Annual Review of Plant Physiology 11:  191-238  (June 1960)
VERNALIZATION AND ITS RELATIONS TO DORMANCY^1
1The survey of literature pertaining to this review was concluded September 1959.
P. CHOUARD
Laboratory of Plant Physiology, the Sorbonne, University of Paris, Paris, France

(continued)

MECHANISMS OF VERNALIZATION

Up to now we have reviewed the main experimental work that determined the characteristics of vernalization in various species. We will now examine what processes were inferred in attempting to rationalize the mechanisms involved. [215]

THE STADIAL DEVELOPMENT THEORY AND THE REVERSIBILITY OF VERNALIZATION

From embryogenesis to flowering, a plant grows through successive stages. The identification of these stages is useful in further analyzing every mechanism in their ontogeny. The "stadial theory" however, is exacting; it holds that the same stages occur in all plants and that they are fundamentally irreversible. For further information refer to Lysenko who proposed the theory (155 to 159) and to commentators on the theory (279). If these stages are believed to be ontogenetic natures or properties defined according to an a priori postulate and unaltered by experiment, they are to be considered as expressing a faith that does not fall within the scope of scientific experimental analysis. If we hold to mere facts that can be repeated in experiments we speak another language. We indeed recognize stages in plant development, but these are neither universal nor irreversible. For example, we cannot say that the vernalization process, or "thermostage," is universally imposed merely as a temperature-controlled period of development. Bidens radiatus reacts to photoperiods as soon as its cotyledons turn green, without any previous thermostage (37) ; also, the "photostage" is not always necessary for flowering. Flowering primordia exist in mature peanut seeds before any light treatment of the seedling. When a thermostage is required, it may be experimentally reversible in certain cases, as we have seen for experimental successions of devernalization and revernalization. What is the obligate requirement for new vernalizing chilling if not the need for a new thermostage, the former having been fully eliminated? Bud regeneration returns the plant to most of its juvenile stages. In experiments with proliferating flowers (44) the partly differentiated floral apex is reversibly returned to a vegetative apex of a shoot which again will have to go through all the previous stages before flowering. As far as the words are concerned, we prefer to consider purely empirical stages as observed for each species in our experiments and their reversible or irreversible sequence, and to analyze the various regulating mechanisms, whether independent or correlated, that control the sequence of the development.

HEREDITARY TRANSMISSION OF DEVELOPMENTAL MECHANISMS

A requirement for vernalization is an hereditary property of certain species and concerns the transmission of a physiological characteristic. We have seen that a single species often includes several strains that display various kinds of vernalization requirements. We even found wild species that seem to have a simple vernalization requirement but in experiments reveal that they are a mixture of mutants with differing requirements. When hybrids can be obtained from plants with different vernalization requirements, the studies (169, 207) have shown that the vernalization characteristics are borne by a few Mendelian factors in which the various alleles are often selected by an adaptation to climate, e.g., one gene only in the case [216] of the two black henbane strains. But vernalization was used as a stepping stone to demonstrate the transmissibility of acquired characters. Mitchourin's followers state that the growing of spring wheat in the autumn wipes out the inheritance of the "spring" character after a severe winter of growth; from the seeds of the few surviving individuals a second autumn crop would show a new "winter" heredity instead of the "spring" one (158, 159). In certain countries, vernalization is mostly studied from this point of view. Since such experiments have not been successfully repeated in any country where properly controlled experiments and pure strains were used, and since they do not concern the physiological processes involved in vernalization, there is no need to discuss them here.

However, the stadial theory led several physiologists to study so-called "lasting modifications" that were found to be reversible in a few generations. For example, Séchet (239) reports the "remnant effects of vernalization" in Camelina sativa, Cicer arietinum, Vicia faba, and in oats. Highkin (119) reported a nonidentical but related phenomenon in the training of peas for surviving detrimental thermal treatments.

DESCRIPTIVE FORMULATIONS OF VERNALIZATION

Several authors have established formulation of the facts observed. Lang & Melchers (144) express their observation on henbane as follows:

I occurs in the cold, II takes place at normal temperatures and in the presence of oxygen, III is heat devernalization without oxygen, B is not necessarily the precursor substance of C, but may also be an agent involved in the synthesis of C.

Gott, Gregory & Purvis (94) and Purvis & Gregory (214) having corrected their previous formulation (212) of vernalization in rye Petkus now represent it as:

which contains in brief all the above-cited observations.

