20 RECORDS OF OBSERVATIONS ON SIR WILLIAM MACGREGOR § 
discolor; but as Vaccinium Teysmannii has also no tubules to its anthers, the 
separation of our plant from Vaccinium would not be justifiable merely on that ground. 
Of the several species of Diplycosia, described by Beccari, no specimens are available 
here for comparison; but that distinguished naturalist placing them unreservedly 
into Vacciniacez, indicates a position in Vaccinium itself, according to perigynous 
insertion of stamens and corolla; whereas the genuine species of Diplycosia, as 
originally defined by Blume and subsequently confirmed by Bentham and J. Hooker, 
would best form a section of Gaultiera according to the hypogynous insertion of 
corolla and stamens, and the necessarily thus only basal fixture of the ovulary, the 
bracteoles counting merely for a sectional position of these plants in their genus. 
The anthers of our new plant, as regards their form, remind of those of Wittsteinia. 
Vaccumum parvulifolium. 
Branchlets thin, densely beset with very minute hairlets; leaves quite small, 
short-stalked, almost obovate or somewhat orbicular, nearly flat or shghtly bent, 
slabrous, devoid of any denticulation, shining on both sides, their venules beneath 
slightly prominent; flowers axillary, solitary, very small, glabrous throughout ; 
pedicels thin, shorter than the flowers, without any conspicuous bracts and bracteoles ; 
adnate portion of the’ calyx semiovate, several times longer than the deltoid lobes ;- 
corolla of about the lenoth of the calyx, hardly longer than broad, its lobes much 
shorter than the tube ; stamens at least half as long as the corolla; filaments very 
narrow; anthers somewhat longer than the filaments, provided with tubules about 
as long as the cells and with very minute acute posterior appendicles ; style stoutish, 
longer than the stamens; stigma truncate; fruit very small, globular-semiovate, 
smooth, five-celled ; ovules several in each cell. 
Mt. Musgrave, at elevations from 7000 to 8000 feet. 
Leaves rigidulous, the majority from a quarter to a third of an inch long. 
Flowers only seen on this occasion in a still unopened state, the corolla then a 
sixth of an inch long. Fruit seemingly not becoming succulent, but two only 
available not fully matured and possibly somewhat deformed, the cells apparently 
to a small extent subdivided. Seeds, well ripened, not accessible for examination. 
This plant stands evidently in near relation to V. obovatum, but the calyx is 
shorter lobed, the filaments are very conspicuous, the anthers glabrous and their 
appendicles hardly perceptible. Comparison with V. microphyllum, which remained 
by autopsy unknown even to Miquel, could here not be instituted; the brief diagnosis, 
given of it by Blume, is quite insufficient for recognition of that species, at any rate 
when obtained from beyond the Sunda-Islands. V. Rollinsonii is distinguished by 
few-flowered racemes, longer calyx-lobes, some vestiture on the inside of the corolla 
and on the filaments, as well as by absence of conspicuous anther-tubules. 
