One striking characteristic of some bee-pollinated Macaronesian
Lotus species is their ability to modify flower color after
anthesis (Sandral et al., 2006). Flower color change is not unique
to the Macaronesian Lotus species, as this phenomenon has been
observed in other continental Lotus species (Weiss, 1995). Observations
in other groups suggest that color change is cued by
pollination, although unpollinated flowers will also change color,
although more slowly (Jones and Cruzan, 1999). We have been
able to make extensive observations of color change, both of wild
plants in the field in the Canary Islands, as well as of cultivated
plants in nurseries and gardens. Flowers after anthesis may modify
coloration from yellow/cream (prechange) to brown, pink, orange,
purple or red (postchange) depending on the species (Fig. 1D–H).
In this study we investigate first whether the four putatively birdpollinated
Lotus evolved in lineages with the capacity to modify
flower color after anthesis and second, the pigment modifications
involved in the transition. In particular we wish to answer the
following questions: (1) Is the ability to modify flower color associated
with the evolution of red/orange flowers in this group?
(2) Are prechange and postchange flowers likely to be perceived
differently by pollinators in bee-pollinated species? Similarly, (3)
are the putatively bird-pollinated flowers likely to be less discernible
by insects than bee-pollinated flowers from the foliage
background? (4) What were the modifications in pigment composition
and expression of anthocyanin genes during the evolutionary
transition in flower color and how does this compare with developmental
post-anthesis flower color change? The answers to these
questions will then be used to discuss the more general hypothesis
that developmental flower color change is a pre-adaptation for the
evolutionary transition to bird pollination syndrome.