Van de Sande Bakhuyzen (9) expresses his findings by the following formulation:

[217]

A "thermophase" occurs under cold conditions releasing an enzyme "vernalase." Then during an "interphase" with warm temperatures and possibly short days, vernalase results in "vernaline" which, in a final "photophase," gives rise to "florigen" B; the latter, according to the proportions in which it combines with photosynthetic products A, either does or does not induce flowering. This diagram is meant primarily to express simultaneously the mechanisms of short- and long-day plants.

Napp-Zinn (186) groups his studies on Arabidopsis thaliana "Stockholm" strain in the following diagram:

in which A is a thermostable precursor; A₁ is the first thermolabile intermediate that can be shifted toward X by a heat treatment at 30°C. soon after sowing and causes anti-vernalization; A₂ is a thermostable intermediate demonstrated by the possibilities of revernalization after five days of devernalization; A₃ is the second thermolabile intermediate, demonstrated by the partial shifting towards X' at 20°C.; A' is the third thermolabile intermediate demonstrated by devernalization at 30°C., leading to Y, and corresponding to the thermolabile intermediate of Purvis and Gregory; and B is the final thermolabile intermediate of vernalization, leading to flowering. It may be directly attained from A by a shunt at 20°C.

These formulations are handy to memorize and they stimulate further investigations on hypothetical substances; but they provide no more clarification than the authors' descriptions of their own results. Further, they may require adjustment for each new discovery and they also change for each plant type that behaves in a particular way and does not fit the particular representation.

REGULATORY MECHANISMS OF DEVELOPMENT

From a simpler point of view, I think that the experiments permit us to analyze separately some mechanisms found either naturally separated or interrelated. For example, the complex mechanism of vernalization is not always linked to that of photoperiodism. Table I shows that among the various species of plants there are all possible combinations of various vernalization and photoperiodic requirements (see Table I) (34, 36, 38, 39, 41,42).

Before constructing a general theory of reproductive development-if one is possible-we have to analyze carefully one after the other, every elementary mechanism involved in it and choose the best possible example of each mechanism where it constitutes the limiting factor. Then we will consider the various relationships between these mechanisms. For example, as we have just seen, vernalization may be either independent from or associated [219] with photoperiodism. The partial or total replacement of chilling by short days to achieve vernalization is illustrated by the rye Petkus, Campanula medium (Wellensiek's strain), and by one strain of Scabiosa succisa but not by most other plants. Partial or total replacement of vernalizing chilling by long days is exemplified by certain Chrysanthemums and not others. Heat devernalization, obtained by warm temperatures applied immediately after vernalizing chilling, although very common, is ineffective for Chinese mustard and Geum. Late devernalization after fixation of the vernalized state seems impossible with the rye Petkus, henbane, Geum, etc. but seems most readily attained with Oenothera and beets.

The analysis of the substitution of chemicals, particularly gibberellin, for chilling in order to achieve vernalization shows, as we will see later, that the vernalizing process probably consists of several stages; and some of them may be absent in related strains.

After the broad eco-physiological analysis of vernalization presented in the second part of this review, there remains the imposing task of interpreting each elementary process at the physico-chemical level.

COMMON PROPERTIES OF ALL VERNALIZATION PHENOMENA

The variety of elementary processes must not overshadow the characteristics common to all vernalization phenomena. Some of these are ecological, others physico-chemical.

Properties related to environmental conditions required for vernalization (42, 98, 173, 174, 273).—We have reviewed these in detail for the rye Petkus and for henbane; with reference to other plants we can summarize the results here. Chilling between the extreme limits of -6° and +12 to 14°C., normally from +1 to +5 or 6°C., is the fundamental agent of vernalization. It is only effective when applied for several days, weeks, or months in the presence of oxygen to moist enough tissues containing enough carbohydrate to support adequate respiration. Concomitant growth may be either moderate or very slow; if it stops completely, vernalization does not occur. Vernalization is directly perceived by stem or bud primary meristems; it is sensed at a certain age which is very early (immature embryo) or much later (leafy plant) according to the species. It is transmitted by more or less stable or prolonged autocatalysis or self-perpetuation to the buds arising from the vernalized ones. It consists in the attainment of a new functional ability (the flowering ability) that is later "revealed" by flowering under special conditions (long or short days, various temperatures, etc.) which are characteristic for each species and variety.

