9 
and referred it without doubt to the Ternstreemiacez, and Pro 
Oliver followed him in placing it in this order. ow, with the fruit 
before me, I cannot suggest any alternative, but it is not closely allied 
— nape are exceptional in the genera Marila and Haplocla- 
i 1 
thi the 
in itt sleenas, and a fruit ages septicidal dehiscence from the base, 
ach 
exceptional in n Archytea. In case the other characters are very 
different from those of echt gyne. It has also been suggested that 
this genus mig elong to the Guttifere, but the anatomical 
characters, the seed, and the fruit are not those of that order 
Mr. . Boodle, Assistant in the Jodrell Laboratory, Kow, who has 
partially studied the a anatomy of Medusagyne, finds that it differs 
from the Gutéifere in possessing no secretory cavities, and in the type 
of the stomata; the latter being usually surrounded by four to six 
epidermal cells. This type of stomata does not mater ially differ from 
sent. The presence of cortical bundles in the stem of Me dusagyne is i 
an important deviation from the anatomical characters of both (utti- 
fere and Ternstremiacee, but in this point it agrees with the neighbour- 
ing Dipterocarpacee. From the last-named order Medusagyne differs i in 
Consequently the embryo remains unknown ; but the probabilities 
are that the seed is Seeslbsainons —W. Borrine Hemstey,. 
Fig. 1, a partially expanded flower; 2, a fully riggers flower; 3, the — from 
which the sepals and petals have been removed ; different view of the same; 5, 
longitudinal section of a gynzceum, hoot the ave ag t of the ovules ; 6, cross 
section 0: ame, showing the thick central axis a erous carpels, which 
are almost free from each other, at this stage, up to the mepehon s; 7, an ovule; 
, an oyule attached; 9, dehisced fruit; 10 and 1], different views of the same Sine 
ffe 
a ‘seed ; 13, the same. All exc cept 7, 9, and 12 enlarged ; the adits natural siz 
