All posts by Jonathan Crowther

UV Microscopy – High resolution imaging poroids in Pinnularia diatoms

A quick update for this post, looking at some work I’ve been doing with UV microscopy recently. I’ve been imaging a strew slide by Neville Bradpiece made in 1998 which has a wide range of diatoms on it from Toome Bridge, Northern Ireland. Today’s post looks at some Pinnularia diatoms, and imaging of very small features – poroids – in their structure using 365nm UV light.

First an unstacked image of a Pinnularia from the slide, taken using my modified Olympus BHB microscope and a 365nm LED light source (image is shown at original pixel resolution).

Pinnularia from Bradpiece strew slide

Going in close to the top right of the image starts to show a hint of of the poroids as dots in the structure.

Poroids in the Pinnularia image

Using ImageJ, I measured 929nm between 6 of them, giving a spacing of 186nm centre to centre. Poroid imaging in Pinnularia is seen as a huge challenge, given how tiny these features are, and it is normally recommended to use circular oblique lighting and cross polarization. This image was done using bright field with slightly (linear) oblique lighting using a Lomo OI-14 Aplanat condenser, and no polarization. The objective was a 63x Leitz Pl Apo NA 1.40, with oil immersion. The condenser also used oil immersion. The benefit here comes from using 365nm UV light, with the shorter wavelength offering improved resolution.

Even in my optimistic frame of mind would accept that the image above barely has enough resolution to see the poroids, however I did image are few other examples of Pinnularia on the slide, one example of which is shown below.

Fragment of a Pinnularia

Going in close and doing some measurements with ImageJ gave the following.

Imaging poroids in a Pinnularia diatom

With this one, the spacing worked out as 190nm centre to centre, and they are a bit clearer here. Other examples gave poroid distances of 235nm and 251nm so there was a bit of variation between samples.

As always, an image of the slide made in 1998.

Slide by Bradpiece with a strew of diatoms from Toom[e] Bridge, Northern Ireland

The slide uses Sirax as the mountant and quotes a refractive index of 1.8(!). Sirax refractive index is quoted between 1.65 and 1.81 depending on sources, and I’ll be getting more into this when I get round to writing a full article on this slide. High refractive index is great for imaging diatoms due to the improved contrast it offers. There are lots and lots of different diatoms on this slide, and I think it warrants a proper paper showing more images from it. Another article for the list of ones to write…..

As always, thanks for reading, and if you’d like to know more about my work I can be reached here.

Microscopy – Some slides from the Quekett Microscopium event 2023

On the 12th August 2023 I attended Microscopium which is an annual event held by the Quekett Microscopical Club where members and visitors meet for the buying and selling of surplus microscopy related items. There were some fascinating items for sale, including some absolutely amazing microscopes, and I came away with a couple of boxes of microscope slides of diatoms (and other things) which I am now gradually working through. Today I’d like to share some of the initial images of the slides.

First is a slide by Frederick Marshall of three Coscinodiscus oblongus diatoms. This slide is beautifully decorated with an engraved pattern all over its surface.

Frederick Marshall slide of Coscinodiscus oblongus

The diatoms from the slide are shown below.

Coscinodiscus oblongus diatoms from the Frederick Marshall slide

Unfortunately at some point in its life, the coverslip has been damaged, as shown in the composite image below.

Composite image showing the coverslip damage

It turns out that these engraved Frederick Marshall slides are quite rare and desirable, and I got lucky with this one as it was in with a box of 100 diatom slides. For more about Frederick Marshall see here.

Next is a Brightwellia coronata diatom from a strew slide from Jackson’s Paddock, Oamaru, New Zealand.

Brightwellia coronata diatom

This is a strew slide with a lot of material present, as can be seen from a wider field of view below (B. coronata in the middle of the image).

Strew slide from Jackson’s Paddock, Oamaru

I’ve yet to find an intact one of these on a slide, and this is about the most complete one I have come across so far. Apparently the ‘LI’ on the right hand side stands for ‘Little Imp’. An image of the slide is given below.

Jackson’s Paddock strew slide

The next one is an interesting curiosity, and also from a Jackson’s Paddock strew slide, although this one was made by Steve Edgar as one of his Meakin Collection slides.

