The results given in tables 27 and 28 are remarkable. Neither in the DR × R nor the DR × DR crosses did the tail fail to develop. The tailless condition, that I had strongly suspected of being recessive and expected in 25 per cent to 50 per cent of the offspring, never once appeared. The only point of variation in the uropygium of the chicks derived from the back cross or from F1's bred _inter se_ was that in some the uropygium seemed distinctly smaller than in the others. This small uropygium was as a matter of fact recorded chiefly in chicks that failed to hatch, but it was occasionally noticed in older birds, being then usually associated with a slight convexity of the back. In some of the families the uropygium is recorded as small in suspiciously close to 25 per cent of the offspring. There is little doubt in my mind that this small uropygium represents in some way the "absence" of tail that was expected. The next step was to cross the other rumpless bantam (No. 116), to see if he behaved like his father. Accordingly, in pen 653, I replaced the cock No. 117 by 116, the hens remaining the same, and got the result shown in table 29. TABLE 29.--_Heterozygotes mated with No. 116._ +------+------------+----------+----------------------------------------+ | | | | Condition of uropygium in offspring. | |Serial| Father's | Mother's +----------+--------+---------+----------+ | No. | No. | No. | Present. | Small. | Absent. | Per cent | | | | | | | | absent. | +------+------------+----------+----------+--------+---------+----------+ | 27 | 116 | 508 | 5 | 2 | 10 | 59 | | 28 | 116 | 577 | 3 | 0 | 3 | 50 | | 29 | 116 | 587 | 3 | 1 | 4 | 50 | | 30 | 116 | 652 | 4 | 0 | 2 | 33 | | 31 | 116 | 705 | 3 | 1 | 5 | 56 | | 32 | 116 | 713 | 1 | 0 | 2 | 67 | | 33 | 116 | 760 | 4 | 0 | 2 | 33 | | +----------+--------+---------+----------+ | Totals (55) | 23 | 4 | 28 | 51 | +------------------------------+----------+--------+---------+----------+ Here we get a result almost exactly in accord with Mendelian expectation. Having, now, obtained rumpless hens, it became possible for the first time to test the inheritance of rumplessness in both parents. The result is shown in the table 30. TABLE 30.--_Rumpless fowl mated inter se._