,514 LX. MELASTOMACE^I. (C. B. Clarke.) [Osbeckia. 



[With regard to the following generally accepted sections of Osbeckia founded pri- 

 marily on the 4-fid or 5-fid flowers it should be premised that several OsbecJcias which 

 have the flowers normally 4-fid carry also a few 5-fid flowers on the same plant ; and 

 that sometimes a strong plant may produce all or nearly all the flowers 5-fid. This 

 occurs eminently in 0. crinita and in 0. cupularis.} 



SECT. I. Amfolyanthera. Perianth normally 4-inerous. Anthers not 

 beaked. Herbs or small shrubs. 



1. O. truncata, Don ms. ; W. A. Prodr. 322 ; hairs on the stem patent, 

 flowers capitate often with two or four leaves close under the head, bristles on 

 the apex of the ovary 16-20, fruit ^-f in. broad 8-ribbed ovoid-oblong. Wight 

 Ic. t. 375 ; Triana in Trans. Linn. Soc. xxviii. 53. 0. muralis, Naud. in Ann. So. 

 Nat. Ser. 3. xiv. 56. O. Leschenaultiana, Dalz. fy Gibs, Bomb. Fl. 92. 0. 

 zeylanica, Steud. in Herb. Hohenack. No. 577. 



Throughout the DECCAN PENINSULA, alt. 0-4000 ft. ; extending plentifully to CHOTA 

 NAGPORE. EAST BENGAL, Mudhopoor ; C. B. Clarke. 



Stems 4-16 in., annual, 4-angled. Leaves 1 in., elliptic, subacute, 3-nerved, hairy 

 on both surfaces, drjang yellow ; petiole ^ j n . Bracts ovate or oblong, glabrous on 

 their backs. Calyx-tube and teeth with stalked stellate hairs ; teeth erect, subper- 

 sistent. Petals less than in., purple. Capsule occasionally 5-celled. 



VAR. Kurzii ; plant scarcely 2 in. with very small leaves and flowers, capsule 

 ovoid ribs altogether obscure. Parasnath Summit, Behar, alt. 4200 ft. ; Kurz. 



2. O. cupularis, Don ins.-, W. $ A*Prodr. 323; flowers subcapitate- 

 often with two or four leaves close under the head sometimes more lax, bristles 

 on the apex of the ovary very numerous, fruit rarely less than \ in. broad 

 obscurely many-ribbed ovoid. Naud. in Ann. Sc. Nat. ser. 3. xiv. 61 ; Triana 

 in Trans. Linn. Soc. xxviii. 55. O. brachystemon, Naud. 1. c. 67 and xiii. t. 7, 

 f. 1 ; Triana 1. c. 63 0. truncata, Arn. in Hook. Comp. Bot. Mag. ii. 308 j 

 Naud. 1. c. 57. 0. confertiflora, Naud. I. c. 59. O. Leschenaultiana, Wight Ic. 

 t. 996 ; Thwaites Enum. 104 partly ; not of DC. O. Wightiana, Benth. in 

 Wall. Cat. 4074 (not in Wall. Cat. 4060). 



Mountains of the southern part of the DECCAN PENINSULA and of CEYLON ; Wight 

 No. 1148 and 1095 (and No. 1093 with 4-fid and 5-fid flowers). Mysore; Lobb. 

 Bababoodun Hills; Law. Nilghiris ; Hohenacker No. 963 (named 0. Leschenaultiana). 

 (MERGTJI ? Herb. Griffith). Ceylon ; Gardner. 



Annual and biennial, drying yellow. Stems 8-20 in., tetragonous, glaucous up- 

 wards, with simple ascending often closely-adpressed hairs. Leaves 1-2 in., elliptic, 

 subacute, 3- (or sub- 5-) nerved, hairy on both surfaces ; petiole ^-\ in. Bracts ovate 

 or oblong, glabrous on their backs. Calyx-tube and teeth with stalked stellate hairs ; 

 teeth patent subpersistent. Petals f-f in., dark-purple or nearly white. 



VAR. erythrocephala, Naud. in Ann. Sc. Nat. ser. 3. xiv. 58. 0. Leschenaultiana, 

 Thwaites Enum. 104. Ceylon, alt. 6000 ft. ; Thwaites No. 1576 and 284, Gardner, 

 Walker. Turns black in drying ; stem below with deflexed hairs ; stellate hairs 

 about the inflorescence a rufous-brown ; fruit sometimes large somewhat glaucous 

 and slightly produced at the vertex; flowers in this variety seem always 4-fid. 

 Triana distinguishes (under the name 0. cupularis) Wight's No. 1093 which has 

 sometimes 5-fid flowers, the stamens 8, 9 or 10, from Wight's No. 1148 which he takes 

 as 0. brachystemon: but the two seem identical. Mr. Thwaites makes but one 

 .species here. Triana arranges the main sections of Osbeckia according as the 

 tlowers are 4-fid or 5-fid ; he thus places 0. cupularis in the 5-fid section at a great 

 distance from 0. brachystemon ; then he says (under the head of 0. Leschenaultiana) 

 that 0. cupularis has 4-fid flowers. 0. Leschenaultiana is the name generally found 

 in herbaria for this variable species, but Triana is right in saying that DC. meant by 

 O. Leschenaultiana a very different plant. 



