tion about Neottioideae and NKerosphacroideae is virtually 
lacking. 
The known basic chromosome numbers for genera 
within each group are also included on Plate NIY. 
Hoffman, Dunean and Blumenschein have given the 
basic number of the Orchid family as x =20, calculated 
on the occurrence of this number in the majority of 
species of horticultural importance (Ophrydoideae and 
Kerosphacroideae). Such a conclusion is, however, un- 
warranted, because the best we can hope for, with our 
present knowledge, is to suggest only the basic number 
for each of the groups. 
The presence of an uninterrupted aneuploid series from 
n=10to n=22 in the Neottioideae, and extensive poly- 
ploidy in the Ophrydoideae and to a lesser extent in 
Kerosphacroideae, as well as the occurrence of euploid 
polyembryony in various groups, indicate a great genetic 
complexity in the family and, if this be fully exploited 
through further investigation and experimentation, it 
will contribute materially to our understanding of the 
origin and phylogeny of the Orchidaceae and probably 
go far towards full clarification of the many attendant 
problems. 
CONCLUSION 
The picture of the Orchidaceae which I have tried to 
present clearly demonstrates the great complexity of this 
family. The various degrees of specialization observable 
within each of the five groups make it rather difficult to 
trace the possible course of evolution. In the whole 
orchidaceous complex, Neuzwicdia is, perhaps, the most 
primitive member, but even this taxon has to be con- 
sidered as derived from more than one ancestor. In many 
respects, Neuwiedia is closely allied to the ypoaidaceae 
and Burmanniaceae, with both of which it shares a pos- 
[ 84 | 
