Microscope Template spreadsheet

[Jonathan Pineault]

Hello everyone,

I got some questions regarding the Microscope Template Elaine has sent to us.

The example I will be talking about is in attachment. It's a assessment I did this morning for a client.

StDev
My first question is about Standard Deviation. Has I remember, Elaine was talking of a StDev of maximum 20% of the Mean. If we exceed 20%, it means that the data is not accurate to have strong conclusions. In the example attached, I've got a Mean of 0.11 fungal strand/field, and a StDev of 0.08. Does it means that there is not fungal biomass? Is that I did not prepare my sample the good way? I think that there is some fungal biomass and I see some. If the StDev is still less than the Mean, can we still give some conclusions?

Nematode
In the spread sheet, the Mean of the nematode line is based on B29:F29. This makes 5 fields of view. Is there any reason for that? Why not 20 fields like others? Has I record, we should scan the whole slide to assess nematodes. Is there a reason it is divided by 5 (to give the Mean) before multiplying by the dilution factor?

Those examples may not be so clear, but I had the same problems with a compost that I believe to be fungal dominated. I can see some really good fungal strand when I take it with my hands. We can see some sprouting mushrooms head out of the compost pile. It smells really good. But when I prepare my sample, it seems that I cannot get good data. The StDev is often exceeding the Mean, so I should conclude that the is no significant fungal biomass.

Could it be the chlorine of my water? Should I sieve well the compost before taking a sample? I realize that a lot of practice I needed before having good microbiological assessments.

I really want to push the Soil Foodweb approach with my company and here, in Quebec, almost everybody is skeptical about that approach. I understand that microscope observation is the corner stone of that approach so I'm seeking some help to improve my observations.


Jonathan Pineault
Québec, Canada

--

[jamesdpinner]

Stick to the standard deviation rule when counting bacteria.

The standard deviation usually includes zero on fungi counts. You would have to have so much fungal biomass that statistically every field you look at will have fungi and that is rarely the case. You can still use the mean for calculating fungal biomass just know that the calculation is probably not exact. Try counting 20 fields of fungi 3 times. Then find the mean and standard deviation of those three points. 

Elaine got us started. Shes got us all looking in our microscopes. Now its up to us to start experimenting. I encourage everyone to question the method and experiment to see what works best. Repeat tests till you are confident. Try new methods to see how they compare.

As for the nematodes when you scan the entire slide you know how many are in one drop. So multiply that number by your dilution then multiply by 20 for the number of drops in a ml. So you are right about the formula. Elaine may have put that there in case you want to look at 5 drops for nematodes. It would give you a more accurate nematode count. The nematode count will work if you only put one number in. It divides by the number of numbers entered. So if you put one number in you get your number divided by 1 which changes nothing. 

Hope that helps :)

[Jonathan Pineault]

Thank you again James,

you really are a White Knight of Compost!

I will make some testing with what you propose.

Additionally I may make a new compost spreadsheet based on multiple drops observation and the Mean of Means, with StDev percentage...

I have 22x22mm cover slips so I will count how many fields I got with my microscope.


More to come!


Jonathan Pineault
permaculture designer
Québec, Canada

Le 09/02/2013 20:34, James Pinner a écrit :

If I look at 5 fields of bacteria and have a mean and standard deviation of 30 and 3 ( 10% of mean) then estimate the rest of the fields would have some where between 27 and 33 bacteria in each field. With fungi you may have fungal strands in 5 of the 20 fields. You would have a standard deviation of more than 100% the mean, but if you found 5 is that significant? Yes...because you found fungi you know its there. Would you bet that the next field you look at will have 1? No... because your odds based on what you have already observed are 1 in 4 fields contain fungi.

Now let's say i look at 20 fields for fungi and repeated that 5 times for the same sample. Based on the mean I estimate fungal biomass using the spreadsheet and receive the following results.