Curious fragment in a Jackson’s Paddock slide

I’m not entirely sure what this is, and I initially thought this was something artificial and a contaminant, although having spoken with a diatom expert it could be part of a sponge spicule. It looks really cool in dark ground imaging.

Dark ground image of the fragment

It reminded me of a punched card and I went back and did a high resolution image using a 63x Leitz Pl Apo NA 1.4 objective and a stack in Zerene.

Stacked image of the fragment

It still amazes me as to how artificial this looks. Very curious and I shall look out for similar examples in other slides. Speaking of which here’s the slide it came from.

Jackson’s paddock slide with the curious fragment

Next is a locality slide from Moulton, California made by Northern Biological Supplies (NBS).

Locality type slide from Moulton, California

And the slide itself.

NBS locality slide from Moulton, California

The image I took of this was not stacked, and I’ll return to it when I have more time to do a better one.

Finally for now, back to Oamaru, and a 4 form slide (unknown maker).

4 form slide from Oamaru

Here’s the slide (if you know the maker drop me a message please).

4 type slide from Oamaru

These were some of the nicest slides I have found so far from the Microscopium event. Most of the ones I got were strews, and they aren’t all like these, but even so I’m more than happy with what I managed to find there. I also came away with a box of slides of different Bacteria which it will be interesting to look through as they are completely different to my usual subjects.

The Microscopium event was great and although you don’t need to be a Quekett member to attend I can well recommend membership to the club. There is an incredible depth of knowledge in the club and the USB copy of the old Quekett Journals is well worth getting. For a summary of this years Microscopium event see here. One thing I would say about Microscopium it is not for people to go along and pick up bargains to then subsequently sell, this is for those passionate about microscopy.

As always, thanks for reading and I hope you enjoyed the images. If you’d like to know more about my work I can be reached here.

Microscopy – effects of mountant refractive index on diatom visibility

It’s well known in diatom microscopy that the refractive index (RI) of the mountant has a huge impact on successful imaging. But why is this the case? Today I’ll show some examples of Amphipleura pelludica – a particularly challenging diatom to image – mounted different ways, and how the final images look.

I’ve chosen three different mountants with different RIs. First a dry mount by Watson. Second another Watson slide with Hyrax used as the mountant. Third is something a bit more unusual, a Horace Dall slide which uses titanium dioxide (TiO2) as the mountant. All images were done on my modified Olympus BHB microscope, using 365nm light from a Zeiss 50W HBO mercury xenon lamp. A 63x Leitz Pl Apo NA 1.4 objective was used with oil immersion. A simple Olympus Abbe condenser was used, again with oil immersion, to give brightfield images. A 2.5x Nikon CF PL photoeyepiece was used to project the image to a monochrome converted Nikon d850 camera. I’ll share the images and then an explanation of what is going on. Images have been reduced in resolution for sharing here.

Dry mount

The dry mount is where we begin, and the name itself is a little bit confusing. Dry mounts are typically thought of as there being nothing between the coverslip and the slide, but that is not always the case. With diatoms there is typically a thin adhesive layer on the coverslip holding the diatoms in place, although they can also be ‘burned on’ by using high temperatures. After a chat with a few people on the Diatom Images Facebook page, it seems as though these ones are likely attached with a thin adhesive layer rather than being burned on. Here’s an image of one of the diatoms on the slide.

Image from the dry slide

Here’s the slide (by Watson).

Watson dry mounted slide

The outline of the diatom is clearly visible, although overall contrast on the original image was a little low. The structures within the diatom cannot be resolved here.

Hyrax mount

Next is a Hyrax mounted slide, again by Watson.

Image from the Hyrax mounted slide

And the slide.

Watson Hyrax mounted slide

The diatoms on this slide are larger than the ones on the dry mount, and at this resolution it is s little hard to see the details, so a crop of the main image is given below.

Crop of the Hyrax mounted diatom

In the crop the striae (lines) in the diatom can easily be seen. Measurement of the spacing between them using ImageJ gave a distance of 2.603 microns for 10 lines, or 260 nm each which is in keeping with the expected values. Overall, while the lines are visible, the contrast between the diatom and its surroundings was still relatively low but better than the dry mount.

TiO2 mount

Finally is the TiO2 mount.

Image from the TiO2 mounted slide

The TiO2 forms a layer over the coverslip and diatoms present, and a small hole in the layer can be seen as a white area in the image above. The diatoms were again smaller in this slide, and the striae are very clearly visible. Other images from the slide are below.

