Some Thoughts on Remontancy
Ryck Birch
I'd like to share some thoughts I've been mulling over on remontancy in
Roses. I would like to thank my correspondents who have patiently
endured, encouraged, and thoughtfully responded to me.
Some Observations:
1) There are two sources of remontancy in our hybrid roses. One line
comes from the Autumn Damasks/Damask Perpetuals [DPs]. The other from
the Chinese roses. Both of these lines started being actively developed
at roughly the same time [the DPs got a head start; some work was being
done with them--not much--before the Chinese roses hit Europe].
2) The Hybrid Perpetuals [HPs], which arose out of the combination of
the Chinese and European rose stocks, were remontant, that is, they
rebloomed, but were not continuous bloomers in the sense that the Tea
roses were or our modern Hybrid Teas [HTs], Floribundas [Fs], et al.,
are. In fact, they show a considerable range of ability in rebloom:
from the scant flower in the fall to generally consistent and reliable
cycles of bloom (like our modern roses).
3) Edward LeGrice, the English breeder, derived a number of 'unusually'
colored roses from crossing Floribundas and some of the dark red and
purple colored "old roses". In particular, Tuscany was used. LeGrice
remarked that the progeny of such crosses were marked by blooming on
second year wood [a dominant trait] and yet retained remontancy.
Some Analysis:
The reblooming characteristic is generally considered to be a simple
Mendelian recessive trait. Crossing a reblooming rose with a
non-remontant rose yields non-remontant roses in the first generation
and these must then be back crossed with remontant roses to recover the
expression of the recessive reblooming characteristic. In most cases
(we'll return to the LeGrice observation shortly), this appears to be
borne out in actual breeding and certainly should be used for general
planning of breeding programs.
What intrigues me about the HPs and their remontancy is this: they
should have pretty cut and dried expression of reblooming factors under
classical Mendelian inheritance theory, when in fact, they don't. The
HPs have shadings and degrees of reblooming ability that make them
challenging materials to work with in hybridizing.
In short, it looks to me that remontancy in the HPs appears to follow
something more like a quadrivalent inheritance expression or pattern for
the reblooming factor. Instead of the genes for remontancy being either
expressed or not, the genes for remontancy are expressed with cumulative
force as the number of remontant genes increases over the sets of seven
basic chromosomes found in roses. In tetraploid roses, there might be
none, one, two three or four copies of the remontant gene being
expressed. Hence, all those shades and degrees of rebloom.
Now this is what strikes me as problematic: How do we get from observed
classical Mendelian inheritance patterns--bivalent inheritance of
traits--to the observed quadrivalent inheritance patterns so common in
tetraploid garden roses [such as blossom length, glossiness of leaves,
mildew resistance, doubleness of blossoms] and seemingly implicated in
the HPs with respect to remontancy?
There are a number of different possibilities. Let me offer one
conjecture based on that report of LeGrice and some biology.
If we take LeGrice at face value, it suggests that something else is
also involved with the expression of the remontant factor in our roses.
To me, it suggests that perhaps we should view remontancy as a composite
of two traits: 1) that of reblooming, and 2) a dominant co-trait, the
recessiveness of the expression of reblooming. Further, let us suppose
that each of these capacities follows quadrivalent inheritance patterns.
Consider the genetic locus for remontancy as being filled with a gene
that allows for remontancy, or, filled by a gene that does not. The
co-trait of the recessiveness of remontancy would act solely on the
non-remontant gene and have the effect of amplifying the remontancy
suppressing capabilities of the non-remontant gene's expression.
Look at it this way. In nature, a plant in the temperate world does not
want to keep flowering on into winter. It blooms and then ripens its
fruit. This takes time and energy--typically more than what is required
to make additional vegetative growth. The plant's general growth cycle
includes the sequence of bud-sprouting, growth, flowering, and ripening
of seed. Under normal circumstances, the plant undergoes a response to
the signal that it's the right time to grow and flower. The flowering
genes switch on, initiate flowers and then switch off.
If conditions are favorable, a temperate adapted plant will indeed
continue to grow after flowering but typically not reflower. This is a
nice balance between the time and energy necessary to flower and ripen
seed and the opportunity to get more physical growth in. This balance
requires that the general growth cycle must be modulated so that
flowering and fruiting do not occur late where the plant would waste
energy in lost seed killed by the oncoming winter. So, this would be
the function of our dominant, remontancy repressing gene [the co-trait
mentioned above]: to suppress flowering on any additional growth during
the season.
Remontancy, in general, would then be a malfunctioning in one, the
other, or both of these growth regulating mechanisms. Such malfunctions
would center on 1) the gene(s) for flowering not shutting down when
finished with its(/their) activity in response to the signal "It's
Spring and time to flower", and, 2) on the inability of that(/those)
same gene(s) to respond to the signal to remain quiescent for additional
growth occurring during that season.
