All specimens where caught by net sweeping. The whole net contents is killed in fumes of ethyl acetate and transferred to airtight glass jars and stored in a deep freezer at -22°C for later examination. Single specimens were put immediately to a deep freezer. After unfreezing the specimens look like freshly caught.

I usually first make a photo of the whole body of the fresh specimen, preferably in the pinned state. Proper orientation of pinned specimens is done with the "entoball" from stefan.ober@entoball.com. To adjust the brightness of the background I found it sufficient to use a black, grey and white card, depending on the brightness of the specimen. Whole body images ususally are focus stacks consisting of ~10 exposures, photos of details consist of fewer exposures or are not stacked at all. As I found it more convenient to remove the terminalia and other parts in the unpinned state, I use folded pieces of card (black, grey, white) for orientation. Small gnats and midges which dry and distort too quickly, and material intended for barcoding is sometimes shown in alcohol, where dusting patterns are sometimes impossible to show.

Clearing of terminalia, legs, ... is done by leaving the parts overnight in 10% KOH, then they are tranferred to pure water (~10 min), to glacial acetic acid (~10 min) and finally to glycerol. When the streaks visible after moving the parts have almost disappeared (~ 1h), they are transferred to a slide with a rounded excavation into a drop of glycerol about 1/3 the diameter of the excavation. I wait again for ~10 min and then start to dissect and to orient the part under the binocular and make a photo under the compound microscope. To be able to work without a coverslip, I use a 20x and 40x Objective from an inverse microscope with a long working distance (and lower resolution). Parts too large to be placed under the compound microscope, I'm shooting with the binocular in a watch glass with incident light.

Clearing the whole specimen, after having documented colours and dusting, is sometimes useful. The wings should be kept in alcohol, because they are damaged by KOH and before transferring the specimen to acetic acid head, legs and terminalia should be separated to avoid breaking by the quickly forming CO2 bubbles. To remove gas bubbles the parts are placed in isopropyl alcohol and then back to glycerol.

For the best possible resolution the dissected parts are embedded in Malinol (artificial Canada Balsam). Before they are placed for ~1h in 2-propanol (isopropyl alcohol) washing out all hydrophilic substances.  Very small parts I transfer with a glass pipette from the alcohol to the slide, add an appropriate drop of mountant, and, after a few minutes, orient the part and add the coverslip. Care must be taken that the alcohol is not evaporated before mountant is added.

After a few days the viscosity of Malinol is high enough to orient the part by moving the coverslip and pressing it down at one side. This way one can show arbitrary orientations provided enough mountant has been added. Using too much mountant degrades the quality of the photos, because high resolution objectives are calculated for thin samples/coverslips. (The refraction index of Malinol is very similar to the index of the coverslip.)

Most of the image processing and creation of html stuff has been automatized by some simple scripts. The additional software used is Gimp, two components from the Hugin suite (align_image_stack and enfuse), and exiftool.

For more details see Basic Introduction and Working at Higher Resolution.