600 um/g 

700 um/g 

550 um/g

750 um/g

625 um/g

Mean = 645

Standard Dev = 79.8

Now we are within the 20% of the mean. 

We could feel pretty confident if you had done a sixth fungal count it would between 565 to 720 um/g

When counting nematodes you may want to look at 5 different drops to get a more accurate count. Just because if you found 1 the first time doesn't mean you will find it the second time. Same thing if you didn't find any you couldn't say there were none at all. 

When counting bacteria I use somewhere between 100 and 5000 dilutions. You want to count about 50 bacteria per field or quarter field or half field. Usually that makes things more accurate. So if you count 50 in a quarter field then multiply by 4 to get to 200. Only count the things that are moving and look like bacteria. You may count a soil particle or some oil from your fingers from time to time don't beat yourself up over it. 

For your fungi question look at my post on the google group for counting fungi more accurately. It is a detailed answer to your question. 

Cover slips!! the formulas on the spreadsheet have 1600 in them. This relates the field to the entire slide. If you have an 18x18 your good. If you don't. Use your 40x objective and measure the edges by counting fields. Then multiply the number of field on the length and width of the coverslip. Then replace the 1600 in the formulas on the spreadsheet. 

On Fri, Feb 8, 2013 at 4:50 PM, Jonathan Pineault <jona...@ecomestible.com> wrote:

Thank you so much James!

Your words are most illuminating :

"Elaine got us started. Shes got us all looking in our microscopes. Now its up to us to start experimenting. I encourage everyone to question the method and experiment to see what works best. Repeat tests till you are confident. Try new methods to see how they compare."

I was beginning to figure that, indeed!

So here is a summary of what I understand:

- To have precise results, the standard deviation should never be more than 20% of the Mean. It means that if the standard deviation is above, the calculation based on the Mean is not accurate. If the StDev equal 100% or higher of the Mean, it means that no significant biomass is present, or that more fields should be observed.

- You should not enter any zero on nematodes "field of observation", because then it will divide by the number of fields to find the Mean.


Next, some new questions:

- First the multiplying factor concerning the size of cover slip. Here are my basic calculation of the surface cover by one field of view, based on the multiplying factor given on the spreadsheet :
Coverlips = surface total @ number of fields --> calculated surface/field
22mmx22mm = 484mm2 @3600fields --> 0.1344mm2/field
22mmx18mm = 396mm2 @2400fields --> 0.165mm2/field
18mmx18mm = 324mm2 @1600fields --> 0.2025mm2/field

I don't understand why the surface of the field would be different. Has I understand, if the water drop is spread out on less surface, you should see more organisms in one field of view. Then again, if you use always the same cover slips, the difference would remain the same on all assessments.

- Second, the average fungi diameter. Should we estimate it, or should we calculate it based on each fungi length and diameter on each field of view?

- Lastly, some observation questions. To observe bacteria, do you really find it easier at 1:1000 dilution? I got hard time counting them all, and some particles seem not to be bacteria (with deformed shapes for example). Is that easier to take photos of each field and few of 'em for focus purpose? Any hint on that one?


Hoping it's not too much questions. An advanced microscope observation workshop would be awesome!

Seeking some advice,


Jonathan
permaculture designer

--
You received this message because you are subscribed to the Google Groups "Life in the Soil" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lifeinthesoi...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Ingham, Elaine]

Look in the body of the message below for my response, please

Practice makes perfect and I really enjoy the interest!

Elaine R. Ingham
Chief Scientist
Rodale Institute

-----Original Message-----
From: Jonathan Pineault [mailto:jona...@ecomestible.com]
Sent: Thu 2/7/2013 12:54 PM
To: Life in the soil Group; Ingham, Elaine
Subject: Microscope Template spreadsheet

Hello everyone,

I got some questions regarding the Microscope Template Elaine has sent
to us.

The example I will be talking about is in attachment. It's a assessment
I did this morning for a client.

