Free 3D software

 
Fiji (is just ImageJ) https://fiji.sc/
Your best friend for general image manipulation (e.g. cropping, adjusting, and orienting image stacks), editing, and basic analysis. Many many plugins to do almost anything, some quite sophisticated. Open-source, easy multi-language scripting.

Dragonfly https://dragonfly.comet.tech/
Dragonfly is my current top choice as a free replacement for Amira/Avizo. Free for non-commercial research (license on request), well-documented and extensively used, easy to get started with video tutorials. Many of the manual segmentation tools also work directly in the 3D view. Includes a convolutional neural networks Deep Learning based tools for complex segmentation. Windows and Linux only. 

Drishti https://github.com/nci/drishti
Can make amazing visualizations, including key-frames-based animations. Powerful segmentation editor (DrishtiPaint). Steep initial learning curve, but the online tutorials are helpful. Drishti Prayog (https://github.com/nci/drishti#drishti-prayog ) makes spectacular interactive touch-screen presentations.

3D Slicer https://www.slicer.org/
An "open source software platform for medical image informatics, image processing, and three-dimensional visualization...built over two decades through support from the National Institutes of Health and a worldwide developer community.” Well-developed and widely used, with extensive documentation and training tutorials. Reasonably easy to get started. 

SlicerMorph is a Slicer extension with extensive tools for 3D analysis and morphometrics (https://github.com/SlicerMorph).

Tomviz https://tomviz.org/
A "cross platform, open source application for the processing, visualization, and analysis of 3D tomographic data. Here the full pipeline of data processing steps from reconstruction to visualization to analysis of 3D data can be presented, saved, and restored." https://tomviz.readthedocs.io/en/latest/visualization/

ITK-SNAP http://www.itksnap.org/pmwiki/pmwiki.php 
 Designed for 3D image segmentation, "emphasizes interaction and ease of use, with the bulk of the development effort dedicated to the user interface."

Microscopy Image Browser http://mib.helsinki.fi/index.html 
A "high-performance Matlab-based software package for advanced image processing, segmentation and visualization of multi-dimensional (2D-4D) light and electron microscopy datasets."

napari https://napari.org/stable/index.html  
 A "fast, interactive viewer for multi-dimensional images in Python is a fast, interactive, multi-dimensional image viewer, with a vibrant plugin ecosystem that expands its capability to tackle various domain-specific visualization and analysis needs." Open source.

InVesalius is a free package for reconstruction of CT and MRI images. Can import DICOM or Analyze files, export files to the STL, OBJ, and PLY formats. Volume rendering, and manual or semi-automatic image segmentation. Win, Linux, Mac.
https://invesalius.github.io/ 

ImageVis3D is a new volume rendering program developed by the NIH/NIGMS Center for Integrative Biomedical Computing (CIBC). The main design goals of ImageVis3D are: simplicity, scalability, and interactivity.  https://sci.utah.edu/software/

Seg3D is a volume segmentation and processing tool developed by the NIH Center for Integrative Biomedical Computing. It combines a flexible manual segmentation interface with powerful higher-dimensional image processing and segmentation algorithms from the Insight Toolkit. It has a strong emphasis on manual and semi-manual segmentation, where filtered data guides the user in making decent segmentations of the data. The program is mainly intended for, but is not strictly limited to, segmentation of biological/medical data. BioMesh3D (part of SCIRun) can be used to turn segmentations into quality meshes which can be used to simulate biological processes on the segmented images. https://sci.utah.edu/software/

ilastik https://www.ilastik.org/ 
"ilastik is a simple, user-friendly tool for interactive image classification, segmentation and analysis.... has workflows for automated (supervised) pixel- and object-level classification, automated and semi-automated object tracking, semi-automated segmentation and object counting without detection."

CDeep3M is a containerized tool, using deep learning for large-scale image segmentation tasks. It is an open source development and the software is free to use. You can run CDeep3M on your local platforms, on cloud providers, on GPU clusters or with free GPU resources on this website. https://cdeep3m.crbs.ucsd.edu/cdeep3m

MeVisLab https://www.mevislab.de/  
A "powerful, modular framework for image processing research and development with a special focus on medical imaging."