Physico-chemical properties of vernalized tissues (76).—A great many investigations have been devoted to this question which consists essentially in comparing treated tissues at the end of the vernalizing treatment with nonchilled tissues, either to check on the completeness and effectiveness of the vernalizing process or to try to observe the intimate nature of the [220] process. The second aim has not yet been fully attained, for in fact, we only observe either the ordinary consequences of growth in the cold or the metabolic consequences of vernalization itself, which are simply concomitant with the acquisition of the ability to flower and are not the agents conferring this ability (75, 76).

Thus, it is observed that carbohydrates are depolymerized, resulting in an increase of soluble sugars, sometimes not very great (62, 63, 66, 75); depolymerisation of reserve proteins occurs with an increase in soluble nitrogen compounds (62, 65, 69) ; and there is a fall in the lipid content (64, 73 to 75). These are typical consequences of a low temperature metabolism and probably not specifically related to vernalization (38, 87). Moreover, during germination, a phase occurs in nonvernalized seeds in which the respiratory rate increases with a respiratory quotient approaching 1.0. In vernalized seeds, this phase is characterized by a low respiratory rate and the respiratory quotient only slowly increases to 1.0 (179).

Also observed are a reduction of the auxin level (57, 242), a shift of the isoelectric point of proteins toward acidity (283 to 286), a slight increase in the coagulability and permeability in the protoplasm (30, 83), and a modified pH of the cell liquids (214). After vernalization cells of the vegetative point are less readily coloured blue by ferric chloride and potassium ferrocyanide, showing, as well as by other colour reactions, a possible change of pH (11, 20). The serological properties of a vernalized plant do not seem modified compared to the control (1, 2). Cholodny's former theory (32), involving a transfer from endosperm of auxin and other hormones such as a blastanin, is no longer entertained (112, 242).

Enzyme activities are markedly modified (57, 69, 112, 120, 224, 225, 227, 242) e.g., relative increases in the activity of sugar hydrolases, in particular amylase and invertase (yet there are contradictory results), and of phosphatase, lipase, catalase, etc. Recent Russian authors (75, 246, 247) emphasize the alteration occurring in enzyme adsorption on protoplasmic surfaces, with tendencies toward depolymerization after chilling, following a loosening of the links between enzymes and protoplasm. There is also a reduced activity of cytochrome oxydase and succinic dehydrogenase and increased activity of ascorbic acid oxydase and of malic and citric dehydrogenases (247), even though, in general, the overall respiratory rate is not particularly changed at the end of vernalization. Changes in the vitamins C and B contents take place during (239) and after (189) vernalization.

When vernalization is complete, the ensuing metabolism is soon similar to that of the control. However, certain authors think that the above enzyme modifications persist after chilling and that they are like the enzyme characteristics attained directly by nonchilled annual strains (246).

In addition, it was reported that after vernalization certain plants (radish) show either increased resistance (240), or decreased (239). Chilling by itself is a process of training a plant to resist frost (259). Vernalized [221] cereals respond more to water than do nonvernalized controls, but they also suffer more from drought (134). Their capacity for mineral and water absorption is changed (191) and the anatomical status looks more mature (25).

All these interesting facts still do not solve the problem: What is the biochemical nature of the process of vernalization?

HORMONAL TRANSMISSION OF VERNALIZATION

I emphasized that the autocatalytic transmission of vernalization is restricted entirely to those meristematic cells derived from actually vernalized ones. We should not overlook the fact, however, that vernalization may also be transmitted as if it were effected by some substance diffusing from cell to cell. The Melchers and Lang graft experiments on henbane (168 to 172, 178) elicited much enthusiasm and were interpreted as pointing to some general process. In actual fact, examples of vernalization induction from donor to receptor, by way of a graft union are few in number. Most of these are listed by Lang in his 1952 Review (140). These are: henbane as a receptor for several donors, beets, cabbage scion upon annual Brassica or upon mustard stock, carrot upon dill and, more recently, late peas upon early peas and grafts of different strains of sweet William, but almost nothing more. In other cases, vernalization is strictly localized to the chilled part of the stem (47, 289). An hypothesis based on a mobile vernaline holds only for a few special cases.

