Need a suggestion for an experiment
Night23
I'll be growing some plants indoors from seeds soon. I'd like to
perform an experiment to see if I can produce genetic variantion in a
certain plant's progeny.
Usually mutations are induced by bombarding seeds with low level
radiation or/and with chemical mutagens.
I would like your suggestion as to what household chemical could serve
as a chemical mutagen. Also some suggestions as to how to organize the
experiment would be nice. I plan to subject a set of plants to a host
of factors and if it produces interesting results, propagate its seeds
in the next generation.
Some thoughts I have had is developing a plant so that it can tolerate a
high level of salinity. e.g. adding table salt little by little
gradually ramping up the amount and propagating the individual plants
which can survive the ordeal. What do you think of this idea? How many
generations would it take rougly to see a plant which could withstand
very high salinity and still grow? (i know it depends on the species of
plant..).
Thanks for any help provided.
Jeremy Harbinson
also people. It is very unlikely that any mutagen would be allowed in
domestic products, basically they would act as powerful carcinogens and so
would probably be illegal. Also you can but mutagenized seed of some plants
for research purposes, and the progeny are largely pretty sick.
all the best,
Jeremy Harbinson
Neil McRoberts
Night23, I'd have to agree with Jeremy Harbinson
on this one. You don't want to be mucking about
with known mutagens unless you have ALL the
necessary safety equipment. In any case, you
should be able to perform the sorts of selection
experiments you mentioned on just on the basis
of selection from within normal populations. If
I were you I'd check out Rapid Cycling
Brassicas. These were developed at Univeristy
of Wisconsin and are available from there or
from an organisation called SAPS (Science and
Plants for Schools) in the UK - try doing a Web
search on either "Rapid Cycling Brassica" or
"SAPS".
Cheers
Neil
---
Night23
I checked out Rapid Cycling Brassica. $8.75 Pack of 50 seeds at
www.carolina.com (hidden somewhere in their online catalog). Flowers in
13 to 17 days - wow! Supposedly the result of 18+ yrs of selective
plant breeding.
Other than selective breeding, is there no artifical and safe way of
causing/introducing mutations in a plant's genes? I know that most
mutations are destructive however in a population of 50 seeds grown in a
tray, surely some will survive with their altered traits.
I had actually wanted to try it out on some exotic flowering plants
grown from seed purchased on the web to see what variation in
flower/leaf color, type might be produced. However the amount of time
per generation is just too long. Plant breeders are usually hammering
away at this their whole life selectively breeding plants to produce
bigger, more fragrant flowers of different varieties.. etc. I just
can't wait that long.
I am actually a computer programmer and was trying to think of
experiments on plants I could do to combine both fields. I recently
heard of an interesting experiment where a yucca plant was hooked up to
a computer with sensors on its leaves measuring slight changes in
electrical conductivity. The computer in turn was hooked up to an
online stock market. If the plant produced a certain electrical impulse
resulting in the computer buying a stock that made money, it was
rewarded with water and light. If not, it got no water and the light
went out. At last count, it was up 13% against the bourse which was
used at a benchmark!
Imagine having a host of these plants hooked up and then selecting the
progeny of the ones that made the most money. You'd have the next best
thing to money growing on trees.
Jeremy Harbinson
physiological process or condition is an area of active research for us (and
others), though using techniques based on chlorophyll fluorescence, and
light-induced absorbance changes. The general approach is called the
'talking plant' (dreadful name) and has numerous applications in glasshouse
environment control, dynamic control of controlled atmosphere storage etc.
Bugger all use for playing the stock market, though it might still work
better than less-than-gifted insights, certainly cheaper.
all the best,
Jeremy
Beverly Erlebacher
Night23 <nig...@nightynight.com> wrote:
>
>Other than selective breeding, is there no artifical and safe way of
>causing/introducing mutations in a plant's genes? I know that most
>mutations are destructive however in a population of 50 seeds grown in a
>tray, surely some will survive with their altered traits.
You used to be able to buy seeds that had been zapped with radiation for
school projects from places like Carolina and similar suppliers.
Selective breeding can't introduce mutations, it can only manipulate
what's already there. You can also strike lucky by finding a plant
with a desirable single-gene altered trait, e.g. in flower color,
and propagate from it. Or you can select over generations for a
multigene trait like plant height. Or you can cross plants with
different genetic backgrounds, even different species, and look for
interesting traits in the following generations.
In an earlier post you mentioned selecting for resistance to salinity.
This is important work, as agricultural lands are being destroyed at
an appalling rate in many parts of the world by build up of salts
from irrigation water. Unfortunately, you can't just take a handful of
wheat seed and select for generations, hoping to get saline resistance.
