I’ve had a bit of a catch up this weekend and looked at some slides that have arrived recently. They are both pretty and interesting (for different reasons), so I thought I would post them here. All have been reduced in size for sharing.

First is a Pinnularia slide by Arthur Cole.

There was a single example of the diatom on this slide. The mount was amazingly clear and needed very little cleaning up. Imaging was simple bright field, using 450nm light. Date for this is likely the late 1800s.

Next we have a Triceratium scitulum slide by Thomas Russell of London.

A nice arrangement of 4 diatoms. Again the mount was very clear and needed very little cleaning up. Imaging was brightfield with 450nm light. Likely date for this one, late 1800s.

I returned to this slide the next day and took a picture of one of the diatoms using the Watson Holoscopic condenser to give a dark ground image.

Using the Holoscopic condenser has brought out the dot features present within the polygonal structures which were barely visible in the bright field image. The ‘dots’ look to be about 500nm across.

Now an arrangement of Actinocyclus by Eduard Thum, Leipzig.

A typical Thum arrangement using two large circular diatoms to mark where the arrangement is. Imaging was oblique brightfield with 450nm light. Likely date for this one, again late 1800s.

Last but not least is a Bacteriastum furcatum slide by Watson.

Imaged using 450nm light and a Watson Holoscopic condenser (discussed here) to give the dark ground image. Single images, no stacking.

There are always lots of these interesting old diatom slides available on places like Ebay and in my experience they offer great value for money for those wanting to image slides, so don’t discount them because they are old. As always, thanks for reading and if you’d like to know more about this or other aspects of my work, I can be reached here.

A special slide to share today. I’ve shared a few images from slides made my Horace Dall before, and this is another one. Dall was an optics genius, and experimented with using titanium dioxide and aluminium layers to improve the visibility of diatoms. This slide is from 1960 and is marked up as being ‘Pleurosigma Balticum‘. Here are some images from it taken on my modified Olympus BHB microscope using 450nm light. All have been reduced in resolution for sharing here.

The aluminium coating has the effect of improving the contrast for the diatom, making the features more visible. These were imaged with simple bright field illumination from below. I’m not happy with my stack with the 40x objective, as there are artefacts in there, so I shall redo that at some point.

Here’s the slide itself.

The slide says it is ‘best seen with vertical illumination and with oil imm[ersion].’. However my system is not really setup for illumination from above, which is why I lit it from below. This slide appears in the Quekett Microscopical Club book, “Microscopical Mounts and Mounters” by Brian Bracegirdle (plate 13, D). It’s an early example for one of Dall’s aluminium coated slides, and looks different to the others I have which are shiny like mirrors. I suspect it is an early experiment using the technique.

EDIT – after a chat with an expert, while the slide is marked as Pleurosigma balticum, (new name, Gyrosigma balticum) it is not that, and is likely to be one of very big Pleuorsigmas like Pleurosigma formosum. This highlights an issue I have had with a few different diatom slides – that of naming. I’ve had a few slides where the name on it did not match the subjects.

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

A while ago I shared an image from a diatom which had been sputter coated with gold to help provide increased contrast (see here). Since then I have been working on a paper about metal and metal oxide coated diatoms for microscopy which I am aiming to publish later this year. In the mean time I’ve been talking with a nice gentleman who coats diatoms with gold using sputter coating and he has sent me a few slides to examine. I’d like to share some of the images from those slides today.

For background, my modified Olympus BHB microscope was used for these, and the objective was a 60x Olympus Splan Apo NA 1.4 oil immersion one. 450nm light was used to illuminate the samples from below. Stacking was done using Zerene. Other details are shared for each image and the image sizes have been reduced for sharing here.

Please forgive me if the names of some of these are wrong, I am still learning the complexities of diatom nomenclature. First one, a Stephanopyxis diadema Ehr. from a fossil diatom strew from Dunkirk, Maryland.

This has an amazing 3d structure, and looks to be hemispherical. There were 13 images in the stack. The tiny depth of field from the objective can be seen if you look at a video where the stage is moving.

