Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 39

The inner epidermis of the

onion bulbs cataphylls


(the onion skin)
Easy and not so easy methods to work with

Walter Dioni - Cancn, Mxico

First part preparing the epidermis, live cell structure, fixing and
staining with iodine

JUSTIFICATION
This work started as an attempt to make a preparation of onion skin
without the annoying air bubbles that almost always bother the
observation. You can work with them under your objective, but this
is annoying, isnt it? I was then guided well beyond my initial
purpose, by a concatenation of suggestions dictated by studying the
material in itself, my memory... and Internet. Although, at a certain
point, work seemed to show a very serious unbalance in the "cost in
time / benefit" quotient, but the intrinsic fun generated by this
"light research, that even drove me on joyful journeys through the
Web for many days, tempted me to continue. I believe that at the
end I had a lot of fun, I learnt several things, and in between there
were registered several points which may be of interest to other
amateur microscopists, biology students, or even biology teachers.
WARNING! Those who start reading this long series of articles must be advised that they
will be subject to the viewing of very similar images, with no artistic or novelty pretentions.
These are just witnesses of the use, on the same plant tissue, of many different techniques.
So, what is important, is the comparison between otherwise very similar images, to verify
(as long as a digital testimony can allow) the result of the different treatments.
Most pictures have been taken with a Canon Powershot A75 camera, handheld over the
10x ocular of my microscope. The 4x objective of the microscope plus the camera system
gives, depending on the intensity of illumination, a central hot spot (which is not evident
visually) even with the most carefully adjusted Khler style illumination. Due to vignetting*,
images have 2Mpx out of the native 3.2 Mpx of the sensor. Most of the time I crop a 1024 x
768 px image, which for sake of better reading in the published article I reduce to 900x 600
px. If necessary for publication, I adjust the size additionally. My only interest is to better
show which I consider a significant testimony of the action of the reagents.
*Vignetting can be eliminated using the optical zoom of the camera. But I work until now
with a power supply, and in this configuration the camera does not admit its use.

INTRODUCTION
As part of 'another project' Im working on, I needed a few onionskin preparations. Who does not need them some times?
But of what onion-skin (epidermis) are we talking about? For me it
was clear that the onion bulb offers us, for

our observations at least, 3 or 4 different types of epidermis. To


define and select the one to work with, was the subject of the
article How many onion skins are there? (MICSCAPE, February
2011).
And ... seeing the results, I asked myself who, when, and why,
the epidermis, specifically the innermost of one of the scales in
the onion bulb, a modest food ingredient, was proposed as a
successful model of plant tissue, and unbeatable elementary
demonstration of the organization of a cell.
Of course these are not big existential problems. This violent world
in which we live will not be better because I found an answer; but I
surrendered to my curiosity, and this was the starting point of the
article Who invented the 'onion skin' biology
lab? (MICSCAPE, March 2011) which retells the long history of this
practice, now inevitable in any course on elementary biology.
I recommend reading these two preliminary articles before trying
this one.
So, after that divertissement, I elected to work with the, for more
than a century, favorite inner epidermis of the middle of the
package of scales that made an onion, and try to work out a
technique satisfactory for me, and possibly useful to others, for not
to have to endure the uncomfortable and unsightly air bubbles.

THE FIRST OF MANY POSSIBLE WET MOUNTS OF AN ONION


SKIN
For the first mounts I made, I followed, with some
adjustments, the standard method, a thousand times
described on the Internet and in any elementary biology
book, exactly as recommended in 1890, or any of several
modifications tirelessly repeated by all those who, like me,
want to take advantage of the facilities offered by this
material.

Everyone who uses this technique must face the first


problem: in the majority of onions the epidermis is very thin,
and when you manipulate it, it folds in capricious forms, or
screws up as a tube. The use of needles and tweezers,
recommended many times to extend it, is difficult, tedious,
and can be ineffective, or cause a disaster in the thin
membrane. Manipulation, folds, the type of mounting
medium, and other details, make up nasty preparations.
The following technique, although laborious, helped me to
solve this problem. I don't recommend it, I describe it. And
of course it also has its own limitations.