*StDev*

My first question is about Standard Deviation. Has I remember, Elaine
was talking of a StDev of maximum 20% of the Mean. If we exceed 20%, it
means that the data is not accurate to have strong conclusions. In the
example attached, I've got a Mean of 0.11 fungal strand/field, and a
StDev of 0.08. Does it means that there is not fungal biomass? Is that I
did not prepare my sample the good way? I think that there is some
fungal biomass and I see some. If the StDev is still less than the Mean,
can we still give some conclusions?

No matter what the SD, you still have results. It is just if the SD is GREATER THAN the mean, it means it is likely that the actual fungal biomass is most likely not significantly different from zero. Fungal biomass is low.

If SD is less than the mean, then good, the reliability of your value is good, and fungal biomass is significantly greater than zero. Yay!!!

If you are worried about sample prep, do a second drop of your sample on the slide. Do the values come out similarly?




*Nematode*

In the spread sheet, the Mean of the nematode line is based on B29:F29.
This makes 5 fields of view. Is there any reason for that? Why not 20
fields like others? Has I record, we should scan the whole slide to
assess nematodes. Is there a reason it is divided by 5 (to give the
Mean) before multiplying by the dilution factor?

The equation should only divide by the number of fields with actual data in them.

Thus the spreadsheet allows for samples with high numbers of nematodes PER FIELD. Or, if you have low nematode numbers, you get onereading from your sample, and thus should only have your one value divided by 1.

Those examples may not be so clear, but I had the same problems with a
compost that I believe to be fungal dominated. I can see some really
good fungal strand when I take it with my hands. We can see some
sprouting mushrooms head out of the compost pile. It smells really good.
But when I prepare my sample, it seems that I cannot get good data. The
StDev is often exceeding the Mean, so I should conclude that the is no
significant fungal biomass.

Can you shake more gently? Also realize that when you have mushrooms,the fungi associated with that body is way down in the pile or soil some where, not typically up where you see the mushroom.

Could it be the chlorine of my water? Should I sieve well the compost
before taking a sample? I realize that a lot of practice I needed before
having good microbiological assessments.

Try to pick out the strands you think are fungi and lay those on the slide. You may discover they are actinobacteria......

I really want to push the Soil Foodweb approach with my company and
here, in Quebec, almost everybody is skeptical about that approach. I
understand that microscope observation is the corner stone of that
approach so I'm seeking some help to improve my observations.


Jonathan Pineault
Québec, Canada

--

<http://ecomestible.com>

[Ingham, Elaine]

Larger coverslip, the liquid is spread out more, so you have more fields to look at.

I sometimes do a quick average, if lots of variability, I actually do an average diameter in each field, and put in total length in the field, then let the computer run that through for total length, and mean diameter.

If you dilute to high, high levels, you do seem to get stuff in the field. Can't count lumpy, bumpy stuff as bacteria.....round cocci, rods with smooth sides, all that...... not clay or mineral things with rough, lumpy sides.

Elaine R. Ingham
Chief Scientist
Rodale Institute





-----Original Message-----
From: Jonathan Pineault [mailto:jona...@ecomestible.com]

Sent: Fri 2/8/2013 4:50 PM
To: lifein...@googlegroups.com
Cc: jamesdpinner; Ingham, Elaine
Subject: Re: Microscope Template spreadsheet

Thank you so much James!

Your words are most illuminating :

"Elaine got us started. Shes got us all looking in our microscopes. Now
its up to us to start experimenting. I encourage everyone to question
the method and experiment to see what works best. Repeat tests till you
are confident. Try new methods to see how they compare."

I was beginning to figure that, indeed!

*So here is a summary of what I understand:*

- To have precise results, the standard deviation should never be more
than 20% of the Mean. It means that if the standard deviation is above,
the calculation based on the Mean is not accurate. If the StDev equal
100% or higher of the Mean, it means that no significant biomass is
present, or that more fields should be observed.