Next is a Stauroneis (perhaps a phoenicentron) from Mud lake, Ottawa, Ontario.

This was mounted in Pleurax and was a lovely slide to image as there was virtually no debris around it, and the diatom was almost flat. Only 3 images were used in this stack.

The small features don’t show up too well in the file which has been resized for sharing, but this is what they look like at original pixel resolution.

Final one is a Heliopelta of some description, again from a fossil diatom strew from Dunkirk, Maryland.

Why do this process? The gold sputter coating of the diatoms makes the silica of the structure darker and so improves their contrast, so that even simple bright field imaging (as used here) can be used successfully.

As always, I remain fascinated by these amazing structures and the worlds that can be viewed with a relatively simple light microscope. Thanks for reading, and if you’d like to know more about my work I can be reached here.

As it was Easter recently, I had a bit of spare time to do some microscopy. This time I returned to one of the diatom slides I had made using a quartz slide and coverslip so it could be used in short wavelength UV light. I thought this would be a good time to try out a trio of Leitz UV objectives (16x NA 0.25, 40x NA 0.65 and 100x NA 1.20) for imaging at 313nm. However as we shall see, things didn’t quit go to plan.

For the imaging I used my UV modified Olympus BHB microscope, and a monochrome converted Nikon d800 camera. Condenser was a Zeiss quartz one (simple brightfield illumination). Light source was a Zeiss mercury Xenon lamp. Edmund Optics 313nm, 10nm bandpass filters (2 stacked together) were used to filter the light going to the camera. The subject was a diatom slide which had a Gyrosigma (balticum, maybe) along with 2 other diatoms. Images have been reduced in overall size for sharing here.

First, a stacked image using the 40x objective, showing part of the Gyrosigma (stacked using Zerene).

Plenty of detail to be seen with the 40x objective. Even being NA 0.65, the short wavelength light results in improved resolution compared with visible light imaging under the same conditions. As I had the system setup, I thought next I would try the 100x NA 1.20 objective (again as a stack). Unfortunately the depth of focus this objective has meant that I could not move the stage enough to get all of the diatom in focus, and the image only shows part of the structure in focus.

This highlights the issues with these high magnification, high NA objectives, i.e. very shallow focus depth. This make imaging thick samples a problem. Perhaps a thinner coverslip would work here, although likely with a bit of loss of image quality (more on that later). Final image, using the 16x objective, not stacked this time.

The 16x image shows the 3 diatoms on the slide. Speaking of the slide, this is it.

Here are the three objectives.

These objectives are interesting and I have been able to find out very little about them. They seem to be the Leitz equivalent of the Zeiss Ultrafluars and are very well made. The 40x and 100x are designed for glycerine immersion, and as I am writing this I have just had a thought. I use a 50:50 mix of glycerine and water for my immersion as this was what I got from a microscope supplier when I started out. This will have a different refractive index compared with pure glycerine. I wonder if this reduced the depth of focus for the 100x objective? I will park that idea and check later. The only mention I have found of ‘Leitz UV objectives’ is in a brochure from 1985 which I have shared before (here). Thing is in the brochure the coverslip thickness is quoted as 0.17mm, not the 0.35mm ones written on the objectives themselves. All very odd. Perhaps a thinner coverslip (0.20 or 0.25mm) might be an option and get me to image slightly thicker samples, although whether this would drop the resolution with the 100x objective would need to be checked.

A good friend asked me recently what was the actual benefit of these quartz based optics and UV light over conventional methods for improving resolution. Funnily enough I have written a bit about that before (here) and there isn’t much between say UV imaging at 313nm and imaging with blue light and a high NA objective and oblique illumination. I do it because I enjoy it and it was a scientific challenge to make the microscope. However where it will get more interesting is below 300nm, as the shorter the wavelength the greater the resolution. It is proving challenging that far into the UV though, and more work needs to be done. The microscopists of the early 1900s managed it though, so I know it can be done.

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

With microscopy it is important to know how big the object is that you are imaging. The simplest way to do this is to use a stage micrometer (also called a graticle or graticule). These are small rulers and designed to be imaged using the same optics setup as being used to image the subject. The ‘pixels per micron’ can then be calculated and used to make a scale bar on the image of the subject. Today I’ll show you a couple of the ones I use. But first a nice image from an old Leitz one showing their name as well as 250µm of the micrometer.

