y THE ORCHID REVIEW. (APRIL, 1917. 
state, where it grows intermixed with O. gloriosum, luteopurpureum, and q 
Lindleyanum, and, in another locality, with O. Hunnewellianum. Messrs. — 
Veitch long ago remarked that these natural hybrids were indefinite in q 
number, but hiyhly deserving of attentive study, though it was at present | 
hopeless to attempt any satisfactory explanation of their origin. And they 7 
added :— : 
‘From the first it has becn observed of these natural hybrids, that it is — 
an extremely rare occurrence for any two appearing in different importations ~ 
to be identical, although apparently derived from the same two species, and — 
sometimes even sufficiently like each other to come under the same name. 
Nor ought this to be wondered at; the agency by which these hybrids q 
and polymorphisms have been produced, has been in operation for ages past, q 
and it cannot but have happened that a large number of these forms, both 4 
of those that are known, and of those hereafter to be brought to light, are q 
not the immediate offspring of two recognised species or more primitive 7 
types, but are descended from their mixed progeny, further complicated by ~ 
an occasional cross with one or other of themselves. Hence we already q 
possess a number of forms that are “ confluent in series,” of which the two 7 
extremes are too widely separated for the interval to be covered by one, two, — 
or even more intermediate forms.”’ @ 
It was naturally hoped that some of these doubtful points would be- | 
cleared up when the possibility of raising Odontoglossum seedlings in 
quantity in gardens was demonstrated, but although some progress has — 
been made it is far less than was anticipated. The reason has been largely ' 
due to the possibility of using showier things, which happen not to grow — 
with O. crispum, as O. Harryanum and O. Pescatorei, or something that 
promised a new development of colour, as O. Edwardii, and outside the 
limits of the genus such fine things as Cochlioda Neetzliana and some of 
the Miltonias. The experiments have been largely upon parallel lines to 
those met with in nature, with the difference that we have records of many 
of the steps, and, of course, that the experiments have only been carried 
out for a very few generations. 
Such experiments have already indicated some of the difficulties that 
are inseparable from ascertaining descent from experiment, for the wide 
range of polymorphism and the amount of reversion seen among batches of © 
secondary hybrids could hardly have been foreseen. And such results 
must inevitably occur in nature, whatever the species. As regards 
improvement, the hybridist has an immense advantage over nature, in 
being able to select desirable forms as parents at every stage, instead of 
having his improvements swamped by a host of inferior forms. 
