As explained in Takhtajan’s preface this book is not a mere translation of his ‘The origin of Angiospermous plants’ (1961, in Russian), but an entirely new book. I find this true and not true. Comparing it with the Origin (1958 translation of the 1954 Russian version) the essence of the new book was there given in a nutshell. In size, chapter subjects, argumentation, and bibliographic documentation, the work is very much extended and it makes very interesting reading indeed. The sequence of the chapters is logical, almost always leading to distinct synthesis. Properly it is a critical commented survey of many opinions — Takhtajan being clearly in complete command of the huge literature on the subject — but from which the author follows his own line of choice and judgement, accepting or rejecting with brief but clear comments. The whole argumentation is admirably concise and rouses admiration for covering this vast subject, comprising taxonomy, plant distribution, morphology, palynology, genetics, population dynamics, flower biology, anatomy, paleozoology, etc. Major questions are embodied in subsequent paragraphs: polyphyletism is rejected; ancestors must be sought among heterosporous ferns or fern-like plants followed by pteridosperms and certain gymnosperms, although direct ancestors cannot be indicated; the basal flower type of angiosperms was bisexual. Takhtajan attaches great importance to occurrence of plants in small populations, especially in mountain plants, facilitating chance variations and genetic drift, rapid spread of mutant genes, which is important for evolution. This entails that missing links are almost never fossilized. Micro-evolution is equalized with macro-evolution. Neoteny (on which Takhtajan devoted a former work) can lead to despecialisation through which phenotypic simplification the complexity of the genome remains intact; it may provide for a maximum phenotypic effect by a minimal genotypic change. Primitive wood structure of early Winteraceous angiosperms is understandable by neotenic origin. Evolution of angiosperms was not only rapid, but also discontinuous as a result of neoteny. Developing in the mountains ‘in many ..... small ..... populations ..... the earliest angiosperms found themselves under conditions most favourable to evolutionary radiation. And if we bear in mind that their evolution was closely tied to the evolution of insects and was based on the complex and peculiar mechanism of mutual selection, then the extraordinary speed of their initial differentiation becomes even more readily understandable.’ Protection of the ovules arose as a selection against damage by ‘early pollinating insects’; this made simplification of their structure possible which led to smaller ovules (loss of thickened integuments, sclerotesta, etc.) and enabled the angiosperms to observe the greatest economy of material in construction of the ovules and ♀ gametophyte, and it also made possible the perfection of the process of pollination. ‘The acquisition of the stigma was undoubtedly a very great event in the evolutionary history of seedplants.’ ‘The primitive insects searched for pollen (beetles), nectar searching ones were a further perfection; this again led to a very great advance in cross-pollination; and as a corollary to a greatly increased rate of evolution, which still continues.’ ‘Isolation of a population is well known to be a prelude to the formation of a new species.’ The question of the hypothetical reconstruction of the first flowering plants is approached by the ‘hypothetico-deductive method’. Knowing the basic evolutionary pathways of angiosperms and the main lines of specialisation of their organs and tissues, we may by extrapolation extend these lines mentally into the past to the lowest possible level of specialisation’, but somewhat further on he writes ‘This reconstruction of the ancestors of the living angiosperms depends on the truth of the assumption that they combined in one plant all the most archaic characters that are now found distributed among the living fossils.’ I have italicized in the citations two words that are in contradiction; furthermore I would like to point out that whereas each plant we know possesses both primitive and derived characters, we cannot make an exception for an ancestral plant; one which would contain all the archaic characters must logically be an idealized fiction.