Image from the TiO2 mounted slide
Image from the TiO2 mounted slide

And the wonderful slide (I am a fan of Horace Dall’s work).

Horace Dall TiO2 mounted slide

Mounted in TiO2, the diatoms have much better contrast with the surroundings. The final diatom image above shows some interesting features where it looks like the diatoms have only been partially coated.

Measurement of the striae distances from this slide gave spacings of between 260 nm and 264 nm, so again in keeping with what would be expected for A. pellucida.

What is going on here? There seems to be a clear difference in the appearance of the diatoms between the different mounts (keep in mind all the other settings were kept the same).

The key thing here is the difference between the refractive indices (RI) of the diatom and what it is mounted in. The bigger the difference between the diatom and its surroundings, the more visible it becomes. This is known as the ‘visibility index’ and is discussed further in “Special Methods in Light Microscopy” by Robert McLaughlin.

The RI of silica in diatoms is quoted as 1.434. The adhesive layer on the dry mount slide is likely something gelatin based with a relatively low RI of around 1.5. With relatively little difference between the the diatom and it’s surrounds, the visibility will be low, resulting in low contrast and making it hard to discern details such as the striae.

Next is Hyrax and I have seen this quoted as having an RI of between 1.65 and 1.70. This bigger difference to the silica of the diatom, increases this visibility index, and a higher contrast is seen making it easier to see the fine striae.

The final on is TiO2. It’s RI is very high, and is written on the slide as 2.90. This is hugely different to the diatom silica, and the result is a much higher contrast image, making the diatom and its structures easier to see.

As you can see, going to a higher refractive index is beneficial for imaging diatoms as it improves contrast and helps with visibility. It should be remembered that what is being enhanced is contrast and not resolution (which is defined by wavelength of light and NA of objective and condenser). The resolution of the setup is the same for all these slides, but changing the mountant varies the contrast and therefore the visibility of the features. Interestingly, although perhaps not visible on the images at the resolution on here, the lines on the TiO2 coated slide were sometimes a little less well defined than the Hyrax slide. I think what is happening here is that the TiO2 is forming a layer over the diatom which is ‘a few’ nm thick and with minor variations in thickness. I think it is these variations in thickness which are being imaged and resulting in a slight waviness of the lines.

If you’ve made it this far, well done, and I hope you found it interesting. As always, if you’d like to know more about my work I can be reached here.

Microscopy – Four diatom slides by W.A. Firth

Today’s post contains the images from four diatom slides by the slide maker William Allott Firth (W.A. Firth) who is widely renowned as one of the finest slide makers there has been. For more information about W.A. Firth, take a look at the excellent writeup here. These were taken using my modified Olympus BHB microscope and 450nm LED light and have been stacked using Zerene. All have been reduced in resolution for sharing as the original file sizes are too big to upload efficiently.

Opephora schwartzii

The first one is Opephora schwartzii. This was imaged using dark ground, using a 100x Leitz Pl Apo NA 1.32-0.60 objective with oil immersion, and a Reichert Neo 1.42/1.18 dark ground condenser. The iris on the objective was closed to make the dark ground image. There are 2 examples of the diatom on the slide.

Opephora schwartzii diatoms

These were small diatoms, but even so the original has been reduced in size for sharing, and some of the details are lost in the image above. Below is a crop of the image shown at the original pixel resolution.

Crop of a Opephora schwartzii diatom

The dark ground image has done a good job of bringing out some of the sub-micron structures in the diatom. Got to love pushing the limits of optical microscopy. Before I move to the next one here’s the slide.

Opephora schwartzii diatom

Aulacodiscus orientalis

Next is Aulacodiscus orientalis. This was imaged using brightfield with an Olympus Aplanat Achromat condenser (oil immersion), using a 40x Leitz Pl Apo NA 1.00 objective with oil immersion.

Aulacodiscus orientalis diatoms

Here’s the slide.

Aulacodiscus orientalis slide

This one confused me a bit at the beginning as it looked like it said ‘Hulacodiscus’ for the name.

Surirella macraeana

The next slide is Surirella macraeana. Back to a smaller diatom, and a 100x Leitz Pl Apo NA 1.32-0.60 objective was used (with oil immersion), and the iris closed down slightly to make circular oblique lighting. A Leitz Heine condenser was used along with oil immersion.

