A simple method has been developed that allows frozen thin sections of fresh-frozen tissue to be cut on a virtually unmodified ultramicrotome kept at room temperature. A bowl-shaped Dewar flask with a knifeholder in its depths replaces the stage of the microtome; a bar extends down into the bowl from the microtome's cutting arm and bears the frozen tissue near its lower end. When the microtome is operated, the tissue passes a glass or diamond knife in the depths of the bowl as in normal cutting. The cutting temperature is maintained by flushing the bowl with cold nitrogen gas, and can be set anywhere from about -160°C up to about -30°C. The microtome is set for a cutting thickness of 540–1000 A. Sections are picked up from the dry knife edge, and are placed on membrane-coated grids, flattened with the polished end of a copper rod, and either dried in nitrogen gas or freeze-dried. Throughout the entire process the tissue is kept cold and does not come in contact with any solvent. The morphology seen in frozen thin sections of rat pancreas and liver generally resembles that in conventional preparations, although freezing damage and low contrast limit the detail that can be discerned. Among unusual findings is a frequent abundance of mitochondrial granules in material prepared by this method.

Christensen, A. K. (1971). FROZEN THIN SECTIONS OF FRESH TISSUE FOR ELECTRON MICROSCOPY, WITH A DESCRIPTION OF PANCREAS AND LIVER. The Journal of Cell Biology, 51(3), 772–804.