1953] 



REVISION OF BRACHYOTUM 



361 



Type Collection and Locality: Ruiz and/or Pavbn s.n. (presumably at MA; 

 probable isosyntypes without locality BR, G-BOIS, G-DEL, P); "Huassahuassi, 

 Panao, Chaclla, et Muna montibus, copiose in Sancti Dominici et Llamapanaui 

 collibus." The first of these localities is in Dept. Junln, Peru, the others in 

 Dept. Huanuco (Ruiz 1940). 



Type Photographs and Illustrations: F 16716 and Gleason 27-2 (destroyed syn- 

 type at B); Ruiz & Pavbn, Fl. Per. & Chil. 3: pi }21, f. b (1802) (as Rhexia quin- 

 quenervis); Trans. Linn. Soc. 28: pi 3, f. }3c (1871) (as Brachyotum quin que nerve). 



Distribution: central to south central Peru, alt. 1500-3200 m. 



Huanuco: Chinchao, McLean s.n. (K), Rivero 95 (P), herb. Ruiz & Pavon s.n. (F); 

 Carpish divide between Huanuco and Tingo Maria, Asp lund 12861 (S), Ferreyra 1231 (USM), 

 Ferreyra 1713 (USM), Ferreyra 1732 (USM), Ferreyra 2113 (NY, US), Ferreyra 8078 (USM), 

 Sandeman 5169 (K); Acomayo, Ferreyra 8182 (USM); Panao, Asplund 13512 (S), Ferreyra 

 1795 (USM), Pearce s.n. (K); Tambillo southwest of Panao, Scolnik 1052 (NY), Macbride 

 3572 (F, G-DEL, NY, S, US); Playapampa, Macbride 4858 (F, S, US); Yanano, Macbride 

 4940 (F, NY). Junin: Carpapata near Huacapistana, Ferreyra 3759 (NY), Killip & Smith 

 24448 (NY, US); near Huacapistana, Ferreyra 3607 (NY), Killip & Smith 24131 (F, NY, 

 US), Sandeman 97 (K), Sandeman 4368 (K); between Punto and Andamarca, Raimondi 8784 

 (USM); near "Andimarca," Mathews 1170 (K, NY, W). Cuzco: Machupicchu, Balls B6813 

 (GH, NY, UC, US), Herrera 3208 (F, NY), Herrera 3224 (F), Sandeman 3586 (K), Vargas 

 801 (F, NY), West 6422 (GH, UC); San Miguel in the Urubamba Valley, Cook & Gilbert 

 1171 (US); Cedrobamba in the Urubamba Valley, Herrera 1559 (F, NY, US); Huayna Picchu, 

 Scolnik 836 (NY); Punto Real in the Urubamba Valley, Tutin 1318 (BM); Urubamba basin, 

 Herrera 1964 (F, NY); between Lares and Calca, Raimondi 9578 (USM). Ayacucho: Ccar- 

 rapa between Huanta and Rio Apurimac, Killip & Smith 22272 (F, NY, US), Killip & Smith 

 22333 (NY, US). 



Vernacular Names: Cachiquis (Ruiz & Pavbn, Fl. Per. & Chil. 3: 1802); Huay- 

 chuy ( Ferreyra 1795); Masuca ( Cook & Gilbert 1171). 



lb. Brachyotum quioquenerve var. pusillum Wurdack, var. nov. 



Kypanthiorum ramulorumque trichomata laevia vel minutissime muriculata (sub 

 lente 90x). Petioii 2-6(-9) ram. longi. Foliorum ramorum principium laminae 10- 

 20 (raro ad 35) x 6-12 (raro ad 20) mm. 



Type Collection and Locality: Pennell 15112 (HOLOTYPE PH); Peru, Dept. 

 Amazonas, along Rio Sonche, west of Molinopampa, dry sandy barren, 2400 m. 

 alt., 8 Jul. 1948. "Shrub. Petals anthracene violet." 



Distribution: northern Peru, alt. 2400-3200 m. 



Piura: Tadene between Provinces of Huancabamba and Jaen, Raimondi 2312 (USM). 

 Cajamarca: northwest of Socota, Stork & Horton 10141 (F, G-DEL, UC); Cutervo, Raimondi 

 3841 (USM), Raimondi 4111 (USM); Llama, Sandeman 4163 (K). Amazonas: Yambrasbamba, 

 Mathews 1257 (K); Chachapoyas, L. Williams 7572 (F, NY, US), Mathews 1258 (BM, K). 



B. quinquenerve is closely related on one hand to the complex including B. 

 huancavelicae and B. grisebachii, on the other to B. campanulare . Its 5-7-nerved 

 leaves and usually well-developed paniculate inflorescences serve as distinc- 

 tions from the small-leaved Peruvian relatives. The McLean Chinchao specimen 

 cited here was placed by Triana and Cogniaux under B. campanulare , but has the 

 inflorescence and sepals of B. quinquenerve , although nearly smooth trichomes; 

 the label on this specimen states "Ex Herb, de R. and P. Lima" and the sprig is 

 matched exactly by a lanceolate-leaved sprig on the herb. Ruiz & Pavbn Chinchao 

 sheet (F) and the Rivero collection (P), so probably this collection has been mis- 

 credited to McLean. 



The small foliage of var. pusillum gives it quite a different appearance from 

 var. quinquenerve-, the inflorescences are also less well-developed than in the 

 typical variety. Unfortunately, no collections from areas between Cajamarca and 

 Huanuco have been seen, so the geographical disjunction, if any, and the varia- 



