The comparison of certain aspects of both techniques can be summarized as follows. The differences in the information and contrast at the CT and NT images result from the differences in the attenuation properties and detection techniques of X-rays and neutrons. Additionally, image quality and contrast in both techniques depends strongly on calibration of the instruments and their resolution. For optimal comparison, the resulting 3-D data volumes have to be co-registered. In the example of the sauropod vertebrae, NT analyses are strongly influenced by the resins used for preservation and preparation of the scanned objects. The special combination of materials like fossil bone (apatite) and sediment matrix (silica) and a high amount of glue and polyester resin significantly decrease the quality of the images. In contrast, information about the distribution and content of glue within fractures and cavities of the bone are quickly available without recalibration of the instruments. Furthermore, glue-filled pneumatic canals that were not visible in the CT images were visible in the NT images. The medical CT scans of the sauropod vertebrae are only minimally affected by preservation and preparation of the scanned objects, but information like glue distribution was diminished. Comparing images of both techniques therefore helps to collect information about the scanned objects.

The time effort necessary for the CT scans depends strongly on the cross-section of the objects, the type of CT scanner and the available instruments and software. A CT scan together with reconstruction from the raw data can take between 5 and 60 minutes. In contrast, NT scans require a larger amount of time, because many radiographs are taken with the camera and the object has to be rotated on a table. Together with the reconstruction from the raw data, between 30 minutes and 5 hours might be required for one complete NT scan. The costs of a CT and an NT scan both depend on the object size and the scanner, and for this study, ranged from 500 and 1000 Euro. It should also be mentioned that medical CT is broadly available in many hospitals, whereas NT is bound to the presence of a neutron source, and therefore available only in few places.

The decision of which technique to use will therefore be dictated by the research questions, the preservation and material properties of the object. From the comparisons undertaken in this work, it seems that in most cases, X-ray CT will yield the best image results. Special cases, in which NT would be an adequate or even better technique, could be:

• to investigate objects containing metal inclusions;
• to investigate unprepared fossil bones to analyze the distribution of sediment in their internal cavities, including specimens containing lithologies other than the marls described here; and
• if there are fossils that have to be investigated in a forensic or historical context, for example to determine the nature of historical preparation of a museum specimen (e.g., identifying the parts that have been glued or modelled with resin materials).

At the moment, NT has not been extensively utilized in vertebrate palaeontology. To fully leverage this technique, future research might focus on identifying what types of sediment-bone taphonomic modes yield optimal contrasts for determining fossil vertebrate anatomy.