LILIACE®. 209 
membranous beak much shorter than the rest of the spathe. Flowers 
very numerous, in a dense globose or shortly ovoid umbel, never 
intermingled with head-bulbules. Perianth leaves connivent, with a 
subscabrous or smooth keel. Stamens exserted, nearly half as long 
again as the perianth, the 3 interior filaments 3-cuspidate, with the 
antheriferous cusp usually nearly as long as the undivided part, but 
sometimes shorter, and the lateral cusps usually shorter than the 
central one. Capsule ovate-subglobose, bluntly trigonous. Seeds 2 in 
each cell. 
On sands in St. Aubin’s Bay, Jersey. Once plentiful, but now be- 
coming scarce from building being carried on. Formerly on ledges 
of St. Vincent’s Rocks, Gloucester; but Mr. T. B. Flower tells me it 
has disappeared from quarrying operations. 
England (extinct). Channel Islands. Perennial. 
Late Summer. 
Bulb flowering when about the size of a cherry-stone, and rarely 
larger than a small nutmeg, enclosed in thin white coats enveloping 
the unequally-stalked bulbules which project above the chief offset 
like bunions on the base of the stem: sometimes the outermost coat of 
the bulb is fuscous. Stem 9 inches to 3 feet high. Leaves 6 inches to 
1 foot long, appearing at the close of autumn and decaying before the 
fruit is ripe, from the thickness of a crowquill to that of a swanquill 
at the base, more or less flattened or channelled above, at first with 
the ribs rough, but afterwards these become nearly smooth. Spathe 
generally tinged with purple, especially towards the apex. Umbel 
with very numerous close flowers; pedicels spreading in all direc- 
tions, slender, } to 1 inch long; the upper ones frequently longer, so 
that the umbel becomes ovoid; all erect and rather stiff in fruit. 
Perianth bright dark purplish-red; the leaves about 4 inch long, 
elliptical-oblong, obtuse, the three outer sometimes subapiculate. 
Anthers dark purple. Capsule about the size of a large hemp-seed, 
with three very blunt angles. Seeds shaped like one of the divisions 
of an orange, black, rugose. 
I have both the Jersey and St. Vincent’s Rock plants in cultivation, 
from roots received from Mr. T. B. Flower; and judging from these, 
I have come to the conclusion that there is no constancy in the leaf 
being channelled above or not. In weak plants the leaf has scarcely 
any channel, in stronger the channel is very.apparent, and in very 
luxuriant examples the channel flattens out, so that the upper side of 
the leaf is nearly flat. I therefore, at present, see no reason to be- 
lieve that A. Deseglisii is a distinct species or even subspecies. 
Round-headed Garlic. 
French, Ail a téte ronde. German, Rundhipfiger Lauch. 
VOL. IX. EE 
