A simple but complicated question about selective breeding and saving seeds

My question is so simple it feels like a stupid question. But my best efforts at research have not produced a simple answer.

When selecting one or more fruits or vegetables from a single laden plant in order to save seeds, does it matter which one(s) of the fruits or vegetables you select?

Say I've got a heroic lemon cucumber plant that came from cheap dollar-store seed. (Which I do.) It made a 9-foot tower of vines and produces prodigiously. I want to save some seeds, because I know the next packet of dollar store seeds that says "lemon cucumber" on the pack probably won't be all that similar or produce as well.

But when I walk out to my plant, I've got a wide variety of cukes on there. Some are larger than a baseball, some are the size of ping-pong balls. Some of them are bright yellow, some are russeted, some have a hard skin that has to be peeled, some are softer and more pleasant to eat without peeling, some have lots of spines and some have few. There are also some variations in seediness and flavor.

My problem is that I don't understand all the sources of variation. The genetics of the original seed I planted are common to all the fruit, then we get epigenetics and their interaction with environmental factors (changes in temperature and moisture over time, bug damage to leaves and stems and fruits, variable amounts of exposure to bacterial, viral, and fungal blight pathogens, probably lots more I don't know about or can't measure), and finally, the influence (if any) from the pollen source by which each female flower was pollinated.

I know the next generation will take genes from the mother plant as well as from the pollen source, which may or may not have been a different plant depending on the plant type in question. But if I pick a tiny little cuke and a large pretty cuke from the same plant, is there any reason to think that choosing the big one will lead to bigger cukes in the next generation? Or should I save my best ones for eating and use the middlin' ones for seed because it doesn't matter?

I guess this is a long way of saying I'm hopelessly confused about genetics and heritability. High school bio was decades ago, Carol Deppe's books are a heavy slog for me though I tried, and I am at sea. I'm clear on the virtues of saving seeds from the best plant, but when I've only got one hero plant that survived my black thumb, I am uncertain how much (if any) effect I'll see from exercising further selection pressure between and among the fruits/veg on that plant. Does it matter, or am I overthinking this?

All feedback appreciated.

Dan Boone: Your post hits very close to home for me. That's because I am a plant breeder and I sell just enough vegetables at the farmer's market to support the plant breeding effort. I always save the best of the best for seed and take the seconds to market. How sad.... That I never get to show off the best that I can grow.

My reasoning for saving the best for seed includes the following speculations:

It seems to me like fruits that grow more vigorously and are more free of bug blemishes are likely to have had more resources available to produce larger seeds than fruits that struggled for whatever reason: water, rain, sunlight, humidity, bugs, shading, timing, daylength, microbes, etc.... I can't envision all the things that might affect the growth of seeds, but I can envision that fruits from the same plant that are larger are likely to have devoted more resources into storing more nutrients into the seed. Seeds with more nutrients get a faster start on life, and often spend the entire season growing more vigorously than plants started from smaller seeds. The difference is sometimes dramatic on plants like beans, where some seeds can have perhaps 10 times more energy stored in them than other seeds produced by the same plant.

I often sort seeds so that I can plant the largest seeds produced by a plant. I believe that results in more vigorous seedlings.

If I am trying to shorten the days to maturity of a variety, then I will often save fruits only from the earliest fruits to mature. I figure that way I am saving seeds from both early fruiting plants and from early pollinating plants.

My experience is that packets of seed are not as uniform as the seed companies would like us to believe... For example, last year I planted about 20 cowpeas. Only 4 of them produced seed for me. I find similar things happening with nearly every commercial variety that I plant. Yes, there could be variations in the soil or environment that affects individual plants. It could also be genetic variability. I saved seed from the 4 cowpeas and replanted. This year, every plant produced seed for me. That seems to indicate a strong genetic component to the differences I noticed last year.

Another thing that I notice frequently in online forums is that once plants get fully grown that people have a hard time distinguishing one plant from another... They do things like plant a hill of squash, and then wonder why one plant is producing different colored fruits (or even different species of fruits) when what is really going on is that there are several plants growing together in the same place.