Fig. 2 A classic preparation, 10x objective, oblique illumination, electronic colouration.


Living cells. The boxes made up by the cell walls enclose the textured cytoplasm with small
granulations and the discoid nucleus near the cell wall. Some of them in the upper layer,
some on the bottom one, due to the difference in focus.

Picking up the fresh onion epidermis

I take an onion bulb, I cut it in halves, and if it is large, into


quarters. I work immediately, because if the scales dry, even
slightly, the epidermis breaks off with thin layers of mesophyll
adhered to its surface, which then prevents the preparation being
flat enough; and which, if dyes were used, would be coloured,
making useless much of the surface of the used peeling".

Fig 3 This is a stained peeling showing the stained mesophyll under the epithelium.

I cut a piece of scale (cataphyll) (usually the third or fourth from the
outside of the bulb, to have a thickness which facilitates
manipulation) of 1 to 1.5 cm wide (to fit comfortably on the slide)

all the length of the scale, and do a cross cut from edge to edge, on
the convex side of the cataphyll, 2 cm from one end.

Foto 4 a,b,c,d cutting the wedge, selecting a middle cataphyll, cutting dorsal epidermis

I break the piece, and with a slight effort I break loose a sheet of
epidermis, which has the width of the piece, which I give a length of
about 2 or 3 cm at least. Folding over the piece, with care, and
cutting off the excess, I reject the intermediate piece of scale (now
without the inner epidermis) and cut the other end of the cataphyll
piece to have a length of other 1.5 or 2 cm. Its necessary to work
with care but it is not difficult. Only needed is a more or less sharp
scalpel, or similar cutting tool.

Foto 5a,b,c,d breaking the slice, peeling the epidermis, discarding the peeled portion,
cutting the second bar.

The piece of skin is now supported at both ends by two scale "bars".
With fine pointed tweezers I can take one of the handles and
manipulate the piece, which will be extended by the weight of the
other "bar". At first I cut the cataphyll handles of only 0.5 cm, and
this worked with some fluids. But I learned that their weight was
not enough with other liquids, so I decided to use the big pieces
which guarantee an easy manipulation in all cases.

Fig. 6a finished piece, ready to be fixed or wet mounted, Fig. 6b two prepared slides,
each with a typical piece of epidermis, fixed and stained.

The epidermis, subtended between the two pieces of scale, can be


moved and safely transferred between several reagents, if
necessary, holding one of the bars with tweezers.
When I finish working with a piece, I mount it over a slide and cut
out the bars.
Perhaps this will be not a technique to propose for use in
classroom practices, but allows me to try comfortably
different working techniques of my interest.

Making the simplest onion peel mounting


I make wettable a perfectly clean slide (pass it through the flame of
a spirit lamp, or a Bunsen burner, a couple of times on both sides)
put a drop of water (or stain) in the center, and lay on it my sample,
if possible with the face that was attached to the mesophyll

downwards. It's a little tricky, but, with care, you can do it. With
high powers this assures a more clear focus of the epidermic cells.
Add a small drop of water (or stain) so that it does not dry out.
Holding one bar with my tweezers, I use a Gillette or a sharp scalpel
to cut with care next to the bar, freeing that end of the peel. Turn
slide and repeat the maneuver.
The result is a peeling of a very good size which is also extended
almost without pleats (usually the cut edges bend a little). Carefully
apply the coverslip with common precautions and if it seems
necessary, absorb a little liquid so that the weight of the coverslip
flattens the epidermis.
This is the first of many types of wet mounts I can apply to
my peel and it provides a good square or rectangular piece of
epidermis, usually more than 1.5 cm wide x 2 cm long,
without any chemical modifiers applied, if I so wish.
Of course I got, as so often in more than 60 years of attempts,
epidermis peels mounted in water or diluted Glycerin, strewn with
air bubbles (some more, some less, but all with bubbles). You must
seek and select fields which have the minimum amount of bubbles.
It can be almost impossible with the 4x, and, depending on luck,
also with the 10x. It is only easy using high magnifications (400x
for example).

