Pollination study

TZ-OH6 · August 2, 2008

[I couldn't get the new pics on the old bee post so that one can be deleted if moderators think appropriate.]

OK, I did a little bit more work on this, took more pics, and took some data on blossom activity. The results are interesting... but first a note on the photography.
Good closeup pictures are not due to someone having a trained "photographers eye" when the picture is taken. the photographers eye comes in when sorting through all of the (mostly crappy) shots later.

I have a small background in closeup nature photography, with a few published pics in textbooks and magazines. I jumped in head-first as a broke graduate student having almost no interest in photography beforehand. The most important thing I learned is that for every good picture you see, the photographer took up to one full roll of film of that one subject (one bug on a leaf etc). Half of these were unusable (focus and/or lighting), and most of the other half were uninteresting (poor placement of the subject etc). Digital made things worse in a way because it is easy to get sloppy and take a lot of pics without thinking about altering each shot, but digital is wonderful in that you can crop and enlarge to make something really good out of an OK shot.

I have a bunch of high end film camera equipment for macro work, but the following images were taken with a little silver matchbox-looking pocket point and shoot CanonPowerShot A540 [6 megapixel, 4x zoom]. I bought this camera for my mother, but did check out the macro capabilities in the store first. If I can't take a clear closeup picture of a dime on the store countertop I pass.

Once I got the settings right my success rate [focused pictures] on these bee pictures went from 10% to 75%-80%. Pictures were taken hand held from about two feet away with the following settings.

Automatic
Red-eye OFF
Autofocus OFF
Digital Zoom On
Picture quality HIGHEST

You can barely see the bees in the original pics, but with a 6 megapixel image you can crop down to a good closeup using whatever image program comes with Windows etc.

About the study:

Tomato flowers do not produce nectar. They have a flowers structure with semifused anthers (male) forming a cone around the pistil (female). Pollen exits through pores on the inside surface of the anthers (in the cone). Pollen matures first at the lower end of the anthers. The pollen then falls down past the stigma and out at the tip of the anther cone. The stamen is the receptive female end of the pistil. It is coated with sticky stigmatic fluid = no pollen sticking during hot dry weather.

Flower strategy is male dominated. Fertilize as many other plants' seeds as you can while minimizing (or neglecting) your own chance of getting pollinated. It only takes one pollinator contact (see pic 4) to fertilize more than all the ova in a flower. Over the long term, evolutionary processes favor the most efficient flower type for outcrossing, which leads to my previous discussion with Carolyn about deformed megablooms being more or less prone to cross pollination. From an evolutionary standpoint they should be less prone to cross pollination because of their shape, or they would be the norm not the exception (you will note in the pictures how well the normal flower shape interacts with the bee‘s buzz posture for transfering pollen).

Certain bee species are known as "buzz pollinators" they contract/vibrate their flight muscles to loosen and release the pollen. They then "mass provide" for their young; they stock pile a big mass of pollen and then lay an egg and seal up the brood chamber. So these bees are adapted to get as much pollen as possible from a flower using as little energy expenditure as possible (reproductive ecology is a study in energy economics, evolution is a study in mathematical probabilities). Someone mentioned on a forum that halicted bees (sweat bees) do not buzz pollinate like bumble bees do, but rather chew into the anthers. My observations show that this is not true. They are bite-buzzing to transfer the vibrational energy to specific points on each anther, probably corresponding to the maturation of the pollen, because the bite position moves up the anther over the course of the day.

Once the pictures showed me what the bees were doing I set out to see what effect the bees were having on the flowers in order to get some idea of pollination efficiency and cross pollination probability. Two flowers (one Anna Russian, one Black Plum) were observed for three, one hour intervals (approx 10-11AM, 1-2 PM, and 4-5 PM). Anna Russian data is given, The black plum blossom became more receptive about an hour later so had slightly less bee encounters during the morning observation period, but more at the end.

The following were recorded: number of landings, landings with buzz, and number of inflight hesitations (fly-by looks) near the blossom. This was a pilot study and I was trying to do a lot of things the first hour so many of the landings were probably buzz landings but not recorded as such, so I grouped all landings the first hour as buzz landings here. Bees seemed to be able to tell at a glance without landing if pollen was available. Possibly by color spectral changes. More data on the timing of fly by looks will help with this hypothesis.