Surirella macraeana diatom

There were 6 examples of this diatom on the slide.

6 examples of the diatom

And of course the slide.

Surirella macraeana slide

Stephanopyxis corona

The final one today is Stephanopyxis corona. This was imaged with a 40x Leitz Pl Apo NA 1.00 objective, oil immersion, and an Olympus Aplanat Achromat condenser, slightly oblique, again with oil immersion.

Stephanopyxis corona diatoms

There was actually 3 examples of the diatom on the slide, but one of them had gone walkabout almost to the inner edge of the ring, as can be seen with this lower magnification image.

Stephanopyxis corona diatoms, one having moved quite a distance

I had a bit of trouble selecting an objective to images this slide. They were quite small, and my original choice of 100x and 63x Leitz ones did not have sufficient working distance to reach the diatoms, and a 40x Nikon UV-F NA 1.30 couldn’t didn’t get the full depth of the diatom. Hence I ended up using the 40x Leitz, which had a slightly longer working distance. The diatoms must be quite thick (dome shaped) and this can be more easily seen if you look at a video of me moving the microscope stage.

It also proved difficult to capture the two rings of spikes in the structure in the flat photo, and they can be more easily seen in the video.

Here’s the slide.

Stephanopyxis corona slide

I really enjoyed looking at these slides as their quality was so good. Amazingly, they were relatively cheap for such good quality slides – I paid about £15 for them. As always, I hope you enjoyed the images, and thanks for reading. If you’d like to know more about my work I can be reached here.

Microscopy – Diatom Arrangements by Klaus Kemp

Diatoms make great subjects for microscopy given the fine details in their structure. In addition to slides with a strew of material and single examples of given species some slide makers have produced amazing arrangements either to show what species are present in a given area, or to create fantastic works of art. While more common in the Victorian era, the production of these artistic arrangements has continued into the 20th and 21st century with a few very skilled practitioners. Today I’d like to share a few examples from the amazing Klaus Kemp (now sadly passed away) which I feel very privileged to have in my collection. All images were taken on my modified Olympus BHB microscope, using 450nm light, and with a Leitz Heine condenser. They have been reduced in resolution for sharing here, so some of the fine detail present will have been lost.

First up an exhibition rosette with 136 diatoms.

Klaus Kemp 136 form exhibition rosette

This slide has unfortunately had an accident at some point in its past. However it will be well looked after and at some point I may try repairing it.

Klaus Kemp 136 form slide

Next, another exhibition rosette, this time a 145 form one.

Klaus Kemp 145 form exhibition rosette

And the slide.

Klaus Kemp 136 form slide

A smaller example now, a 25 diatom exhibition star.

Klaus Kemp 25 diatom exhibition star

And the slide.

Klaus Kemp 25 diatom exhibition star slide

Next is a circular arrangement, again with 25 diatoms.

Klaus Kemp 25 diatom circle

And the slide for the circle.

Klaus Kemp 25 diatom circle slide

For the final couple of examples, we’re moving away from the more artistic arrangements to ‘Type Slides’. Both of these are with 50 forms, and both images are much smaller than the original file sizes to allow me to upload and share (for example Type Slide 2 was made from a composite of 4 stacks of images and the original file size was 24750×5504 pixels!!).

Klaus Kemp 50 form Type Slide 1
50 form type slide 1
Klaus Kemp 50 form Type Slide 1
Klaus Kemp 50 form Type Slide 2

These slides really are works of art, crafted by a hugely skilled diatomist. Thank you Klaus for the beauty you brought to the world.

As always, I hope you enjoyed the post, and if you’d like to know more about my work I can be reached here.

Microscopy – Amphipleura pellucida with polarized light

This post finishes off something I started yesterday (see here) where I tried pushing the resolution I managed to get on a couple of diatom slides while only using 450nm light. The N.B.S. slides I imaged had diatoms with small features on them, but one really difficult diatom to photograph using visible light is Amphipleura pellucida, as this has features down around 250nm in size. While I have done some work with this using UV light (for example here), my challenge was to try using 450nm blue light in combination with circular oblique illumination and even cross polarization.