MeshLab http://www.meshlab.net/   
Famous freeware for working with surfaces. 


The Scientific Community Image Forum at https://forum.image.sc/ is a discussion site for software-oriented aspects of scientific imaging, particularly (but not limited to) image analysis, processing, acquisition, storage, and management of digital scientific images.

Fiji plugin to open Xradia microCT files

XRM Reader https://doi.org/10.5281/zenodo.7124262

This is a java plugin for Fiji (ImageJ) that reads an *.xrm, *.txrm, or *.txm image file from Xradia microCT systems (in their proprietary format, a form of OLE container) and opens it as an image stack, along with a text window displaying some metadata parameters. The latest version sets the voxel size for the ImageJ stack using the value from the Xradia file. 

To add the plugin to Fiji, download and extract the .zip folder and just drop the two jar files (poi-3.7.jar and XRM_Reader.jar) one at a time onto the Fiji main window. The program should put them in the right places. Restart Fiji and you should find it under Plugins > XRM Reader. If not, manually place poi-3.7.jar in Fiji.app/jars, and the XRM_Reader.jar in Fiji.app/plugins, then restart Fiji.

This code is publicly available and free for anyone to use. If you use it in a publication, please cite the Zenodo doi above. Current versions can also be found here:

https://ucloud.univie.ac.at/index.php/s/RSJ05Nb9FTViCVK

(I modified the above plugin from this one: https://github.com/mrsutherland/XRM_Reader/releases by mrsutherland, 14 Nov 2017)

Macro to make thumbnails for all your XRM files 

I have also included an ImageJ macro (XRM_files_thumbnails.ijm) that makes a preview image (PNG) and metadata file (.txt) with the same filename base as each unpreviewable *.xrm, *.txrm, or *.txm image file (examples below). This is very useful for quickly looking over your stored scans and seeing what each Xradia file contains.

It is possible to select single files, multiple files, or combinations of files and folders as inputs. You can choose whether the script automatically processes every Xradia-format image in all subdirectories, and whether it will skip files that already have .txt and .png file with the same base file name (i.e. Xradia files already processed by this macro).

This is in the ImageJ macro language and requires the XRM Thumbnails plugin, also included in the same archive. This plugin isn't meant to be run on its own, so I like to put XRM_Thumbnails.jar in the folder Fiji.app/plugins/Utilities, so that it does not appear in the Plugins menu, where it is confusing to see next to the XRM Reader plugin.

I like to put the .ijm macro in either Fiji.app/plugins/Scripts/File or Fiji.app/scripts/File, and then it appears in the File pull-down menu.

The macro makes an XYZ montage for each reconstructed stack and a 0° & 90° mugshot for each projection series (really first and middle projections – for a 360° scan it will be front and back images; my machine doesn't do full rotations). It now includes the date/time from .txrm files. (Fun fact: the container files *.xrm, *.txrm, or *.txm can be extracted to a bunch of hex files using 7‑Zip. I opened some of these with Hex Fiend or HxD to get the format of the date entry.)

If you have questions or comments, feel free to contact me.

Brian Metscher

Vienna, Mar. 2026

brian.metscher[at]univie.ac.at

 

>> Note that Xradia .txm files can also be opened directly in Amira 6.4 and higher (Windows), and also in Drishti (https://github.com/nci/drishti). (Also in ORS Dragonfly Pro, but not in the free Dragonfly version.)

TXM-Wizard by fmeirer, liuyijin can open Xradia files also:

https://sourceforge.net/projects/txm-wizard/

https://pubmed.ncbi.nlm.nih.gov/22338691/

>> Real progress toward a Python solution can be found here: https://pypi.org/project/xrmreader/

It's based on the dxchange Python code, which seems to cover the reading of the xrm container files better than the Java parser: https://github.com/data-exchange/dxchange

 

Sample outputs from the preview macro XRM_files_thumbnails.ijm:

MicroXCT end of support

Please enter any points for discussion, questions, whinges, etc. as comments here. 

