346 ROBINSON. 
1 mm long^ lamina -mcmbranaceouSj oblanceolate or oblong-oblanceolate, 
12 to 26 mm long, 2 to 7.5 mm wide, the base acnte or obtnse, the margins 
entire, the apex blunt; venation obscure, except somewhat the costa. 
r 
Luzon, Province of Cagayan, Sanchez Mira, Bur. Sci. 7^10 Ramos. The genus 
contains from 70 to 80 species, all from Australia, Tasmania, or New Zealand, 
with the exception of the Philippine plant and P. hrevituha Fawc., of Timor, 
which is its closest ally, the latter having longer sepals and stamens. They 
are referable to the section ThecantheSf but are somewhat intermediate between it 
and Calyptrostegia, having the united involucral bracts of the former, but tlie 
receptacle circumscissile as in the latter. The locality of this exile is at the 
extreme north of Luzon, the elevation was not noted by the collector, but 
according to his recollection, it grew in light forest or jungle, at less than 100 
m elevation. *^, 
elIeagnaceae. 
ELAEAGNUS Linn, 
Elaeagnus philippensis Perr. in M€m. Soc. Linn. Paris 3 (1824) 114. 
Elacagnus penotietii Schlecht. in DC. Prodr. 14 (1857) 613. 
Elaeagnus cumingii Schlecht. 1: c. 
Elaeagnus alingara Schlecht, 1. c. 615. 
Elaeagnus angustifolia Blanco Fl. Filip. (1837) 74, non Linn. Sp. PI. 
(1753) 121. 
Elaeagnus latifoUa Auct., quoad Philip., non Linn. 1. c. 
Realization of the fact, which ahould never have escaped my attention, that 
Perrottet had collected plants of this genus in the Philippines, has caused me 
to reexamine his description of E. philippensis, which I had stated showed his 
plant to be a Capparis.^ There is room for doubt that my statement was wrong, 
for wliile some of the less important characters do point to a Capparis common 
near Manila, all the more essential facts are otherwise. The Philippine species 
of Elaeagnus, exhibits considerable variation, but none of the difTercnees are of 
sufficient importance to warrant segregation. It ranges from the extreme north of 
Luzon, to Mindanao, but is represented by comparatively few collections. Per- 
rottet's name, which has been neglected, is much the oldest, and should be used. 
MYETACEAE. 
DECASPERMUM Forst. 
Decaspermum fructicosum Forst. Char. Gen. (1776) 74. 
This, as represonted by Vaupel 201, from Samoa, has much similarity to both 
Philippine species. Tlie nearer alliance is with D. hlancoi Vidal, but the leaves 
of the latter are distinctly less acuminate. In this respect, the closer resemblance 
is to D. paniculatum Kurz, but the flowers of D. fruticosum are larger, and the 
inflorescence less branched than is usual in D. paniculatum. 
EUGENIA Linn. 
Eugenia balerensrs nom. nov. 
^ Eugenia hrunnea C. B. Rob. in Philip. Journ. Sci. 4 (1909) Bot. 372, non 
(Berg) Niedenzu in Engl. &. Prantl Nat. Pflanzenfam. 3^ (1893) 81. 
A new name would seem to be required for the above, but the original will 
again become valid under the Vienna Code if some future monographer should 
consider the original E. Irunnea to be untenable. 
"This Journal 3 (1908) Bot. 305. 
