104 THE JOURNAL OF BOTANY 



Diclis umbonata, sp.n. Herba parvula inflorescentia glan- 

 duloso-puberula excepta glabra ut videtur perennis, caudice sub- 

 ligneo, caulibus tenacibus subtus pallidis super herbaceis et pallide 

 viridibus ascendentibus tetragonis semipedalibus vel ultra dimidio 

 inferiore folioso superiore minus foliato, foliis oppositis anguste 

 ellipticis vel lanceolatis apice obtusis basi breviter petiolata sub- 

 sessilive plus minusve angustatis vel fere rotundatis firme herbaceis 

 utrinque viridibus subtus subpallidioribus margiiie denticulatis vel 

 subintegris 6-10 mm. longis l-25-3-5mm. latis, petiolis brevissimis 

 latiusculis anguste decurrentibus, racemis terminalibus brevibus 

 densisque vel subtus laxioribus atque circiter 44 mm. longis pluri- 

 floris, pedicellis in bractearum axillis orientibus unifloris ebracteo- 

 latis, inferioribus fere ad 25 mm. longis subgracilibus rectis patulis, 

 superioribus brevioribus, bracteis foliorum similibus sed minoribus 

 et glanduloso-puberulis, calycis segmentis ovato-ovalibus obtusis 

 glanduloso-pilosulis 2-2-5 mm. longis, corolla subpurpurea bilabiata, 

 labio posteriore trifido 6 mm. longo lobis rotundatis, labio anteriore 

 bifido 8 mm. longo lobis semi-ellipticis media basi umbone aurantiaco 

 puberulo pr^ditis, palato pulverulento umbonibus 2 aurantiacis 

 breviter barbatis sub labii anterioris eis praedito, calcare e basi 

 conica anguste oblongo obtuso parum curvo 4-5 mm. longo, fila- 

 mentis glabris nitidis latiusculis, longioribus 1-25 mm. longis, brevi- 

 oribus '625 mm. longis, antheris aurantiacis 1 mm. longis. 



Habitat colonize Orange Fdver in regione Kalahari ; legit anno 

 1900 Lieut. Pateshall Thomas ! 



The habit of this plant, with its comparatively narrow leaves, 

 suggests the genus Xemesia, but the corolla is that of Diclis ; the 

 specimens unfortunately do not supply ripe fruit. 



BRITISH HIERACIA. 



As we announced in our last issue, Messrs. Linton have issued 

 the sixth fascicle of their useful and admirably prepared series of 

 British Hieracia, thus bringing their original undertaking to a close. 

 They think, however, that a supplementary fascicle of forms not 

 represented in the set may be forthcoming later. The present 

 fascicle contains the following forms which are believed to be 

 endemic : — H. anylicum var. calcaratwn ; H. Griffitldi ; H. Leyi ; 

 H. Schmidtil var. eustomon : H. caledunicum ; H. ruhicundum var. 

 Boswelli ; H. argenteum var. septentrionale ; H. Sommerfeltii Yen'. 

 spleyidens ; H. saxifragum var. orimeles ; H. rivale var. subhirtum ; 

 H, murorum var. pulcherrimum ; H. murorum var. luciduhim; H. 

 miiroriim Y&v. sanguineum ; H. murorum var. subidatidens ; H. orca- 

 dense ; H. Orarium var. erythrmim ; II. duriceps ; H. vidgatum var. 

 amplifolium ; H. vulgatum var. mutahile ; H. surreianum ; H. steno- 

 phyes var. oxyodus ; H. gothicwn var. Stewartii ; H. rigidnm var. 

 longiciliatum ; H. cantianum var. suhrigidum ; H. zetlandioum forma ? ; 

 H. vidgatum var. sejunctum. 



The notes accompanying the specimens may be of interest to 



