Blake Revision of Poly gala 15 



Linnaeus's Polygala diversifolia was based solely on a Cliffortian 

 specimen, now in the British Museum, of Securidaca volubilis L., 

 of which it becomes a synonym. 



9. P. stipitata, nom. nov. Frutex ramosus 1.3-2 m. altus, ramu- 

 lis dense strigillosis. Folia ovata apicem versus saepissime abrupte 

 breviterque acuminata vel acutata apice calloso-mucronata inter- 

 dum emarginata basi acute cuneata supra plus minusve nitentia 

 subsparse strigillosa subtus vix pallidiora sparse strigillosa aetate 

 subglabrata coriacea, nervis lateralibus 5-7-jugis, 4-6.3 cm. longa 

 2.1-3.3 cm. lata, in petiolis appresse puberulis 2-3.5 mm. longis. 

 Racemi axillares pauciflori, axe 1.5-2.5 mm. longo, pedicellis frugi- 

 feris 2-2.8 mm. longis. Flores ut dicitur " viridescentes." Sepala 

 orbiculari-ovata rotundata ciliolata ceterum subglabra 1 mm. longa 

 lataque. Alae consimiles 1.5 mm. longi. Carina in lateribus et 

 basi pubescens 2.4 mm. longa. Petala superiora obovata infra pu- 

 bescentia carinam subaequantia. Fructus transverse oblongus 

 sparse ciliolatus 8-9 mm. longus 11-14 mm. latus, in stipem 2-2.5 

 mm. longum desinens, vix per quintam partem longitudinis lo- 

 batus; ioculi patentes (uno saepe abortivo). P. diversifolia Chod. 

 Monog. ii. 10 (1893), in part, not L. Badiera cubensis Britton! 

 Bull. Torr. Club xxxvii. 362 (1910), as to description and specimens, 

 in part; Britton, 1. c. xlii. 495 (1915); not Polygala cubensis Chod. 

 Monog. ii. 62 (1893). Cuba: Chuico Azal (?), Bahia Honda, 23 

 Sept. 1863, Wright 1913 p.p. (cotype coll.: G); savannas, 

 Embarcaden di Bacunagua, San Cristobal, 6 April, Wright 1913 

 p. p. (cotype coll.: G). T. 1. Fig. 4. 



Badiera cubensis Britton was originally based on Wright 1913, 

 with which were placed Wright 3496, and 115 pro parte. In his 

 1915 revision (p. 495) Dr. Britton says, under B. cubensis: " C. 

 Wright's Cuban 1915 consists, apparently, of three species; it is 

 to the specimen with abruptly acuminate leaves cuneate-narrowed 

 at the base, 4-6 cm. long, that the name should be restricted. As 

 shown by specimens from Bahia Honda, Pinar del Rio (P. Wilson 

 9429) the leaves vary to obovate." On the next page, in describing 

 B. propinqua sp. nov., Britton says: " The description is drawn to 

 include the part of Wright 1913, and Wright 3496, previously re- 

 ferred by me to B. cubensis. The part of Wright 115, included by 

 me in the original description of B. cubensis, is perhaps referable 

 here also." There is no essential difference in the descriptions of 



