A culture of unicellular organisms growing at because events in different clones are not in phase maximal rate under constant conditions contains with each other. cells at all stages of the division cycle. The relaWhen conditions are not constant, as when the tive frequencies of cells of different ages-i.e., the stationary phase is approached, growth need not age structure of the population-is calculable if be balanced, but there is no logical reason it could one knows the distribution of individual doubling not be. Neither need it be balanced if some one times. The calculation assumes only that the age extensive property is increasing exponentially. structure remains constant from one moment to The concepts of logarithmic growth and balanced the next. This review is concerned with reports of growth are thus independent. situations in which the age structure deviates Synchronization. We will not attempt so precise from the calculable one, and varies with time in a definition of synchronization or degree of syna somewhat periodic manner. chronization. The limits of their utility will become clearer as the data are discussed. Roughly, DEFINITIONS synchronization denotes any process in which the Balanced growth. The concept of unbalanced events in different cell lines are put into phase growth has been used by Cohen and Barner (7), with each other. A completely synchronized culto describe situations in which the normal ratio ture would be one in which, at any given moment, of synthesis of nuclear to cytoplasmic components every cell was doing exactly the same thing. is disturbed. For our purpose, it will be convenReal systems lie somewhere between this ideal ient to say that growth is balanced over a time and the completely random situation of a culture interval if, during that interval, every extensive in balanced logarithmic growth. It is desirable to property of the growing system increases by the have a quantitative expression for the extent to same factor. which this ideal is approached. The term degree Consider a population which has been growing of synchronization has sometimes been used for under constant conditions for a very long time. this purpose. This is best achieved in a continuous culture The usual procedure is to pick some stage of device (29), but is probably approximated by a the division cycle which ordinarily lasts for only batch culture which has been in the logarithmic a small fraction of the doubling time and to phase for several generations. It is logically measure the percentage of cells which are in that necessary that, so long as the constant conditions phase. Thus, yeast might be said to be "60 per prevail, growth will be balanced over any time cent synchronized" if 60 per cent of the cells are interval, either finite or differential. budding at once (2); and Tetrahymena could be Now consider one cell within the population. considered 85 per cent synchronized because 85 As time proceeds, it will give rise to a clone of * descendants. Over any small interval of time, the pertcnt ells are dividin atione t e growth of the cell will, in general, not be balanced. Unortunatel, theterm isthe number of cells It might, for example, synthesize all of its nucleic choetoi consiei the number of cel acid at one stage of the division cycle, and its undergoing dvson, the number at for instance, protein uniformly throughout the cycle. If one anaphase, we would record a smaller degree of assumes that the doubling time is precise [clock synchroimzation. mode of growth (14)], the growth of the clone will This ambiguity does not alleviate the need for be balanced only over intervals which are integral some such expression, however. The fact that multiples of the doubling time. The growth of 85 per cent of the cells are dividing at one time the population which is the aggregate of all these places them in a group which are rather close to clones is balanced over arbitrary time intervals, being completely in phase with one another. The