2o8 The; cactaceiae;. 



Subtribe 8. RHIPSALmANAE. 



Mostly epiphytic cacti, generally growing on trees but sometimes clambering over rocks or 

 pendent from them, much branched ; branches alternate or often in whorls, slender, terete, angled or 

 flat and thin, spineless, except in Pfeijfera and Acanthorhipsalis; flowers regular, mostly small, rotate, 

 and without any tube or with a very short tube ; stamens usually few, attached to disk or near base 

 of flower- tube; style usually short; fruit a small juicy berry, white, red, or purple; seeds minute. 



We have placed this subtribe at the end of our monograph because it appears to us 

 to represent the most extreme differentiation within the family. It is indeed difficult to 

 explain to most people that its species are really cacti. 



We recognize eight closely related genera. 



Key to Genera. 



Flowers with a short definite tube. 



Joints terete i. Erythrorhipsalis (p. 208) 



Joints flattened, ribbed, or angled. 



Joints and flowers terminal 2. RMpsalidopsis (p. 209) 



Joints and flowers normally lateral. 



Joints with spiny areoles; ovary and fruit with areoles subtended by scales. 



Joints ribbed; fruit-areoles spiny 3. Pfeiffera (p. 210) 



Joints flattened or 3-winged; fruit-areoles not spiny 4. Acanthorhipsalis{p. 211) 



Joints not spiny; fruit mostly without areoles. 



Upper joints normally flattened; areoles not pilose 5. Pseudorhipsalis (p. 213) 



Upper joints flattened or 3-angled; areoles long-pilose 6. Lepismium (p. 215) 



Flowers without tube. 



Petals erect; ends of same joint unlike; flowers and branches always terminal 7. Haiiora (p. 216) 



Petals usually widely spreading; ends of same joint usually similar; flowers and 



branches lateral or terminal 8. Rhipsalis (p. 219) 



1. ERYTHRORHIPSALIS Berger, Monatsschr. Kakteenk. 30: 4. 1920. 



Epiphytic, with slender terete stem and branches, often pendent; branches dichotomous or 

 sometimes in whorls of 3 to 6 ; areoles scattered, small, all bearing several bristles ; flowers terminal, 

 regular, diurnal, white to rose-colored with a short but definite tube; ovary and fruit bristly, the 

 latter red; seeds much larger than in Rhipsalis. 



Type species: Rhipsalis pilocarpa Lofgren. 



The generic name is from epvdpo^ red, and Rhipsalis, referring to the red fruit and 

 to the resemblance of this genus to Rhipsalis. Only one species is known. 



The genus resembles in habit some of the species of Rhipsalis with round stems but 

 has a distinct flower-tube, on the top of which the stamens are borne. It also differs from 

 Rhipsalis in having a long exserted style, exserted even in the bud; in its slowly opening 

 flower (requiring several days to expand) ; in its very fragrant flower; in having its ovary 

 and fruit bearing areoles, each with a cluster of bristles; and in its larger seeds. 



Lofgren, when he described Rhipsalis pilocarpa, was inclined to think that it might 

 belong to Pfeiffera. In his latest treatment of it (Arch. Jard. Bot. Rio de Janeiro i : 68) 

 he referred it and Pfeiffera ianthothele to Rhipsalis under the subgenus Pfeiffera. 



At the place cited above, Berger proposed that Rhipsalis pilocarpa should be regarded 

 as a new subgenus of Rhipsalis but at the same time he incidentally made it the type of a 

 new genus. Mr. Berger, who has written most interestingly of it, says in part: 



"In 1903, lyofgren made known Rhipsalis pilocarpa (Monatsschr. Kakteenk. 13: 52 to 57) 

 which formerly had a fairly wide distribution in our collections. I received it from various sources, 

 the finest specimens coming from the Botanic Garden in Bremen, from which place it was sent for 

 naming. The plant grew well in my hothouse but appeared to prefer it cooler and sunnier. The 

 habit picture in the Monatsschrift, above cited, is not exactly right. The plant is striking because 

 of its beautiful bristles ; it is very odd. In general it does not differ from the rest of the species of 

 Rhipsalis. Because of its beautiful bristles one is persuaded to put it into Ophiorhipsalis. Meantime 

 the habit, the flowers, and the fruit show themselves to be a fundamental obstacle. 



" In all the species of Rhipsalis which I have had experience with, the ovary and later the fruit 

 are entirely naked; at the most there is at times a little scale. In these plants, however, the ovaries, 

 which in form remind one of those of Cereus, bear a mass of small tubercles with little scales, in whose 