This forum will remain active for continuing use by the MicroXCT community. 

Our first Zoom meeting:

Topic: MicroXCT support group
Time: May 6, 2022 04:00 PM Vienna

Email me ( brian.metscher [at] univie.ac.at ) or Lidija Korat ( lidija.korat [at] zag.si ) for the meeting invitation. 

Repositories for 3D image (and other) data

 Please send me corrections, updates, and additions!

Registry of Research Data Repositories http://re3data.org/	Listing of many repositories in all subject areas.
FAIRsharing https://fairsharing.org/	"A curated, informative and educational resource on data and metadata standards, inter-related to databases and data policies." Lists lots of repositories, standards, and knowledge bases.
Zenodo https://zenodo.org/	Hosted by CERN. "All fields of research. All types of research artifacts." Open data for open science. Up to 50GB per record, or by arrangement. My general favorite.
BioImage Archive https://www.ebi.ac.uk/bioimage-archive	A free, publicly available online resource for biological images that are either associated with a peer-reviewed publication, or of value beyond a single experiment.
Figshare     https://figshare.com/	Choice of CC license.
Giga Data Base (GigaDB)   http://gigadb.org/	Uses CC0 (public domain) licensing.
iDigBio https://www.idigbio.org/	The National Resource for Advancing Digitization of Biodiversity Collections (ADBC) funded by NSF. Data and images for millions of biological specimens.
MorphoSource http://morphosource.org/	Duke Univ. Especially for museum specimens. (Boyer et al 2017)
Brain Image Library   https://www.brainimagelibrary.org/	"... national public resource enabling researchers to deposit, analyze, mine, share and interact with large brain image datasets."
Morph·D·Base   https://www.morphdbase.de/	Morphological Description Data Base. Not easy to access.
MorphoBank    http://www.morphobank.org/	"Homology of phenotypes & a database of peer-reviewed morphological matrices"
Digital Fish Library (DFL) http://www.digitalfishlibrary.org/	"...explores the morphological diversity of fishes using magnetic resonance imaging (MRI)"  Not open to external contributions
Digital Morphology (Digimorph) http://digimorph.org/	Univ. Texas microCT library. Lots of fossils and other vertebrate samples, stacks, movies and more.
Dryad      http://datadryad.org/	Uses CC0 (public domain) licensing.
Phenome10k        http://phenome10k.org/	CT and surface scans of biological and palaeontological specimens (skulls).
Harvard Dataverse https://dataverse.harvard.edu/	All kinds of data. Set up your own dataverse collection, up to 1TB.
MorphoMuseuM (M3) https://morphomuseum.com/	Mainly surface models. A peer reviewed, online journal that publishes 3D models of vertebrates, anatomy atlases, and 3D datasets. Also direct submission.
Aves 3D      http://aves3d.org/	Bird bones and skeletons
heidICON https://heidicon.ub.uni-heidelberg.de/search	Image and multimedia database. Die Heidelberger Bilddatenbank, is the "Virtual Slide Collection" in progress of Heidelberg University.
Sammlungen Göttingen https://sammlungen.uni-goettingen.de/	Wissenschaftliche Sammlungen der Georg-August-Universität Göttingen.
Image Data Resource (IDR) https://idr.openmicroscopy.org/about/	A public repository of reference image datasets from published scientific studies. (Williams et al 2017)
Phaidra (UniVie) https://phaidra.univie.ac.at/	Phaidra is the repository for the permanent secure storage of digital assets at the University of Vienna.
GitHub       https://github.com/	The place for open software
New Mexico Decedent Image Database (NMDID) https://nmdid.unm.edu/welcome	provides researchers with access to whole human body computed tomography (CT) scans and a rich body of associated metadata.
FaceBase https://www.facebase.org/	Comprehensive craniofacial data (including 3D imaging datasets) from model organisms (mouse and zebrafish) and humans.
MorphoBrowser http://morphobrowser.biocenter.helsinki.fi/	‘MorphoBrowser’ database and interface is a 3D visualisation and searching tool for mammalian teeth, accessible over the web.
Phenome10K http://phenome10k.org/	A free online repository for 3-D scans of biological and palaeontological specimens.
Genetics of craniofacial shape in Mus https://osf.io/w4wvg/	High-resolution 3D microCT head scans of a mouse panel between C57BL/6J and A/J mouse strains and associated genotype data. Contains mCT scans of ~500 mice heads and associated cranial landmarks.
Digital Morphology Museum of Kyoto University (KUPRI) http://dmm4.pri.kyoto-u.ac.jp/dmm/WebGallery/index.html	DMM provides a large collection of CT and MRI tomography scans of various primates.
The Open Research Scan Archive https://www.penn.museum/sites/orsa/Overview.html	(formerly Penn Cranial CT Database) contains high resolution (sub-millimeter) scans of human and non-human crania from the Penn University Museum and other institutions.
GB3D http://www.3d-fossils.ac.uk/search.cfm	Fossils Online project, aims to develop a single database of the type specimens, held in British collections, of macrofossil species and subspecies found in the UK, including links to photographs and a selection of 3D digital models.

