In the last 40 years, pre-breeding works induced, in more and more centers of maize breeding, full-sib reciprocal recurrent selection programmes to identify some heterotic pairs which can be sources for obtaining performance inbred lines. The aim is to identify the heterotic pairs with the best results according to the yield potential of maize, the breaking and falling resistance, and the grains moisture at the harvesting time. The creation programme of A and B composite population started at ARDS Turda in 1985. Inside of A composite came the next inbred lines: B73, A632, M117, TC209, T291, being from the B SSS heterotic group, and inside of B composite came the inbred lines Mo17, C103, TC 208, T248, W633, appreciated by us or being related to Lancaster Sure Crop heterotic group. The experimentation was done in two orientation comparative cultures, each one with 49 variants, in 4 repetitions; the comparative culture was a balanced quadratic grid of 7x7 type. From each culture were chosen the first six variants, which were evaluated according to the next characters: production potential, breaking and falling resistance, grains moisture at harvest. The presented results are a part from the second cycle of full-sib reciprocal recurrent selection. Test crosses and self-pollinations were made on plants from the two composites which had two cobs; on the first cob from A Composite realised the cross with the corresponding plant from the B Composite, and from the plant panicle of the B Composite was collected pollen to pollinate the chosen plant from the A Composite. At the both plants from the crossing, the second cob was self-pollinated and kept in reserve until 2010, when the test crosses was experimented and were selected the pairs with the best results according to the above characters. Using the full-sib reciprocal recurrent selection, we can successfully harnessing, simultaneously, the additive and non-additive gene effects.
