Cryo-scanning electron microscopy observations of vessel content during transpiration in walnut petioles. Facts or artefacts?

The reliability of the cryo-SEM technique for stabilizing and quantifying e mbolisms in vessels of transpiring plants has recently been criticized, on the grounds that the embolisms observed are artefacts of the freezing. One of the tests used was a comparison of the embolisms found in rachises of tr anspiring walnut leaves (Juglans regia L.) frozen intact on the tree, with both cryo-SEM images of vessel contents and the measured hydraulic conducti vity of similar samples whose xylem pressure had been returned to atmospher ic pressure by cutting the leaves off under water. Embolisms found in intac t frozen rachises were not present in cut rachises, and the high rachis hyd raulic conductivities indicated a similar absence of embolisms. We show tha t the authors' conclusions are wrong because their test produces a differen t artefact. When the petiole is cut under dye solution it is obvious that t he immersing solution is drawn into the embolized vessels and fills them. T hus, the cryo-SEM images and the high hydraulic conductivities of specimens prepared by this technique do not indicate the contents of the xylem in th e intact plant. In fact, this artefact may perhaps be used (with the dye) t o measure the embolisms in the intact plant. The embolisms seen in the cryo -SEM are most unlikely to be artefacts. The published work that also shows embolisms and their refilling in transpiring plants by techniques involving no freezing is reviewed. (C) 2001 Editions scientifiques et medicales Else vier SAS.