 

Other sources for 3D images:

The Visible Human Project https://www.nlm.nih.gov/research/visible/visible_human.html		Full-body images of a male and female adult
Thingiverse https://www.thingiverse.com/	Loads of cool 3D models for 3D printing
Phaidra (UniVie) https://phaidra.univie.ac.at/	Phaidra is the repository for the permanent secure storage of digital assets at the University of Vienna.

 

A lot of these repositories and many others for various kinds of data and documents are listed at PUBLISSO (itself a repository): 
https://www.publisso.de/en/research-data-management/publishing/publisso-repository-finder/

Also good to know is DataCite: https://datacite.org/ 

PTA colour change, agarose and destaining

On 18/03/2019 21:38, Gonzalez, Brett wrote:

Hi Sarah…

My name is Brett Gonzalez and I am a postdoc at the Smithsonian National Museum of Natural History, previously a Ph.D. student with Katrine Worsaae. I am not sure if you remember, but last year I emailed you regarding some general advices towards integrating CT work into my research. Here at the museum we have a newly installed GE nanoCT and since the technicians are still technically new, I am hoping you could potentially assist once again with my questions.

I work on scale worms and since they soft bodied and fragile, I wanted to integrate alternative methods for scanning aside from just placing drying them or leaving in ethanol or other liquid. Several papers, including some where you have worked on, have used low-melting agarose to imbed the animals prior to scanning. The agarose percentages I have seen range from 0.5%-1.5% with very few other specifics. I have now tried twice, the most recent being with a 0.5% agarose embedded animal, and the entire pre-scan viewing is opaque or nearly. The animal cannot be seen. Can you think of anything in the embedding process that I am doing wrong that would prevent the X-rays from penetrating the agarose and the specimen? The agarose is prepared in 1% TAE buffer mixed with di-water. 

My only thoughts are that when putting the specimen in the agarose, the warm temperatures are causing the PTA to come out of the animal and disperse among the agarose. Could this be the case or is the agarose maybe wrong brand or age or something else? I would really like to use agarose so that specimens without chaetae don’t move during the long scans.

The only other question I have is that I had a specimen turn from ivory color (in ethanol) to blue/brown after a scan, but only in the portion being scanned. The specimen eventually turned back to the original ivory color upon upon placing in new ethanol. Have you seen this before and is this somewhat normal in liquid mounted specimens or is it an energy issue when running the scan? I have not been able to see any literature or mention of this either.

Sorry for such random questions but would really like to keep going with this technique in order to investigate muscular innervations in swimming scale worms and other annelids. Any help on the issue is greatly appreciated.

Thank you for your time.

