Micro-menagerie

[Guest rubidium]

Advertisement (gone after registration)

Diatoms are a major group of single-celled algae, that are encased within unique cell walls made of silica. These walls show a wide diversity in form, some quite beautiful and ornate. Usually the walls - referred to as "frustules" or "tests" - consist of two symmetrical sides with a split between them, one of which overlaps the other like the lid of a box or petri dish. (hence the group name.) The term "tests" arises because the patterning of the cell walls is generally fine enough to enable assessment of the resolution limits of an optical system used to observe them.

 

The first full-frame photograph below was made with the R9/DMR and a Leitz 6.3x fluorite objective lens. The full image width and height are about 0.86 mm and 0.57 mm, respectively.

 

The second full-frame photograph explores the fine detail of a single cell, and is made with the R9/DMR and a Leitz 40x apochromatic objective. The full image width and height are about 135 microns and 90 microns, respectively. The clearly resolved "dots" in the silica walls of the cell are spaced about 0.57 microns apart. That's about 1 wavelength of light in the middle of the visible spectrum! Very often we hear of Leica APO lenses as being nearly "diffraction limited" in performance, but how this performance is seldom exploitable (or irrelevant) in "normal" photographic situations. Here's a chance to see such performance at play. The fact that the "dots" can be resolved is indicative of a resolution of at least 1750 line-pairs/millimeter. The diffraction limit is about 2700 lp/mm.

 

Cheers,

Jim

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[larry]

Jim,

 

These images are mesmerizing. I find it fascinating how unfamiliar reality, such as we see in photo-microscopy, carves out its own aesthetic niche.

 

Thanks,

 

Larry

[wbesz]

Thanks, very interesting to see the results of challenging imagery.

[martinop]

Interesting to see the results of your efforts. #1 is my personal favorite.

[billh]

The images are beautiful, and I very much enjoyed the explanation accompanying them. Are these very uniform “dots” the magnifyied silica structure?

 

I would love to see (and hear about) more of your work.

[Guest rubidium]

Yes Bill, the "dots" are very fine punctures (pores) in the cell walls that are made of silica. Since the specimen is being illuminated from below, the pores appear a bright spots.

 

The pores actually have a 3-dimensional funnel-like structure to them, as opposed to simply being holes in a flat plate. However, the limited depth-of-field under these circumstances does not allow this feature to be resolved in a single frame. For the image on the right, I estimate the depth-of-field to be approximately 2 microns (!!!), so any structure of a pore that extends beyond that is blurred out. It is possible, however, to "stitch" together multiple frames shot at various focusing depths to create a composite image that exhibits greater depth-of-field. This is called "Z-stacking." One typically steps through the desired focusing range in increments that are approximately one-half of the longitudinal resolution (that would be about 1 micron intervals in this case.) This is analogous to assembling a panorama in traditional landscape photography.

 

Thanks for looking,

Jim

[Guest rubidium]

Advertisement (gone after registration)

Correction to my last post: The pores are actually the blue spots and not the white. The specimen was not directly illuminated from below as I had originally indicated, but rather illuminated from below using an annulus of light having a diameter that exceeds the frame dimension; thus the dark background. This is done to enhance contrast.

 

Sorry for the slip-up,

Jim

[dkCambridgeshire]

Superb ... is this an example of dark field transillumination?

 

dunk

[Guest rubidium]

Dunk: Yes indeed, the above shots were made with darkfield transillumination. Specifically, the voltage on a 100w tungsten source is set to a calibrated value to give a known color temperature (3200K in this case). Later, this calibration can be used to set the white balance. The light is collimated through a set of apertures and passed through a condenser that is designed to form a hollow cone of light that illuminates the specimen from below. The light at the apex of the cone is focused at the plane of the specimen, and as this light moves past the specimen plane it spreads again into a hollow cone. The objective lens sits in the dark hollow of this cone above the specimen, so that the light travels around and past the objective lens, and no rays enter it. The entire field therefore appears dark when there is no sample on the microscope stage (thus the name darkfield illumination.) Alternatively when a sample is on the stage, the light at the apex of the cone strikes it, and the image is made only by those rays scattered by the sample and captured by the objective lens. The image thus appears bright against the otherwise dark background. This situation can be compared to the high contrast appearance of dust particles in a dark room illuminated by strong shafts of light coming in through a side window. The dust particles are very small, but are easily seen when they scatter the light rays.

 

 

An illustration of "Z-stacking": Given the abysmal depths-of-field achievable in these situations, it is possible to capture multiple images at different planes of focus, and integrate them to form a composite image having greater depth-of-field. The full-frame image below is an example of combining 14 frames from the DMR, taken at progressively altered focus planes spaced 0.5 micron apart. The subject is another diatom, and the results of the Z-stacking allow visualization of the structure of the peripheral and internal ridges of silica expskeleton, and also portrays the central pore (blue dots) in each exoskeleton fustule (larger white bead-like structures). R9/DMR, Leitz 40x apochromatic objective lens; full image width and height about 135 microns and 90 microns, respectively. Z-stacking is a tough business, and I am essentially a novice at it.

 

Cheers,

Jim

[dkCambridgeshire]

Jim,I use a similar technique but on a much larger scale for photographing translucent and semi translucent shells and shell sections... using snooted tungsten spots for the backlighting and a Schott fibre optic projector plus another tunsten spot and reflector for frontal fill in. But I have not yet got round to buying a scanner ... when I do will be posting some pictures. Subject support is as challenging as the lighting. Alfred Blaker's books and William White's books have been a big help ... the latter illustrates the Dynaphot light scanning method for 'stacking' images ... no doubt it is a very powerful tool now when combined with digital stitching to obtain additional depth of field ... but I have not tried such microphotography yet.

