Numbers of recipients are given in parentheses after “r.” Analyses of 13 to 20 recipients of fetal or newborn blood or control BM that are detailed in the caption of Table 3. Total numbers of PHSC are estimated using the binomial equation: N = P (100 − P) ÷ [covariance] where P = mean % donor type, and the covariance = (r) × (SD₁ × SD₂). Such estimates of numbers using covariance become more meaningful as r approaches 1.00 and the SDs approach the same value. The r values are Pearson's product moment correlation coefficients relating percentages of donor type erythrocytes (E) and lymphocytes (L) in the same recipients sampled at time points 2, 3 and 4, or 2 successive sampling times (2 to 3; 3 to 4) for either E or L.

Concentrations of PHSC in blood donors are estimated by subtracting estimated numbers of BM standards (half the number estimated for BM controls) from total numbers of PHSC, and adjusting to conc./10⁶ by multiplying by 2. RU_ave/PHSC is the average number of RU calculated from lymphocytes and erythrocytes in Table 2 divided by the estimated PHSC concentration (both per 10⁶ donor cells).

These correlation coefficients are not significant ( P > .05), probably because of confusion in sample order.