Q2 APPENDIX. 



calyx, not one-fourth or one-sixth the length of the fruit. These 

 are all characters of Salpichroma and not of Lycium, and although 

 certainly we have not positive evidence, we have every lair indi- 

 cation that the plant in question belongs to the former rather 

 than to the latter genus. This subject will be again considered 

 in a review I have prepared of the genus Lyctum. 



Another singular medley of incompatible genera, resulting 

 from the rejection of the character of aestivation, occurs m the 

 genus Juanulloa, which M. Dunal divides into three sections : 

 1. Eujuanulloa ; 2. Physalina; 3. Sarcophysa. In JuanuUo a 

 proper we see a small group of plants, distinguished by their 

 scandent habit, large thick coriacous leaves and conspicuous 

 pendent flowers ; the calyx, covered with dense yellow stellate 

 tomentum, is formed of five distinct sepals, which are connivent 

 by their tomentous margins into a pentagonous tube, with un- 

 dulating prominent angles; this is persistent, increasing but . 

 little with the growth of the berry, which it partially encloses, 

 beinc- generally slit into its original segments by the separation 

 of the adherent edges of the sepals ; the corolla, covered also 

 with orange-coloured tomentum, is twice or three times the 

 length of the calyx, and in the form of an elongated narrow 

 tube, somewhat ventricose in the middle, and contracted in the 

 mouth, with a small border of five almost orbicular lobes, qum- 

 cuncially imbricated in sestivation, being, as well as the tube, of 

 a thick fleshy consistence ; the berry is filled with seeds, having 

 a nearly straight terete embryo : these characters are fully deli- 

 neated in plate 46 of this volume. In Sarcophysa we find 

 a corolla very similar in texture and structure, but larger and 

 more ventricose ; the calyx, of half its length, is fleshy m sub- 

 stance, roundly tubular, ventricose, decreasing m diameter to- 

 wards the mouth, where it is terminated by five short erect teeth ; 

 this scarcely increases in size, but becomes stdl thicker, more 

 coriaceous in texture, and is at length irregularly ruptured on 

 one side, nearly to the base, by the growth of its enclosed large 

 berry. This genus has been shown to be gencrically distinct 

 from Juanulloa {huj. vol p. 42), both being closely a.llied to 

 Solandra ; its characters are delineated in plate 47 of this work. 

 The place in the system of Juanulloa and Sarcophysa, as I have 

 shown, is not among the true SolanacetE, but in the tribe Solandrea 



of the Atropacea. 



The section Physalina of M. Dunal belongs to a very difterent 

 group, which I have described under the name of Cleochroma, a 

 genus' closely allied to lochroma, and therefore belonging to the 

 family of the true Solanacete. The Juanulloa {Physalina) um- 

 bellota, Dun. (Prodr. 530) is again recorded (Prodr. ^491) as 

 tTio Tn^lrnmn p/ihicma. Beuth.. fisrured in 'Bot. Reg. (lo<51) 



