371 





t 





plant are those of some species altogether different from ]>. xativa, 

 Benth. and, owing- to the outline of the capsules, the fruit in 

 Ehret's plant is that of a species quite remote from D. sativa*, 

 Benth. But while it is thus certain that D. sativa, Benth; 

 cannot he the species which Linnaeus accepted as the basis of 

 IK sativa, Linn., we are not brought nearer to the identity of 

 the Ehret plant. 



The suggestion made by Lamarck in 1789 was in essence 

 accepted by Kunth in 1850 and there is no doubt, when examples 

 of D. rillosa, Linn., are compared with Ehret's plate that, as 

 regards the leaves, that plate may be accepted not only as a good 

 but as a singularly accurate representation of this North 

 American species. The difficulty which the plate then presents 

 resides in the fact that the fruiting spike shows the capsules of a 

 | species which we know from experience should contain seeds 



which are winged at the too only, and that they are in shape, 

 size and appearance wholly unlike the capsules of I), villosa, 

 Linn., in which the seeds are winged all round. The two facts 

 mentioned can only be reconciled if we assume that the leaves 

 and the fruit depicted by Ehret were taken from different speci- 

 mens belonging to different species. That the leaves and the 

 the fruit w r ere in fact derived from different specimens we already 

 know. The plant described by Linnaeus (Hort. Cliff, p. 459) had 

 not flowered in 1737; the fruits shown therefore either came from 

 that plant in a later year or belonged to some other plant. An 

 examination of the figure shows that the rachis of the fruiting 

 spike is made to leave the stem nearly but not quite from the axil 

 of the subtending leaf, and that this rachis is made to diverge 

 from the stem at an unnatural angle. It is also noticeable that 

 at the time the fruiting spike was added to the plate a stipulary 

 prickle was added at the base of the subtending petiole but placed 

 at the upper aspect of the petiole instead of at the point where 

 a stipulary prickle would normally be situated. The conclusion, 

 indeed, to which an examination of the plate leads is that both 

 the fruiting spike and the stipularv thorn were added after the 

 main figure was drawn and the style in which these additions to 

 Ehret's figure have been executed induces some suspicion that 

 they may have been the work of another artist. The eircxim- 

 stance to which we have already adverted, that Linnaeus did 

 not refer to the figure of D. foliu cordatis alt emit, caule Jaevi in 

 his text and that both Royen in 1740 (Flor. Leyd. Prodr. p. 527) 

 and Linnaeus himself in 1747 (Flor. Zeyl. p. 170) have avoided 

 the citation of this plate, at least suggests the possibility that 

 there was some doubt in their minds as to the congruity of the 

 foliage and the fruit. 



Fortunately the matter no longer remains doubtful. Th 

 specimen from which the drawing of the fruiting spike shown in 

 the Hortu* Cliff ortianus figure was made is still available for 

 study in the Natural History Museum collection at South Ken- 

 sington. It is in fact the Cliffortian specimen which purport^ 

 to represent the species now under discussion and though it 

 bears no legend in the handwriting of Linnaeus, it had been 

 written up as D. satim, by some early student, before it became 

 the property of the British Museum. The leaves and stem of 



d2 



