According to the boards' case law, a specific combination of elements requiring the selection of elements from two known groups/lists cannot be regarded as disclosed in the art and so fulfils the novelty requirement (cf. T 12/81, OJ 1982, 296).
T 12/81 (OJ 1982, 296), a decision of fundamental importance with regard to novelty in the field of chemistry, is referred to time and again in the case law of the boards of appeal. It states that the teaching of a cited document is not confined to the detailed information given in the examples of how the invention is carried out, but embraces any information in the claims and description enabling a person skilled in the art to carry out the invention. If a product cannot be defined by a sufficiently accurate generic formula, it is permissible to make the definition more precise by additional product parameters such as melting point, hydrophilic properties, NMR coupling constant or the method of preparation (product-by-process claims). From this it necessarily follows that patent documents using such definitions will be prejudicial to the novelty of later applications claiming the same substance defined in a different and perhaps more precise way. Decision T 12/81 related to such a case. Summarising, the board stated that in the case of one of a number of chemical substances described by its structural formula in a prior publication, the particular stereo-specific configuration of the substance - though not explicitly mentioned - was disclosed in a manner which was prejudicial to novelty, if it proved to be the inevitable but undetected result of one of a number of processes adequately described in the prior publication by the indication of the starting compound and the process.
The applicant argued that the novelty of the claimed product was based on a selection. The starting substance was chosen from a list of 20 compounds and combined with one of the five alternative process variants. The board did not share this view, but used the opportunity to comment on this argument and develop criteria for selection inventions that have frequently been adopted in later decisions:
A substance selection can come about if an unmentioned compound or group of compounds having a formula covered by the state of the art is found, in the absence of any information as to the starting substance or substances. The subject-matter in the case in question, however, did not involve a selection of that kind in an area which, although marked out by the state of the art, was nonetheless virgin territory.
However, the disclosure by description in a cited document of the starting substance as well as the reaction process is always prejudicial to novelty because those data unalterably establish the end product.
If, on the other hand, two classes of starting substances are required to prepare the end products, and examples of individual entities in each class are given in two lists of some length, then a substance resulting from the reaction of a specific pair from the two lists can nevertheless be regarded for patent purposes as a selection and hence as new.
The board held that a combination of starting substances and process variants, however, was quite a different matter from a combination of two starting substances, and thus not comparable. At its simplest, if the starting substances were regarded as fragments of the end product, then every conceivable combination of a given starting substance in the first list with any starting substance in a separate second list of additionally required starting substances involved a true substantive modification of the first starting substance, since in every combination it was supplemented by a different fragment of the second starting substance to become a different end product. Each end product was thus the result of two variable parameters.
However, combining a given starting substance from a list of such substances with one of the given methods of preparation did not result in a real substance alteration of the starting substance but only an "identical" alteration. In the case in question, for example, no matter which of the processes described in detail was used, the end product was always the particular starting substance's hydrogenation product, which differed from the starting substance itself only in that it contained two additional hydrogen atoms. The process parameter was thus - seen in terms of the end product - not a variable parameter that would result in an immense widening of the range of possibilities, so that precisely in this case the end product was not the result of two variable parameters (see also T 3/89, T 1841/09).