913 
PrantL **, in which, in the genera observed by Enorrmann®3, 
Irmiscu®4, Kreps** and Lussock*’, the at first arched cotyledon 
finally becomes erect, usually carrying the seed remnants up 
on its tip. Murcraea differs from its allies in the Agavoideae, 
and thus agrees with the majority of Amaryllid genera, in 
that the cotyledon remains arched and does not lift the seed 
above the ground (Pl. XLV), — F. tubiflora, reported as germi- 
nating after the manner of Agave, being in reality a Beschorneria 
as Irmiscn noted when recording its germination. 
Like the related genus Agave, Furcraea has developed both 
marginal and median variegation of foliage, the former in 
F. Selloa marginata and the latter in F. gigantea medio-picta. 
In the first-named the etiolated parts pass from nearly white, 
with a slight rosy shade in youth, into a rather bright yellow; 
while in the other the coloration is scarcely more than creamy 
even in perfectly adult leaves. The structural modification in 
both is closely comparable with what has been observed in 
Agave, the etiolated tissue being otherwise normal mesophyll 
and, in the median-banded form, underlaid by deep-seated 
pale green cells. In the marginate F. Selloa the scape is less 
markedly etiolated; but in the median-banded F. gigantea the 
entire inflorescence is cream-colored. In both, the variegation 
is transmitted to the bulbils of the inflorescence, being well 
marked from the first in the latter but declaring itself clearly 
only after growth has begun, in the former. As in most such 
cases, the record of the origin of these variegated forms is obscure. 
f. Selloa marginata, described by Jacosi®? in 1870 under 
the name Fourcroya Lindenii, was collected in 1867 by Wats 
in the mountains flanking the Cauca valley of Colombia, where 
Scattered small individuals were found. Jacos1's inference from 
this mode of occurrence that it was propagated only by seeds 
and would not be found to be bulbiferous is not confirmed, 
in that it is now known to be freely bulbiferous in cultivation, 
While the] formation of fruit has not yet been recorded. 
The form of F. gigantea with median variegation (Fi. XI, 
below ; PI. XLVI—XLVIII) seems to have received @ horticultural 