I speculate that fruits exposed to sunlight produce more sugars than those hidden under foliage. I speculate that the additional sugars can be converted into larger fruits with more energy stored in the seeds.

I think of genetic influences as the major driver in heritability, with environmental factors playing a secondary role, and epi-genetic factors playing a tiny role.

How selective I am with fruit selection depends somewhat on how desperate I am for seed. If I'm growing a rare or precious variety and it produces only a few fruits, then I will save seed from anything that it produces, no matter how bad looking it is. But if there is an abundance of fruits to choose from then I save from the best examples. Sometimes I'll save two packets of seed from a plant "Best of", and "Everything Else".

Here's an example using tomatoes: This is an F2 hybrid so a rare variety. I saved seeds from both fruits into a common seed lot.

I agree with Joseph.
I feel that if I want to produce the best plants, I need to save seeds from the best fruit.

It has been shown that diseases and pests target the weakest plant in a setting.
Selecting for fruits without signs of pest/disease damage should therefore be selecting those healthiest individuals for the next generation.

Yes, complicated...

Another factor to think about is the size of the genetic pool.

Some species are very picky about how much genetic variability they need to continue to thrive. Corn is the classic example. Your seed corn will peter out if you have less than 40 or 50 plants that help pollinate each other, AND you have to have seed from many different plants to preserve that genetic diversity.

That book you mentioned tells you how many plants you need to keep up the vigor/diversity.

Some species are not picky at all, you could literally just use seed from the best fruit or vegetable, over and over again. But most need more diversity than that.



As a general principle, if in doubt, grow many plants, and get a little seed from many fruits.

Here is a pretty accessible article on losing genetic diversity if you only grow small population numbers.


http://www.nature.com/hdy/journal/v84/n4/full/6887050a.html

Mainly, I agree with Joseph too.
For eggplant, I keep the 1st nice fruit, they self polinate, easy.

For cucumbers.....
Beeeeeeees!
Your plant will be the mother, but your fruit might have many fathers.
Because your neighbours do not grow the same variety.

So you risk to have hybrids next year.
When you are at time to do it, for all cucurbs, hand polinate.
This means protect both female AND male flower from bees before opening, do the job, keep protected at 1st, and then surely keep noted which plant and fruit are securely home made!
Do enough to chose the best fruit as Joseph said.

With personal selection, in a few years you can have a modified variety, according to your orientation in breeding.

John Polk wrote:It has been shown that diseases and pests target the weakest plant in a setting.
Selecting for fruits without signs of pest/disease damage should therefore be selecting those healthiest individuals for the next generation.

Agree for disesases, not sure for pests....
I think they have the same taste as us, they like tender and nutrients rich plants.
All beings look for C= Carb = sugar, for example.

Also, less pests means a plant with natural defenses, that are anti-nutrients, also for the big 2 feet pests called humans!
I you take the example of grains, saponines and more than what I cn think of right now....

Hey everyone, thank you for your thoughtful responses!

Joseph Lofthouse wrote:
It seems to me like fruits that grow more vigorously and are more free of bug blemishes are likely to have had more resources available to produce larger seeds than fruits that struggled for whatever reason: water, rain, sunlight, humidity, bugs, shading, timing, daylength, microbes, etc.... I can't envision all the things that might affect the growth of seeds, but I can envision that fruits from the same plant that are larger are likely to have devoted more resources into storing more nutrients into the seed. Seeds with more nutrients get a faster start on life, and often spend the entire season growing more vigorously than plants started from smaller seeds. The difference is sometimes dramatic on plants like beans, where some seeds can have perhaps 10 times more energy stored in them than other seeds produced by the same plant.

That's a reason I had not thought of! And I agree it's a good one. I know that right now I have some "vine peaches" (mini yellow canteloupe-like melons) that did not finish growing before the vines expired for the season. They ripened on the dead vines and turned nicely sweet and yellow, but their seeds are clearly under-developed upon visual inspection. Luckily I saved some well-developed seeds from a few of the largest ones that ripened before vines succumbed to pestilence and senescence.

