A. Transparent soil is prepared for imaging by saturation with RI-matched solution to achieve transparency (left, fully saturated; right, larger pores are drained). Scale bar = 2.5 cm. B. Optimal RI of nutrient solution for RI matching with Nafion using projected straight line images deformed by the substrate. Curve shows gaussian non-linear regression (R² = 0.38). C. Water retention in transparent soil with 3 different Nafion particle sizes compared to vermiculite [37] and sand [16]. Error bars show standard error. D–E. Comparison of plant growth in transparent soil and other substrates. D. Excavated plants with representative root systems from each substrate type after 2 weeks of growth. Scale bar represents 1 cm. E. Quantification of root system parameters in different substrates. Plants grown in transparent soil had lateral root lengths and densities more similar to plants grown in soil than plants grown in phytagel.

A. Projection image from OPT scan of Nicotiana benthamiana roots. Scale bar represents 1 mm. B. Root tracking algorithm is applied to the reconstructed data to segment and dilate (to improve visibility) the root (green) from the small air bubbles (blue). Scale bar represents 1 mm. C–F. Snapshots of volume renderings of confocal scans. C. Arabidopsis thaliana roots expressing GFP in plasma membranes (grey) in transparent soil with sulphorhdamine-B-dyed particles (orange) where scale bars represents 300 µm. Inset shows root skeletonisation and edge detection applied to scan C to detect roots and particles. D. GFP labelled Escherichia coli 0157:H7 cells and colonies on surface of Latuca sativa (lettuce) root with prominent root hairs. Scale bar represents 30 µm. E. Box shows enlarged region of lettuce root in D with Nafion particles visible in orange. Scale bar represents 100 µm. F. Arabidopsis thaliana root tip with nuclear RFP expression linked to auxin reporter [41]. Inset with box shows enlarged region. Scale bar represents 54 µm.