Introduction
Although several seed dormancy classification schemes have been proposed (e.g. Crocker, Reference Crocker1916; Harper, Reference Harper1957), the one by the Russian seed physiologist M.G. Nikolaeva (Nikolaeva, Reference Nikolaeva1967 [1969], Reference Nikolaeva and Khan1977, Reference Nikolaeva2001; Nikolaeva et al., Reference Nikolaeva, Razumova, Gladkova and Danilova1985, Reference Nikolaeva1999) is by far the most logical and comprehensive. Her system is the only one that uses both names and formulae (or symbols) to designate the different types (sensu Nikolaeva) of dormancy. Baskin and Baskin (Reference Baskin and Baskin2004) published a modified version of the Nikolaeva scheme. They suggested that this scheme could form the basis for an international system of seed dormancy classification, with the realization that it probably would need to be revised/updated from time to time. However, they did not use the Nikolaeva scheme of dormancy formulae, which she has revised several times since first publishing it in 1967 (Nikolaeva, Reference Nikolaeva1967 [1969]). Thus, the purpose of this paper is to: (1) correlate terminology for the types (sensu Nikolaeva) of dormancy and their formulae in the five versions of Nikolaeva's scheme to each other and to the kinds (classes, levels, types) of dormancy in her modified system (sensu Baskin and Baskin, Reference Baskin and Baskin2004); and (2) suggest the adoption of the Nikolaeva (Reference Nikolaeva2001) formulae for the seed dormancy classification system outlined by Baskin and Baskin (Reference Baskin and Baskin2004), with the realization that the scheme may need some further refinements.
Evolution and modification of the Nikolaeva scheme
Terms and formulae used by Nikolaeva in various publications (Nikolaeva, Reference Nikolaeva1967 [1969], Reference Nikolaeva and Khan1977, Reference Nikolaeva2001; Nikolaeva et al., Reference Nikolaeva, Razumova, Gladkova and Danilova1985, Reference Nikolaeva, Lyanguzova and Pozdova1999) for describing what she called types of dormancy, along with terms for classes, levels and types of dormancy suggested by Baskin and Baskin (Reference Baskin and Baskin2004), are presented in Table 1. A major difference in the Nikolaeva scheme and the Baskin and Baskin modification of it is that, whereas Nikolaeva recognizes chemical and mechanical dormancy as distinct types, Baskin and Baskin do not. Nikolaeva (Reference Nikolaeva2004) agreed that ‘… exclusion of mechanical and chemical dormancy from the Nikolaeva classification scheme is probably correct’. Thus, since no seeds with a water-impermeable seed coat (physical dormancy) are known to have an underdeveloped embryo (i.e. a morphological component of dormancy) (Baskin et al., Reference Baskin, Baskin and Li2000), the only possible major combinational (i.e. exogenous type combined with endogenous type, sensu Nikolaeva, see Table 1) category remaining is between physical dormancy (PY) and physiological dormancy (PD), thus class (PY+PD) of Baskin and Baskin (Reference Baskin and Baskin2004).
Comparison of the various versions of Nikolaeva's seed dormancy terminology and formulae with each other and with the dormancy terminology of Baskin and Baskin (Reference Baskin and Baskin2004). Names for types (sensu Nikolaeva) of dormancy in Nikolaeva et al. (Reference Nikolaeva, Razumova, Gladkova and Danilova1985, Reference Nikolaeva, Lyanguzova and Pozdova1999) are the same as those in Nikolaeva, Reference Nikolaeva and Khan1977 and/or Reference Nikolaeva2001; thus, they are not repeated
a These subtypes are not recognized in Nikolaeva (Reference Nikolaeva1967 [1969], Reference Nikolaeva and Khan1977) nor in Nikolaeva et al. (Reference Nikolaeva, Razumova, Gladkova and Danilova1985, Reference Nikolaeva, Lyanguzova and Pozdova1999).
b No names given for these five subtypes in Nikolaeva (Reference Nikolaeva1967 [1969]).
c These types are not recognized in Nikolaeva (Reference Nikolaeva1967 [1969]).
d This type is not recognized in Nikolaeva (Reference Nikolaeva and Khan1977).
e Classes of dormancy.
f Levels of dormancy.
