The already purely externally -- and not considering any functional feature -- decisive criterium of wholeness is not the somehow based togetherness [i.e. a in one or another way based togetherness of elements], but a togetherness as a result of physico-chemical material bonding (in contrast to a purely "mental bond" as that what holds things together, representing the up-to-now view). In the first case, the case of a in one way or another based togetherness, variations or deviations are more or less unimportant, whereas in the second case [togetherness as a result of material bonding] they are almost always destructive. In the first case there is a kind of spontaneity in forming an association and structure, in the second case this formation is an, according to natural law, "formula-adherent" compulsion of the wholeness-form and -structure.
The only valid --perhaps not always directly and easy to find -- criterium of wholeness is the chemical bonding between the extreme boundary points [bonding continuum from head to feet]. As a result, all the many other, partly simple, partly complex, criteria of wholeness -- which, for example, are also about spontaneous formation, about purpose, etc. -- are superfluous, or they have merely the nature of a description, or they are created for rapid determination.
In inorganic molecular wholes [i.e. molecules] the statement that the whole is "more" than the parts, is common sense [chemical compounds generally have properties different from those of their atomic constituents] and [therefore] barely considered [as to its precise meaning and significance]. The question why a known whole could work differently as do its known parts, may, -- from a chemical viewpoint and applied to chemical systems (a compound as compared to its atomic partners) -- by a large number of examples and known rules not, it is true, exactly be solved, but nevertheless be sufficiently clarified : The constructional relationships [in such inorganic molecular wholes] we have established by exact chemical and physical laws.
The constituents of Unimol are wholeness-generating only from an analytical point of view (which here is not valid), in the true sense, however, they are wholeness-maintaining.
From the huge number of definitions of wholeness we select : No divisibility without breakdown. One also can easily distinguish between the whole (as indivisible unity) and the total (as compiled system-sum). Other wholeness conditions are here of no interest. Much has been said about them already.