1896.) Flowers and Insects. 79 
i handl. 1886: 116, 
1 d. K. D. Vidensk. Selsk. For , 
Be ore are, ¢ BaD. Engler u. Prantl. Die ahs ergs 
ih ee 33- 1887. (Just 16': 563).—(23) Bulman, 
illo I : 
en N 
discolor, visitors, etc. : 5 
IS Tourn.—In this genus we find regular Si perap ttre 
tribe flowers. Each sepal, with a stamen and a ae e atin 
is modified. into a form which is almost a RUSTE ARN ee 
ofa specialized flower such as we find in the oh the kind of 
etc. The form of the style serves to ger tpn ollen to 
insect pollinators and secures the application o : E at Tae 
the insect’s back. Kerner (5,23) mentions pees showin 
* Serving to protect the pollen, without, ght ds devel- 
the Probability of this having anything to do w 
opment. 
i ers 
Except in the single case mentioned ain Bi hr 
are adapted to bumblebees, but are also visited y g 
bees, such 
Synhalonia, 
; erved 
rase as yet recorded is that of Aconitum lycoctonum, obs 
by Aurivill; 
ited by bu 
i by a 
Son (28) in an account of the blue-flag, accompanied by 
ae S 
ently of /. versicolor, implies that Jris in emcral ‘4 
eed to bumblebees and large pete Riverine 15 
father logs t Rhingia rostrata. ; 
ris has neeee peed eet as a type of ae aeg 
Cogame Contingente, Delpino 6), but J. séb¢rica has : 5 il ae 
‘© be proterandrous (Dodel Port 24, Loew 30). some ex- 
“en below that I. versicolor is also pr oterandrous eg is said 
nt. Meehan (9) records a case in which /. virgin 
to have Proved fertile under a net. 