Nearly 150 species in the Nikolaeva et al. (Reference Nikolaeva, Razumova, Gladkova and Danilova1985) compendium of the types (sensu Nikolaeva) of dormancy in seeds of >2000 species of gymnosperms and angiosperms are indicated to have combined dormancy. These include, for example, chemical dormancy combined with non-deep PD (A1-B1), mechanical dormancy combined with deep PD (A2-B3), mechanical dormancy combined with deep simple morphophysiological dormancy (MPD) (A2-Б-B3), mechanical dormancy combined with deep complex MPD (A2-БB-B3) and chemical dormancy combined with intermediate complex MPD (A1-БB-B2). Further, in a survey of the types of seed dormancy in about 170 families of angiosperms and eight families of gymnosperms, Nikolaeva (Reference Nikolaeva1988, Reference Nikolaeva1989a, Reference Nikolaevab, Reference Nikolaeva1990) indicated that combined dormancy is represented in about 70 families of angiosperms and in four families of gymnosperms. If it is accepted that chemical and mechanical dormancy are not dormancy types per se, then it would appear that the kind of dormancy could be correctly identified simply by removing A1 (chemical dormancy) and A2 (mechanical dormancy) from the formulae. Thus, in the above examples: A1-B1 becomes B1 (non-deep PD); A2-B3, B3 (deep PD); A2-Б-B3, Б-B3 (deep simple MPD); A2-БB-B3, БB-B3 (deep complex MPD); and A1-БB-B2, БB-B2 (intermediate complex MPD). See Table 1 in Nikolaeva (Reference Nikolaeva1988) for the various combinations possible in an exogenous types of dormancy × endogenous types of dormancy matrix.
Two of the Nikolaeva (Reference Nikolaeva1967 [1969], Reference Nikolaeva and Khan1977) papers on dormancy classification are in English, making the information in them easily accessible to most seed scientists, whereas her other three papers on dormancy classification listed in Table 1 (Nikolaeva, Reference Nikolaeva2001; Nikolaeva et al., Reference Nikolaeva, Razumova, Gladkova and Danilova1985, Reference Nikolaeva, Lyanguzova and Pozdova1999) are in Russian, which means that the information probably is not easily available to the great majority of seed scientists. However, her most recent paper on seed dormancy classification (Nikolaeva, Reference Nikolaeva2001) has been translated into English, and Nikolaeva sent J. and C. Baskin a copy of the translation (manuscript). In the 2001 paper, her symbols for physical (Aph), chemical (Ach), mechanical (Am) and morphological (B) dormancy are unchanged from the 1977 paper; combined dormancy is not discussed. However, formulae for the kinds of PD and of MPD have been revised considerably (Table 1). Thus, using the English translation of the 2001 paper, we will briefly summarize the differences between the 1977 and 2001 formulae, along with some of the reasoning Nikolaeva (Reference Nikolaeva2001) used for making these changes.
As she did in her 1977 paper, Nikolaeva (Reference Nikolaeva2001) used C1, C2 and C3 to represent non-deep, intermediate and deep PD, respectively. However, based on the fact that seeds of some species (i.e. summer annuals, many perennials) require cold stratification to come out of dormancy in nature, while others (i.e. winter annuals, some perennials) require warm stratification (or a warm afterripening period) to do so, Nikolaeva (Reference Nikolaeva2001) recognized two subtypes of non-deep PD (Table 1): subtype a, designated C1a (cold stratification, dry storage or GA breaks dormancy); and subtype b, designated C1b (warm stratification, dry storage or perhaps GA breaks dormancy). Further, C1a contains the Baskin and Baskin (Reference Baskin and Baskin2004) Type 2 (not Type 1, as Nikolaeva indicated) and C1b Types 1 (not Type 2 as Nikolaeva indicated), 3, 4 and 5.
