Data are given as mean ± SEM, where column n is the number of compartments analyzed; the number of ER analyzed ranged from three to seven per dendrite (fewer for other compartments). There were at least two slices for each condition, with dendrites distributed as follows: control ± TTX, 21 dendrites; single train, 25; 4 trains, 31; 4, trains + TTX, 9; 4 trains + recovery, 15; 4 trains + thapsigargin, 28. SE and n values are based on compartmental analyses, because a statistical evaluation of the data indicated that dendrite-to-dendrite and slice-to-slice variabilities were small compared with compartment-to-compartment variability. The quantification procedure provides concentrations in units of millimoles per kilogram of dry weight; conversion to millimoles per liter of hydrated cell volume is described in Materials and Methods.

↵F1-a  Morphologically defined ER consisted of multiple populations with differing Ca uptake capacities, as discussed in the text and illustrated in Figure 5. Therefore, mean concentrations reported in this table are not necessarily derived from normal distributions. Large SEs result when the underlying populations are different, for example, ER Ca concentrations after one or four stimulus trains.

↵*  Significantly higher than corresponding values from resting control cultures, except cytoplasmic K⁺after four trains in the presence of thapsigargin, which is lower;t test, p < 0.05.

↵F1-160  Significantly lower than corresponding values from cultures subjected to four stimulus trains, except Ca in mitochondrial matrices after four trains in the presence of thapsigargin, which is higher; t test, p < 0.05.

↵F1-165  Not significantly different from zero. However, the data are given to indicate the errors of Ca analyses, which in this case are mainly limited by counting times. Negative values, reflecting statistical fluctuations, sometimes occur for populations that are not significantly different from zero.