CHEMICAL OR BIOCHEMICAL VERNALIZATION

Many authors understand "chemical vernalization" to mean the more or less synergic effect of some substance acting conjointly with the chilling of seeds to promote vernalization. In Brassica and in peas (in as much as vernalization here is truly a vernalization as defined earlier) indole-3-acetic acid synergizes chilling whereas gibberellin and 2,3,5-triiodobenzoic acid antagonize chilling (27, 150). See also the several cases of synergism between different chemicals and chilling reported by Séchet (239).

The hypothetical vernaline has never been extracted from plants (176), but some reports indicated that vernalized seeds may yield undefined substances exerting a vernalizing effect when used to soak unvernalized seeds. Purvis and Gregory reported that extracts from cereal seeds being processed for vernalization could replace the effect of chilling, but they have not reported again since 1953 (215). More recently, Highkin showed that if late pea seeds are chilled for a long time they yield water extracts that replace chilling for the vernalization of other peas or even for winter cereals (116, 117). Guanosine may be involved (118).

The problem of vernalization by chemicals has had to be reinvestigated completely since Lang's discovery of the effect of the prolonged gibberellin treatment of henbane (biennial strain). In the green house under long-day conditions, such treatment induces elongation and then flowering (142). This [222] report stimulated numerous experiments (15, 21, 43, 47, 153 to 156, 273, 280). The results are as follows. Most often gibberellin increases the elongation of the treated organ; caulescent stems become longer, "rosettes" elongate into leafy stems, etc. However, among plants requiring vernalization, few flower as a result of gibberellin treatment in the absence of chilling. Replacement of chilling by gibberellin is observed only among rosette plants mainly those flowering on elongating terminal buds such as: henbane, Centaurium minus, fox glove, Oenothera lamarkiana and parviflora (but not biennis), Reseda luteola (slightly), Beta maritima (but not several other Beta), some strains of Scabiosa succisa and Campanula medium (but not other strains), and a few species of Geum (intermedium, etc.). Likewise, gibberellin replaces the photoperiodic induction of flowering in some rosette plants Samolus parviflorus (146) and Polemonium coeruleum (43). In Scabiosa ukranica, under short-day conditions, gibberellin causes elongation and the formation of "inflorescence" primordia although not of "flower primordia." Usually, even repeated treatments will at most cause elongation from the rosette stage but without even initiation of inflorescence primordia (Scrofularia vernalis, S. alata, winter cereals). Frequently, elongation is restricted and results in either temporary or permanent "perched rosettes" (Geum urbanum, Campanula medium, sugar beets). In other plants gibberellin causes little (Dianthus) or no elongation (Saxifraga rotundifolia, S. cotyledon, Eryngium variifolium, etc.). Lastly, among caulescent plants, gibberellin has so far failed to replace the vernalization or photoperiodic requirements for flowering (Euphorbia lathyris, Teucrium scorodonia, etc.) (42, 43). Species vernalizable by seed chilling (Brassica campestris, Cicer, Lens, etc.) are not vernalizable by treatment of the seed with gibberellin (25).

PROBLEMS TO BE SOLVED

Having summarized what is already known about vernalization, we will now mention the many unsolved problems. The duration of the "juvenile period," i.e., the time elapsed before "ripeness to vernalization" is achieved must be determined in many cases. Most plants with an obligate vernalization requirement have embryos within the mature seed that cannot be vernalized (Scrofularia alata is an exception). This should be elucidated to find out whether chilling such seed effects a preliminary part of the vernalization process; also, a determination should be made as to when these plants actually become fully vernalizable.

Vernalization kinetics remain to be clarified. Further, the effects of overchilling should be investigated in as much as it may depress vernalization or conversely prolong the effectiveness unduly.

Anti-vernalization techniques should be elucidated in relation to the efficiency of chemicals (metabolic and growth inhibitors) about which we know very little (28, 48). In this regard, the efficiency of gibberellin as an [223] anti-vernalizant of pea, the effect of metabolic agents, deficiency of food such as sugar (99, 208, 210), and the role of physical agents {heating or leaching previous to chilling (112)], remain to be studied in more detail.

Devernalization should be "explored" in a number of plants which have not yet been studied from this point of view. The effect of devernalization treatments should be studied not only immediately after the vernalization process but subsequently, for instance as in the effect of short days on beets (164) or Oenothera (201, 202). This kind of devernalization is probably very different from the classical devernalization by heat or by anaerobiosis immediately after chilling. Alternating temperatures should be investigated as to whether they delay or accelerate the vernalizing process.