Equipment wise I used my Olympus BHB microscope with the Reichert Neo condenser, in combination with a 100x Leitz Pl Apo NA 0.60-1.32 objective. Kept wide open, this objective has such a high NA, the result is that instead of dark ground, I get a circular oblique illumination. 450nm LED light. No stacking was done, but for the non-polarized image I did try averaging 10 images in camera. Cross polarization was done by adding one linear polarizer just above the field lens, and another on top of the photoeyepiece, rotating the bottom one until the image was at its darkest. The images showing the full diatom have been reduced in size for sharing here, but I’ll also include crops at original resolution to better show the features.

First the non-polarized image, so just using circular oblique illumination.

Amphipleura pellucida with 450nm light and circular oblique illumination

The striae are nice and obvious with this setup, and can be more easily seen by looking at a closeup of the top right of the diatom.

Amphipleura pellucida with 450nm light and circular oblique illumination, cropped

Putting this into ImageJ and doing some measurements, we have 10 striae in 2.659µm (or 37.6 striae per 10µm). This matches pretty well with the slide which says 40 lines per 10µm. Think of it another way, those lines in the image above are 266nm apart.

Cross polarizing in combination with oblique lighting is well known as a way of showing the punctae on this diatom, so I tried that approach. I placed a linear polarizer on the field lens (below the condenser), and then another on top of the photoeyepiece. I rotated the one on the field lens until the image was at its darkest which would be when the polarizers are aligned at 90 degrees to each other, and then took an image of the diatom.

Amphipleura pellucida with 450nm light and circular oblique illumination and cross polarization

Even looking at the whole diatom, the lines appear a bit more broken up, but this can more easily be seen on closeups of the top right and bottom left of it.

Amphipleura pellucida with 450nm light and circular oblique illumination and cross polarization, top right
Amphipleura pellucida with 450nm light and circular oblique illumination and cross polarization, bottom left

With cross polarization, I’m just starting to see a hint of punctae of the diatom (dotting), where the lines are breaking up into individual features, so has been beneficial for the imaging here. For using 450nm light I am happy with this result, especially at the first attempt.

Before I go, here’s the slide.

Amphipleura pellucida slide by N.B.S.

As always, thanks for reading, and if you’d like to know more about my work, I can be reached here.

Microscopy – Two N.B.S. diatom strews with circular oblique illumination

Yesterday I wrote about the use of a Reichert Neo dark ground condenser which I had had a mount made for so it could be used on my Olympus BHB microscope (see here). Today’s post also used the Reichert Neo condenser, but this time in combination with a 100x Leitz Pl Apo NA 0.60-1.32 objective. Kept wide open (or nearly wide open) as this objective has such a high NA, the result is that instead of dark ground, I get a circular oblique illumination.

First slide is an N.B.S. strew test slide of Nitzschia obtusa. It is a stack of 10 images in Zerene and used 450nm light. This is the full image frame, but has been reduced in resolution for sharing here.

Nitzschia obtusa using 450nm circular oblique lighting and a 100x Leitz Pl Apo NA 0.60-1.32 objective

Below is a crop of the top end of the diatom, shown at original pixel resolution to allow the details to be more easily seen.

Crop of Nitzschia obtusa using 450nm circular oblique lighting and a 100x Leitz Pl Apo NA 0.60-1.32 objective

Measuring the striae distance in Image J gave me 3 in 1.058µm, which equates to 28.4 per 10µm. A little below the information on the slide (30 per 10µm), but not too far away. Here’s the slide.

Nitzschia obtusa slide from N.B.S.

Note the refractive index of the mount is high – 1.72 – and that has helped here with the imaging.

After the success with the Nitzscia slide, I had a look at my collection, and found another N.B.S. one with a more challenging diatom – Frustulia rhomboides var. Saxonica – which the slide claimed to be 35 lines per 10µm. Setup was the same as above (other than I left the iris on the objective fully open), but this time it was just 2 images stacked, and the frame has been cropped slightly before resizing for sharing, as the individual diatoms were smaller than the Nitzschia.

Frustulia rhomboides var. Saxonica using 450nm circular oblique lighting and a 100x Leitz Pl Apo NA 0.60-1.32 objective

Again, there is lots of detail in the image, and here is a crop showing part of the lower diatom at original resolution.

