In T 220/83 (OJ 1986, 249) the board ruled that the grounds for appeal should specify the legal or factual reasons on which the case for setting aside the decision was based. The arguments had to be clearly and concisely presented to enable the board and the other party or parties to understand immediately why the decision was alleged to be incorrect, and on what facts the appellant based his arguments, without first having to make investigations of their own. Whilst passages from the literature setting out the state of the art might – provided sufficiently precise reference was made to them in the grounds of appeal – be considered an integral part of the grounds, they could not by themselves form a sufficient statement of grounds. This principle was expressly confirmed in many decisions, for example T 250/89 (OJ 1992, 355), T 1/88, T 145/88 (OJ 1991, 251), T 102/91, T 706/91, T 493/95, T 283/97, T 500/97, T 1045/02, T 809/06, T 2346/10, T 294/11 and T 2001/14 (see also T 12/00, where the case was contrasted with T 220/83 and found to differ on the facts).

In T 22/88 (OJ 1993, 143), the board held that a written statement announcing only that the appellant would complete an omitted act (here the filing of the translations of the revised claims) did not comprise sufficient reasons, and therefore did not constitute a valid statement of the grounds of appeal.

Where, as a result of contradictions and inexactitudes in an anyway very brief statement of grounds, it is left to the board to make sense of it, the appeal is usually inadmissible (T 760/08).

In T 2377/13 the appellant's grounds of appeal had not addressed any of the main reasons for the opposition division's decision, and in particular the finding that experiments D9 and D13 were not suitable to show that a technical effect resulted from the difference between the closest prior art and the claimed subject-matter, and so had not set out why the appealed decision should be set aside.