* 
46 . ILLUSTRATIONS OF INDIAN BOTANY. 
keel is sometimes entire, and then, usually with a crest, sometimes 3-lobed, and without a crest, 
whence the crest is supposed to be the altered middle lobe of a 3-lobed keel. The stamens 
below are usually eight, ascending, combined into a tube adhering to the base of the petals ; some- 
times 4 and distinct ; when combined the tube is split opposite the upper sepal. Anthers clavate, 
erect, l-celled, opening by a terminal pore, rarely, by a longitudinal slit. Disk sometimes large 
usually copious and fleshy ; sometimes reduced to a thin plate, or wanting as in Xanthophyllum: 
Embryo Straight, radical next the hilum, cotyledons usually foliacious. The following further 
explanation of the nature of the irregularity of the flowers I extract from Dr. Lindley’s Natural 
n 
_ _“ The calyx apparently consists of but three pieces,which are usually green, and like sepals 
in their common state; but their real number is 5, the two coloured lateral petal-like bodies, 
a has an appendage of an anomalous character, called technically a crest, and often consisting 
therefore are Suppressed : or in krameria 4 one bei 
that the relative position of the fifth sey ring suppressed. 
Brown.” Denham 3}. sepal and petal respecti 
I may remark, in addition, 
vely, was first indicated by 
per Pei tt ee eet oer wag et 
. 0 change its lace, ’ : 
Polypetala D eCandolle considering It more allied ] Vilas sge Z leriece anae 
Fumariaceae, Papavera - C eit in his class: Rhoeadeae” elon ‘with Residacee 
all; , ceae, Cruciferae, and Capparideae, Lindley préfere placing it in his 
lacee, (horse chesnuts). The follow; ae (Sycamore’s Sapendaceae, 
ge say NS eXposition of the affinities of this order. is extracted 
* Polygalace ati : 
and ne got i an ue cinity of Pe oataalle between Droseracee and Tremandracceé, 
their hypogynous Garin. ¢ y : - cee. With the latter they are related on account of 
iy ase owers, and cucullate stigma; and with Tremandracee 
eo 