The ones I tend to use on a daily basis were made by Graticules Ltd in the UK (now know as Graticules Optics Ltd). I have a 2mm one with 10µm divisions) and a smaller one which has 2µm intervals which I use for for higher magnification objectives. Both of these were bought second hand from ebay for about £30 to £40. Here’s what the 2mm one with 10µm divisions looks like.

Using a 10x Olypus UVFL objective this is what the scale bar from the Graticules Ltd slide looks like.

In the image above, each larger division is 0.1mm (100µm) with 10µm smaller divisions between the larger ones. I also picked up an antique Leitz one, which is also 2mm long and has 10µm intervals. Here’s what that slide looks like.

Interestingly the Leitz one has a coverslip attached which the Graticules Ltd one does not. Here is an image of the micrometer scale on the Leitz slide using the same setup as for the Graticules Ltd slide above.

As can be seen from the images of the micrometer, both slides are giving the same scale – about 830µm from the top of the image to the bottom. The Leitz one also has the makes name on it below the scale as shown in the first image in this post.

I also bought another Leitz one which I thought was a micrometer, see below.

Turns out this wasn’t a stage micrometer at all, it was actually a haemocytometer, used for counting cells (you can read more about them here). Here’s an image from the slide.

I should also point out that there are plenty of new, cheap, stage micrometers available from places like ebay, however I have no idea how accurately made they are, which is why I tend to look out for older, second hand ones. In addition to these types of glass ones, they are also available on metal or opaque substrates for reflection microscopy.

The measurement side of things tends to often go unnoticed when compared with some of the beautiful subjects which can be imaged, but it is an important aspect of microscopy. As always, thanks for reading, and if you’d like to know more about my work I can be reached here.

A short update today. I returned a diatom slide I have imaged before – the Amphipleura linheimerii slide originally imaged here. I wanted to try imaging it at 365nm, but using a Watson Quartz cassegrain condenser that I have on loan (a very special condenser, which I will try and write more about in the future). In summary, the condenser is a high NA dark field condenser, and being made from quartz has great UV transmission. This post shares an image from the slide using 365nm dark field illumination.

First though a quick note about the setup. Microscope was my UV modified Olympus BHB. Light source was a 365nm torch slid into the light port of the microscope. Condenser was a Watson quartz cassegrain one, which was used with glycerine/water as the immersion fluid. The objective was an Olympus 40x NA 1.3 silicone immersion lens, which I used with glycerine/water as the immersion fluid. Photoeyepiece was an Olympus 2.5 NFK. Then I used 2 stacked Edmund Optics 365nm, 10nm FWHM bandpass filters. Finally a monochrome converted Nikon d800 which is UV sensitive.

Ok, the image – this has been cropped slightly, and has been reduced in resolution for sharing.

There are two examples of the diatom on the slide, and part of the other one can be seen in the lower left of the image above. There is plenty of detail in the image, but perhaps that does not show up as well with the lower resolution image being shared here. Here’s a crop, at original pixel resolution.

With the cropped image some of the smaller features of the diatom now become visible. The distance between the small white dots was measured on using ImageJ was was found to be about 380nm (length 0.379 in the image below, as it was calibrated in microns).

A few things to note with this setup. Glycerine/water was used as the immersion fluid, while the objective was designed for silicone. Glycerine/water was used for a few reasons – firstly, I had some, secondly, it is a similar refractive index to the silicone, and finally it is much, much easier to clean. However I have to acknowledge this may impact performance of the objective. Being quite a low magnification objective, the field of view was large. As such the resolution achievable by the objective is getting close to the actual pixel resolution of the camera. I did try taking RAW files and then converting them to monochrome in Monochrome2DNG to get more resolution, and I even tried frame averaging in the camera, but neither of these improved things noticeably and I just stuck with the original single frame jpg captured in the camera in the end. The slide itself was not very UV transparent. I suspect if I measure the transmission spectrum of the slide, it will not be great at 365nm. As such this was not the ideal subject. I am also not sure I got the iris on the objective fully optimized, so perhaps a little more resolution can be squeezed from this setup.

