There is surely something physiologically significant 
in the position of the labellum. It does not seem to be 
a matter of indifference to a species whether or not the 
flowers are upside down; whether the labellum is adaxial 
rather than the lowermost member of the perianth. Just 
recently I repeated an experiment I have often tried and 
by means of a string and anchor bent the inflorescence of 
Calopogon pulchellus so that the tip was directed toward 
the ground. In this species the labellum is normally the 
uppermost member of the perianth. In about a week the 
flowers began to expand and in each one the labellum had 
assumed the position characteristic of the species. These 
flowers had adapted themselves promptly to an unusual 
situation and the labellum was the uppermost perianth 
segment although the raceme was inverted. In the flow- 
ers of Habenaria lacera the labellum is the lowermost 
segment of the perianth. If the inflorescence is forcibly 
inverted so that the apex is directed toward the ground, 
the flowers as they expand will exhibit varying degrees 
of resupination, some of them being in the position they 
would have occupied had the raceme been allowed to re- 
main upright. Goodyera pubescens presents an unusually 
interesting study in connection with resupination. If 
plants with the lowermost flowers fully expanded are in- 
verted, the ovaries of the remaining flowers will continue 
to twist and anthesis being imminent, will for the most 
part cause complete resupination even though no percep- 
tible torsion was observable in the ovaries of the buds. If 
plants with very young buds are inverted, resupination 
is checked and the labellum remains adaxial throughout 
the raceme. (cf. plate on p. 157) If a raceme, interme- 
diate between imminent anthesis and extreme Juvenility, 
is inverted, the flowers may turn variously so that the 
inflorescence appears to be composed of flowers in every 
conceivable stage of resupination. It is as if at the mo- 
[ 155 } 
