CONTEIBUTIONS TO SOUTH AFRICAN ASCLEPIADOLOGY. 269 



Prope Heilbronn in terra Orange Free State, Rev. Wallace 

 (comin. Dr. S. Schdnland) . 



This species belongs to my section Eu-Schizoglossum, charac- 

 terized by the insertion of the caucUcuUt, not on the top, but about 

 the middle, on the inner side of the pollinia. As far as I am at 

 present able to judge, this section seems to be confined to Extra- 

 Tropical South Africa. All the tropical species, as well as the 

 Abyssinian, seem evidently members of the section Lagarinthus, 

 with the exception of a few, which do not belong to the genus 

 Schizoglosswii , but to Gomphocarpus ; of these I may mention here 

 6^. (§ P achy car pus) Grantii Schltr. [Schizoglossum Grantii Oliv.) and 

 G. (§ Pacliycarpus) apathuhdus Schltr, {Schizoglossuni spathnlatnm 

 K. Sebum.), which both seem to me to be true Pachijcarpl. The 

 fruit of the former is wiuged, and thus shows another character of 

 Pacliycarpus ; I feel pretty sure that the fruit also of the latter will 

 prove to be winged. 



I consider Schhoglossuiii. Wallacei a near ally of S. hidens E. Mey. 



24. Krebsia carinata Schltr., n. sp. Herba erecta, 20-35 cm. 

 alta e basi parum ramosa ; ramis flexuosis, subteretibus, unifariam 

 velutinis basi glabrescentibus, dense foliatis ; foliis erectis linearibus 

 acutis utrinque glabris, basi in petiolum brevem angustatis, mar- 

 ginibus revolutis, 2-9 cm. longis, supra basin 0*2-0-5 cm. latis, 

 petiolo vix 0-3 cm. longo ; floribus in umbellis vel capitulis breviter 

 pedunculatis vel sessilibus, extra-axillaribus alternantibus, pedi- 

 cellis filiformibus divaricatis, 0-5 cm. longis, tenuissime puberulis ; 

 calycis segmentis erecto-patentibus lauceolatis acutis tenuiter pilosis, 

 quam corolla dupio plusve brevioribus ; corollte lobis erectis apice 

 marginibusque reflexis, ovato-oblongis subacutis, utrinque glabris, 

 0'4 cai. longis, medio vix 0-2 cm. latis ; coronas foliolis carnosis 

 falcatis linearibus obtusis, dorso more generis carina longitudinaliter 

 donatis, gynostegium paulo excedentibus ; antheris subquadratis, 

 marginibus cartilagineis alteformibus basi oblique truncatis loculos 

 baud sequantibus, appendice hyaline brevi rotundato apice in stigma 

 depresso, marginibus loculorum profunde excisis ; polliniis more 

 generis valde compressis oblique triangulis obtusis, caudiculis fili- 

 formibus basi incrassata glandule supra basin insertis, glandula 

 rhomboidea, polliniis multo minori apice obtusiuscula ; stigmate 

 depresso. 



In udis graminosis prope Kokstad, terras Griqualand Orientalis, 

 flor. Febr. 1883, alt. 5000 ped., W. Tyson, No. 1439; in saxosis 

 prope flumen Umzimhlava, flor. Jan. 1895, alt. 4500 ped., R. 

 Schlechter, No. 6548. 



Although at one time I thought it better to unite Harvey's 

 genus Krebsia with Gout pilocarpus R. Br., I have already stated in 

 another place (cf. Engl. Hot. Jahrh. vol. xx. Beibl. 51, pp. 39-40) 

 my reasons ibr now keeping the genus distinct. It seems to me 

 better to make a number of such small genera in orders as difficult 

 as the Asclepiads, provided that anything like sufficient cbaracters 

 exist to distinguish them, than to unite everything under a mere 

 handful of unwieldy genera. 



The present species is the third of the genus ; it comes nearest 