Diatom imaging in the UV continues to be a fascinating journey, and one i will continue to pursue. As always, thanks for reading, and if you’d like to know more about my work, I can be reached here.

With my UV microscopy, I’ve noticed a dramatic improvement in resolution as the wavelength decreases. A well known effect, but the UV gives a great demo of how big a change can be observed – for example see my work on a test slide here. However converting a microscope to be able to image in the short wavelength UV region is no small task as glass absorbs the light. Is there a benefit to filtering visible light to image in the short wavelength blue region? There certainly is, and again this is well known, and is something I regularly do. However I thought I would share an example to show the effect, and also how to go about doing it in practice.

The test slide was a piperine mounted diatom slide, and the subject a Pleurosigma angulatum that was present on it. I set up my Olympus BHB with a 40x Olympus Dplan Apo UV objective (set the coverslip thickness collar to as thick as it would go, as the sample had a thick coverslip, and this gave the best setting). Camera was a monochrome converted Nikon D800. Light source a white light LED (more on that later). Condenser an Olympus Aplanat Achromat, set to oblique. Single image, no stacking. Everything was kept the same for the images, except a Thorlabs 450nm (40nm FWHM) filter was added above the photoeyepiece for the 450nm image. Images from the white light and the filtered 450nm light were processed the same way, and the full images have been reduced for sharing here.

First the white light image.

Next, the 450nm light image.

At first glance there doesn’t look to be much between the two images, however going in closer on parts of the images tells a different story. Here are a couple of sets of crops from different parts of the diatom. In each crop the white light image is on the left, and the 450nm image on the right.

Note, the crops above were taken as screen grabs, and have a slight warm tinge to them as that is how I like my monitor. It is subtle, but there is a slight improvement in resolution as a result of using 450nm light vs the full spectrum of the white light.

For the 450nm image, I used a 25mm diameter Thorlabs 450nm (40nm FWHM) premium bandpass filter, and simply placed it on top of the photoeyepiece.

The filter is dichroic and has a preferred direction for the light going through it. This is shown by the arrow on the side of it. The LED is use is a cool white one I got from ebay and made an adapter to fit my light. This is its irradiance spectrum (measured on my Ocean Insight FX USB spectrometer).

As can be seen the LED gives a broad wavelength range, but actually a nice strong peak at around 450nm. This is why I chose this filter and not a 405nm or 400nm one. Also using 450nm vs 405nm or 400nm is good for normal unmodified cameras as typically the internal filters on them will start blocking light at around 400nm. As a result 450nm is a good compromise – short wavelength, but detectable by normal cameras. Also the Thorlabs 450nm filter I used had a wide bandpass of 40nm, allowing plenty of light through. A 10nm bandpass filter would give a narrower range of wavelengths, but much less total light.

This can also be done (and was done historically) by putting a blue filter below the stage. I prefer to have the filter as the last stage before the light reached the camera though, which is why I put the filter on top of the eyepiece.

Of course a blue LED could be used, but as the white LED had a strong peak, I went down the route of filtering the light rather than having yet another light source.

A simple modification, but a helpful one when it comes to diatom imaging, adding a blue filter can produce higher resolution images for a given optical setup when compared with normal white light imaging. As always, thanks for reading, and if you’d like to know more about my work I can be reached here. Oh, I nearly forgot. Here is the slide I used for the images. While is says ‘Surirella gemma‘ on it there are plenty of Pleurosigma angulatum as well.

The Fuji IS Pro camera (or to give it its full name the Fujifilm Finepix IS Pro) was a DSLR introduced back in 2007. Based on their S5 camera body (which was based on a Nikon D200) this was advertised as being able to image in the UV, visible and IR and as such was not really aimed at the consumer market but more at the scientific and forensics areas. Indeed it didn’t go on general release and Fuji required buyers to sign a release form saying they weren’t going to use it for ‘unethical purposes’. It followed on from their S3 Pro UVIR version, but even from the beginning it was apparent that all was not well with the IS Pro and its UV sensitivity especially was not that great. I recently got hold of one these cameras and thought it would be fun to measure its spectral response as I have done with my other cameras.