Crop of Frustulia rhomboides var. Saxonica using 450nm circular oblique lighting and a 100x Leitz Pl Apo NA 0.60-1.32 objective

Putting this into ImageJ, I got a measurement of 2.87µm for 10 striae, which equates to 34.8 striae (lines) per 10µm. The slide said 35 lines per 10µm, so I am happy with that.

Frustulia rhomboides var. Saxonica slide from N.B.S.

Overall, I really like this setup for imaging diatoms with small features at high resolution. The circular oblique lighting is a little lower contrast than I would prefer, but certainly shows up the features, and it shows what can be done with relatively straightforward visible light microscopy (instead of my usual UV work). As always, thanks for reading, and if you’d like to know more about my work I can be reached here.

Microscopy – Actinocyclus ehrenbergii diatoms by W.A. Firth and modifying microscopy equipment

A two part post today, first the imaging of a beautifully made slide by W.A. Firth containing 3 examples of the diatom Actinocyclus ehrenbergii, and then a piece on adapting microscope equipment from different manufacturers to be able to use them together.

First the slide. This was imaged on my Olympus BHB microscope, using 450nm LED light. A 20x Nikon Plan Apo NA 0.65 objective was used, in combination with an Olympus Aplanat Achromat condenser set to slightly oblique. A 2.5x Nikon CF PL photoeyepiece was used, and the camera was a monochrome converted Nikon d850 camera. This is a stack of 9 images (Zerene stacker), and has been reduced from the original resolution for sharing (this is 1600 pixels across vs. 4632 pixels for the original) which has obviously impacted viewing resolution. Here’s the image.

W.A. Firth slide with 3 examples of Actinocyclus ehrenbergii, 20x objective

The slide is beautifully made (and W.A. Firth is well known for excellent slides – see here for more information), and oblique illumination has done a good job of showing the features. Here’s the slide.

W.A. Firth slide with 3 examples of Actinocyclus ehrenbergii

Being a bit of an optics geek, I enjoy trying out equipment from different manufacturers, hence the use of a Nikon objective and photoeyepiece on my Olympus microscope. With regards to condensers, very often different manufacturers have different specifications in terms of how they are mounted to microscope, so unlike objectives which are often RMS threaded, trying different condensers can be more of a challenge.

A few weeks back I bit the bullet and contacted a machinist I have used before – Machined Precision Components Ltd, based in Norfolk, UK – to make me some adapters to allow different condensers to be mounted on my Olympus BHB. The first one I’ve got back is for a Reichert Neo 1.18/1.42 dark ground condenser. Here’s the condenser in its custom made mount (condenser at the top – black knurled section and above, mount at the bottom).

Reichert Neo 1.18/1.42 dark ground condenser in its custom mount

Here’s the information about the condenser, from the underside of its original Reichert mount.

Reichert Neo 1.18/1.42 condenser information

What a wonderfully made piece of equipment this condenser is, but then Reichert are well know for high quality microscope components. The custom made mount fits the condenser really well, and is a nice snug fit in the Olympus microscope. The condenser was oiled to the underside of the W.A. Firth slide, and other than that the same setup was used as for the image above (7 images were used for the stack was used this time) to produce a dark ground image shown below.

W.A. Firth slide with 3 examples of Actinocyclus ehrenbergii, 20x objective, Reichert Neo dark ground condenser

The Reichert Neo dark ground condenser produced a really nice dark ground image, which did need a bit of tidying up – a few bright spots in the background, probably from dust on the slide, and a bit of haloing around the bright diatoms (not unexpected). While perhaps not obvious on the images here the dark ground image looks be slightly lower resolution than the oblique illumination image. It is commonly thought that dark ground imaging is a way of getting higher resolution, however as I understand it is more of an ‘apparent increase’ in resolution as a result of improved contrast making things more easy to see. In the oblique illuminated image contrast is already very good, hence the dark ground image actually looks a little softer in comparison.

As a practical man I like to make my own equipment whenever I can, but not having a lathe (or at the moment, the skill to use one) finding a good machinist is a key part of my research. I’d like to thank Machined Precision Components for putting up with my complex demands, and less than perfect sketches when I send through requests, and I look forward to reporting back on the other items that are being made.

As always, thanks for reading, and if you’d like to know more about my work, I can be reached here.