Joseph Lofthouse wrote:I often sort seeds so that I can plant the largest seeds produced by a plant. I believe that results in more vigorous seedlings.

Another very useful idea!

Joseph Lofthouse wrote:My experience is that packets of seed are not as uniform as the seed companies would like us to believe...

One of Carol Deppe's books goes into this in horrifying detail. And my own limited experience bears it out. This is my third year growing lemon cukes, each time planting from a "new" packet of no-brand seed bought at a chain store. In previous years some sort of virus or rust has killed all my plants after they produced just a few fruits. The plants from the packet this year are really ugly -- there's a lot of discoloration on the leaves and many leaves died prematurely -- but they managed to outgrow whatever was attacking them and produce well. Hence my interest in saving these seeds, no matter what else may have pollinated them.

Joseph Lofthouse wrote:Another thing that I notice frequently in online forums is that once plants get fully grown that people have a hard time distinguishing one plant from another... They do things like plant a hill of squash, and then wonder why one plant is producing different colored fruits (or even different species of fruits) when what is really going on is that there are several plants growing together in the same place.

Erm... Now that you mention it, my seven-foot tower of cukes might have two plants in it. I would have to stand on my head in a swarm of bumblebees to confirm, but I might have transplanted more than one seedling into that spot. Would explain some of the variation for sure!

Joseph Lofthouse wrote:How selective I am with fruit selection depends somewhat on how desperate I am for seed. If I'm growing a rare or precious variety and it produces only a few fruits, then I will save seed from anything that it produces, no matter how bad looking it is.

That reminds me, you might be amused to hear that I just finished the second year of a land-race experiment in what I am calling "field Okra", with the goal of breeding an okra that I can plant anywhere around the property that will grow and produce without any irrigation during the hot dry summer months. Okra grows excellently here with irrigation, so last year I planted two varieties in numerous likely spots beyond the reach of my garden hose. (40 acres of land here and I only have 200 feet of hose.) Virtually all the plants died in the deep-summer drought and heat, but I got two tiny pods with seeds. Planted them out this year in about six different places along with saved seed from my irrigated plants, and had much better success -- about eight pods, of which a couple were actually "eating size" before they dried down. Weather permitting (this year was better than last) I have high hopes for production amounts of dryland okra in the next few generations.

For "bad looking" I tried a similar experiment this year growing aggressively neglected canteloupe plants grown from copious distribution of supermarket canteloupe seeds. I actually got *one* ugly bug-ridden canteloupe fruit, but the seeds look viable so I'm really eager to see how they do next year.

Appreciate the inspiration and feedback.

John Polk wrote:I agree with Joseph.
I feel that if I want to produce the best plants, I need to save seeds from the best fruit.

I "feel" that way too, and it's what I'm doing. But I wish I were more certain of whether it's a science-based thing to do, or just this gardener's superstition.

Does anybody specifically know whether the pollen source affects the physical characteristics of the fruits produced in the same generation? If one flower is pollinated with pollen from a nearby plant of the same species, while another one is pollinated from a smelly-fleshed bitter wild gourd, can it make a difference in the cukes produced this year by those two flowers? (Obviously the answer is "yes" for subsequent generations in open-pollinating species.)

I am currently frustrated because I believe there is a biology buzzword for pollen effects on same-generation fruiting, but I can't remember it and can't seem to dredge it up with Google when I don't already have it at the tips of my typing fingers.

The xenia effect is commonly visible in corn. The color and other characteristics of the kernel may be different depending on the pollen donor.

For example, the colored kernels on the bottom cob were pollinated by the plant that produced the top cob.


The opaque/smooth kernels on this cob of sweet corn were pollinated by a flour corn instead of by a sweet corn.


The xenia effect isn't typically visible in beans, or squash. I wonder if a sweet pepper seed pollinated by a hot pepper would develop capsicum? Can't see that, but might be able to taste it.

In my garden, with a large honeybee apiary, I get about 5% cross pollination on squash separated by 100 feet. The corn plants in the above photo were separated by 3 feet and only a few kernels got pollinated by the off-type plant that accidentally got planted in a patch of white corn.