Before we begin with the technical stuff, here’s the camera with a Zeiss ZF-IR 50mm f1.4 lens attached.

Handling it, I like the feel of it. It feels well made and nice and solid. It takes compact flash cards, and has an APS-C sized sensor. Speaking of the sensor, this was a quirky design. A 12.3 megapixel ‘Super CCD’ with two different types of pixels – 6.17 million S-pixels and 6.17 million R-pixels for enhanced dynamic range. The degree to which these are used together can be selected by the user from the menu on the back. Although not a huge resolution, the sensor (when used in the S5) was well regarded by photographers especially with regards to the skin tones it was able to produce.

What’s the spectral response like? Here’s the results of my measurements.

First thing to notice is this is barely UV sensitive, and only at the top end of the UV range from about 370-400nm. Below that sensitivity is essentially zero. The sensitivity it has is also mainly in the blue channel. This is different to the Sony, Nikon and Canon cameras I have looked at which have a little more sensitivity in the red and green channels in the UV, and a little more reach at the short wavelength end (for example see here for my Sony A7III data). While the IS pro could be used for imaging in that 370-400nm region, it is not really a UV camera. At the IR end things are better, and it is behaving here as expected for the other cameras which have been converted for multispectral imaging, although with a bit more emphasis on the red channel vs the blue and green channels in the 650nm to 780nm region.

However the really interesting area for the Fuji is how it splits blue, green and red. If you look at the Sony A7III for example, the red channel shows some significant response in the green and even blue wavelengths. Likewise the green channel spreads well into the blue and red channel regions. The Fuji IS Pro has very clean colour channels – while the overlap at their edges they do not spread over wide wavelength ranges, and the red does not bleed across the blue for example. This might account for its ability to capture skin tones accurately. I can also imagine it would be good at blue subjects – the Bluebell test, for example.

This IS Pro cannot be used simply like a normal colour camera given its IR (and the a lesser extent UV) sensitivity. However by using a filter such as BG38 on the front of the lens could be used for normal colour imaging, although the images would probably need to be carefully white balanced to get the best from them. I have also heard that IR photographers like it for the images it produces, so this is something I will try out.

Over the years, camera manufacturers have played around with different sensor technologies and each have their own little quirks and subtleties with regards to how they capture colours. Taking a look at the Fuji IS Pro has certainly been interesting, end despite its relatively low resolution compared with today’s offerings I can imagine it would still earn a spot in the camera bag for certain subjects. As always, if you’ve made it this far I thank you for your time, and if you’d like to know more about any aspect of my work I can be reached here.

It’s no secret that I enjoy imaging diatom microscope slides. They provide delicate and detailed structures and push me to get better at the imaging process. I tend to buy second hand older ones to image. While there are vendors still making slides, I think the older ones tend to have more character, and many are still surprisingly clear and usable for a fraction of the price of newer ones. Today’s post shows some of the slides to arrive recently and discusses how the images were captured (note that images have been reduced in resolution for sharing here, and all have been tidied up a bit to get rid of dust/dirt from the camera sensor).

The first is a slide with a Triceratum kinkerianum diatom mounted in Styrax.

There is only a single example on the slide, so it took some finding as it is very small. This was lit using oblique lighting from below (Olympus Aplanat Achromat condenser) with the condenser set to still produce a bright field image, but one which has obvious side lighting, as can be seen by the shadows in the image. The objective used here was a 20x Olympus Splan NA 0.46, and the camera my Nikon d800 monochrome conversion. White LED light was used. Here’s the slide itself.

Next up is a Navicula firma slide. This has multiple examples of the diatom and was again imaged using oblique light from below, and with the 20x Splan objective.

The diatoms are quite large but show some lovely features and will be well worth coming back to at a later date to look at in more detail. Here’s the slide.

Next is a diatom and spicule strew made by Eduard Thum (one of my favourite slide makers). This was also imaged using oblique lighting from below and with the 20x Splan objective.

