity as one-tenth part of 

 a revolution, an opisthodromical calyx becomes at once empros- 

 thodromical. Besides, at all times it must be a doubtful test 

 among the Solanace(je and Atropacece^ where the insertion of the 

 peduncle is always more or less extra-axillary and lateral, forming 

 a kind of inflorescence termed centrifugal ; for here the point of 

 the calyx, that under ordinary development would be directed 

 towards the axis of its parent branchlet, is actually twisted away 

 from it, so that in the more bilabiate genera of the Atropacece 

 the two lips cease to be superior and inferior as in the true Scro- 

 phulariacece : this is very clearly manifest in Nierembergia and 

 several other genera when examined in a living state. We 

 should be guided by facts rather than by hypothesis in these 



cases. 



Dr. Sendtner in his work (p. 225) has made some rather ill- 

 natured remarks for my want of attention to what he conceived 

 to be essential characters, quoting in addition that I had not 

 observed the structure of the stamens in Cyphomandra^ and had 

 not noticed the articulation of the pedicel in Cestrum. My de- 

 tails of Pionandray as illustrated in plate 8, were made, and the 



rawin 



only published about the same time as Cyphomandra ; these natu- 

 rally differ in some respects from the dissections of Dr. Sendtner, 

 made from dried specimens, and it is from this cause that this 

 excellent botanist failed to observe the fleshy annular ring that I 

 have depicted. It is true that I omitted to mention the articu- 

 lation of the pedicel in Cestruniy as well as many other ordinary 

 characters .which it was not thought necessary to notice, but it is 

 evident that this feature, which is common to other species, did 

 not escape my observation, for in plate 16 of the same work, 

 every flower of Oestrum organense there delineated will be seen 

 to be distinctly articulated on its pedicel. If an omission of this 

 kind has crept into the descriptions of the desultory nature I had 

 adopted, it is evident that the more learned and systematic work 

 of Dr. Sendtner is not less free from error; for instance, among 

 many others, we may quote his generic character of HyoscyamtcSy 

 where the lobes of the corolla are said to be plicated in aestivation 

 when they are really imbricated — the placentation is stated to be 