Nikolaeva (Reference Nikolaeva2001) states that seeds with the various types (sensu Nikolaeva) or levels (sensu Baskin and Baskin, Reference Baskin and Baskin2004) of MPD come out of dormancy in three stages: (1) PD is broken in the underdeveloped embryo (C1a and/or C1b, C2 or C3); (2) the underdeveloped embryo grows and becomes fully developed inside the seed, what Nikolaeva called embryo postdevelopment; and (3) PD is broken in the seed that now has a fully developed embryo. To make formulae for the five kinds of simple MPD (‘simple’ indicating that the underdeveloped embryo grows at warm temperatures, see Baskin and Baskin, Reference Baskin and Baskin2004) more precisely describe the causes of dormancy, symbols representing the physiological component of dormancy (C1b or C3) were added to the five kinds of dormancy in this category of MPD, and C1 [in non-deep PD (B-C1)] was replaced by C1b (Nikolaeva, Reference Nikolaeva2001, Table 1). For example, non-deep simple MPD [not recognized by Nikolaeva (Reference Nikolaeva and Khan1977), but recognized by Nikolaeva et al. (Reference Nikolaeva, Razumova, Gladkova and Danilova1985, Reference Nikolaeva, Lyanguzova and Pozdova1999) and given the formula Б-B1, which is equivalent to B-C1, see Table 1] in the winter annual Chaerophyllum tainturieri (Baskin and Baskin, Reference Baskin and Baskin1990) is described by the formula C1bB-C1b, which reads as follows. (1) The fresh seeds have underdeveloped embryos with non-deep PD of the C1b subtype, thus requiring warm stratification for dormancy-break; (2) following breaking of non-deep PD (C1b), the underdeveloped embryo, now with only morphological dormancy (B), grows to full size (postdevelopment) at warm temperatures (warm stratification) if exposed to a light stimulus (see below); and (3) the seed, now with a fully developed embryo, germinates at warm temperatures (-C1b).
Another sequence of environmental factors exists for seeds of species with non-deep simple MPD that require exposure to low temperatures to come out of dormancy. An example is Thalictrum mirabile (Walck et al., Reference Walck, Baskin and Baskin1999). In contrast to the warm-temperature requirement for dormancy-break (C1b) in the underdeveloped embryos (B) of C. tainturieri, those of T. mirabile require cold stratification to come out of PD (C1a). However, like the underdeveloped embryos of C. tainturieri, those of T. mirabile undergo postdevelopment at warm temperatures (B), although in spring rather than in autumn, and the seed with a fully developed embryo requires warm temperatures to germinate. Thus, the formula for dormancy in seeds of T. mirabile is C1aB-C1b.
A more complicated kind of deep simple MPD is deep simple epicotyl MPD, such as occurs in seeds of Hydrophyllum spp. and in many species of Viburnum. The formula for this kind of dormancy in Nikolaeva (Reference Nikolaeva and Khan1977) is B-C3e, but based on the requirement for a warm pretreatment (C1b) to break dormancy of the radicle, it was changed to C1bB(root)-C3(epicotyl) in Nikolaeva (Reference Nikolaeva2001). Thus, the revised formula reads as follows: a warm pretreatment is required for dormancy-break, growth and radicle emergence of an underdeveloped embryo [C1bB(root)] whose epicotyl is in deep PD [-C3(epicotyl)], requiring a long period of cold stratification to grow. However, although not included in her classification scheme, Nikolaeva recognized that the epicotyl in seeds of some species may be in intermediate or non-deep PD, thus C1bB(root)-C2(epicotyl) and C1bB(root)-C1a [or C1b] (epicotyl), respectively.