I think of the third year as the magical year in a landrace development project, so I have great hopes for the field okra experiment. The first year I grew okra, less than 1% of the plants produced seed and the plant was up to my ankles when it got frozen. The second year, a few plants produced seeds, and they grew as tall as my knees. The third year, one of the plants was as tall as I am, and I had enough okra to take to the farmer's market. This year the tallest okra plant is about 8 feet.

3rd Year Okra.

Joseph Lofthouse wrote:
The xenia effect...

That's the word! Thank you. Wikipedia: xenia (plants)

Joseph Lofthouse wrote:I wonder if a sweet pepper seed pollinated by a hot pepper would develop capsicum? Can't see that, but might be able to taste it.

On two or three occasions, I have had mouth-burning hot peppers grow on plants from seeds advertised as sweet pepper seeds. In one case, there were hot peppers and sweet peppers on the same plant, that looked the same but differed sharply in capsicum levels. I ascribed this to poor-quality dollar-store seed without thinking about it much.

Joseph Lofthouse wrote:
I think of the third year as the magical year in a landrace development project, so I have great hopes for the field okra experiment. The first year I grew okra, less than 1% of the plants produced seed and the plant was up to my ankles when it got frozen. The second year, a few plants produced seeds, and they grew as tall as my knees. The third year, one of the plants was as tall as I am, and I had enough okra to take to the farmer's market. This year the tallest okra plant is about 8 feet.

Yay! I've taken so much inspiration from your landrace articles. I am a lazy gardener and thus very interested in developing strains that "just grow" in my conditions. I'm convinced I can make the field okra experiment work, but I thought it would take longer than it seems to be taking. Field canteloupes are next...

I'm having good success developing weed tolerant varieties of hard to grow crops like carrots... (Hard to grow cause I couldn't keep them weeded.)

This question continues to plague my brain.

Yesterday I picked two gallons of sweet crab apples from a tree that was growing without care in a public space near me. Fruiting was prodigious, so I was able to pick just the biggest prettiest fruit. Of course I'm planning to plant seeds all over my property.

But it's late in the season. There were mushy rotten fruits in many of the fruit clusters. I'd grab them and get four good crab apples and a handful of mush from the one rotten one that slimed in my hand under the pressure of plucking the other four.

There are probably genetics that govern the apples' hold time on the tree before they start to rot. But once again, I'm uncertain whether there's any selection pressure in my act of picking nothing but late-holding fruit. It seems to me that the genetics governing that should not vary from apple to apple; every apple on the tree has the same genes for that. Their descendants will surely vary depending on the pollination source for each apple, but that's random and didn't effect the apples in the generation I'm picking. Right? I'm so confused...

The grand secret of plant breeding is that seeds grow up to resemble their parents and grandparents... We don't know much about the grandparents of the seeds you are collecting other than that their child produces crab apples.

Pollination is a highly localized event, and as far as I can discern crab apples are self-fertile, so the crab apple tree probably pollinated itself.

The fruits are maternal tissue, produced only by the mother plant. They are unlikely to be influenced to rot based on the genetics of the embryos in the seeds. Each seed in a fruit has the same mother, but each seed in the same fruit could have a different father.

Their descendants will surely vary depending on the pollination source for each apple, but that's random and didn't effect the apples in the generation I'm picking. Right?

Yup.

But what if... The fruits started rotting because one of the seeds in the fruit didn't have a properly functioning dormancy mechanism, and the seed started growing while still in the fruit, causing the fruit to rot? I've occasionally seen that with tomatoes. I'm more likely to blame the rotting fruits on damage from insects, hail, or sunlight.

Thank you! It does help.

Joseph Lofthouse wrote:But what if... The fruits started rotting because one of the seeds in the fruit didn't have a properly functioning dormancy mechanism, and the seed started growing while still in the fruit, causing the fruit to rot? I've occasionally seen that with tomatoes. I'm more likely to blame the rotting fruits on damage from insects, hail, or sunlight.

