494 
Arber and Parkin . — Studies on the 
Parlatore, 1 author of the Gymnosperm section of De Candolle’s 
4 Prodromus ’, divided the Gymnospermae into three groups of equivalent 
rank, viz. the Gnetaceae, Coniferae, and the Cycadaceae ; a conclusion 
which was also adopted by Bentham and Hooker. 2 Here for the first time 
we find the Gnetaceae clearly separated from the Conifers. From this 
period onwards the two genera, with the addition of W elwitschia in 1863, 
have been constantly associated together as a separate group of naked 
seed-plants. The more modern trend of opinion, as we shall see, has been 
to isolate them more and more, and to raise the taxonomic value of the 
group to that of a class, equivalent in rank to the Coniferales and Cycadales. 
Reviewing the main steps by which the Gnetacean genera have risen 
in schemes of classification, it is to be noticed that, on the one hand, their 
inclusion in the Dicotyledons became prohibited when the significance of 
the naked ovule was realized, and on the other hand the presence both 
of floral envelopes simulating perianths, and of male sporophylls, showing 
a likeness to Angiospermous stamens, as well as the dicotyledonous type 
of foliage possessed by Gnetum, precluded their retention in the Coniferae. 
They have thus, in the history of classification, hovered as it were midway 
between the Coniferae and the Dicotyledons. 
Passing from taxonomy to phylogeny, a paragraph in Lindley’s 
‘Vegetable Kingdom’, 3 though written before the publication of ‘The 
Origin of Species ’, may be worth quoting. ‘ There exist, however, a few 
plants, not very similar to each other in appearance, bearing the names 
Gnetum and Ephedra , in which we find precisely the structure and habit 
that would be wished for by a theorist searching for evidence to bring 
Gymnogens into communication with true Exogens ; for one of them has 
all the appearance of a Chloranth and the other of a Casuarina ; and yet 
both retain the true peculiarities of Gymnogens.’ 
Parlatore 4 suggested that a connexion existed between the Gnetaceae 
and the Amentiferous and Piperaceous families. Agardh 5 concluded that 
a close relationship was to be found with the Loranthaceae, a view which 
was received favourably by Henfrey. 6 Hooker 7 also, from his study of 
Welwitschia, showed that this plant, in common with Gnetum and Ephedra , 
presents some very curious points of resemblance to both the Loranthaceae 
and Santalaceae. Strasburger, 8 referring to the functionless ovule of the 
male fructification of W elwitschia, remarked that such a flower calls to 
mind the hermaphrodite flower of the higher Phanerogams, and that 
doubtless here are to be derived certain Angiospermous groups, such as 
the Loranthaceae. 
1 Parlatore (’68), p. 347. 2 Bentham and Plooker (’83), vol. iii, p. vii. 
3 Lindley (’53), p. 232. 4 Parlatore (’67), p. 100, (’68), p. 346. 
6 Agardh (’58), pp. 317-20. 6 Henfrey (’59), p. 299. 
7 Hooker (’63). 
8 Strasburger (’72), p. 243; see also the plate opposite p. 264. 