Among plants bearing leaves, the roles of intensity, duration, and quality of light during chilling appear important and deserve further study. Is light necessary just to supply sugar or does it have specific effect? We do not know for sure whether leaves formed before vernalization are or are not photoperiodically receptive. Data is still scanty as to the role of the root, which when tuberized, intervenes as a sugar donor.

Some data suggest that vernalization may proceed by successive stages that may differ both as to requirements and results, e.g., short chilling, unable by itself to promote vernalization, may become effective by an additional treatment with gibberellin, which by itself would not have been effective. This should be checked (148,201).

What has been termed "vernalization" in chrysanthemum, tomatoes, and peas, may differ from what is termed "vernalization" in henbane and Oenothera. Peas growing from nonchilled seeds produce, below the flowering nodes, axillary buds which tend to develop into abortive flowers. Had the seeds been chilled, flowers would differentiate; in that case, the "assumed vernalization" completes what otherwise would have only been initiated. Yet to be clarified is the chronological sequence of the initiation of inflorescence, the initiation of flowers, the various stages of elongation, and the possible interrelations among these events.

Perennial plants deserve special attention in relation to all the problems alluded to above as well as those peculiar to themselves. How can the perennial condition be preserved? The various alternative methods discussed above have been investigated only for a very few plants.

New researches are urgently needed on the total or partial replacement of "vernalizing chilling" by gibberellin, other chemical agents, extracts from vernalized plants, or by long or short days. The vernalizing effects of short days (apparently complete or partial) when followed by long days (Wellensiek's phenomenon) or in the absence of long days (Petkus rye, Iberis durandii) require more research. We know nothing about the possible similarity to the above of the different photoperiodic requirements in amphiperiodic plants.

Lastly, the biochemistry of vernalization should be investigated with [224] increased vigor. Emphasis should be given to the identification of natural stimulating or inhibiting agents intervening in the various phases of vernalization, rather than to assays involving known, metabolically active substances. The exchange of material between the nucleus and the cytoplasm should be taken into consideration in terms of modern physiological genetics as discussed below.

VERNALIZATION AND DORMANCY

"Dormancy" coincides with a cessation of growth (or auxesis) caused by internal factors within the organism or organ. The growth of any developing organ is arrested since the mechanisms controlling the increase in size are inhibited. Among such mechanisms is production or effectiveness of auxin. Release from dormancy is evidenced by renewed growth, and implies a resumption of previous activities rather than the introduction of new activities. During dormancy the frequency of mitosis decreases correlatively with the inhibition of growth, e.g., a twig of Salix repens growing under long-day conditions forms leaves and internodes; when transferred to short-day conditions the twig forms short internodes, the leaves are reduced to the condition of scales, and the terminal bud goes into dormancy; when subjected to cold, the twig recovers its former activity (38, 40).

The shortening of internodes, termed "brachyblasty," and expressed by the "rosette" configuration, is related to dormancy since it results from a shortening of internodes, but it differs from dormancy in not interfering with the formation and expansion of leaves (43). What is affected is the mechanism of cell elongation in internodes and of cell formation in between leaf bases, but the ring of cells around the apex, from which leaves originate, retains its activity and the enlargement of leaf primordia remains unretarded,

The "nonvernalized state" may be defined as the inability to form flowers and affects the potential activity of the summit of the apex and of terminal or axillary vegetative points. These retain an "anticipating meristematic" condition that is completely independent of the production or nonproduction of leaves or internodes.

Dormancy may be broken by submitting buds to treatments apt to release inhibitions. Among external stimuli, cold is the common one, but similar effects may be achieved using heat (applied as a hot bath), drying, long days, anesthesics, glycol monochlorhydrin, thiourea, gibberellin, or an exogenous auxin. Brachyblasty, when released, is normally followed by the formation of long internodes. The release can be effected by exposure to cold or long days, or by treatment with gibberellin. Other agents that break dormancy are generally ineffective. The nonvernalized state is removed by cold or gibberellin, or rarely by either long or short days. The effectiveness of cold, and less often of gibberellin, is a feature common to the three phenomena: breaking of dormancy, releasing of brachyblasty, and vernalization. [225] These three phenomena, however, can be differentiated in terms of the response of each to other agents.

