p. 7] justly claims that the purpose of any comparison and the task of comparative science is not only to distinguish similar traits, but even more to search for the differences within the similarity. Comparative psychology, consequently, must not merely understand man as an animal, but much more as a non-animal.

The straightforward application of the principle led to a ubiquitous search for similarity. A correct method and reliably established facts led to monstrously strained interpretations and distorted facts when applied uncritically. Children’s games have indeed traditionally preserved many echoes of the remote past (the play with bows, round dances). For Hall this is the repetition and expression in innocent form of the animal and pre-historic stages of development. Groos considers this to show a remarkable lack of critical judgment. The fear of cats and dogs would be a remnant of the time when these animals were still wild. Water would attract children because we developed from aquatic animals. The automatic movements of the infant’s arms would be a remnant of the movements of our ancestors who swam in the water, etc.

The mistake resides, consequentiy, in the interpretation of the child’s whole behavior as a recapitulation and in the absence of any principle to verify the analogy and to select the facts which must and must not be interpreted. It is precisely the play of animals which cannot be explained in this way. “Can Hall’s theory explain the play of the young tiger with its victim?”—asks Groos [1904/1921, p. 73]. It is clear that this play cannot be understood as a recapitulation of past phylogenetic development. It foreshadows the future activity of the tiger and not a repetition of his past development. It must be explained and understood in relation to the tiger’s future, in the light of which it gets its meaning, and not in the light of the past of his species. The past of the species comes out in a totally different sense: through the individual’s future which it predetermines, but not directly and not in the sense of a repetition.

What are the facts? This quasi-biological theory appears to be untenable precisely in biological terms, precisely in comparison with the nearest homogeneous analogue in the series of homogeneous phenomena in other stages of evolution. When we compare the play of a child with the play of a tiger, i.e., a higher mammal, and consider not only the similarity, but the difference as well, we will lay bare their common biological essence which resides exactly in theft difference (the tiger plays the chase of tigers; the child that he is a grown-up; both practice necessary functions for theft life to come—Groos’ theory). But despite all the seeming similarity in the comparison of heterogeneous phenomena (play with water—aquatic life of the amphibian—man) the theory is biologically meaningless.

Thorndike [1906] adds to this devastating argument a remark about the different order of the same biological principles in onto- and phylogencsis. Thus, consciousness appears very early in ontogenesis and very late in phylogenesis. The sexual drive, on the other hand, appears very early in phylogenesis and very late in ontogenesis. Stern [1927, pp. 266-267], using similar considerations, criticizes the same theory in its application to play.

Blonsky (1921) makes another kind of mistake. He defends—and very convincingly—this law for embryonic development from the viewpoint of biomechanics and shows that it would be miraculous if it did not exist. The author points out the hypothetical nature of the considerations (“not very conclusive”) leading to this contention (“it may be like this”), i.e., he gives arguments for the methodological possibility of a working hypothesis, but then, instead of proceeding to the investigation and verification of the hypothesis, follows in Hall’s footsteps and begins to explain the child’s behavior on the basis of very intelligible analogies: he does not view the climbing of trees by children as a recapitulation of the life of apes, but