250 



MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



[VOL. 10 



stems having "villoso" pubescence refers to the type of spreading pubescence 

 characteristic of the lower portion of the stem of P. hirsuta. On the other hand, 

 the isotype of P. stricta Benth., which I have examined at NY, shows the pubes- 

 cence on the lower part of the stem appressed to ascending with smaller narrower 

 lanceolate leaves. This taxon (P. stricta) tends to have stems less branched and 

 with generally fewer, often solitary heads terminating a stem. However, various 

 intergradations exist between this taxon and typical P. hirsuta, for which I have 

 followed Bremekamp 's treatment of reduction of P. stricta to a varietal category 

 under P. hirsuta. K. Schumann (Fl. Bras. 6(6): 90. 1889) misinterpreted P. 

 stricta, placing it as a synonym under P. galioides, whereas, because of its op- 

 posite pairs of leaves, he should have placed it with P. hirsuta. Several collections, 

 such as Maguire, Cowan, & Wurdack 30938, Britton & Britton 2911, Fendler 991, 

 Britton, Britton, & Hazen 74, Baldwin, Jr. 4069, Ule 7636, Cardona 785, and 

 Duss 1450, are somewhat intermediate in approaching the habit and in showing 

 the tendency toward reduced leaves of var. stricta, but have been placed with 

 var. hirsuta because of the spreading pubescence of the lower part of the stem, 

 together with subovate spreading leaves. 



In general the lower half of the stems is covered with spreading hairs in 

 P. hirsuta var. hirsuta, but rarely have few hairs, becoming almost glabrate as 

 in the collections of A. C. Smith 2173 from British Guiana and Coulon 173 from 

 Surinam. These collections, likewise, show leaves wholly or mainly without the 

 characteristic pubescence present in most specimens of P. hirsuta var. hirsuta or 

 var. stricta. This tendency towards a glabrate-leaved condition is also exhibited 

 by Vie 7636, Cardona 833, and Samuels 246. As there are varying degrees in the 

 amount of foliar pubescence present Avhich do not appear to be correlated, and 

 as the reduction of hairs may possibly be correlated with age and maturity of 

 the plant, no separate category has at the present time been given for the spec- 

 imens with the glabrate-leaved tendency. 



Some authors (Schumann, Standley) have considered Mattuschkea hispida 

 H.B.K. as synonymous with Perama hirsuta. However, as is shown under my 

 treatment of P. galioides (H.B.K.) Poir., Humboldt & Bonpland's M. hispida 

 refers to a plant with the ternate-leaved condition characteristic of the P. 

 galioides group, rather than to leaves decussate in pairs, as characterizing the 

 P. hirsuta group. 



6. Perama vellerea K. Sch. Bull. Soc. Bot. France 56 (Mem. 3d) : 364. 1907. 

 Distribution. Known only from Minas Geraes, Brazil. 



BRAZIL. Minas Geraes : Serra do Cipo, Glaziou 19462 (holotype P, photo of 

 type from B in F and NY) ; same locality, Glaziou 19463 (P) . 



This species is doubtfully included in the genus Perama. It differs from the 

 other species of the genus in having 5 instead of 4 corolla-lobes, 5 instead of 4 

 stamens, strigose anther cells densely barbate at base, rose-colored corollas, and 

 flowers solitary in the upper leaf -axils. 



Platycarpum Humb. & Bonpl. PL Aequin. 2 : 81. 1. 104. 1809. 



At the time I treated the known species of this genus (Am. Jour. Bot. 39: 

 418-423. 1952) five specific taxa were indicated. In 1957 Bremekamp added a 

 sixth species, P. froesii, from Brazil (Meded. Bot. Mus. Herb. Rijks. Utrecht 

 141: 374-375. 1957). In this last paper he erects for Platycarpum, along with 