UV Microscopy – Amphipleura pellucida diatom slide using 365nm light

Those that know me, know I have built myself a UV capable microscope primarily for imaging sunscreens at 365nm and 313nm. I also enjoy imaging other things and diatoms are great for UV imaging as the achievable resolution for a given setup improves as the wavelength decreases. However there can be issues with UV imaging – glass absorbs short wavelength UV, so optics and slides need to be made of quartz or fused silica when going deeper into the UV. Normal slide mountants absorb short wavelength UV, so things like glycerin need to be used. UV dedicated objectives tend to need non standard thickness coverslips, which can be expensive and awkward for getting slides made up. However there is a bit of a middle ground – imaging slightly into the UV at 365nm means that often fairly conventional microscope optics can be used, and normal slides are usable (depending on the mountants). In today’s post I return to a favourite of mine – the diatom Amphitpleura pellucida – which has particularly small features, and is a real challenge to image with optical microscopes.

The microscope I used was my modified Olympus BHB. The objective was a 63x Leitz Pl Apo NA 1.4 (oil immersion). The condenser a Watson Quartz Cassegrain. The light source a 365nm LED torch. A 2.5x Nikon CF photoeyepiece was used, and 2x 365nm, 10nm bandpass, Edmund Optics filters were stacked together to filter the light going to the camera. The camera was a monochrome converted Nikon d850 by MaxMax. No stacking at different focus depths was used, however 10 images were averaged within the d850 using its own handy multiple exposure feature. The image is cropped from the full sized one, but is shown at original pixel resolution. Finally after all that, here’s the image showing one end of an A. pellucida diatom.

Amphipleura pellucida, imaging using 365nm light

A. pellucida is a challenging diatom to image as the striations and puncta are so close together (distances of about 200nm apart) with the striations being slightly further apart than the puncta. Usually a combination of oblique and polarized light are needed to resolve both striations and puncta when using visible light. I used a standard (although very high quality), non UV specific objective, a 63x Leitz Pl Apo NA 1.4 in combination with a Watson Quartz Cassegrain dark field condenser which I currently have access to. This condenser has a high NA of about 1.4, and when used in combination with the high NA Leitz objective gives me a circular oblique illumination (not quite full dark field). I did not use any polarization, and just used a cheapy eBay 365nm torch as the light source. The end result was clearly defined striations and puncta and I was able to get some measurements on the distances between them using ImageJ. Firstly, the striae.

Striae measurements

Between 10 striae there was 2.55 microns, corresponding to a 255 nm distance between each one.

Next the puncta.

Puncta measurement

Between 5 punctae there was a distance of 1.081 microns, corresponding to a distance of 216 nm between each one (given how narrow the diatom is I wasn’t able to get a reliable reading between 10 like for the striae, hence I just did 5). So a striae spacing of 255 nm and a puncta spacing of 216 nm, which is in the same ball park as the values I have seen in the literature for this diatom. Let than sink in for a moment, the features in the image above are just over 200 nm apart, that is 250x less than the diameter of a typical human hair (50 microns), 40x less than the diameter of a red blood cell (8 microns), and more like the size of a single virus.

The camera used was my monochrome converted Nikon d850, and I did do something more than usual when capturing the image. The image was taken as a 10 shot average stack done within the camera, where 10 images are combined together and averaged to produce the final image. This was done to reduce noise (even though ISO100 was used, exposure times were about 1s) and to help define the small features in the diatom. I saved it as a RAW file (NEF format). This was then put into Monochrome to DNG which can take a RAW file from a monochrome converted camera and process it to take into account the de-bayered sensor, the results being an improvement in resolution vs. not doing this. Different people claim different amounts of improvement but the effects are certainly noticeable here as I also tried processing the image without do this. After this the DNG file went into Darktable and processed as I would with a normal RAW file, before being exported as a 16 bit TIFF file. This then went into Photoshop for final cropping and processing. All in all it took about 45mins to do this processing, and there was the benefit here of not working with stacking images which can be very time consuming for my old PC.

I have tried imaging A. pellucida before with UV (for example here) and have never been really happy with the results. The image shared here today is the best I have managed to get of it so far, and I love the 3D effect the lighting gave to it combined with the fantastic resolution. What amazed me more though was actually I didn’t use that much in terms of ‘dedicated UV’ optics for this. The objective and the slide are standard not UV specific, as is the photoeyepiece. The condenser is, but there are options for high NA dark field condensers which have reasonable transmission at 365nm. The camera obviously need modifying, and the bandpass filters need buying, but the light source cost me about £30 on eBay.