What would remontancy based on a lack of repression of flowering on
additional growth look like as opposed to remontancy based on a lack of
flowering genes turning off? I would suggest that remontant flowering
based on flowering genes always being on would give us something like
the Chinese style of blossoming where flowers just keep on coming with a
minimal pause in growth. Whereas, the remontancy based on a lack of
additional flowering being repressed would look rather like the style of
old remontant European roses where there is a pronounced pause in growth
before additional flowering occurs.
Generally speaking, I would expect that cold climate adapted remontancy
would take the form of the lack of rebloom suppression while warm
climate adapted remontancy would be based on lack of flowering genes
being shut off. I am particularly thinking of R. rugosa in this
regard. Fast seed ripening presumably allowed such a mutation--no
rebloom suppression--to survive and multiply. I would also imagine that
the Chinese roses may have been more weakly selected for rebloom
suppression given their mild climate. (This would have bearing on the
problematic issue of getting these roses to go dormant those of us in
the North.)
It is intriguing for me to think of the initial fusion of Chinese and
European roses and their representative remontancies with all this in
mind. Chinese roses probably brought remontancy based on continuous
flowering [flowering genes never shut off] and some level of dominant
bloom suppression on additional wood (whose effects would be insensible
to the appropriate mutant gene). The European roses contributed some
level of incomplete suppression of rebloom on additional growth and
fully functioning flowering genes--those which properly shut off after
their activation. It is not surprising to me then that the initial
efforts of hybridization were so problematical in achieving reliable
rebloom.
I certainly think of the HPs in this regard. Brent Dickerson remarked
in a personal correspondence that to his eye, the Bourbons in particular
combined the two styles of remontancy: especially the dwarfer ones.
LeGrice probably hit on the fortuitous combination of a modern
floribunda with both kinds of remontancy and bred with a rose--again,
Tuscany as reported--that by chance had a recessive condition with
respect to rebloom on additional growth. My own hybridizing experience
with old roses and those crossed with modern roses suggest that we still
have a mixed population as far as the types of remontancy are concerned.
Some Last Thoughts:
One of the issues in following this story is that of the conversion of
the chromosome base of our cultivated tetraploid roses--I beg the
indulgence of our diploid friends-- from an alloploid nature to an
amphiploid one.
In a nutshell, simple wild roses are made up of two sets of a base
number of seven chromosomes. Their chromosome compliment is then 7 x 2
== 14. Other roses have higher numbers of chromosomes but these numbers
are almost entirely based on higher even multiples of seven, e.g.,
tetraploid roses are 7 x 4 == 28 chromosomes. Etc. Of course, when the
rose creates pollen and ovules, the rose's base chromosome number is
halved: half for the pollen, half for the egg. The genes go through a
sorting process before this reductive division so you get a random
assortment of new traits when egg and pollen combine.
It is thought (and experimentally supported) that most natural higher
ploidy based species arose from hybridization between lower ploidy based
species. It is a matter of classification in assuming that there were
generally five types of "simple" diploid (7 x 2 == 14) species. These
species gave rise to similar sorts and through hybridization to those
higher order species. The British researcher C. C. Hurst proposed that
these simple five species types had certain traits that were preserved
differentially when they interhybridized and called these stable
collections of traits "Septets". So, it was possible to derive the
species types from which higher ploidic species were came from. These
five species types were arbitrarily assigned the symbolic markers A
through E. Each group was represented in its diploid form as a doubling
of its base symbol: AA, BB, CC, DD, EE. Interspecies hybrids are
represented by combination of their constituent simple septets: R.
gallica is determined to be AACC, as is R. damascena. R. chinensis is
determined to be AA in most circumstances.
[Be mindful that Hurst's theory has a number of issues and that while
predictive in the big picture, there really hasn't been a whole lot of
careful research--at least that's been published--in quite a while.
Hurst published in the late 1920's.]
In alloploidy, the septets are considered to remain generally separate
in 14 pairs of chromosomes (for tetraploids). In amphiploidy, the
chromosomes tend to clump up in higher numbered groupings. When two
chromosomes clump together, the arrangement is known as bivalent. When
four chromosomes clump up, it is known as quadrivalent. Generally, the
greater the clumpiness of the chromosomes, the higher the chance that
the gene sorting will occur in unusual ways. Bivalent inheritance
patterns are the ones Mendel and his peas talked about. Quadrivalent
inheritance patterns are how roses work in a lot of traits. In
quadrivalent inheritance, dominance is expressed as a matter of discrete
degree--where from one to a full four copies of any gene may be found
and subsequently expressed. It is this quadrivalent inheritance that
makes for SO many combinations and rose breeding such fun/a challenge.
It is a matter of extraordinary personal interest to me to think that
all of the breeding and wrangling with remontancy, and the other
beauties of the rose, occurred while this transition was going on: the
changing of inheritance patterns as the chromosomes themselves changed
their basic manner of expression and development. It is also a matter
of great practical significance to those who may want to breed with
species.
Given the complexities involved, I offer all this with a grain of salt
and the hope I have been clear on some of this...
Warm regards to all,
Ryck Birch