You can concentrate all the saline-resistance in that particular gene
pool, if you are careful and lucky, but you will reach a dead end unless
you can introduce more salinity-resistance genes from other populations,
assuming such genes exist and can be found. One of the motivations for
genetic engineering is to bring genes into a population that can't be
obtained by conventional breeding, genes from organisms that are
extremely unrelated, e.g. frost-resistance genes brought into tomatoes
from a species of fish. There has already been some work done to bring
salinity-resistance genes into crop plants from far-unrelated plants that
naturally live in high-salinity environments on seashores or deserts.
>I had actually wanted to try it out on some exotic flowering plants
>grown from seed purchased on the web to see what variation in
>flower/leaf color, type might be produced. However the amount of time
>per generation is just too long. Plant breeders are usually hammering
>away at this their whole life selectively breeding plants to produce
>bigger, more fragrant flowers of different varieties.. etc. I just
>can't wait that long.
Unfortunately, if you "just can't wait that long", plant breeding may not
be the hobby for you. Money can substitute for time to some extent - you
could invest in some metal halide lamps and try to get 2-4 generations
per year of some annual plants.
You'll have noticed that most plant breeding is done with species that
have a relatively short life cycle, like annuals and herbaceous perennials.
If two years is not too long, you can have some fun crossing tomatoes,
squash, or other easy-to-pollinate plants and seeing what you get both
in the first year and after incrossing the progeny from your first cross.
There are a lot of extremely weird tomato cultivars, but you usually have to
get them from a seed exchange. Generally available squash varieties are
much more diverse, but you have to be aware that there are 3-4 species
involved, and most interspecific crosses between them are infertile in
either the first or second generation. It's not too difficult to figure
out which variety belongs to which species, and almost all of them are
either Cucurbita pepo or C.maxima. You could grow your first cross tomatoes
under lights to save a year, since you only need a few fruits to get the
large population of F2's which will show the diversity you can select from.
>I am actually a computer programmer and was trying to think of
>experiments on plants I could do to combine both fields.
You could do some simulation and modelling stuff - this would give you
short generation times, and can be fun to program. It's a great way to
get a feeling for population biology and how selection works.
>I recently
>heard of an interesting experiment where a yucca plant was hooked up to
>a computer with sensors on its leaves measuring slight changes in
>electrical conductivity. The computer in turn was hooked up to a
>online stock market. If the plant produced a certain electrical impulse
>resulting in the computer buying a stock that made money, it was
>rewarded with water and light. If not, it got no water and the light
>went out. At last count, it was up 13% against the bourse which was
>used at a benchmark!
A novel random number generator!
Btw, if you are interested in the way the stock market works, look into
some statistical analyses of performance of mutual funds. Although thse
funds are managed by people who presumably have deep intensive knowledge
of the market, their performance has a very strong random element to it
despite the averaging implicit in mutual funds. I.e. good or poor performance
of a fund compared to others in a particular year is not a very good
predictor of its future performance.
>Imagine having a host of these plants hooked up and then selecting the
>progeny of the ones that made the most money. You'd have the next best
>thing to money growing on trees.
If you want to make money, plant breeding is not the answer!
As far as selecting for money-divining plants, I am reminded of the early
'parapsychology' experiments at Duke U. They gave a lot of people tests
for telepathy, and selected those with the top 10% of scores. When they
retested these people, they found that only 10% got comparable scores, the
remainder formed the rest of the bell curve. Rather than realizing that
this is exactly what you would expect when the individuals you select are
no different than the ones you don't select, they came up with some theory
about people with psychic powers unconsciously suppressing them.
There's also the failed efforts to develop Dalmation dogs with even and
symmetric spot patterns. Selection didn't work, because this 'trait' is
in the eye of the beholder and not the genes. Dogs with 'good' patterns
form a population no different genetically than other Dalmations.
The summary is, you can't select for something that doesn't exist.
duncan vincent
"Jeremy Harbinson" <"Jeremy Harbinson"@users.tbpt.wau.nl> wrote in message
news:3B091233...@users.tbpt.wau.nl...
> The problem with matagens is that they will not only mutagenize plants but
> also people. It is very unlikely that any mutagen would be allowed in
> domestic products, basically they would act as powerful carcinogens and so
> would probably be illegal.
You would think!!! Try coffee or caffeine pills. It is an old school way
of creating new genes. It can temporarally unwind DNA!!
Night23
know it could unwind DNA!
How would I go about doing this if I had a few seeds I wanted to mutate?
I'm thinking of dunking the seeds in a glass of warm coffee and leave it
overnight to seep.
Or "water" the plant with warm coffee periodically after it has
germinated and the roots might absorbe caffein during uptake.
Has it ever been tried? I assume Nescafe (with caffein) should be fine
:-)
Please send me some advice on how to go about this..
Night23
> Or you can cross plants with
> different genetic backgrounds, even different species, and look for
> interesting traits in the following generations.
Lots of good info. This one in particular caught my attention.
