Technical Abstract: In genbanking, seed quality is essential for maintaining the seed viability for extended periods of time. High pre-storage germination is a good indicator of the seeds expected post-storage viability in species that can endure long-term storage. Knowledge of the seeds physical characteristics may offer an insight on germination levels. An X-ray technique allows for a non-destructive evaluation of seed fill, mechanical damage and potential insect injuries. Seeds of new crops germplasm often have a low germination, especially non-domesticated accessions. The objective of the study was to assess physical appearance of seeds for selected accessions from four different new crops germplasm collections using an X-ray technique. A total of 29 accessions from the National Plant Germplasm System collections of Cuphea, Limnanthes, Opuntia (6, 6 and 7 accessions respectively) and Parthenium argentatum (7 accessions plus 3 cultivars from industry) were evaluated. From each accession, four replications of 100 randomly selected seeds were viewed in a XPERT (Kubtec) X-ray cabinet. The seeds were separated into five different categories, as ‘full’, ‘>50% filled’, ‘<49% filled’, ‘damaged’ and ‘empty’. Seeds in each category were germinated according to AOSA germination standards. All non-germinated seeds were exposed to a TZ test that indicated the viability status of the seeds embryo. Data were evaluated by ANOVA (Tukey-Kramer HSD, a 0.05; JMP 7.0, SAS). In the Cuphea accessions the fraction of ‘full’ seeds was between 69.8 to 99.3 % and the ‘empty’ fraction varied from 0 to 25.0 %, and was strongly correlated with the seed germination level. The same trend was observed for the Limnanthes and Parthenium accessions. In the seed group of ‘>50% filled’, the germination was from a few percent to >60%, whereas it was 0 to about 20% in the group of ‘<49% filled’, a few percent in the ‘damaged’ and 0 in the ‘empty’ seed groups. The X-ray non-destructive seed evaluation is a good indicator of the seed germination potential and may be considered as a complementary tool for evaluation of seed sample quality and provide some elucidation on reasons for low germination. This technique may be a good tool in evaluation of seed lot quality.