- You should not enter any zero on nematodes "field of observation",

because then it will divide by the number of fields to find the Mean.


*Next, some new questions:*

<http://ecomestible.com>

Le 08/02/2013 09:57, jamesdpinner a écrit :

> *StDev*

> My first question is about Standard Deviation. Has I remember,
> Elaine was talking of a StDev of maximum 20% of the Mean. If we
> exceed 20%, it means that the data is not accurate to have strong
> conclusions. In the example attached, I've got a Mean of 0.11
> fungal strand/field, and a StDev of 0.08. Does it means that there
> is not fungal biomass? Is that I did not prepare my sample the
> good way? I think that there is some fungal biomass and I see
> some. If the StDev is still less than the Mean, can we still give
> some conclusions?
>

> *Nematode*

> In the spread sheet, the Mean of the nematode line is based on
> B29:F29. This makes 5 fields of view. Is there any reason for
> that? Why not 20 fields like others? Has I record, we should scan
> the whole slide to assess nematodes. Is there a reason it is
> divided by 5 (to give the Mean) before multiplying by the dilution
> factor?
>
> Those examples may not be so clear, but I had the same problems
> with a compost that I believe to be fungal dominated. I can see
> some really good fungal strand when I take it with my hands. We
> can see some sprouting mushrooms head out of the compost pile. It
> smells really good. But when I prepare my sample, it seems that I
> cannot get good data. The StDev is often exceeding the Mean, so I
> should conclude that the is no significant fungal biomass.
>
> Could it be the chlorine of my water? Should I sieve well the
> compost before taking a sample? I realize that a lot of practice I
> needed before having good microbiological assessments.
>
> I really want to push the Soil Foodweb approach with my company
> and here, in Quebec, almost everybody is skeptical about that
> approach. I understand that microscope observation is the corner
> stone of that approach so I'm seeking some help to improve my
> observations.
>
>
> Jonathan Pineault
> Québec, Canada
>
> --
>

> <http://ecomestible.com>

[Jonathan Pineault]

Thank you so much James!

Your words are most illuminating :

"Elaine got us started. Shes got us all looking in our microscopes. Now its up to us to start experimenting. I encourage everyone to question the method and experiment to see what works best. Repeat tests till you are confident. Try new methods to see how they compare."

I was beginning to figure that, indeed!

So here is a summary of what I understand:

- To have precise results, the standard deviation should never be more than 20% of the Mean. It means that if the standard deviation is above, the calculation based on the Mean is not accurate. If the StDev equal 100% or higher of the Mean, it means that no significant biomass is present, or that more fields should be observed.

- You should not enter any zero on nematodes "field of observation", because then it will divide by the number of fields to find the Mean.

Next, some new questions:

- First the multiplying factor concerning the size of cover slip. Here are my basic calculation of the surface cover by one field of view, based on the multiplying factor given on the spreadsheet :
Coverlips = surface total @ number of fields --> calculated surface/field
22mmx22mm = 484mm2 @3600fields --> 0.1344mm2/field
22mmx18mm = 396mm2 @2400fields --> 0.165mm2/field
18mmx18mm = 324mm2 @1600fields --> 0.2025mm2/field

I don't understand why the surface of the field would be different. Has I understand, if the water drop is spread out on less surface, you should see more organisms in one field of view. Then again, if you use always the same cover slips, the difference would remain the same on all assessments.

- Second, the average fungi diameter. Should we estimate it, or should we calculate it based on each fungi length and diameter on each field of view?

- Lastly, some observation questions. To observe bacteria, do you really find it easier at 1:1000 dilution? I got hard time counting them all, and some particles seem not to be bacteria (with deformed shapes for example). Is that easier to take photos of each field and few of 'em for focus purpose? Any hint on that one?


Hoping it's not too much questions. An advanced microscope observation workshop would be awesome!

Seeking some advice,


Jonathan
permaculture designer

Le 08/02/2013 09:57, jamesdpinner a écrit :

--