Could you tell me how I could do this? Would I just rub the pollen from
the stamen of one species with the pistil of another? e.g. poppy pollen
with roses?
> There has already been some work done to bring
> salinity-resistance genes into crop plants from far-unrelated plants that
> naturally live in high-salinity environments on seashores or deserts.
I'm curious to know, what is the process by which they snip the gene
from one species and insert it into the DNA of another? I don't know
about the snipping part but I heard a certain virus implanted with the
modified DNA is used to infect the species... or something along those
lines. I don't suppose I would be able to do that myself at home :-)
> Money can substitute for time to some extent - you
> could invest in some metal halide lamps and try to get 2-4 generations
> per year of some annual plants.
When I look at the variety of dogs around, i am constantly amazed at the
diversity. Little ones, big ones, shaggy ones, short-fured ones..etc
etc. Dog breeding seems to have produced a huge variety of wierd
looking shapes and sizes in the span of what 500 yrs? Assuming a breed
has been bred continuously for that time - a new generation every 2 to 3
yrs. This means in under 250 generations, so much change has occured
such that if we were to find these animals in the wild for the first
time, they would be categorized as a different species altogether! It
makes me think then that my hopes of producing something interesting
(note I'm not aiming towards producing a different species here, just
something different), is not all that unrealistic.
> If two years is not too long, you can have some fun crossing tomatoes,
> squash, or other easy-to-pollinate plants and seeing what you get both
> in the first year and after incrossing the progeny from your first cross.
Again, could you tell me how to do this please. I have just got a bunch
of tomato seeds out of Roma Tomatoes I bought at the grocery store and
have planted them today.
> Generally available squash varieties are
> much more diverse, but you have to be aware that there are 3-4 species
> involved, and most interspecific crosses between them are infertile in
> either the first or second generation.
In lay man's terms, that means if I were to cross-pollinate say two
different varieties of sunflowers, the seeds which would develop would
be sterile?
> A novel random number generator!
Not necessarily. Supposing there is a pattern to the rise and fall of
stocks on the stock market. Evolution would find this pattern through
countless of selective breeding experiments (although how many
generations that would be I cannot say). Eventually if you were to hook
up the plant to control all stocks on the stock market and selectively
breed it, many generations down the line you could have a plant
spreading the risk and "investing" evenly in all stocks (much like
mutual funds) since economic growth is generally positive from year to
year. By doing so, it would thus ensure its survival.... until it hit a
time of economic depression when the economy contracts and its
water/light supply is cut.
Now if there was no pattern to the stock market, then it would just be a
random number generator. But given that people hand billions of hard
earned dollars to 'fund managers' who claim to be able to predict the
movement of stocks, I don't see what's wrong in performing such an
experiment to determine if a pattern exists. After all, the fund
manager is looking for a pattern and so is the plant.
Beverly Erlebacher
Night23 <nig...@nightynight.com> wrote:
>Beverly Erlebacher wrote:
>> Or you can cross plants with
>> different genetic backgrounds, even different species, and look for
>> interesting traits in the following generations.
>
>Lots of good info. This one in particular caught my attention.
>
>Could you tell me how I could do this? Would I just rub the pollen from
>the stamen of one species with the pistil of another? e.g. poppy pollen
>with roses?
The plants have to be related, usually within the same genus, occasionally
different genera in the same family. Many long-domesticated plants, such
as roses and lilies, are already complex hybrids of several species.
The mechanics of performing a cross depend on the species. If the plant is
self-fertile, you have to prevent self-pollination. You have to prevent
undesired pollination by wind or insects, too. You need to do the cross
when teh pollen is mature and the stigma is receptive. But basically, yes,
you just put pollen from one plant on a stigma of the other.
>> There has already been some work done to bring
>> salinity-resistance genes into crop plants from far-unrelated plants that
>> naturally live in high-salinity environments on seashores or deserts.
>
>I'm curious to know, what is the process by which they snip the gene
>from one species and insert it into the DNA of another? I don't know
>about the snipping part but I heard a certain virus implanted with the
>modified DNA is used to infect the species... or something along those
>lines. I don't suppose I would be able to do that myself at home :-)
There are a number of different methods, I don't know enough about any
of them to describe them accurately. Given money and skill, you could
probably do some of it at home. A lot of people do relatively sophisticated
stuff like tissue culture at home for fun.
>When I look at the variety of dogs around, i am constantly amazed at the
>diversity. Little ones, big ones, shaggy ones, short-fured ones..etc
>etc. Dog breeding seems to have produced a huge variety of wierd
>looking shapes and sizes in the span of what 500 yrs? Assuming a breed
>has been bred continuously for that time - a new generation every 2 to 3
>yrs. This means in under 250 generations, so much change has occured
>such that if we were to find these animals in the wild for the first
>time, they would be categorized as a different species altogether!
