In 1902 Gottlieb Haberlandt attempted to maintain and to investigate surviving angiosperm cells removed from the plant body. Haberlandt’s success was limited by the techniques at his disposal, and his observations were restricted to minimal signs of maintained vital activity which could be observed with the light microscope, as summarized in a composite figure which has recently been reproduced (Steward 1968, p. 468) and which also appears in a complete translation of Haberlandt’s text (Krikorian & Berquam 1969). Nevertheless, Haberlandt made farsighted forecasts. It is often attributed to Haberlandt that he suggested that it should be possible to grow plants even from such isolated surviving cells as those that he used (e. g. the parenchyma cells of Labiates or the hair cells of Tradescantia ). But in a relatively recent re-reading of that oft-quoted paper (Steward, Blakely, Kent & Mapes 1963) it became clear that Haberlandt’s forecast was even more acute, for he prophesied that ‘out of cells like these we ought to be able to make artificial embryos '. Thus the concept of the essential totipotency of all the living cells (derived by equational divisions from the zygote of angiosperms, whatever their location or function), and their presumptive growth in culture essentially derives from Haberlandt in 1902. Even so, Haberlandt’s remarkable forecast was not really vindicated until about 60 years later. However, the events here documented, which bear upon this question, arose from techniques developed and experiments performed with very different ends in view. It is of some interest to show how this came about. The subject area which was to lead the author to the study of morphogenesis in free cells was a seemingly very different one: it concerned the problem then known as the accumulation of solutes by plant cells. Throughout the 1930s and early 1940s, this problem had been prominently investigated, using very thin (0.75 mm) disks cut from various plant storage organs (potato tuber, carrot root, Jerusalem artichoke tubers, etc.) and adopting a technique devised for the purpose (Steward 1932). The methods used in that period identified what may now be (figure 1 c ), and they also synthesize protein at the expense of their stored nonprotein nitrogen compounds (figure 1 b ). Even when such cells are in contact with moist air, they turn starch into sugar (figure 1 a ) and display an elevated rate of respiration and of all concomitant metabolic events. Thus, the pace of all the events that depend upon energy use (e. g. protein synthesis and ion accumulation) is regulated by temperature (Steward, Berry, Preston & Ramamurti 1943), by oxygen partial pressure (figure 1 d ), by potassium concentration, etc., in ways which demanded an explanation of the causal link between them. It was, however, the experimental system devised to illuminate this essentially cell physiological problem which led, unexpectedly, to the work on growth induction and morphogenesis that is now to be described. This came about as follows. Several problems were posed for investigation when work was resumed after the Second World War. These problems were: (1) To devise a system, more versatile than that provided by the cut disks of potato tuber, on which to study the metabolism, respiration, growth and ion intake of cells. For this it was desirable to find a common, quiescent, parenchymatous tissue that could be stimulated, easily, into the most rapidly growing state possible. (This objective was achieved by the use, under aseptic tissue culture conditions, of explanted pieces (2.5 mg) of the secondary phloem of carrot root and by the growth inducing effect, over and above ordinary nutrients, of coconut milk added to the external solution.) (2) To devise new means to study in greater detail the protein synthesis which occurred concomitantly with progressive ion intake. (This second objective was met by adopting the then new procedures of paper chromatography and radioautography.) (3) A third, and seemingly more distant, goal was to begin the study of nutrition and metabolism at the beginning, i.e. with the fertilized egg or zygote and, hopefully, to remove such cells from the ovule and induce them to grow in isolation. (Paradoxically, as will be shown, it is now far easier to make isolated, free, mature carrot cells recapitulate embryogeny than it is to remove, surgically, single eggs from ovules and to grow them in isolation.)