In a recent review on the influence of particle size on catalytic properties of supported metals [l], evidently a most important aspect of heterogeneous catalysis by any account, the authors end on a fairly discouraging note since they suggest doing even more work — with ever more sophisticated techniques and methods as well as normalized procedures — on top of an already formidable amount of literature. At such a state of the art it may be appropriate to reflect and to ask whether full pursuit of reductionism as scientific method needs fresh thinking in order to progress more effectively. Perhaps the difficulties are rooted in emphasis on mere data collection, the lack of a good empirical guiding principle, inappropriate models; in brief, the frequent simplification of a vastly complicated object — the real catalyst — into the sum of phenomena of better-understood idealized subsets. It might even be argued that a thorough knowledge of a well-characterized prototype or a particular model catalyst would be actually of little help in seeing the real world where catalysts are of irregular shape, suffer sintering, become poisoned; in other words, lose their ideal state after some time of operation. Also, we realize that a change in “quantity” often engenders a change in “quality”: For instance, a material with a BET area of hundreds of square meters per gram can hardly be considered to have a surface like a huge tennis court or ball field: it is rather “convoluted” in some very irregular fashion, having a shape that is difficult to realistically construct as a linear combination of planar portions.