Fig. 7 - 4x Obj. fresh peeling mounted in water. No staining. This and the following images
had been subjected to a contrast enhancement, to make more visible the structural details.
As most of all living tissues the fresh onion skin is very transparent, as you know.

Fig. 8 - 10x obj. Phantom nuclei. No staining. Colours were a not programmed, unexpected
bonus, from the illumination and the recording camera set up.

Fig. 9 - 40x obj. A nucleus attached to the lateral wall of the cell.

Fig. 10 - 40x obj. A frontal view of two nuclei, surely embedded in the upper layer of the
cytoplasm. You can even see a nucleolus in the upper one. Strips of cytoplasm and included
mitochondria are visible.

With difficulty, you see in these fresh preparations, that the


cytoplasm is arranged in a delicate layer applied to the cell wall,
with many granulations included, which are mostly mitochondria.
This you can discern at least at 400x by slowly focusing through the

depth of the cell, going from the upper cytoplasm layer, through the
sap of the central vacuole to the lower cytoplasm layer. Included in
one of the cytosol layers you find the nucleus, visible in frontal view
as a disk, or in lateral view, when applied to the lateral wall, as a
shallow dome. An accumulation of cytosol and mitochondria is
discerned around the nuclei. And bands and strips of cytosol
link the nucleus through the central vacuole to other parts of
the cell. If you persevere in your efforts to optimise your
illumination, focus with care, and study many cells, you could
discern all these details. Of course your work would be facilitated if
you have a phase contrast system, and would be almost perfect if
you have DIC. But the loose structure of the cytoplasm and its
layers and strips can be visualized even with the 10x objective,
using a dark background stop (fig 11).

Fig. 11 Darkfield (a technique easy to install in any modest microscope). Fresh onion skin,
10x Obj.
Search the MICSCAPE library for technical suggestions on easy and cheap darkfield
installations.

Fig.12 the spongy tri-dimensional structure of cytoplasm is better seen in this resized
image of the cells.

On YouTube there is an amazing video of the mitochondria streaming


through the living cytosol. See it, because is very difficult to have this
experience oneself. http://www.youtube.com/watch?v=VXbQpRpUDmQ

The fresh preparation we have reviewed is


our reference. It shows what the live cell
structure is. Remember this.
Staining the peeling
But it is normal to recommend adding some reagent to stain cell
nuclei and make more visible the cytosol, its inclusions, and cell
membranes.
The most recommended colouration, the more ancient, the most
simple, and the cheapest, is a drop of a diluted pharmaceutical
iodine-iodide solution (black iodine, iodine tincture)
Note: One of the black iodine solutions sold in Cancun denounce on its label this formula,
but can be diluted 5 or more times its volume for use.
Metallic Iodine

2.0 g

Potassium iodide

2.5 g

Water

c.s.p.

100 ml

Beware! The iodine solution stains indelibly many fabrics ... and skin, for a long period!

In a small capsule I mix 2 drops iodine solution, and 10 drops


water, put 1 drop of this mix over the slide, apply the epidermal
strip, one more drop of the solution, and a cover. Allow 1 or 2
minutes for the colour to develop, absorb any excess solution, and
see the result.
As commercial solutions could be different, dilute your own solution
until it is a beautiful, transparent yellow colour. Darker solutions
stain nuclei dark, and obscure its texture. Make some trials.
Iodine precipitates the proteins of the protoplasm, and stains them
a light brown colour, the nucleus in a more intense colour, and
nucleoli even darker, as well as cell walls.

Take care, use a new, or a relatively fresh bottle of reagent. Active


Iodine concentration drop with time.
If possible, if yours is like mine, dont use Iodopovidone (iodinepolyvinylpyrrolidone). It is a pharmaceutical aqueous solution at 10-11%,
equivalent to 1% of iodine. The stains that it can produce are washable. It's
the favorite of homemakers. But, in my experience, it is dissolved in a dense
vehicle that causes plasmolysis in my preparations.