Dogs have been domesticated for many thousands of years, and even the
earliest records show that the major types were already in existence
2-3000 years ago, e.g. mastiffs, sight hounds, herding dogs, etc. Note,
however, that most of the weird traits, like the short legs of dachshunds
and bassett hounds are single-gene mutations. Dogs are only superficially
diverse, even very different breeds differ from each other in very few genes.
In 'The Origin of Species' Darwin uses the example of fancy pigeons to
show how rapidly artificial selection can produce dramatic visible
changes in an animal species.
>It
>makes me think then that my hopes of producing something interesting
>(note I'm not aiming towards producing a different species here, just
>something different), is not all that unrealistic.
It's not at all unrealistic. There are thousands of amateur plant breeders,
especially of easy to work with plants like daylilies. I'd encourage
anyone who likes plants to mess around crossing them, or planting seeds
that set by themselves, especially of plants with diverse forms. You will
get something new, maybe something you like better than what you've already
got, maybe just a bunch of spare plants to give to your friends. You are
not at all likely to get something with commercial value, but you're not
likely to win the lottery either, and this costs less. :-)
>> If two years is not too long, you can have some fun crossing tomatoes,
>> squash, or other easy-to-pollinate plants and seeing what you get both
>> in the first year and after incrossing the progeny from your first cross.
>
>Again, could you tell me how to do this please. I have just got a bunch
>of tomato seeds out of Roma Tomatoes I bought at the grocery store and
>have planted them today.
Well, suppose you decide that you want a yellow Roma-type tomato. You
would get a yellow tomato plant, cross it with your Romas, and plant
the seeds next year. I think yellow is a recessive trait, so next year's
tomatoes will be red, and I don't know enough about the genetics of shape
in tomatoes to predict that. Let those tomatoes self-pollinate, and in
the following year, you will get a diverse selection of plants, about a
quarter of them bearing yellow fruit, some of thsoe more or less Roma-shape.
Pick the ones that most resemble what you are aiming at, let them self-
pollinate, and plant the seeds the following year. It shouldn't take long
to get to your goal, depending on how many seeds you have space to grow out.
Btw, I think yellow Romas will not be in demand, since yellow tomatoes
have a rather unappetizing color when cooked into sauce.
There's a really excellent book called something like "Breed Your Own
Vegetable Varieties" by Carol Deppe, that has a lot of excellent info on
working with a wide range of plants. It's aimed at amateurs, and is very
well written. It really makes you want to get out in the garden with a
roll of masking tape, some colored yarn, a notebook, a pair of tweezers...
Much of the info is relevant to non-vegetable plants too. This book explains
the mechanics of crossing tomatoes as well as other species.
Btw, on the 'just getting lucky' front, one year I bought a bushel of
peppers to freeze, and found a couple of orange ones among the reds.
I saved some seed, and found out that not only did they breed true, but
they were early and productive in my not very pepper friendly climate.
I offered them on a seed exchange, and recently saw the variety, which
I had named 'Toronto Orange', on the list of a small seed company. I
find this rather charming.
>> Generally available squash varieties are
>> much more diverse, but you have to be aware that there are 3-4 species
>> involved, and most interspecific crosses between them are infertile in
>> either the first or second generation.
>
>In lay man's terms, that means if I were to cross-pollinate say two
>different varieties of sunflowers, the seeds which would develop would
>be sterile?
It means that you can't count on a cross between members of different
species producing fertile seed. If the sunflowers are the same species,
you should get good seed, given that growing conditions are good. I
wouldn't recommend working with sunflowers and related plants unless you
a steady hand and good eyesight. Squash and daylilies, on the other hand,
can be cross pollinated by clumsy people who can barely focus.
>> A novel random number generator!
>
>Not necessarily. Supposing there is a pattern to the rise and fall of
>stocks on the stock market. Evolution would find this pattern through
>countless of selective breeding experiments (although how many
>generations that would be I cannot say). Eventually if you were to hook
>up the plant to control all stocks on the stock market and selectively
>breed it, many generations down the line you could have a plant
>spreading the risk and "investing" evenly in all stocks (much like
>mutual funds) since economic growth is generally positive from year to
>year. By doing so, it would thus ensure its survival.... until it hit a
>time of economic depression when the economy contracts and its
>water/light supply is cut.
Like I said before, selective breeding only works with what is already there.
If you could just breed for anything in anything, we'd have wheat plants
composing poetry and rice programming computers. More to the point, there
would be pigeons flying faster than light.
>Now if there was no pattern to the stock market, then it would just be a
>random number generator. But given that people hand billions of hard
>earned dollars to 'fund managers' who claim to be able to predict the
>movement of stocks, I don't see what's wrong in performing such an
>experiment to determine if a pattern exists. After all, the fund
>manager is looking for a pattern and so is the plant.
Go ahead, but I'm not going to invest in it!