PHILIPPINE URTICACEAE, II. 
307 
stipules narrowly oblong, 4 to 6 mm long, acuminate, with cystoliths on 
the costa, ciliate or glabrous. 
Luzon, Subprovince of Benguet, Mount Tonglon, in mossy forest at about 
1,900 m elevation, Phil. PI. 812 Merrill. With this belongs Merrill 7644, from 
a more shaded place at a slightly lower elevation on the same mountain. The 
leaves average slightly larger, -and are conspicuously thinner, with fewer marginal 
teeth. Otherwise the two collections agree down to the details of the flowers, 
.and the stipules and venation are the same. E. purpureum has as much the 
general appearance of E. lagunense Merr. as of any other Philippine species, 
but the staminate bracts are less corniculate, the leaves smaller, and the 
stipules shorter. Probably the closest alliance is with E. lithoneurum Stapf. 10 
43. Elatostema scapigerum 0. B. Rob. in Philip. Journ. Sci. 5 (1911) 
Bot. 542. 
While additional collections have in general confirmed the statements previ- 
ously made regarding the position of the staminate receptacles of this species, 
a single small plant of Vanoverbcrgh 1236 has the peduncle arising from one 
of the upper leaf-axils, and on another plant the peduncle is about half-way 
up the stem. In rare cases, therefore, the species might fail to be recognized 
by the characters assigned to it in the key, but it is still very different from 
any other of our species. On the whole, it may best be compared with E. 
longipedunculatum Elmer, but the structure of 'the staminate receptacles is 
very different, resembling more that found in the ordinary pistillate receptacles 
of' the genus. By dissecting a young receptacle, also from Vanoverbergh 1236, 
which had attained a diameter of about 8 mm, or about half that of the average 
mature receptacle, it was easy to trace the 6 bracts typical of the staminate 
receptacles of the genus, but the two pairs which ordinarily are at least 
partially covered by the outer pair were at this stage already lateral to the 
latter. It is probable that still younger stages might show an even more typical 
structure. These two outer bracts were ovate, 2.5 to 3 mm long, the apex 
rather acuminate-mucronate than corniculate, for the apical projection is formed 
by the costa, which bears no spur. The other two pairs of bracts are about 
2.5 mm long, also ovate, less acuminate. Still another receptacle of the same 
collection has a peduncle over 13 cm long. All suspicion that the form might 
prove to be a monstrosity has been removed, and its range has been extended to 
Benguet, at Baguio, Merrill 7882, and on Mount Tonglon, Merrill 7637, 7645. 
2. Elatostema variabile 0. B. Rob. in Philip. Journ. Sci. 5 (1911) Bot. 514; 
1. c. 6 (1911) Bot. pi, 1. 
20. Elatostema banahaense C. B. Rob. 1. e. 5 (1911) Bot. 526. 
28. Elatostema carinoi W. R. Shaw ex C. B. Rob. 1. c. 5 (1911) Bot. 532. 
All of these have been supplemented by additional collections, some of which 
are as typical as could possibly be desired, while others have presented very 
difficult problems. As regards E. banahaense, there are two additional collections 
from Mount Banajao, Merrill 7516 and one from the herbarium of the Philippine 
College of Agriculture, obtained by 8. Asuncion, the former certainly and 
the latter probably gathered within a very short distance from the exact type 
locality. Both have more mature staminate receptacles than those of the type, 
and while the bracts are corniculate, this character is so slightly developed 
that it is inconspicuous in the mature receptacles, so that the species is better 
placed near E. carinoi than with E. viridescens and its allies. Moreover, the 
See above, page 297. 