Before I go, here is the slide.

A. pellucida slide by Watson

It’s a Watson slide, and is a strew of diatoms mounted in Hyrax. As always thanks for reading, and if you’d like to know more about my work, I can be reached here.

Microscopy – Triceratium nitescens diatom by R.I. Firth

The microscope images today come from a Triceratium nitescens slide by the maker R.I. Firth. The slide is marked as ‘Barbados’ and ‘Very rare’, is mounted in Styrax and was made in 1943. There is a single example of the diatom on the slide. The microscope used was my modified Olympus BHB. The condenser was an Olympus Aplanat Achromat, set to slightly oblique and oiled to the underside of the slide. The objective was a 63x Leitz Pl Apo NA 1.40, oiled to the top of the slide. The photoeyepiece was a 2.5x Nikon CF. Lighting was 450nm LED (white LED source with a Thorlabs 450nm, 40nm bandpass filter). 17 images were stacked in Zerene stacker. Camera was a monochrome converted Nikon d850 done by MaxMax. The image has been reduced in resolution for sharing here, although a crop of the main image at original resolution is also shared. Here’s the final image.

Triceratium nitescens diatom from R.I. Firth slide

And a crop of the image shown at original pixel resolution.

Triceratium nitescens diatom, crop at original resolution

A friend asked me about why I do stacking, and it reminded me that I have written something which briefly covered the reason before (see here). Essentially though, at high magnifications with high NA objectives, the depth of field of a single image is tiny – way less than a micron in the setup used here – making it impossible to get the whole diatom in focus in a single photograph. So I take multiple photos, moving the stage very slightly between each one, and then stack them together using software called Zerene which takes the in focus parts of each image and combines them. Live view on the camera is very helpful here, as I can see what is in focus with each image and make the movements accordingly to capture everything. After stacking, the image still need some cleaning up which I do in Photoshop – making the background smoother, and removing artifacts from dirt on the sensor coverglass. Microscopy will reveal dirt that you never knew was there before in normal photography – the shorter the wavelength of light used, and higher the objective NA, the smaller the features that will be revealed and will need editing out. I tend to capture my images in the camera as Jpeg files, and after stacking output the stack a 16 bit Tiff for subsequent work (outputting the stack as a Jpeg gave me more artifacts especially in the background of the image). Ideally I would do everything as Tiff files, but my computer is old and stacking Jpeg files is a tough enough job for it.

Before I go to cover some UV work I did with this slide, here is the slide itself.

R.I. Firth slide of Triceratium nitescens

While the slide was on the microscope, I also did some experiments with 365nm UV (the camera is monochrome converted and sensitive to UV, visible and IR light). Keeping the same objective, the condenser was swapped for a Watson Quartz Cassegrain (which I just realised I’ve not written about before so must rectify that), the light source was a 365nm LED torch, and I filtered the light going to the camera by a stack of 2 Edmund Optics 365nm, 10nm bandpass filters. Here’s a photo of the diatom made up from a stack of 4 images with this setup.

Triceratium nitescens diatom, imaged using 365nm light with a Watson Quartz Cassegrain condenser

The effect wasn’t quite what I was after, but I think it still looks nice and shows the features in a different way. Given the high NA of the objective the effect is produce circular oblique lighting when using the Watson Quartz Cassegrain condenser.

Before I go, one last thing to mention, when dealing with what is effectively extreme macro imaging on the microscope, the settings under which the images are taken can have a huge impact on the quality of the final photograph. I use 3s exposure delay and electronic front curtain shutter, but found out something yesterday which is worth sharing. Initially I tried ‘S’ (Single shot) and the image was quite blurry. I changed to ‘Mup’ (Mirror up) and the image was much sharper. Comparison below using 365nm LED light, crops shown at original pixel resolution.

Dial in ‘S’ position
Dial in ‘Mup’ position

I tried this a few times, and got the same result each time, so will be using Mup from now on with the d850. I seem to recall Nikon changed how they deal with Mup with the d850 so it can now be used with live view unlike earlier Nikon cameras, so perhaps it is a d850 specific consideration.

Slightly longer post than usual today, but I thought it would be useful to cover off some other things in addition to showing the main image. As always, thanks for reading, and if you’d like to know more about my work I can be reached here.