sible origin, on the one hand with Curculigo and Hypozis, 
and on the other with the leafy species of Burmannia(e.g. 
B. longifola). Apostasia is hardly to be considered as a 
descendant of Neuwiedia, since its floral aspects and other 
morphological characters suggest rather a parallel course 
of development for both taxa. Curculigo and Hypowis 
both have globular seeds with a sclerotic seed coat, as 
does Apostasia; Neuwiedia is characterized by an ad- 
vanced type of seed with a loose testa. In addition to this 
characteristic, the non-saprophytic members of the Bur- 
manniaceae show other primitive characters, such as axile 
placentation, monocolpate microspores, etc., which are 
also present in the Apostasioideae, Cypripedioideae and, 
to some degree, in the Neottioideae. 
The presence of connecting trends between the Orchid- 
aceae and its allies indicates that the family has been de- 
rived from other similarly complex groups and not, as 
often proposed, from any given family. 
It is safe to assume that the Orchid family originated 
somewhere in the Asiatic tropics, possibly in Malaysia, 
because those species which possess primitive character- 
istics, and also the allied families, are native to that area. 
The prototype of orchids, as Rolfe aptly wrote, may 
be visualized as ‘‘terrestrial monocotyledons, with an in- 
ferior ovary, numerous minute seeds having a reticulate 
seed coat and rudimentary embryo; the stamens and pis- 
tils are not yet aggregated into a column. The flowers 
were doubtless fertilized by insects, which on visiting the 
former, would become dusted with the pollen grains from 
the anthers as in the case of other entomophilous mono- 
cotyledons. We may also infer that the ancestral orchids 
were natives of great tropical region. The characters men- 
tioned are found in the Malayan genus Neuwiedia, the 
most primitive of the existing orchids. The species of 
Neuwiedia have short, erect stems with a tuft of plicate, 
[ 85 ] 
