106 



MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



[VOL. 10 



Additional Material: COLOMBIA: Van pes: Cerro Yapoboda, Rio Kuduyari, Schultes 4" 

 Cabrera 14249 (NY, US); Cerro Kanenda, Rio Kubiyu, Schultes $ Cabrera 1S339 (NY, US), 

 18389 (US); Mesa de Yambi, Rio Karuru, Schultes $ Cabrera 19127 (NY, US); R.'o Parana 

 Pichuna, Schultes $ Cabrera 19927 (NY); Rio Vaupes, Mitu, Humbert 27324 (P). 



The Munich specimens apparently were never annotated by Cogniaux or 

 Triana. None of the sheets bear de Candolle annotations but perhaps were seen 

 by him before being mounted. All three Munich sheets bear Schrank diagnosis 

 slips, although the diagnosis attached to the "Rhexia bicuspis" sheet is for 

 Leandra quinquenodis (DC.) Cogn. Drs Baehni and Weibel supplied the infor- 

 mation that there is only one herbarium sheet of Pachyloma in the de Candolle 

 herbarium, with a packet containing leaves and a flower along with a label 

 written by A. P. de Candolle "Rh. pachyloma L. Bresil M. Martins 1827"; 

 two other labels, not attached to any plants and not in de Candolle 's script, bear 

 brief French diagnoses of "R. amdzonum a ovatae" and "R. amazonum (3 lanci- 

 folia." 



Baehni and Weibel also supplied the following information: Martins kept 

 up a close correspondence with A. -P. de Candolle and called him "my dear 

 friend." In one of his letters to DC, dated March 13, 1827, he wrote; "Je viens 

 de vous envoyer par la messagerie le second cahier dn 2" volume de mes Nova 

 Genera." There is one thing of great interest in the Memoires et souvenirs de 

 Augustin-Py ramus de Candolle (issued in 1862). De Candolle says that he went 

 to Munich in Fall 1827 to attend a meeting of the German men of sciences; he 

 was received there with great kindness by M. de Martins; he writes (p. 377): 

 "Pendant les quelques jonrs que j'y ai passes, j'avais vu Les portions de Pherbier 

 de Martins relatives aux families dont je m'occupais pour le troisieme volume 

 du Prodromus." 



The single-branch stamen dimorphism in the Schules & Cabrera collections 

 is marked. Markedly dimorphic antesepalous and alternisepalous stamens are 

 found on 14249 (NY), 18339 (NY), 18389 (US), and 19127 (NY). Snbiso- 

 morphism is seen in H249 (US) and 18339 (US). The NY sheet of 19927 has 

 one isomorphic branch and a separate anisomorphic branch. An unmounted I S 

 sheet of 18339 has dimorphic stamens but nearly glabrous hypanthinm and 

 glabrous style; the NY sheet of this number has moderately <>landnlar hypanthia 

 and o-landular styles. The hypanthial and pedicellar glandulosity is so variable 

 that there seems to be no reason for maintaining De Candolle 's varietal distinc- 

 tions. On two of the Martins sheets at Munich, long connective prolongations 

 are visiblf ; the sheet of P. coriacevm var. glaberrimum now has no intact 

 stamens. 



2. Pachyloma huberioides (Naud.) Triana, Trans. Linn. Soc. 28: 64. 1871. 



Urodesmiu/m huberioides Naud. Ann. Sci. Nat. TTL 15: 338. 1851. 

 Pachyloma scandens Ducke, Arch. Inst. Biol. Veg. Rio 2: 65. 193."). 



Weak shrub 0.5-2.5 m, the terete branches glabrous. Petioles 2-8 mm. gla- 

 brous; leaf-blade 4-10 X (0.6-) 1-5 cm, lanceolate to ovate-elliptic, the apex 

 acuminate, the base rounded to subcordate, the maruins obscurely to notably 

 denticulate when young (obscurely so with age) and not in;-rassate (the blade 

 otherwise glabrous), 3-nerved or weakly 5-nerved, the nerves impressed above 

 and prominulous below, the secondary nervation obscure to finely prominulous, 

 sparsely punctate beneath at maturity. Pedicels 0.6-12 mm long below the 

 bracteoles, 1-3 mm above, the caducous linear to oblanceolate bracteoles glabrous 

 or sparsely glandular-setulose. Hypanthinm 6-12 mm long, suburceolate to long- 



