The pore complexes of the nuclear envelope presumably regulate macromolecular exchange between the cytoplasm and the karyoplasm (7-9, 11, 14-16, 18, 22, 24-27, 30) . We have previously presented quantitative data, obtained by negative staining (29, 32) and thin-sectioning, (31) on the number, size, and fine structure of nuclear pores in the macronuclear envelope of the ciliate protozoon Tetrahymena pyriformis GL in different physiological states . We now relate these results to those obtained by freeze-etch electron microscopy, a technique which allows observation of large areas of partially hydrated biomembranes . We also describe the influence of glutaraldehyde prefixation on the freeze-etch appearance of nuclear pores .
MATERIAL AND METHODS
CulturesStock cultures of Tetrahymena pyriformis (amicronucleate strain GL) were maintained axenically at 15'C in 2% proteose peptone containing 0 .47 7 0 liver extract . We have examined cells of cultures in the logarithmic, as well as in the stationary, growth phase, and also those of heat-shock synchronized cultures at the end of the synchronization treatment (EHT) and at the division maximum (29, 30) .
2Freeze-Etching and Electron Microscopy Cells in the four different physiological states were centrifugally concentrated into 2 ml of culture medium. Fixation, when performed, was done by addition of 2 ml of 3-4% glutaraldehyde, in 0 .1 M cacodylate (pH 7 .2) .After 15 min the cells were thoroughly washed with buffer, and transferred through buffered solutions of increasing glycerol concentration to a level of 20 0]0 glycerol. Unfixed cells were directly incubated in the glycerol solutions for 1-3 hr. Small samples of packed Tetrahymena cells were frozen on gold-alloy specimen holders in Freon 22 (E . I . duPont de Nemours & Co ., Inc., Wilmington, Del.) . The frozen specimens were replicated according to Moor and Mühlethaler (21) on a Balzers model BA 360 M freeze-etch device (Balzers A . G., Liechtenstein) . The etching time was usually 1 min at -100°C. In some experiments, etching was minimized (see reference 28) by keeping the temperature of the specimen holder at -150°C and by replicating the freshly fractured surface in less than 5 sec after fracturing . Replicas were examined in a Siemens Elmiskop IA, the magnification of which was routinely calibrated by using a carbon grating replica .
MeasurementsAll measurements were made on calibrated positives . Pores per square micron were determined by direct counts on fracture faces with large areas of membrane. 100 pores of different macronuclei from