60 BOTANICAL GAZETTE [JANUARY 
30-50 high, fuscous; teliospores catenulate, adhering laterally to 
form a cylindric column, ellipsoid or cuboid, 12-14 «13-16y; wall 
pale cinnamon-brown, uniformly thin, 1 yu or less, smooth. 
On an undetermined Bignoniaceous vine, La Seiva Valley, 
April 1913, II, iii, Thaxter 38. 
The host has large trumpet-shaped flowers and large, ovate, pointed leaves, 
pale with dense pubescence beneath. The rust is unusually minute. The 
presence of telia was pointed out by THAxtTEr, who supplied a microscope 
slide with sections showing the telial structure. 
Maravalia, gen. nov. 
Cycle of development includes telia, and possibly pycnia. 
Telia subepidermal, erumpent, somewhat indefinite. Telio- 
spores free, one-celled, with apical germination, pedicellate; wall 
colorless, thin, smooth. 
6. Maravalia pallida Arthur and Thaxter, sp. nov. 
O. Pycnia unknown, probably not formed. 
III. Telia hypophyllous, numerous in circular areas 5-10 mm. 
across on somewhat larger yellowish spots, at first pulvinate, 
roundish, o.2-0.4mm. across, becoming larger, elongate and 
branched, somewhat confluent, early naked, yellowish becoming 
white, velvety by germination, ruptured epidermis erect or some- 
what overarching; teliospores elongate-oblong, clavate-oblong, or 
cylindroid, 13-22 58-67u, rounded at both ends or narrowed 
below, germinating upon maturity; wall colorless, uniformly very 
thin, about 0.54, smooth; pedicel slender, 8-1oy in diameter, 
20-35 long, colorless. 
On Pithecolobium latifolium (L.) Benth. (Zygia latifolia St. Hil.), 
Maraval Valley, April 1913, Thaxter. 
The genus differs from Chaconia, established by Jur. for a white rust on 
Pithecolobium divaricatum (Bong.) Benth., from Paraguay, by the mode of 
origin of the spores. Both are short cycle species; in Chaconia the spores are 
sessile and clustered on a large basal cell, while in Maravalia they are pedicelled 
and arise directly from the hymenial layer of hyphae in the same manner as 
is usual in the large majority of rusts (fig. 1). The genus is apparently a short 
cycle condition of the genus Spirechina. M. pallida much resembles in gross 
appearance S. epiphylla, a Texan rust on Rubus. In its teliospores it also 
much resembles S. epiphylla, and there is even a closer resemblance to those 
of S. Pittieriana, another Rubus rust from Costa Rica. Figs. 1 and 2 well 
illustrate the structure and habit of the rust. 