Cheers,

Brett

--
Brett C. Gonzalez, PhD.
Postdoctoral Fellow

Smithsonian Institution
National Museum of Natural History

Hi Sarah & Brett,

Random questions are sometimes the best ones. Starting with the colour change: I often see PTA-stained regions change to blue-green-brown under X-rays, then revert after some time to the original whitish. I have assumed that this is caused by an oxidative change in the tungsten. I have not seen any effect on the scan quality. 

PTA can leach out into the agarose (where it can also turn green), but usually does not if the sample has been rinsed after staining. PTA binds strongly to proteins under acidic conditions, and this seems to be permanent if the pH stay low. PTA staining can be removed with a slightly alkaline buffer, or even with PBS, as the PTA polyacid molecule dissociates into smaller tungstate species at higher pH. 

Which brings us to the agarose: I always use agarose in water (usually 0.5%-1.0%), unless I want to keep the tissues in a buffer for some reason. I have seen PTA staining fade in agarose in PBS; iodine staining is OK. I use low-melting temperature agarose, so that it can cool to below 37°C before immersing the specimen (it gels around 33-35°). 

So I imagine the problem is the TAE, which has a pH of 8 or so as I recall, plus a chelating agent (EDTA). My guess is that your agarose effectively dissociated and dispersed the PTA more or less uniformly. Aqueous agarose might solve the problem. 

Another fun trick for embedding fragile samples is to use CyGel, a thermoreversible gel which solidifies at room temperature and melts in the fridge. This avoids the problem of removing agarose from delicate specimens: just wash in cold buffer or ethanol. However, it's really expensive. (http://www.biostatus.com/CyGel/)  

Another way to remove agarose is to drop 6M potassium iodide over the specimen while brushing off the agarose as it dissolves (this helped with a centipede holotype - lots of breakable legs... Akkari et al. 2018) 

Hope this helps. With your permission, I will also post you message and this reply to my blog (http://microtomography.blogspot.com/). 

Best,

Brian 

Akkari N, Ganske A-S, Komerički A, Metscher B. (2018). New avatars for Myriapods: Complete 3D morphology of type specimens transcends conventional species description (Myriapoda, Chilopoda). PLoS ONE 13(7): e0200158. 

https://doi.org/10.1371/journal.pone.0200158

Destaining: PTA

Actually, PTA staining can (mostly) be removed after scanning. I had a student (Hannah Schmidbaur) do some studies on this, and I have done some more experiments. The short answer is to wash out the PTA with a slightly alkaline buffer, e.g. PBS with 0.01M NaOH (figure below; third row). The destaining takes about as long as the staining did (I think), and you should make sure there is enough destaining buffer (at least 10X the volume of the tissue). And of course the only way you can see if the PTA is gone is using X-ray imaging.

Hannah's presentation from the Bruker MicroCT user meeting 2015:
https://www.bruker.com/fileadmin/user_upload/8-PDF-Docs/PreclinicalImaging/microCT/2015/uCT2015-21.pdf

The 2018 Euro Evo Devo meeting!

http://www.evodevo2018.eu/ehome/index.php?eventid=232389

Zeiss/Xradia users meeting!

It's going on now!
Check out the tweets at
https://twitter.com/zeiss_micro

My ToScA talk

Here is a pdf of the slides from my ToScA presentation.

https://www.dropbox.com/s/nxql9qu5tvcky9c/Metscher_ToScA-2017c.pdf?dl=0

and the video collage is now on YouTube at

https://youtu.be/85wocaNifP4

Plant CT

Several people asked about published work on contrast-enhanced microCT for plant specimens. Here is an article from a group in Vienna, and a couple of more recent ones, as well as a couple of pictures I made using vascular contrast agents on wild and domestic cereal plants (details on request). 

