Leg muscles
The glass tardigrade now has its leg muscles, about fifty per side, each a single cell, and they move when it walks. We named every one and fitted it into the X-ray animal from Vladimir Gross and Georg Mayer's confocal tracings, which also show which muscles in different legs are the same muscle repeated.
The data
We worked from the four leg stacks and tracings of Gross & Mayer 2019, released on Dryad under CC0: muscle filaments (F-actin) stained with phalloidin, whole animals imaged with an Airyscan confocal microscope, and every leg muscle of one side traced in Amira. Their scheme names the muscles by leg and number (iM1 is muscle 1 of leg I, iiM1 of leg II) and groups them into 14 serially homologous sets, M1 to M14.
Naming without names
We named every muscle from its colour. As with the X-ray, the tracing files hold numbers, not names, but the colour of each label matches the colour code of the paper's Figure 2, where each homologue M1 to M14 has its own colour. The set of colours in each leg matches the paper's Table 1 exactly: 14 muscles in leg I, 12 in II, 11 in III, 10 in IV, with a few muscles split across two labels. So every label could be named, and every muscle in the viewer carries its proper name, such as iiM5.
A depth that didn't add up
We measured the stacks' real depth from the muscles themselves. The files claim 0.0425 µm per pixel in every direction: right across each image, but it would make each stack only 8 µm deep, too shallow for a leg. Muscle strands are round in cross-section, so we found the slice step that makes them round: about 0.125 µm, consistent across all four legs (0.10 to 0.15). It is an estimate and is marked as one.
Fitting another animal's muscles
We fitted each leg's muscles into the X-ray animal by their attachments: the lowest point (at the claw) on that leg's claw gland, and the top (on the back) on the body wall above it, at one shared scale for all legs. The confocal stacks are squashed in depth under the coverslip, so we laid each muscle's depth inward from the X-ray animal's own body wall. Because the muscles come from a different animal, some strands ran through this one's storage cells, gut or ovary; we nudged them around, moving each piece of strand as a unit so it bends rather than splits.
In the viewer
Turn the muscles on and off under "Body wall and muscles", hover to see each name, and click one to see its serial homologue in every leg. When the animal walks, the muscles bend with the legs.
What's uncertain
- Positions are approximate: two landmarks per leg from a different animal, not a registration. The shapes and names are Gross & Mayer's; the placement is ours.
- The slice step is estimated.
- Muscles don't yet contract, and body-wall muscles (not in this dataset) are missing.
Get it
The placed muscle meshes (GLB, STL) are on Resources. For the original stacks and tracings, go to Gross & Mayer's Dryad record.
How this is built
- Data
- Gross & Mayer 2019's four leg stacks and their Amira tracings (phalloidin-stained muscle, Airyscan confocal), CC0 on Dryad, with the homologue colours of their Figure 2 and the counts in Table 1. The body they sit in is Gross et al. 2019's nano-CT (CC BY 4.0).
- Built with
- Our Python script
build_muscles.py. It names each label by its palette colour, estimates the slice step from round strands, meshes each muscle with marching cubes, and places each leg by two landmarks at one shared scale. It lays depth inward from the body wall and nudges strands out of the organs; the right side is a mirror. three.js draws the 100 meshes and bends them with the walking legs. - Checked
- The colours in each leg match Table 1 exactly: 14, 12, 11 and 10 muscles. The slice step comes out at 0.10–0.15 µm in all four legs. The placement is not yet tested against data.
- Limits
- Two landmarks per leg from a different animal, not a registration. The slice step is our estimate, not a measured value. No contraction yet, and no body-wall muscles.
- More
- How the whole project is built. The figures below come from the figure library, each with its own credit and licence.
Muscles, from the papers