This slide is a strew so contains a wide selection of material to image. It seemed to lend itself very well to the oblique lighting, creating a great 3D effect. Here’s the slide.

The final one for today is a Niztschia species slide by A.C. Cole. I spent a bit of time with this one. First a couple of images done with the 20x Splan objective.

Both the images above where taken with the same condenser with light coming from below – an Olympus Aplanat Achromat. In the first image the lower part of the condenser was moved to one side. A small movement to the side produces oblique lighting and nice shadows, however a larger movement (or closing down the iris further) as done here, produces a dark field effect with a black background. This can be very useful for highlighting features in the diatoms. In the second image the lower part of the condenser was returned to the middle, and normal bright field imaging was done. The resultant image is relatively low contrast, and it is harder to see some of the features of the diatoms. How you light the sample has a huge impact on how it appears.

Next I imaged the same slide with a 40x Olympus Dplan Apo UV NA 0.85 objective. This has a coverslip thickness correction collar, and as the mounting of the sample had suffered over the years, the coverslip had tipped up slightly on one side. So I adjusted the correction collar to it’s max thickness (about 0.22mm). I also used 450nm filtered light to try and squeeze a bit more resolution out, and I set up the condenser for slightly oblique illumination. I did a stack of 7 images in Zerene stacker. This was the final image.

The higher resolution from using a higher NA objective and 450nm light is evident. The keen eyed amongst you will also have spotted the diatoms have been rotated by 180 degree from the images with the 20x objective. I took the slide off during imaging and must have put it back on the other way around when I returned it.

Stacking diatom images can be a challenge, as when the focus is changed different parts of the structure not only come in and out of focus, but look different as a result of the way light passes through the diatom structure. This can be seen in the video clip below, where I moved the stage upwards as the video clip was captured.

Here’s the slide.

This is an interesting slide, with the label on the back, so effectively the slide needs to be turned upside down to image (in the photo above the sample and coverslip are on the underside of the slide).

There are always plenty of diatom slides for sale on places like eBay, and while some go for high sums of money, others can be had for a few pounds. Some of the historic microscopists were capable of producing fantastic slides and as such these are a great way to get some beautiful slides to image. With imaging lighting is key, and changing the lighting setup can have a huge effect on the final image. As always if you have made it this far thanks for reading, and if you’d like to know more about my work I can be reached here.

Microscopists have become very inventive when it comes to finding mounting materials to make the subjects they are imaging more obvious. One way of doing this with diatoms is to use something with a high refractive index (RI). While organic based mounts can provide some improvement, some of the inorganic mounts have very large RIs, and as such are able to produce slides with very good contrast. Metal and metal oxide slides are one approach (such as the work by Horace Dall which I have discussed here and that of John Dale, see here). Another approach is the use of arsenic sulfide (known by mineralogists as Realgar). I am currently in the process of writing a couple of research articles, one on metal and metal oxide mounts, and another on Realgar mounts, which I hope to publish later this year. For now though I wanted to share some images from a historic Realgar slide made by Henri Van Heurck.

The slide was likely made sometime around the 1880s, and is a strew of a number of different diatoms mounted in Realgar. I’ll show a picture of the slide later, but for now here are some images from it, captured on my home built Olympus BHB microscope.

The images above were captured using a 40x Olympus Dplan Apo UV objective (NA 0.85) and with oblique illumination from below. The high contrast the mount provides is evident in the images. They were produced using white LED light and the mount itself is a yellow colour, as can be seen below.

Realgar slides are delicate and light sensitive, so effort must be taken to store them in the dark and only expose them to light during imaging. Very often the slides show cracking of the Realgar and this is evident here as well.

It should be mentioned that making the slides is a hazardous process, and should only be attempted by someone familiar with the use of dangerous chemicals. This is probably one of the reasons why the technique is no longer done.

I have a number of Realgar slides in my collection now, and as mentioned above aim to write a paper about these fascinating historical items later this year.

As always thanks for reading, and if you’d like to know more about this or other aspects of my research, I can be reached here.