What is the population density of bees? At any one point in the morning there is usually one bee for every plant or two, but you have to look close. There are three species visiting my plants, a very common emerald green, a common bronze green about the same size, and a little black one. Blossom anther cone discoloration is the easiest way to estimate bee activity. Bagged blossoms do not looked chewed up.

You might say: Tomatoes are not native to Ohio, so why all the bees, and why should your theories hold water? We have grown a dozen or more tomato plants along side the house here for twenty years or so. Lots of tomato pollen = lots of food for baby bees every year. There are some species of nightshade growing around here with tomato-like flowers, so the native bees are adapted to the flower type.

tzbee buzz0288.jpg

Picture 1. Bee buzz pollinating blossom. Notice the wing tip movement, which lasts for a half or third of a second. (Before I watched the bees closely I deleted a few pics of buzzing bees thinking they were out of focus.) The bee will land, walk around a it. Bite down on an anther, buzz, release, move to the next anther and repeat. The first bees of the day may proceed around the anther cone 4-6 times. Later on bees only make one revolution or partial revolution indicating pollen levels are very low.

tzbee 0290.jpg

Picture 2. Same bee after buzz moving pollen to pollen baskets on legs. Wings are still.

tzbee on pp 0296.jpg

Picture 3. Bronze green species on Pruden’s Purple. Note the enlarged exerted stigma seen on many of my beefsteak heirlooms.

tzbee0304.jpg

Picture 4. Bee coated with white pollen.

tzbee0305.jpg

Picture 5. Same bee in bite-buzz posture. Notice the contact with the stigma due to the characteristic posture needed to catch released pollen on the body. I think that the unusual/uneven shape of megablooms would inhibit this contact, not enhance it.

tzcomp textweb.jpg

Picture 6. Data summary for pollination. Note how quickly the stigma is fully exposed due to the bite damage to the anther cone. Pollen tube formation is slow enough that a couple of hours delay in cross pollination (self pollen on non-exerted stigma vs. cross pollen on exerted stigma) shouldn’t make a difference. A Potato Leaf Pruden’s Purple is growing among these RL plants, and will give me some idea of cross pollination percentage based on the bee activity.

From this data I estimate that each blossom gets around 50 bee-pollination contacts per day with mixed pollen from any/all blooming plants in the area (the garden). If cross pollination is low in bee areas it is most probably due to pollen release before the flower opens fully, with self pollen out racing cross pollen. This mechanism would suggest that nearly all fruits have some percentage of crossed seed. When the Pruden’s Purple fruits mature I’ll sow the seeds have some quick data on this hypothesis.

tznightshade0425.jpg

Picture 7. Native nightshade

tzbeeflying_0292.jpg

Picture 8. Cool pic, lucky shot.

salix · August 2, 2008

Very cool pics, and probably not a "lucky shot" - you had to be there and were well prepared! Thank you for taking the time to do the study and post the info for the rest of us. Thanks also for the technical info re: cameras - I really need to get more familiar with mine...

**Lee** · August 4, 2008

Great pix and report Todd! Thanks for sharing this information.
I would never have thought the flowers get that many visits..

Lee

dokutaaguriin · August 5, 2008

I was finally able to capture a bee buzzing my tomato flowers in a format small enough to be uploaded here. Previous videos I took were much larger in size as I chased the bee around the greenhouse going from plant to plant in a sort of random manner. The bee seemed not to go to the larger blossoms but rather the nice perfectly formed flowers.

Jeff

TZ-OH6 · August 6, 2008

Nice little video,

I finally saw a couple of big bumbles working the flowers here, and they did the same thing the little bees did except rather than go around the cone biting each anther in turn they only bit the cone in one spot, buzzed once and took off for better pickins.

TZ-OH6 · October 14, 2008

Here is cross pollination data for first fruits, which are supposed to be "safe" from bee activity. All of these Potato-Leaf plants were adjacent to/surrounded by Regular-Leaf plants for the most part.