Staedler YM, Masson D, Schonenberger J. (2013).
Plant Tissues in 3D via X-Ray Tomography: Simple Contrasting Methods Allow High Resolution Imaging. PLoS ONE 8(9): e75295.
http://www.ncbi.nlm.nih.gov/pubmed/24086499

Saoirse R. Tracy, José Fernández Gómez, Craig J. Sturrock, Zoe A. Wilson and Alison C. Ferguson. 2017. Non-destructive determination of floral staging in cereals using X-ray micro computed tomography (µCT). Plant Methods 13:9
https://doi.org/10.1186/s13007-017-0162-x

David Rousseau†, Thomas Widiez†, Sylvaine Di Tommaso, Hugo Rositi, Jerome Adrien, Eric Maire, Max Langer, Cécile Olivier, Françoise Peyrin and Peter Rogowsky. 2015.
Fast virtual histology using X-ray in-line phase tomography: application to the 3D anatomy of maize developing seeds. Plant Methods 11:55
https://doi.org/10.1186/s13007-015-0098-y

ToScA Workshop (Life Sciences): Microtomography for life sciences research

This workshop is offered twice on Wed. 6 Sept. and will introduce some methods for enhancing x-ray contrast in non-mineralised tissues and techniques for mounting biological samples for microCT imaging, especially embryos and other soft tissues, insects and other invertebrate specimens, and any samples of particular interest to the participants.

We will begin with some principles of x-ray imaging, discuss various types of samples and applications, and then work with your own interesting specimens.

You are invited and encouraged to bring your own samples to the workshop! Fixation and staining can take days, so you might want to prepare your samples ahead of time.

Please also bring pertinent questions, including issues concerning image analysis, publishing, and archiving!

Details of stains etc. are given in the accompanying blog entry. And you may of course email me with questions: brian.metscher@unvie.ac.at

1) General advice about sample preparation for microCT imaging  

Fixation:

The best fixation for microCT is the one that preserves the features you need to see. I have had good results with most of the common fixation procedures, but the properties and effects of the fixative must be taken into account for contrast staining. Usually most relevant are shrinkage, decalcification, removal of lipids or carbohydrates, and protein condensation or precipitation.

Preservation:

Samples can usually be stained effectively after storage in 70% ethanol or in formalin. If you want to use an aqueous stain, transfer the sample back to an aqueous solution; likewise for alcoholic stains. Dry samples are easy (they're dry).

2) Some tips for preparing different sample types

Insects & other arthropods:

I have had good results from alcohol-fixed crawlies by re-fixing them in alcoholic Bouin's (1:1 Bouin's:ethanol/IMS) for a few hours or longer and then dehydrating to ethanol (absolute but not anhydrous, i.e. 96-100%), followed by staining in I2E (below).

Others have made excellent images of critical-point dried insects (better than HMDS; Sombke et al. 2015).

Dry insects can be scanned easily, but the internal anatomy is dodgy. Chitinous structures usually come out beautiful. Pins can be a challenge, but scans can be done with pinned insects.    

Embryos and other squishy samples:  

My favourite fixative for soft stuff is 4F1G (4% formaldehyde and 1% glutaraldehyde in phosphate buffer, or other appropriate buffer, like PBS, or whatever your samples are happy in). The actual concentrations of the two fixatives are not crucial; I usually just add glutaraldehyde to 10% NBF and I'm done. The glut must be high-grade and fresh - otherwise it polymerises and becomes less effective.

PTA and iodine both give good results. Shrinkage can be a problem...   

Little fish and anything that sort of resembles a little fish:  

Mostly the same as embryos, but pay attention to whether you want to see e.g. brain, visceral organs, muscles, bones, teeth, etc. Staining with PMA can allow nice distinction of mineralised tissues and soft tissues; PTA in methanol has given good images of hearing structures (Schulz-Mirbach et al. 2013a, b).

Samples in methanol:

Samples preserved for nucleic acids work are often stored in methanol, typically after aldehyde fixation. These can be stained easily and effectively with  PTA in absolution methanol (van Soldt et al. 2015).

3) Mounting samples for microCT

The sample must be immobilised and held on a vertical rotation axis for the duration of the scan. I often use 0.5-1.0% agarose to embed (not infiltrate) samples in narrow plastic tubes or micropipette tips. Other schemes can work also, and may work better for some kinds of objects: bits of sponge and soda straws have helped on occasion.  Other friends of sample mounting include Legos, Parafilm, UHU Patafix (Blu Tack), and a hot-glue gun.

