Dr Graham's Description ofNetc or Rare Plants. 369 



are a little reticulated, and channelled above. Peduncles scattered, oj)- 

 posite, but never immediately, sometimes at a considerable distance from 

 the leaves, single-flowered, longer than the upper, but shorter than the 

 larger leaves. Flovwrs nodding, rather powerfully, and, as some think, 

 l)leasantly perfumed. Calyx green, 5-parted, segments patent, cordato- 

 ovate, acute, retlexed in their sides, subraarcescent. Petals (I inch long) 

 reflexed, navicular-hatchet-shaped, cucuUate, glandular- pubescent, but 

 sparingly except at the claws, reticulated, ciliated along the keel, alter- 

 nating with white, erect, truncated, obcordate scales, hollow, and open- 

 ing longitudinally on their inner surface, flat at the apex, there trans, 

 versely marked w'ith elevated stripes, and each extended into two short 

 rose-coloured points in the centre of the flower, the stripes reddish-brown 

 on the outside, and more and more greenish-yellow towards the centre. 

 Stamens very numerous, about half as long again as the claws of the pe- 

 tals, and lodged, as in the genus, in their cavities, and advancing in succes- 

 sion ; filaments white, glabrous ; anthers yellow, becoming brown, sliort, 

 bilobular, bursting along their sides, pollen granular, small, ])ale yellow. 

 Two abortive, subulate, hooked, jmbescent filaments, shorter and broader 

 than the others, are placed on the inside of each of the hollow scales, and 

 are at all times erect. Germen unilocular, turbinate, inferior, covered 

 with pungent hairs, its upper surface flat, becoming conical, and rising 

 above the calyx, but there empty, trifid ; ovules numerous, attached to 

 three linear j-iarietal placentae. Styles 3, cohering, their lower half, as 

 well as the upper surface of the germen, covered with hairs. Stigmas 

 very small. 



I cannot see the propriety of changing the specific name of Linnseus to 

 the ei^ually objectionable one of Jacquin. nor could I have thought the 

 reasoa assigned by Jussieu for departing from this last sufficient, even 

 though he had not taken a third, which in its turn may be considered 

 inappropriate. It is very true, that all the species of Loasa are hispid, 

 and all are stinging ; but if such be considered a reascm for changing a 

 specific name, endless confusion would arise in almost every old genus 

 in which we have lately become acquainted with many new species. I 

 wish specific names, when once given, were considered mere arbitrary 

 and immutable terms. Trattinnick, it appears, has considered L. urens, 

 Jacq., and L. ambrosiafolia, Juss. as distinct, and is followed by Spren- 

 gel ; but as I am not acquainted with the reasons for this opinion, as 

 Jussieu is perfectly explicit that they are the same, as his figure agrees 

 sufficiently, and as the dififerences noticed in his description appear to 

 me to be trifling, I must follow his opinion, supported as it is by Persoon 

 and Lamarck. 



The seeds of this most beautiful species were received at the Botanic Gar- 

 den in spring last, from my invaluable correspondent Mr Cruckshanks, 

 whose additions to our stock of cultivated plants I take delight in re- 

 cording. They were gathered in the valley of Lima. The specimen 

 described was raised in a hot-bed, and flowered in the greenhouse in the 

 end of July. It has also flowered in the Royal Botanic Garden at Glas- 

 gow, and I understand also in the neighbourhood of London. 



Palavia rhonibifolia. 



T.rhomltifoUa ; foliis rhomboideis, lobato-crenatis, ad venas sporsim stel- 

 lato-pilosis, pedunculo brevioribus ; stipulis subulatis ciliatis viridi- 

 bus ; petalis obovato-cuneatis, oblique emarginatis ; caule prostrato, ra- 

 mose, parce stellato-piloso. 



Description Annual ? Stem prostrate, branched, sprinkled very loosely 



with raiher rigid hairs, which are single or stellate. Leaves {\^ inches 

 long, 14 broad,) alternate, petioled, soil, bright green above, jialer below ; 

 rhomboid, glabrous, sublobate, or deeply and unequally crenate, nearly 

 entire at tlie base, O-ncrved, veined, the nerves and veins prominent be. 

 low, channelled above, and both above and below, but especially below, 