40 seeds were planted per fruit (40 because that was the max number from one of the fruits).

I will soon be growing out seeds from mid season fruits pollinated during the period of the above bee-visit counts. Plants were pinched out shortly after the counts were done so I know approx. when the flowers were pollinated.

Plant code is: Variety-plant#-truss-fruit [e.g. Prudens Purple plant 1, 1st truss- second fruit] plant sets 1 and 2 were in different gardens at least 50 ft apart and with different light and temperature conditions.

% Crossed Seed
RL-PL/# Seeds
Early test - 40 seeds- No bag

0%

0-30/40

Brandywine Red PL 2-1-1

0%

0-40/40

Green Giant 1-1-1catface

12.5%

5-35/40

Green Giant 1-1-2

0%

0-40/40

Japanese Black Trifele 1-1-1

0%

0-35/40

Japanese Black Trifele 1-1-2

0%

0-38/40

Lucky Cross 1-1-1

0%

0-39-40

Prudens Purple 1-1-1 all seeds

8.6%

3-32/40

Prudens Purple 1-1-2

5.4%

2-35/40

Prudens Purple 2-1-1

0%

0-39/40

Prudens Purple 3-1-x

Seedlings were planted out 10 days after recommended last frost date for Central Ohio. This was long after apple and cherry trees finished blossoming, and wild flowers/weeds had been blooming for a while, so pollinators must have been active. The argument that there are fewer tomato flowers open early in the season so cross pollination should be extremely low is not very compelling in my opinion because the active pollinators are going to be repeatedly visiting all the few flowers that are open, and the chances are high that much of the loose pollen on their body is from another plant rather than from another flower on the same plant. Poor weather that lessens bee activity is a better arguement.

I am surrounded by a good bit of natural/wild vegetation, presumably with good areas for ground nesting bees to make nests, but if you are in town surrounded by manicured lawns, or out in the country surrounded by miles of corn fields pollinator numbers may be very low early in the season.