Dormancy may not coincide with the nonvernalized state and may intervene as flowers are being differentiated in the bud, which then ceases growth and forms leaves. The breaking of dormancy in this case has nothing to do with vernalization. The example of Salix repens is most illuminating (38, 47). Axillary catkins form following photosynthetic activity in the leaves. Transfer from long to short days may then be arranged to induce dormancy either before or after the catkins have differentiated. Thus, breaking dormancy may not be required for differentiation of catkins, and may therefore be confounded with vernalization or sharply separated from it. These alternatives clearly distinguish the breaking of dormancy from vernalization.

Likewise, dormancy in the seeds of Geum may be broken by low temperatures or by excising the integuments: neither of these treatments causes vernalization of the seeds.

Brachyblasty is closely related to the nonvernalized state, although not always. Most plants requiring vernalization are of the rosette type. Cold has a dual effect: elongation of internodes, and vernalization. Thus, vernalization can be confused with release of the inhibition of internode elongation. However, although they may be concomitant, the two phenomena are distinct, as shown by the following comparisons.

1. Many plants that require vernalization are caulescent and never have a rosette phase; their vernalization is not related to the length of internodes [Euphorbia lathyris, Teucrium scorodonia (43, 47)].
2. Gibberellin may shift the brachyblastic form to the caulescent type, without causing any vernalization [Oenothera biennis (201), Scrophularia alata (148)].
3. Long days sometimes cause elongation of internodes from the rosette stage without effecting vernalization [Salvia silvestris (47)].
4. Flowering does not necessarily depend on the elongation of the flower-bearing stem; thus, Lavauxia (Oenothera) acaulis develops flowers while it is a rosette, as does Rudbeckia bicolor (with high temperatures and short days) (226).
5. Initiation of flowers may occur deep in the rosette with the flower stalk elongating later (Spinacia oleracea) (176), or the converse may occur [Oenothera sp., some strains of Scabiosa ukranica (47)].
6. In henbane, cold-induced vernalization first entails inflorescence initiation and then elongation, whereas the situation is reversed when the inducing agent is gibberellin (228).
7. In peach trees, seed dormancy is partially released with time if cold does not intervene, but the seedling can form only a rosette. Elongation is made possible following chilling or treatment with gibberellin (10). Later, in due time, flowering occurs and such a release of brachyblasty has been confused with vernalization. Flowering requires a proper equilibrium [226] of mild temperature and nutrition such as can be obtained with properly developed leaves. Then flowering may occur either on long twigs or on rosette-like twigs ("bouquets de Mai"). Whenever a nonvernalized state is concomitant with brachyblasty or the rosette stage, it may be said that vernalization requires at least as much, and frequently more, cold (or gibberellin) as would be required to release brachyblasty.

Vernalization and brachyblasty have common features but must be, considered as different phenomena unless we recognize various types or stages of vernalization. One such stage would be completely separate from the release of brachyblasty while the others might be strictly associated with its release. One or several stages would be needed according to the species, varieties, or even genotypes. Further research is needed to solve this problem.

ELUCIDATING THE INTIMATE PROCESS OF VERNALIZATION

Vernalization fundamentally implies: (a) an aftereffect resulting in active mitoses spreading to the very apex of the vegetative point, which then develops into the initial inflorescence or flower; and (b) irreversible, or reversible, autocatalytic preservation of the after effect.

Self-perpetuating new properties are expressed either through particulate release of DNA from the nucleus—to induce autocatalytic multiplication of cytoplasmic RNA particles as carriers of the new property—or through displacements of an equilibrium involving several pre-existing and self-perpetuating enzymes (68).

Cooling may control such a displacement of equilibrium while it slows down the rate of some enzymatic reactions and speeds up others; thereby triggering a cellular reaction which becomes autocatalytic. Cooling may even directly affect nuclear properties at the time it confers the capacity for the increased rate of mitoses. Gibberellin is endowed with properties similar in many respects to those of cold. It directly controls mitosis in primary meristems. This control is independent of that exerted on enlargement via auxins (216). Similar vernalizing properties of chilling and sometimes of gibberellin may be linked with similar properties affecting nuclear activity.