Contrast staining for soft tissues

Probably the most versatile microCT contrast stain for soft tissue is diluted Lugol’s solution (aqueous iodine: Metscher 2009a,b; Degenhardt et al. 2010; Gignac et al. 2016). There is more than one formulation of "Lugol's;" the “IKI” solution I published in 2009 is actually 20% Lugol’s, and my “10% IKI” is 2% Lugol’s. Always be clear about the actual concentrations of iodine and iodide you are using (don't rely upon the term "Lugol's" to specify the composition unambiguously).

IKI  (Metscher 2009a, b)

2% (w/v) potassium iodide (KI) + 1% iodine (I2) aqueous solution. e.g. dissolve 2.0 g KI in 100ml distilled water, and then add 1.0 g I2 (elemental iodine, "iodine metal"). The iodide dissolves easily, and elemental iodine only dissolves in water along with iodide. Keeps indefinitely at room temperature as far as I know.

Fix fish, embryos, or whatever in your favourite aqueous fixative. Rinse samples in water or buffer (they can go directly to stain solution). Stain overnight or longer. Change the solution when it looks thinner. Wash in water.  Some iodine will still leach out; this is usually not a problem. Can be scanned in water or buffer, or mounted in agarose.

Note that some plastics (notably pieces of sponge used for bracing the specimen) will absorb some iodine, but not usually enough to be a problem for scanning.

For specimens already stored or fixed in alcohol, an alcoholic iodine solution works well. This stain is especially good for arthropods.

I2E, I2M  (Metscher 2009a, b)

1% (w/v) iodine metal (I2) dissolved in 100%  ethanol (I2E) or methanol (I2M) I2 dissolves readily in alcohol.

Take samples to 100% alcohol. Stain overnight or days or even weeks for larger specimens. Rinse in alcohol. Scan in alcohol.

Iodine does not seem to work in 70% alcohol, only 0 or 100. Anyone know why? In fact, storage in 70% ethanol usually removes most of the iodine staining.

PTA is my other favourite stain. Used in some standard histological staining methods, PTA is known to bind proteins and is commonly used in electron microscopy. The chemistry of phosphomolybdic acid appears (to me anyway) to be mostly similar, and PMA gives good staining also. And it's green. I have had some preference for PMA as a counterstain in dual-energy imaging for differentiating materials in a sample (see Handschuh et al. 2017).

PTA, PMA  (Metscher 2009a, b; Metscher 2011)

Works well on tissues fixed in formalin, 4F1G, glyoxal, Bouin’s, or alcoholic Bouin’s.

Make a 1% (w/v) phosphotungstic acid solution in distilled water.  Mix 30 ml 1% PTA solution + 70 ml absolute ethanol to make 0.3% PTA in 70% ethanol. Keeps indefinitely. The pH must be on the acidic side: the above solution comes out around 2.9. The affinity of PTA for different proteins is pH-dependent (Nemetschek 1979, Scott 1971, Silverman 1969), but I have not tested this systematically for use in microCT imaging.

Take samples to 70% ethanol. Stain overnight or longer. Change periodically for larger samples. Change to 70% ethanol. Scan samples in 70% - 100% ethanol. Store in 70-100% ethanol.

Gives high general radiopacity and excellent contrast among tissues and structures in vertebrate embryos and soft-bodied invertebrates. Penetration is vaguely 1-2 mm per day, so overnight is usually sufficient for samples no thicker than about 3-4mm.   Staining is stable for months if not years. Samples can be embedded and sectioned for histology afterward.

Osmium tetroxide (Johnson et al. 2006; Metscher 2009a, b)

Standard EM post-fixation, binds abundantly to lipids.

Same as routine EM processing. Osmium-stained samples can be scanned in resin blocks, with some loss of contrast.

Gives very high radiopacity and no better tissue contrast than PTA. Osmium tetroxide is volatile and toxic, but your local EM lab is probably already set up to work with it.