Water relations of Robinia pseudoacacia L.: do vessels cavitate and refill diurnally or are R-shaped curves invalid in Robinia?

Since 2005, an unresolved debate has questioned whether R-shaped vulnerability curves (VCs) might be an artefact of the centrifuge method of measuring VCs. VCs with R-shape show loss of stem conductivity from approximately zero tension, and if true, this suggests that some plants either refill embolized vessels every night or function well with a high percentage of vessels permanently embolized. The R-shaped curves occur more in species with vessels greater than half the length of the segments spun in a centrifuge. Many have hypothesized that the embolism is seeded by agents (bubbles or particles) entering the stem end and travelling towards the axis of rotation in long vessels, causing premature cavitation. VCs were measured on Robinia pseudoacacia L. by three different techniques to yield three different VCs; R-shaped: Cavitron P50  = 0.30 MPa and S-shaped: air injection P50  = 1.48 MPa and bench top dehydration P50  = 3.57 MPa. Stem conductivity measured in the Cavitron was unstable and is a function of vessel length when measured repeatedly with constant tension, and this observation is discussed in terms of stability of air bubbles drawn into cut-open vessels during repeated Cavitron measurement of conductivity; hence, R-shaped curves measured in a Cavitron are probably invalid.