Re: Do You Know, well do you???

pumpkins@mallorn.com wrote:
> 
> Harold,
>   Is it true  that a pumpkin self pollinated will have more genetic 
> variance (a weakened > strain) than a pumpkin crossed with a 
> different variety?

  In genetics, one is always talking about probabilities. If one selfs
any plant (self pollination) he is probabily working with a smaller
number of alleles. Therefore, the probability of getting two good
alleles is better in a cross than in a self simply because you have more
to pick from. However, by selfing you may expose and get rid of some
weak genes (alleles). 
  Allele is a good word to understand. Firstly, a given point on a
chromosome is called a locus (point where a gene is located}. There may
be many versions of a gene which can reside at that locus. Those
differing versions are called alleles. Allele and locus are two good
words to understand and use because lots of laymen  use gene
interchangeable for locus and allele causing beginners to get confused.
  The usual job of a gene is to make an enzyme. For example, one or more
enzymes acting one after the other might change a colorless substance to
an orange color to the delight of a Pumpkin breeder. 
  Back to the original question, my experience has been that when I
begin inbreeding any crop, I get mostly inferior plants because there
are so many genes coding for defective enzymes (etc) in the population.
In corn after 2 selfings, I commonly have mostly little seedless plants,
many bright purple, and that ends the work. (If you get no seed, you
can't go any farther). Any good genes are lost because the plant was not
able to make seeds. 
  I took that problem to experienced corn breeders and they told me this
problem was discovered in the first few years of corn work about 1900.
Those early breeders found a better method was to plant a dozen or two
seeds from the original ear and remove the worst plants and then use
pooled pollen from the best plants on the best plants. I tried that and
it works. Some of the plants will be selfed and get discarded for
weakness. Some will be selfed with two good alleles at a locus and be
kept. It is still very random, but I think it works. 
  Inbreeding (selfing) is a way to find and discard inactive genes. In
corn, we have gotten rid of so many of the poor alleles that corn yields
have increased. About 15 years ago, I recall we heard talk that we might
begin planting purebred corn instead of hybrids because some of the
inbreds were of such high quality. 
  Even when one is crossing pumpkins rather than selfing, he stands a
chance of finding and discarding undesired traits such as green skin.
Eventually, we will be rid of the "green skin alleles" and can be sure
that orange x orange with produce only orange offspring. I assume that
has already happened. I use it as illustration of what lies in the
future for other more complex traits. 
  Pumpkins, cows, and strawberries have so many loci (plural of locus)
that we never know very much for certain about the alleles present in
the two parents we are contemplating. 
  I do not want to lead anyone up a blind alley, but my personal plan is
to pick two good blackberry parents which have as many of the desired
traits as possible and mate them. As far as I can see, it seems all
breeders and geneticists do that.
  Sometimes it is possible to run chemical tests or microscopic
examinations and learn details of the parent's alleles which are not
visible to the naked eye. A well known example is sickle cell aneamia in
human. Via the microscope one can distinguish those red blood cells
which carry the sickling form of hemoglobin. 
  One assumes the size of a pumpkin depends upon many loci. The root
hairs need to extract nutrients from the soil and many loci probably
affect that. The minerals must flow to the leaves and fruit and surely
many loci affect that. You need good food manufacture in the leaves and
many genets affect that. 
  There is one special area. The mitrochondria, main producers of ATP
(energy), are found only in the cytoplasm of the egg and therefore come
only from the female. The sperm is too small to contain much cytoplasm
and hence no mitochrondria. I suppose a person trying to breed a huge
pumpkin would try to learn what traits, if any, are carried only by the
egg. In snails, the direction of twist in the shell is determined by the
egg. When cell divisions of the snail zygote begins the spiral of the
pile determines the future twist--or so I have heard.  


> The best crosses for stability are from grandfather to grandchild,
> this will make repeatable the plants characteristics?

  This may be one model breeding system, but I have no knowledge whether
it is useful or not.  I have not heard of this breeding rule. I have
just spend a little time with pencil and paper but have not come to any
quick conclusion. Most of my life has been at the molecular level and
not in practical plant breeding. I know very little about accepted,
proven, mating systems. I have conducted insect research in labs which
were trying to develop such practical breeding systems. They were mainly
interesting finding useful systems quickly with insects so that poultry
and livestock breeders could test them. 
  Since pumkin bears egg and sperm on the same plant, using a livestock
breeding system for pumpkin might not be wise. A system proven in plants
might yield faster results. Also keep in mine that the genetics of corn
endosperm is different than the genetics of a pumpkin shell. Lessons
from soybean breeding might be more useful to the pumpking breeder. If
you can find any academic breeding work on pumpkin, that work would be
more applicable to huge pumpkins. 

> I self pollinated many of my best pumpkins last year. If I plant 
> these seeds am I less likely to get a big pumpkin this year?

  On average I am "guessing the selfs will tend to be runts", but if you
are lucky to get a huge pumpkin with few bad traits, then I would
treasure that seed as it is less likely to have as many inferior alleles
hidden (masked) by good alleles. 
  If I wanted to get beyond the crossing the best with the best, I would
go to the library and look for the books that chicken and corn breeders
study. I would look for the results of breeding systems which have been
followed many years. At Purdue they used mice and Tribolium castaneum in
the Population Genetics Institute. I saw interesting work in progress
there in 1959-1966. Now they have 30 years more results. That allows
time for 300 beetle generations. I would look at work like that for
clues. Much of their work is highly mathematical and hard for me to
understand. 
  If you decide to look at the academic literature, some of it speaks of
model breeding systems. They pick a model, such as biggest with biggest
and repeat that for dozens of generations. We had insects 3 times as
large after a dozen or so generations. I guess pumpkin breeders have
done better than that. 
- 
Harold Eddleman Ph.D. Microbiologist.       i*@disknet.com 
Location: Palmyra IN USA; 36 kilometers west of Louisville, Kentucky
http://www.disknet.com/indiana_biolab



---------------------------------------------------------------------
To sign-off this list, send email to majordomo@mallorn.com with the
message text UNSUBSCRIBE PUMPKINS
Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index