PUNICA. 643 
visible at a somewhat later period, when the ovula have just 
commenced to appear. . 
The affinities remain to be pointed out. These are most 
marked with Soneratia and with Leptospartion Lagerstree- 
mia grandiflora of Roxburgh. These three genera form a group 
uniting Lythrariee with Myrtacez, which group is at once 
known by its valvate calyx. They all agree in this, in the 
Structure of the petals, aud ovaria, and in habit. 
They differ from Myrtacez in almost every point, but the 
adhesion of the calyx with the ovarium. 
No pellucid dots exist in Punica, neither are there any 
traces of their existence, the pellucid portions which do 
exist are not of a glandular nature, and appear to depend on 
obliteration of green parenchyma taking place irregularly, I 
find however that in many instances, pellucid markings do 
exist, but they are unaccompanied with any aroma. There 
LE isa tendency towards an intromarginal vein, but to this I 
~ attach little importance. : | 
.. In habit and especially in the shortness of the branchlets, 
there is much analogy with Pomacee, as likewise in the 
tendency to become spinous, Compare this with Mespilus 
Japonia, | 
| , With Lythrariea they have most points in common agree- 
. hg with them, especially in the valvation of the calyx, the 
Structure of the petals and ovula, differing essentially in the 
union of the calyx with the ovarium, or in the inferior fruit. 
~ With Myrtacea they agree in opposite leaves without sti- 
E. Pulee, in the inferior fruit, differing in the valvation of the 
= calyx, the nature of the petals, stamina, and plura carpellisa- 
tion of the ovarium. 
. . Punica agrees with Memecylon in the convolute cotyle- 
E dons, and among Lythrariez above all others with Grislea, 
= the coloration of the flowers being in both the same, and 
So far as the calyx is concerned of an unusual nature. . 
What I have said about the greater permanence of the 
SMemata over the lamina of the carpellary leaf is on 