Fig 13 - Typical iodine mount 4x, many (and this is a carefully selected area) air bubbles.
Are they not a nuisance?

Fig. 14 - Iodine mount 10x. Well delineated cell walls, nucleus with two to four nucleoli, air
bubbles!

Fig 15 - Cytoplasm streaming freezed by iodine. A stack of 3 Canon handheld pictures


combined with CombineZP
Roll mouse over image above for labels.

Fig. 16 - Iodine mount, 40x. The folds and grooves, that indent the nucleus, are seen in this
image. They are channels of endoplasmic reticulum which some times tunnel through the
entire nucleus.
See http://www.sciencedaily.com/releases/2001/06/010614064109.htm
Cytosol nicely fixed, see the strips of mitochondria. Cell walls well evident. Net image of
nucleoli

Fig. 17 - A good iodine fixed and stained preparation . 40x Obj.

18 - Cytoplasm streaming freezed by the iodine a stack of 3 Canon handheld pictures


combined with CombineZP.

Fig 19 - Fresh mount, 100x. A typical nucleus, with one nucleolus. The image shows what is
normal: the cytosol, evidenced by his charge of mitochondria, is concentrated around the
nucleus. Which seems strange to me is the nucleolus clearly divided in four parts. The image
is a stack of four original ones, combined in CombineZ.

Fig 20 - Three nuclei from the same epidermis (Obj. x 100) some are fairly circular, but
many have folded surfaces. The third one is a lateral view of one included in the cytosol
layer adhered to the cell wall.

I think it was fortunate that the iodine solutions have been used
since the 19th century. Certainly it was used in a much diluted
form to detect starch (which iodine stains blue), but it was soon
found that both protein precipitation and their strong colouration,
which not only detects the cytosol, but also their inclusions, as
leucoplasts and mitochondria, made it ideal for staining the onion
epidermis temporarily. In the words of Bastin (1877) "Iodine
solution also rapidly kills protoplasm without dissolving it, and is
therefore useful as a fixing agent". Even if I could not find clear
statements about this, it seems iodine is useless for permanent
preparations because, exposed to light, it finally fades.

I must say, loud and clear, that, apart from the nasty
bubbles that I hate, and which really are very
unsightly, as you have seen, the preparation, exploited to
its full potential, can fulfill its task:

To show with very little effort, to elementary biology


students, with a very simple and very cheap technique,
what a plant cell is, and all their components even in
very good detail! Giving them a polite apology for the
bubbles, of course!
Most teachers use this technique, and it is a good
practice, of course.

But to eliminate the bubbles was the original goal


of this effort... wasnt it?
Not yet fulfilled, of course. More efforts, next
month.
I promise that my next articles will be shorter than this one.
http://www.microscopy-uk.org.uk/mag/artapr11/wd-onion3.html

[ Share:

odine reacts with the starch present in onion cells, producing a coloration that makes
the cells easily visible under a microscope. Onion cells are naturally transparent, so it is
difficult to properly visualize them without using a solution to increase contrast.
CONTINUE READING
KEEP LEARNING

What is an onion cell lab?

How does iodine react with starch in water?

What is iodine used in?

Credit: Ed Reschke Photolibrary Getty Images


FULL ANSWER

To obtain a clear image of onion cells, a wet mount microscope specimen must be
prepared first. The person preparing the specimen must place a drop of water on a
clean microscope slide, then place a piece of onion skin on it. To ensure contrast, a
drop of iodine solution is added. The specimen is then protected with a cover slip and
observed under the microscope. If the specimen contains a higher amount of starch,
adding iodine creates a blue coloration.

