Inheritance of the P.T.C. Taste Character in Man: An Analysis of 126 Rátrhi Brahmin Families of West Bengal

Summary 1. P.T.C. taste threshold data for 127 families have been given, of which one family has been rejected on account of illegitimacy. 2. A x2 test by the ‘2×2’ contingency table indicates that the matings in the families studied were random (x2= 097, P>0.3). 3. Analysis of the families by Fisher's method following Race & Sanger (1954) indicated a ‘good fit’ (P > 0.1) between the observed and the expected numbers of T × T as well as T×t (including t×T) families. 4. Fisher's test following Smith (1956) indicated a ‘good fit’ (P>0.1) for the totals of families of each of the types T×T and T × t (also t × T)–each family containing at least one non‐taster child. But a significant discrepancy (P<0.01) was found for the family size 4 of T × T type mating type considered singly. 5. Comparison of x2 calculated on the totals of the observed and the expected numbers of non‐tasters in sibships produced by T × T as well as by T × t (including t×T) matings, each of which contained at least one non‐taster indicated a significant discrepancy (0.05>P>002). 6. Discrepancies have been noticed between the observed and the expected numbers of non‐taster children in the combined family sizes 2 and 3 out of T × T matings, and the family size 4 out of T×t (and t×T) matings–each producing at least one non‐taster child (P<0.01 and P < 0.05, respectively). 7. Totals of x2 values for families of all sizes (calculated separately for each size) in respect of the numbers of non‐taster children indicated a significant discrepancy (P< 0.02) for T×T families, but a satisfactory agreement (P>0.05) for the T×t (including t×T) families–each family containing at least one non‐taster child. 8. Summing up the x2 values calculated on totals under the four different tests following Smith (1956), a single x2 of 13–38 with 4 D.F. was obtained, which finally indicated a significant divergence (P< 0.01) between the theory and the data taken as a whole. 9. It has been suggested that an assumption of incomplete penetrance (say, about 90 %) of the dominant taste gene might interpret the entire P.T.C. data (including the sib‐pair data of Harris & Kalmus, and Das referred to already) more satisfactorily than the simple hypothesis as it stands to‐day.