NOTES ON SOME CURIOSITIES OF ORCHID BREEDING. 



475 



Cattleya, and Sophronitis, forming as it were a connecting link 

 between them and the reed-like Epidendra. Indeed, recent 

 results of hybridisation seem to show that the second group of 

 Epidendrura is more closely allied to Cattleya, Laelia, and 

 Sophronitis than to the first group. Thus E. ciliare of the second 

 group when crossed with Laelia anceps and L. autumnalis pro- 

 duces ordinary hybrids fairly intermediate in character (Epi- 

 Laelia x Hardyana and E.-L. x belairensis respectively) ; 

 whereas when E. vitellinum of the second group is crossed with 

 E. radicans of the first group the result is similar to crossing 

 E. radicans with Sophronitis, Cattleya, and Laelia — namely, a 

 slightly modified E. radicans (the only traces of the mother 

 parent E. vitellinum being a slightly glaucous tint of the leaves, 

 a tendency of the reed-like steins to enlarge at the nodes, 

 together with irregularly modified flowers E. x radico- vitellinum 

 O'Brien, Gard, Chron. July 10, 1897, p. 16). 

 To sum up briefly we find that — 



(1) Cattleya, Laelia, Sophronitis, and the pseudo-bulbous 

 Epidendra when intercrossed produce normal hybrids inter- 

 mediate in character. 



(2) The reed-like Epidendra when united with one another 

 also produce ordinary hybrids intermediate in character. 



(3) But when the reed-like Epidendra are crossed with the 

 pseudo-bulbous Epidendra or with Cattleya, Laelia, and Sophro- 

 nitis abnormal hybrids are produced having the essential 

 characters of the reed-like Epidendra. 



And, having regard to recent researches into the nature of 

 fertilisation, we are almost bound to conclude that Cattleya, 

 Laelia, Sophronitis, and the pseudo-bulbous Epidendra contain 

 in their germ-cells some nuclear particles of the reed-like 

 Epidendra. 



These remarkable curiosities in Orchid breeding therefore 

 go far to prove that not only are the pseudo-bulbous Epidendra 

 (like E. vitellinum) descended from a reed-like Epidendrum 

 ancestor, but that it is highly probable that Sophronitis grandi- 

 flora, Laelia purpurata, and Cattleya Bowringiana are also so 

 descended. 



This opens up a wide field for the student of evolution — a 

 field into which we cannot enter now, being beyond the scope of 

 these notes. 



M 2 