August 2, 2008	#1
TZ-OH6 Tomatovillian™ Join Date: Feb 2008 Location: Mid-Ohio Posts: 847	Pollination study [I couldn't get the new pics on the old bee post so that one can be deleted if moderators think appropriate.] OK, I did a little bit more work on this, took more pics, and took some data on blossom activity. The results are interesting... but first a note on the photography. Good closeup pictures are not due to someone having a trained "photographers eye" when the picture is taken. the photographers eye comes in when sorting through all of the (mostly crappy) shots later. I have a small background in closeup nature photography, with a few published pics in textbooks and magazines. I jumped in head-first as a broke graduate student having almost no interest in photography beforehand. The most important thing I learned is that for every good picture you see, the photographer took up to one full roll of film of that one subject (one bug on a leaf etc). Half of these were unusable (focus and/or lighting), and most of the other half were uninteresting (poor placement of the subject etc). Digital made things worse in a way because it is easy to get sloppy and take a lot of pics without thinking about altering each shot, but digital is wonderful in that you can crop and enlarge to make something really good out of an OK shot. I have a bunch of high end film camera equipment for macro work, but the following images were taken with a little silver matchbox-looking pocket point and shoot CanonPowerShot A540 [6 megapixel, 4x zoom]. I bought this camera for my mother, but did check out the macro capabilities in the store first. If I can't take a clear closeup picture of a dime on the store countertop I pass. Once I got the settings right my success rate [focused pictures] on these bee pictures went from 10% to 75%-80%. Pictures were taken hand held from about two feet away with the following settings. Automatic Red-eye OFF Autofocus OFF Digital Zoom On Picture quality HIGHEST You can barely see the bees in the original pics, but with a 6 megapixel image you can crop down to a good closeup using whatever image program comes with Windows etc. About the study: Tomato flowers do not produce nectar. They have a flowers structure with semifused anthers (male) forming a cone around the pistil (female). Pollen exits through pores on the inside surface of the anthers (in the cone). Pollen matures first at the lower end of the anthers. The pollen then falls down past the stigma and out at the tip of the anther cone. The stamen is the receptive female end of the pistil. It is coated with sticky stigmatic fluid = no pollen sticking during hot dry weather. Flower strategy is male dominated. Fertilize as many other plants' seeds as you can while minimizing (or neglecting) your own chance of getting pollinated. It only takes one pollinator contact (see pic 4) to fertilize more than all the ova in a flower. Over the long term, evolutionary processes favor the most efficient flower type for outcrossing, which leads to my previous discussion with Carolyn about deformed megablooms being more or less prone to cross pollination. From an evolutionary standpoint they should be less prone to cross pollination because of their shape, or they would be the norm not the exception (you will note in the pictures how well the normal flower shape interacts with the bee‘s buzz posture for transfering pollen). Certain bee species are known as "buzz pollinators" they contract/vibrate their flight muscles to loosen and release the pollen. They then "mass provide" for their young; they stock pile a big mass of pollen and then lay an egg and seal up the brood chamber. So these bees are adapted to get as much pollen as possible from a flower using as little energy expenditure as possible (reproductive ecology is a study in energy economics, evolution is a study in mathematical probabilities). Someone mentioned on a forum that halicted bees (sweat bees) do not buzz pollinate like bumble bees do, but rather chew into the anthers. My observations show that this is not true. They are bite-buzzing to transfer the vibrational energy to specific points on each anther, probably corresponding to the maturation of the pollen, because the bite position moves up the anther over the course of the day. Once the pictures showed me what the bees were doing I set out to see what effect the bees were having on the flowers in order to get some idea of pollination efficiency and cross pollination probability. Two flowers (one Anna Russian, one Black Plum) were observed for three, one hour intervals (approx 10-11AM, 1-2 PM, and 4-5 PM). Anna Russian data is given, The black plum blossom became more receptive about an hour later so had slightly less bee encounters during the morning observation period, but more at the end. The following were recorded: number of landings, landings with buzz, and number of inflight hesitations (fly-by looks) near the blossom. This was a pilot study and I was trying to do a lot of things the first hour so many of the landings were probably buzz landings but not recorded as such, so I grouped all landings the first hour as buzz landings here. Bees seemed to be able to tell at a glance without landing if pollen was available. Possibly by color spectral changes. More data on the timing of fly by looks will help with this hypothesis. What is the population density of bees? At any one point in the morning there is usually one bee for every plant or two, but you have to look close. There are three species visiting my plants, a very common emerald green, a common bronze green about the same size, and a little black one. Blossom anther cone discoloration is the easiest way to estimate bee activity. Bagged blossoms do not looked chewed up. You might say: Tomatoes are not native to Ohio, so why all the bees, and why should your theories hold water? We have grown a dozen or more tomato plants along side the house here for twenty years or so. Lots of tomato pollen = lots of food for baby bees every year. There are some species of nightshade growing around here with tomato-like flowers, so the native bees are adapted to the flower type. tzbee buzz0288.jpg Picture 1. Bee buzz pollinating blossom. Notice the wing tip movement, which lasts for a half or third of a second. (Before I watched the bees closely I deleted a few pics of buzzing bees thinking they were out of focus.) The bee will land, walk around a it. Bite down on an anther, buzz, release, move to the next anther and repeat. The first bees of the day may proceed around the anther cone 4-6 times. Later on bees only make one revolution or partial revolution indicating pollen levels are very low. tzbee 0290.jpg Picture 2. Same bee after buzz moving pollen to pollen baskets on legs. Wings are still. tzbee on pp 0296.jpg Picture 3. Bronze green species on Pruden’s Purple. Note the enlarged exerted stigma seen on many of my beefsteak heirlooms. tzbee0304.jpg Picture 4. Bee coated with white pollen. tzbee0305.jpg Picture 5. Same bee in bite-buzz posture. Notice the contact with the stigma due to the characteristic posture needed to catch released pollen on the body. I think that the unusual/uneven shape of megablooms would inhibit this contact, not enhance it. tzcomp textweb.jpg Picture 6. Data summary for pollination. Note how quickly the stigma is fully exposed due to the bite damage to the anther cone. Pollen tube formation is slow enough that a couple of hours delay in cross pollination (self pollen on non-exerted stigma vs. cross pollen on exerted stigma) shouldn’t make a difference. A Potato Leaf Pruden’s Purple is growing among these RL plants, and will give me some idea of cross pollination percentage based on the bee activity. From this data I estimate that each blossom gets around 50 bee-pollination contacts per day with mixed pollen from any/all blooming plants in the area (the garden). If cross pollination is low in bee areas it is most probably due to pollen release before the flower opens fully, with self pollen out racing cross pollen. This mechanism would suggest that nearly all fruits have some percentage of crossed seed. When the Pruden’s Purple fruits mature I’ll sow the seeds have some quick data on this hypothesis. tznightshade0425.jpg Picture 7. Native nightshade tzbeeflying_0292.jpg Picture 8. Cool pic, lucky shot.