As I process these I am finding enough fruits where rot has started to maybe shed light on that. All the fruits with "rotten spots" I've found have had the rotten spots concentrated on the upper shoulder of the fruit, under unblemished and unbroken skin. That makes me think of hail or sunlight damage. We haven't had a hailstorm during the lifetime of these fruits, so sun damage seems a reasonable theory.

I was told that apples are polygenus? Meaning the seed will not produce the tree it wad taken from.The best way to get that same tree is to graft a branch onto scion root. Quicker too. But I still plant the seeds. Saving seeds is really fun, and I have and still am having amazing things going on in our garden. I save the best producing, best tasting and the earliest,since I live in a short season area. What amazes me is the new varieties we get and then end up isolating and trialing them. Watch what your plants are trying to tell you by what they produce. We are now trialing some of them for a small local seed company.

Apples don't breed true from seed. That's just a genetic fact of life.


Having said that, there is a variety that originated in russia, and somewhat in poland, that is far more likely to produce offspring from seed that are very similar to the parent.

It's a tart, cold hardy apple called Antonovka. It also makes great rootstocks that impart winter hardiness to many varieties of apples.

https://en.wikipedia.org/wiki/Antonovka

Yeah, I know. But I can scatter 500 crab apples without spending a dime. Any trees I get will be more fruitful (literally) than the scrubby saplings I'm clearing/thinning. If I don't like the fruit, the turkeys will.

Historically, most apples weren't for eating out of hand. They were for cider. You can use a huge spectrum of apples in cider and the cider comes out awesome. Especially compared with cider made from all Delicious apples--bland.

Not that it's entirely up to me, but my preference if possible would be that this thread not wander off into the endless conversation about the good and bad parts of growing apples from seeds. I feel pretty well briefed on that. The genetics conversation, on the other hand...

I've been thinking about this question in regards to okra... It is known that genes can be activated, or deactivated based on the environment: . And that some of those on/off states can be inherited by the offspring (keyword epigenetic). Cold stress, heat stress, drought, flooding, etc are common environmental conditions that can turn genes on or off. Even whole segments of genes. So what does that have to do with okra? I've been harvesting okra this fall as it matures. I might harvest a dry pod every couple of weeks from the same plant. Some of those pods matured in warm (for my garden) weather, while some matured in very cold weather... I wonder if the seeds that matured in very cold weather will have genes switched on for handling cold that the earlier harvests from that same plant don't have...

There is also a genetic component to this question... As the season got colder and colder some plants self culled, and I chopped some plants out for doing poorly in cold weather. So I can expect that the later fruits may have been pollinated by plants that were growing better in colder weather... Not quite the same scenario as an apple tree where every flower opens at about the same time. But I'm cussing myself that my seed saving strategy didn't include saving sibling groups with labels including an ID for the mother and a harvest date.

Good point! My okra has slowed down production dramatically, but is still chugging along. And I have been saving seed about once every two weeks throughout the season, though I didn't think to segregate and label.

Thank you for an excellent thread. How would pollen that is mostly from native plants affect sweet corn? Last year our corn was sweet but when cooked it was like corn to feed animals. I thought it was the dollar store variety, but now I wonder?

Anne Miller wrote:Thank you for an excellent thread. How would pollen that is mostly from native plants affect sweet corn?

Unless you have feral maize or wild teosinte near you, native plant pollen won't affect your corn -- it'd need to be the same species.

Thanks for the reply. So that is a possibility since teosinte is in Wildlife seed mixes or is that teosinte not wild? I don't know what was in our seed mixes or what is in my neighbors' mixes. I picked up some corn looking plants in our driveway but thought they were millet. I'll have to keep a eye out for corn looking plants when I am out and about. Since we came out of a drought the grass has grown so much that it is hard to see what all is growing around us.

Anne Miller wrote:Thanks for the reply. So that is a possibility since teosinte is in Wildlife seed mixes or is that teosinte not wild? I don't know what was in our seed mixes or what is in my neighbors' mixes. I picked up some corn looking plants in our driveway but thought they were millet. I'll have to keep a eye out for corn looking plants when I am out and about. Since we came out of a drought the grass has grown so much that it is hard to see what all is growing around us.