Assuming that flowering is induced by some interaction of nuclear DNA with an autocatalytic cytoplasmic system, flowering regulation would then be investigated either as a release of nuclear particles, or as an inhibition of stimulation of the proliferation of such particles once they had been released into the cytoplasm, or in the stimulation of inhibition of the activity of their products. Such effects might be, in some cases, mediated via mobile and diffusible hormonal compounds. Hereditary properties would be expected to intervene in forming stable or labile linkages between active particles and nuclei. The preceding discussion features vernalization as a self-perpetuating process involving certain "cellular lines" and focuses attention on some nucleo-cytoplasmic interrelationships. [227]

Heretofore, biochemical investigations have concentrated on the consequences rather than the causes of vernalization. New lines of research are now possible on nucleo-cytoplasmic relations based on information from the ecophysiological studies reviewed here. The available information permits a screening to select the material most likely to yield valuable information on the fundamental problems of organogenesis.

(See the Addendum following Literature Cited section.)

ADDENDUM

During the 9th International Botanical Congress, held in Montreal late in August of 1959, the problem of vernalization was dealt with during several sessions. We would like to report the data emphasized there and make some comments.

In winter cereals, the vernalization effect of chilling also acts on leafy plants as long as flower initiation has not occurred spontaneously without chilling (310).

The vernalization requirement appears quite variable in different races of red clover for example (311). As regards Oenothera, the differences between the author's findings and those of others resulted from the fact that, under the same name, they were speaking of different plants: the genetically controlled O. biennis, from the Munich Botanical Gardens (309) is not the same plant as the wild O. biennis from the Berlin area (307). Alternating temperatures have been confirmed as very effective and precise in the true Oenothera biennis (309).

Vernalization is effective on germinating seed of stock (301), endive (300), and chickory (297). In endive (302), after insufficient chilling, vernalization can be attained with short days, though the quickest flowering is caused by long days given as supplementary incandescent light.

The requirements for vernalization may vary according to the bud concerned: in Limum alpinum (293), the main shoot requires vernalization, but axilary shoots that appear after the head is removed have either low requirements or none at all.

Vernalizability varies with plant age; in Lunaria biennis (313), the seed does not seem vernalizable and yet it records a chilling effect that later appears in the smaller chilling requirement of the adult plant for vernalization. In the Oenothera of Berlin (307), vernalizability starts at 30 days of age and shows 2 maxima, one at 56 days and the other at 180 days. In Arabidopsis thaliana (308), a minimum vernalizability is observed at 45 days of age under low light, and at 7 days under intense light. Illumination during the vernalizing treatment increases the effectiveness of the treatment in plants less than 50 days old. When applied in that way, vernalization is nearly wholly fixed at the end of the treatment, whereas it requires a 4‑day fixation at 20°C., with light, when it has been applied in the dark.

Several plants, mostly crucifers, show the same phenomenon as recorded in Brussels sprouts: for example, in stock (301), chilling is not only a "preparatory" condition, but is also required for the realization of the process of flowering. The first flower buds can only form in the cold; then later, the others can keep on blooming in a warm temperature. Something similar occurs in Lunaria biennis (313) at 6‑8°C.

Whereas tubers are necessary for chilling efficiency in the beet and in carrot buds, leaves are useless, but they later become necessary for further flowering in a warm temperature even if tubers are present (303).

During germination of cereal embryos, indole‑3‑acetic acid increases the ascorbic acid production at the expense of sucrose, more so at a low temperature than at a high one, an observation supposedly related to the chilling vernalizing effect (292). However, when vernalization occurs below 0°C. (at ‑2°C., for example) mitoses completely stop (299).

The action of gibberellin has often been reported, extending or confirming what has been said above, either in general (295, 304, 306), in brachyblast shoots of a half‑dormant peach tree (298), in Oenothera (309), in Geum (305), or in various seeds that are vernalized by cold and not by gibberellin (291). Gibberellin is certainly not the direct cause either of flowering or of the production of flowering ability, though it may restrain flowering in certain cases, when no elongating occurs during which it would act. The various factors in vernalization may result from an inability for elongation (necessary to flowering), or from an inability for flowering, or from both (295). Czajlahjan (296) thinks gibberellin is the growth factor that is lacking under short‑day conditions in long‑day plants, that "anthesine" is the flowering factor that is lacking under longday conditions in short‑day plants, and that they both constitute the proper florigen. This tempting hypothesis cannot be applied to long‑day caulescent plants under short‑day conditions nor to caulescent biennials before chilling, in which gibberellin cannot ever cause flowering (295).

A few more suggestions were made (294 to 312) concerning the above work schedule. It was supported in its general form and it seems likely that in about 10 years, through one or another of the suggested ways, that the study of vernalization will make considerable progress.