August 4, 2008	#3
Lee Tomatopalooza™ Moderator Join Date: Jan 2006 Location: NC-Zone 7 Posts: 2,188	Great pix and report Todd! Thanks for sharing this information. I would never have thought the flowers get that many visits.. Lee __________________ Intelligence is knowing a tomato is a fruit. Wisdom is knowing not to put one in a fruit salad. Cuostralee - The best thing on sliced bread.

August 2, 2008	#2
salix Tomatovillian™ Join Date: Dec 2007 Location: north central B.C. Posts: 2,310	Very cool pics, and probably not a "lucky shot" - you had to be there and were well prepared! Thank you for taking the time to do the study and post the info for the rest of us. Thanks also for the technical info re: cameras - I really need to get more familiar with mine...

August 6, 2008	#5
TZ-OH6 Tomatovillian™ Join Date: Feb 2008 Location: Mid-Ohio Posts: 847	Nice little video, I finally saw a couple of big bumbles working the flowers here, and they did the same thing the little bees did except rather than go around the cone biting each anther in turn they only bit the cone in one spot, buzzed once and took off for better pickins.

October 14, 2008	#6
TZ-OH6 Tomatovillian™ Join Date: Feb 2008 Location: Mid-Ohio Posts: 847	Here is cross pollination data for first fruits, which are supposed to be "safe" from bee activity. All of these Potato-Leaf plants were adjacent to/surrounded by Regular-Leaf plants for the most part. 40 seeds were planted per fruit (40 because that was the max number from one of the fruits). I will soon be growing out seeds from mid season fruits pollinated during the period of the above bee-visit counts. Plants were pinched out shortly after the counts were done so I know approx. when the flowers were pollinated. Plant code is: Variety-plant#-truss-fruit [e.g. Prudens Purple plant 1, 1st truss- second fruit] plant sets 1 and 2 were in different gardens at least 50 ft apart and with different light and temperature conditions. % Crossed Seed RL-PL/# Seeds Early test - 40 seeds- No bag 0% 0-30/40 Brandywine Red PL 2-1-1 0% 0-40/40 Green Giant 1-1-1catface 12.5% 5-35/40 Green Giant 1-1-2 0% 0-40/40 Japanese Black Trifele 1-1-1 0% 0-35/40 Japanese Black Trifele 1-1-2 0% 0-38/40 Lucky Cross 1-1-1 0% 0-39-40 Prudens Purple 1-1-1 all seeds 8.6% 3-32/40 Prudens Purple 1-1-2 5.4% 2-35/40 Prudens Purple 2-1-1 0% 0-39/40 Prudens Purple 3-1-x Seedlings were planted out 10 days after recommended last frost date for Central Ohio. This was long after apple and cherry trees finished blossoming, and wild flowers/weeds had been blooming for a while, so pollinators must have been active. The argument that there are fewer tomato flowers open early in the season so cross pollination should be extremely low is not very compelling in my opinion because the active pollinators are going to be repeatedly visiting all the few flowers that are open, and the chances are high that much of the loose pollen on their body is from another plant rather than from another flower on the same plant. Poor weather that lessens bee activity is a better arguement. I am surrounded by a good bit of natural/wild vegetation, presumably with good areas for ground nesting bees to make nests, but if you are in town surrounded by manicured lawns, or out in the country surrounded by miles of corn fields pollinator numbers may be very low early in the season.