150 
CONTRIBUTIONS TO BOTANY. 
fig. 43), we see the same apparent cylinder separating by the 
application of force into a spiral cord, formed of “ des sortes de 
vaisseaux fascicules rameux et tissu cellulaire.’^ Griffiths says 
of it {loc. cit. p. 304), “ to the upper portion and to one side of 
the cavity (in the albumen) is attached the embryo by means of 
an enormously long, tortuous, and spirally but irregularly twisted 
cellular funiculus, the cells being very much elongated and 
twisted; its length varies, when moderately pulled out, from 
3^ to 5 inches, the length of the fruit being 1 inch. This funi- 
culus, as well as the extremely similar one in Cycas, has the 
property of contracting when immersed in water ; when in situ, 
it is tolerably closely packed ; it is dilated towards its attachment 
with the embryo.’^ 
From Gaudichaud’s account we may infer that this suspensor 
IS composed of cellular tissue loosely compacted round bundles 
of very elongated pollen-tubes which become spirally twisted into 
a hollow cylinder, moulded into the size’and shape of a cavity of 
the albumen. Griffiths’s drawing seems to confirm this ; for 
when drawn out into a spiral thread, it appears like a bundle of 
such tubes. From its absence in Brongniart’s elaborate ana- 
lysis of Gnetum Gnemon^, we may infer that it is not a constant 
or essential feature in the organization of the seed. Roxburgh, 
who minutely describes the structure of the embryo in two spe- 
cies of Gnetum, makes no mention of any such suspensor; and 
Griffiths affirms the ‘^very rare and partial development of 
the funiculus ;” for, after examining abundance of fully-formed 
fruits of two species, he found only two instances where its 
partial formation, it lying loose in the cavity of the albumen, 
was observable {loc. cit. p. 302). Some additional light will be 
thrown on this subject in the following remarks. 
It has been the general opinion of botanists that the Gnetacece 
are intimately allied to the Coniferce and Cycadacece, some in- 
clining to their affinity with the former, others to the latter 
group, which two families are placed by Endlicher widely apart 
in his systematic arrangement; while others, again, congregate 
the three families into one class, Gymnosperms or Gymnogens. 
The principal argument in support of the relation to Coniferoe 
has been founded on the assumption that in the Gnetacece the 
seeds are naked — a doctrine which, it appears to me, ought 
never to have been applied to this family. In other respects 
there is no analogy ; for the Coniferce have a very different habit 
and another system of inflorescence, which is amentaceous, with 
achlamydeous flowers and pistils generated upon the under side, 
or on the margin of phylloidal bracts, forming altogether a de- 
Voy. Coq.,Bot. vol. ii. tab. i. figs. 6, 7, 8, 9, 10, 13, 14, 15. 