Method[edit]
1. First add a few drops of water or solution on the microscope slide to avoid dryness and
wilting
2. Take a small piece of onion and using forceps (tweezers), peel off the membrane from the
underside (the rough side).
3. Lay the membrane flat on the surface of the slide

4. Add a drop of Iodine solution to the onion epidermis.


5. Using a pin, lower the thin glass cover slip or cover glass onto the slide. Make sure there
are no air bubbles (Fig. 1).
6. Make sure the lowest power objective lens (the shortest lens if there are several present) is
in line with the optical tube, and the microscope light is turned on. Then place the prepared
slide onto the stage of the microscope.
7. Looking from the side (NOT through the eyepiece), lower the tube using the coarse focus
knob until the end of the objective lens is just above the cover glass. Do this carefully so as
not to crack the cover glass (and possibly damage the objective lens).
8. Now look through the eyepiece and turn ONLY the smaller, fine focusing knob to move the
optical tube upwards until an image comes into focus. The cells should look something like
lizard skin.
9. Swap the objective lens for a higher powered one so that you can see the cells at greater
magnification. You should be able to make out a nucleus in each cell.
10.Be very careful; these dyes can stain your skin and clothes.Could be dangerous if it is on
you.

Proper use of the microscope - intended to prevent damage to the objective lenses - requires that
the following techniques be followed:

Never use the coarse focus knob while looking through the eyepiece. The point of focus will
be very near the cover glass. Looking from the side, lower the optical tube until the objective
lens is as close as you can get it to the cover glass without actually touching it. Starting with the
low power objective lens is the fastest way to achieve proper focus.

Initially, slowly focus back (turn the fine focus knob to raise the optical tube) while looking
through the eye piece. Once the specimen comes into focus, you can make fine adjustments up
or down with the fine focus knob without fear of damaging the slide or the microscope.

If the specimen does not come into view (does not focus), raise the tube a little with the
coarse focus knob and attempt to focus again with the fine focus knob. Once the object is in
focus, switching objective lenses (to a higher power) should be possible without any further
coarse adjustments.

Tissue from an onion is a good first exercise in using the microscope and viewing plant cells. The
cells are easily visible under a microscope and the preparation of a thin section is straight forward.
An onion is made of layers, each separated by a thin skin or membrane. In this exercise you will
make a wet mount on a microscope slide and look at the cells of the onion membrane magnified by
the high power, compound microscope.

Method[edit]
1. First add a few drops of water or solution on the microscope slide to avoid dryness and
wilting
2. Take a small piece of onion and using forceps (tweezers), peel off the membrane from the
underside (the rough side).
3. Lay the membrane flat on the surface of the slide
4. Add a drop of Iodine solution to the onion epidermis.
5. Using a pin, lower the thin glass cover slip or cover glass onto the slide. Make sure there
are no air bubbles (Fig. 1).
6. Make sure the lowest power objective lens (the shortest lens if there are several present) is
in line with the optical tube, and the microscope light is turned on. Then place the prepared
slide onto the stage of the microscope.
7. Looking from the side (NOT through the eyepiece), lower the tube using the coarse focus
knob until the end of the objective lens is just above the cover glass. Do this carefully so as
not to crack the cover glass (and possibly damage the objective lens).
8. Now look through the eyepiece and turn ONLY the smaller, fine focusing knob to move the
optical tube upwards until an image comes into focus. The cells should look something like
lizard skin.
9. Swap the objective lens for a higher powered one so that you can see the cells at greater
magnification. You should be able to make out a nucleus in each cell.
10.Be very careful; these dyes can stain your skin and clothes.Could be dangerous if it is on
you.

Proper use of the microscope - intended to prevent damage to the objective lenses - requires that
the following techniques be followed:

Never use the coarse focus knob while looking through the eyepiece. The point of focus will
be very near the cover glass. Looking from the side, lower the optical tube until the objective
lens is as close as you can get it to the cover glass without actually touching it. Starting with the
low power objective lens is the fastest way to achieve proper focus.

Initially, slowly focus back (turn the fine focus knob to raise the optical tube) while looking
through the eye piece. Once the specimen comes into focus, you can make fine adjustments up
or down with the fine focus knob without fear of damaging the slide or the microscope.