 

dunk

[billh]

Yes Bill, the "dots" are very fine punctures (pores) in the cell walls that are made of silica. Since the specimen is being illuminated from below, the pores appear a bright spots.

 

The pores actually have a 3-dimensional funnel-like structure to them, as opposed to simply being holes in a flat plate. However, the limited depth-of-field under these circumstances does not allow this feature to be resolved in a single frame. For the image on the right, I estimate the depth-of-field to be approximately 2 microns (!!!), so any structure of a pore that extends beyond that is blurred out. It is possible, however, to "stitch" together multiple frames shot at various focusing depths to create a composite image that exhibits greater depth-of-field. This is called "Z-stacking." One typically steps through the desired focusing range in increments that are approximately one-half of the longitudinal resolution (that would be about 1 micron intervals in this case.) This is analogous to assembling a panorama in traditional landscape photography.

 

 

Jim,

 

How do you adjust the focus in one micron increments - visually, or is there a more precise, reliable method?

 

Some years ago we were using an SEM. I remember being fascinated by the views!

 

I hope you’ll continue to share these images with us.

[fotografr]

The science invloved with these images is not within my realm of knowledge but just taken as visually interesting images, these are fantastic. I feel the same about many of the images taken by the Hubble telescope (unfortunately not Leica lenses) of activity in other galaxies that occurred billions of years ago.

[Guest rubidium]

Bill,

Focusing is accomplished as follows. On the Leitz Orthoplan (circa 1965) microscope that I use, the all of the optical components remain fixed while the specimen is affixed to a stage that can be translated in 3 dimensions (x,y,z) with an arrangement of rack and pinion gears having essentially zero-backlash. The z translation (coinciding with focusing) is unique in that it has a "coarse" adjustment knob along with a coaxial "fine" adjustment knob coupled to the "coarse" via a reducing planetary gear system. One full rotation of the "fine" knob corresponds to 0.1 mm of specimen travel along the optical axis. The "fine" knob has a large barrel that is graduated with 100 marks throughout its full circumference, each of which then corresponds to 1 micron of specimen travel. Since this control has essentially "no play", one can confidently interpolate still further about one-fourth of a graduation interval.

 

The microscope head has 3 viewing ports: 2 for the occulars (eye pieces) and 1 for a camera tube. One of the occulars is fixed and the other has a diopter correction that you adjust for your particular eyesight to that the image is parfocal in both eyes. Analogous to that adjustment, the length of the camera tube is machined to be "about the right length", but has a short-travel fine-threaded length adjustment feature to enable the camera to be made parfocal with the occulars. The output port of camera tube is fitted with an R-style bayonnet, which in my case attaches to the R9/DMR. By taking a series of test shots one fine tunes the camera tube length so that what you see in the eyepieces is what you get at the DMR sensor plane.

 

Despite the robustness of this instrument, vibration coupled with long exposure times is a recipe for failure. However, I have been successful making up to 2-second exposures if I first lock up the mirror on the R9 and then use the internal timer to fire the shutter hands-free from the camera. Exposures about 1/24-second and shorter are a piece of cake without having to take these measures. The aperture priority modes of the R9 work like a charm for single images, but when "Z-stacking" multiple images it is necessary to shoot the series in manual mode with a fixed shutter speed (just like you need to do with a landscape panorama to get the images to match properly).

 

Cheers,

Jim

[leicamann]

Very interesting notes, topic and images.. man oh man do I miss my microscope..you've reminded me how much I used to enjoy this type of stuff...maybe I should get back into it?

 

 

Thanks you for sharing

 

Regards, Leicamann

[Guest rubidium]

John, you're starting to sound like I did this past spring. Somehow I began evoking fond memories of having a "student" microscope as a kid (I think it was part of my Gilbert Chemistry Set.) Then I began coupling these thoughts to my enjoyment of photography and my 25+ years of being a "Leica addict." What emerged was a desire to have one of those "serious German research-grade instruments." A modern Leica research-grade microscope costs as much as a new Porsche - and I wasn't THAT serious. However, after some further investigation, I discovered it was possible to get something that was around 40 years old and lacked the modern appurtenances of the digital age but otherwise was optically and mechanically a top-class instrument - for several thousand dollars. Thus I was led to the Leitz Orthoplan from 1965.

 

The challenge was finding one that had been cared for properly. Fortunately, there is a fairly significant level of trade among these items, and a healthy fraction of these are from laboratories or semiconductor cleanrooms where they were handled by professionals who had respect for them as precision instruments. Occasionally, you'll see some frames that look as if they were dug out from under a pile of rusted engine blocks in a salvage yard, or some optics that look as if they were cleaned with sandpaper. But with care and patience, one can get both entire instruments and parts in like new condition for fair prices. Case in point: About a month ago, I obtained a Leitz APO objective lens that had originally listed for $3300US for a little over $200US on eBay. It was in mint condition to the point that I could see no evidence that it had ever been used.

 

Another interesting thing is that these instruments are relatively timeless. So long as they haven't been abused, then time doesn't take very much of a toll on them. After 40 years, the controls on the Orthoplan are as precise and buttery smooth as the focusing mount on my newest Leica R lens. I guess the fact that nothing moves without literally hundreds of ball bearings intervening, and that everything is precision-machined metal has a lot to do with that just as it does in an old Leica rangefinder that's as old or older. Plus it was essentially trivial to couple my R9 to something 40 years its senior with a $25 bayonnet fitting still made by Leica.

 

Cheers,

Jim

[lambroving]

Really neat! I haven't a clue what you are talking about but I enjoy these images, textures and colors.