A. candidissimum

[Mellard, David]

>Are seeds ever offered for sale of the various varieties?
>(although if they are unstable as I mentioned above then that'd be pretty
>pointless as there'd be no better chance than normal seed).

Hi Paul,

This gets into the wonderful realm of genetics.  If Mendelian genetics hold
true (that is, if color resides in the chromosomes as opposed to
non-chromosomal factors that affect color (e.g., plastids, etc)), then you
could cross a white cand. to a white cand and get

1.      all white ((if both parents are homozygous for the gene(s)
that give white, that is, both sets of genes on each chromosome(s) in each
plant are "white genes"), or
2.      you could get white and pink flowers ((if one or both sets
of parents are heterozygous (that is, the genes in at least one parent have
"white and pink" genes on the same chromosome(s).
3.      you could get white, pink, and green ((based on the sample
principle as in #2, its just that all three color genes are present in one
or both parents)
4.      Dominance and recessive (that is, which gene and hence color
is dominant over another  gene, which is then designated as recessive) also
comes into play in deterimining the number (ratio) of plants that come out
different color.

I doubt if you bred white to white if you would get all pink or all green.
To muddy this even more, I had to add an "s" because sometimes it's more
than 1 gene that controls color.

It gets even muddier when non-Mendelian genetics get involved in color and
its probably beyond me to explain further since all my training is in
mammalian genetics and that seems to be a life-time ago.

David

P.S.  Sometimes I like to think of myself as a mammalian toxicologist living
vicariously in the plant kingdom.