If the specimen does not come into view (does not focus), raise the tube a little with the
coarse focus knob and attempt to focus again with the fine focus knob. Once the object is in
focus, switching objective lenses (to a higher power) should be possible without any further

coarse adjustments.
https://en.wikibooks.org/wiki/School_Science/How_to_prepare_an_onion_cell_slide

To Prepare Stained Temporary Mount of Onion Peel


Materials Required

Real Lab Procedure

Pour some distilled water into a watch glass.

Peel off a leaf from half a piece of onion and using the forceps, pull out a piece of
transparent onion peel (epidermis) from the leaf.

Put the epidermis in the watch glass containing distilled water.

Take a few drops of safranin solution in a dropper and transfer this into another watch glass.

Using a brush, transfer the peel into the watch glass containing the safranin solution.

Let this remain in the Safranin solution for 30 seconds, so that the peel is stained.

Take the peel from the Safranin solution using the brush and place it in the watch glass
containing the distilled water.

Take a few drops of glycerine in a dropper and pour 2-3 drops at the center of a dry glass
slide.

Using the brush, place the peel onto the slide containing glycerine.

Take a cover slip and place it gently on the peel with the aid of a needle.

Remove the extra glycerine using a piece of blotting paper.

Place this glass side on the stage of the compound microscope and view it.

Observations

There are a large number of regularly shaped cells lying side by side and each cell has a
distinct cell wall.

A distinct nucleus is present on the periphery of each cell.

Lightly stained cytoplasm is observed in each cell.

A large vacuole is present at the centre of each cell, and is surrounded by the cytoplasm.

Conclusion
As cell walls and large vacuoles are clearly observed in all the cells, the cells placed for observation
are plant cells.

Precautions

Use a brush to transfer the peel from one apparatus to another.

Staining of peel should neither be too dark, nor too light.

Extra glycerine stain should be removed using blotting paper.

To Prepare Stained Temporary Mount of Human Cheek Cells


Materials Required

Real Lab Procedure

Gently scrape the inner side of the cheek using a toothpick, which will collect some cheek
cells.

Place the cells on a glass slide that has water on it.

Mix the water and the cheek cells using a needle and spread them.

Take a few drops of Methylene blue solution using a dropper and add this to the mixture on
the slide.

After 2-3 minutes remove any excess water and stain from the slide using a blotting paper.

Take a few drops of glycerine using a dropper and add this to the test mixture.

Take a clean cover slip and lower it carefully on the mixture with the aid of a needle.

Using a brush and needle, press the cover slip gently to spread the epithelial cells.

Remove any extra liquid around the cover slip using a blotting paper.

Place this glass side on the stage of the compound microscope and view it.

Observations

A large number of flat and irregular-shaped cells are observed.

The cells do not have a cell wall. However, each cell has a thin cell membrane.

A deeply stained nucleus is observed at the centre of each cell.

No prominent vacuoles are observed in the cells.

Conclusion
As the cells observed do not have a cell wall, nor a prominent vacuole, the cells of the specimen on
the slide are animal cells.

Simulator Procedure (as performed through the Online Labs)

The Select view drop down list allows you to select either the Microscope or Binocular View
(It is the 'Binocular view' through which you can see the cell structure as viewed through the
microscope).

To select the sample, you can use the Select sample drop down list.

To change the power of the lens, you can choose a lens from the Select objective lens drop
down list.

For coarse adjustments, you can either click on the up and down arrow of Coarse
adjustment knob, or click on the left and right arrows of 'Course Adjustment' seen on the left
controls panel.

For fine adjustments, you can click on the left and right arrows of Fine adjustment seen on
the left controls panel.

You can redo the experiment by clicking on the Reset button.

Precautions

Ensure toothpick used to scrape the cheek is clean, so it does not cause infection to the
cheek.

Extra glycerine stain should be removed using blotting p

http://amrita.olabs.edu.in/?sub=79&brch=15&sim=125&cnt=2

You might also like