(i) Partial hydrolysis cuts some of the peptide bonds, and each fragment keeps its order. Read off every run of two and three neighbouring residues:
| Peptide | Dipeptides | Tripeptides |
|---|
| T1 = Gly-Ala-Gly-Ala | Gly-Ala, Ala-Gly | Gly-Ala-Gly, Ala-Gly-Ala |
| T2 = Ala-Gly-Ala-Gly | Ala-Gly, Gly-Ala | Ala-Gly-Ala, Gly-Ala-Gly |
(ii) She is not right. The two lists contain exactly the same fragments: Gly-Ala and Ala-Gly, and Gly-Ala-Gly and Ala-Gly-Ala. A sequence with a repeating pattern can be read in either register, so overlapping the fragments gives both Gly-Ala-Gly-Ala and Ala-Gly-Ala-Gly. The composition does not help either, because both tetrapeptides contain 2 Gly and 2 Ala.
What decides is the identity of one end residue. If the N-terminal residue (with the free –NH2 group) is glycine, the peptide is T1. If it is alanine, the peptide is T2. Knowing the C-terminal residue would also decide.

This is why peptide sequencing never relies on overlapping fragments alone. The two ends of the chain are chemically different, and T1 and T2 are different compounds with different primary structures.