Teosinte is usually only wild in Mesoamerica. Your location isn't listed and I didn't want to assume. It's not something that would usually pop up unexpectedly. Millet-type grains won't cross with corn.

TO: Dan Boone
FROM: Eric Koperek = erickoperek@gmail.com
SUBJECT: Visual Selection in Plant Breeding
DATE: PM 4:05 Friday 6 May 2016
TEXT:

(1) "Breed the best with the best and hope for the best". I learned this from Burbank before you were born.

(2) Visual selection can be successful for simple plant traits, usually those governed by 1 or 2 genes. Visual selection is not successful for complicated plant traits (like yield) that are governed by many genes.

(3) For example: It is easy to select for ear number in corn. This is a simple trait that is easy to see and has high heritability. It is not easy to select for yield because this trait is not obviously visible = standing out in a corn field, which of many thousands of plants have "good" genes for yield? You can't tell because yield is a highly complicated trait and heritability is relatively low.

(4) You can't predict heritability unless you have direct experience breeding for particular traits (or you ask another breeder or dig this information out of an agricultural library).

(5) Plant breeding is a numbers game. The more accurate your selections, the more crosses you make, and the more progeny you test greatly increases your chances of sifting out the right combination of genes to produce a successful commercial variety.

(6) Be very careful about saving seed. You need to preserve seed from at least 40 different parent plants in order to prevent inbreeding in cross pollinated crops.

(7) Visual selection can be very effective way to identify plants with "good" genes. For example, if you want to breed disease resistance in tomatoes, plant hundreds of varieties of tomatoes in a heavy clay soil in a cold damp climate then turn on an irrigation system to keep the leaves wet. You are creating ideal conditions for fungal and bacterial infections of leaves and roots. Most of the plants will die. Save the best for breeding then replant their progeny in the same field. (No crop rotation! You want insects and diseases to build up). Repeat cycles of selection, breeding, and progeny testing for 8 or 10 generations. The survivors may not be edible but will have greatly improved disease resistance. Now you select and cross for fruit quality and hope that you get a variety that is both disease resistant and tastes good.

( It takes 15 to 20 years to develop a new variety using conventional, old-fashioned plant breeding methods. Sometimes you get lucky and can release a new variety in 10 years. Lots of time is needed because new varieties need at least 3 years of field testing before seed can be released for increase and eventual retail sale.

(9) The simple answer to your question is YES. Save the best looking fruits from the most healthy plants. Over many, many generations you will develop a "landrace" = farmer's variety. Heritability is very low for most traits but over hundreds of generations small improvements (and occasional large improvements) can be made. This is how ancient peoples developed their crops.

ERIC KOPEREK = erickoperek@gmail.com

end comment.

Dan Boone wrote:This question continues to plague my brain.  

Yesterday I picked two gallons of sweet crab apples from a tree that was growing without care in a public space near me.  Fruiting was prodigious, so I was able to pick just the biggest prettiest fruit.  Of course I'm planning to plant seeds all over my property.  

But it's late in the season.  There were mushy rotten fruits in many of the fruit clusters.  I'd grab them and get four good crab apples and a handful of mush from the one rotten one that slimed in my hand under the pressure of plucking the other four.

There are probably genetics that govern the apples' hold time on the tree before they start to rot.  But once again, I'm uncertain whether there's any selection pressure in my act of picking nothing but late-holding fruit.  It seems to me that the genetics governing that should not vary from apple to apple; every apple on the tree has the same genes for that.  Their descendants will surely vary depending on the pollination source for each apple, but that's random and didn't effect the apples in the generation I'm picking.  Right?  I'm so confused...

Another possibility is that those particular fruits were attacked by some pest and died.

Anne Miller wrote:I don't know what was in our seed mixes or what is in my neighbors' mixes.  I picked up some corn looking plants in our driveway but thought they were millet.  

In the usa, the corn looking plants that often grow from birdseed mixes is typically milo/sorghum. It looks a lot like corn, except that the seeds are carried high on the top of the plant (where a corn tassel would be).

Teosinte is so wild, that it's not suitable for growing as a grain