A case in point is a study by Karlsson et al. (Reference Karlsson, Hidayati, Walck and Milberg2005), who concluded that Viburnum tinus ‘… does not have epicotyl dormancy’ (p. 323, abstract) and ‘… does not have deep simple epicotyl morphophysiological dormancy’ (p. 329, text). Even at the most favourable temperature regime for germination (20/10°C), root emergence was delayed for a minimum of 10–12 weeks, suggesting some physiological dormancy. Further, depending on temperature regime, including that simulated for the natural habitat of V. tinus, there was about a 6–15 week delay in epicotyl growth in seeds with emerged radicles. Thus, the phenology of germination fits very well that of epicotyl dormancy, as Karlsson et al. (Reference Karlsson, Hidayati, Walck and Milberg2005) acknowledge. Neither radicles nor epicotyls emerged at constant 5°C, and cold stratification was not required to break dormancy in epicotyls in seeds with radicles that had emerged at the higher temperature regimes. While we agree that seeds of V. tinus do not have deep simple epicotyl MPD, we do not agree that they lack epicotyl dormancy. Thus, based on the results of Karlsson et al. (Reference Karlsson, Hidayati, Walck and Milberg2005), we suggest that seeds of V. tinus have non-deep simple epicotyl MPD, i.e. C1bB(root)-C1b (epicotyl). One also wonders if the level of PD in seeds with double dormancy (e.g. Trillium spp.), designated by Nikolaeva as deep simple double MPD [C3B(root)-C3(epicotyl)], also could be non-deep (C1a) or intermediate (C2) simple double MPD.
However, seeds of at least some ‘white oaks’ (Quercus subgenus Leucobalanus) do not require a warm pretreatment for radicle emergence. In this group, seeds are dispersed in autumn, and shortly thereafter the radicle emerges. Otherwise, the recalcitrant seeds die upon drying. We suggest the formula for this kind of epicotyl dormancy (sensu Baskin and Baskin, Reference Baskin and Baskin1998) should be Cnd(root)-C3(epicotyl), signifying that the embryo is fully developed (i.e. no B), the root is non-dormant (i.e. Cnd) and the epicotyl is deeply dormant, requiring a long period of cold stratification to come out of dormancy [-C3(epicotyl)].
In the three kinds of complex MPD, both dormancy-break and postdevelopment (growth) of the underdeveloped embryo occur at cold temperatures (indicated by ‘complex’, see Baskin and Baskin, Reference Baskin and Baskin2004). Then, after the embryo has grown to full size (i.e. completed postdevelopment), the seed, now with a fully developed embryo, requires cold stratification to come out of dormancy (C1a, C2 or C3). Thus, Nikolaeva (Reference Nikolaeva2001) replaced the general symbol for physiological dormancy (C) in the first part of the formulae (BC) with symbols for the levels of PD, i.e. C1a, C1b, C2 or C3 and reversed the order of B (representing an underdeveloped embryo) and C (Table 1). The only other change in complex MPD between the 1977 and 2001 papers was replacement of C1 in the second part of the formula for non-deep complex MPD with C1a.
Nikolaeva (Reference Nikolaeva2001) discussed seeds of Osmorhiza longistylis as an example of non-deep complex MPD (Baskin and Baskin, Reference Baskin and Baskin1984). She did not have this category of dormancy in her 1977 paper. However, Nikolaeva et al. (Reference Nikolaeva, Razumova, Gladkova and Danilova1985) did have it in their scheme, as БB-B1, which is equivalent to BC-C1. Nevertheless, the formula was revised to C1a1bB-C1a in Nikolaeva (Reference Nikolaeva2001). It indicates that dormancy-break in the underdeveloped embryo requires warm and cold temperatures and that after growth of the embryo is complete, dormancy-break in the seed (now with a fully developed embryo) requires an additional short period of cold temperatures (the second C1a, stage 3). However, we suggest that since seeds with this kind of dormancy, e.g. O. longistylis, O. claytonii and some Erythronium species (Baskin and Baskin, Reference Baskin and Baskin1998), need a short period of warm (C1b) followed by a period of cold (C1a) temperatures for embryo dormancy-break and postdevelopment, the formula should be C1b1aB-C1a, rather than C1a1bB-C1a.
Frasera caroliniensis is an example of a species whose seeds have deep complex MPD. The formula for this kind of dormancy was changed from BC-C3 (Nikolaeva, Reference Nikolaeva and Khan1977) to C3B-C3 (Nikolaeva, Reference Nikolaeva2001). C3B indicates that the underdeveloped embryo (B) is in deep PD (C3), requiring a long period of cold stratification for dormancy-break and growth. The second C3 (i.e. -C3) indicates that the seed (now with a fully developed embryo) requires an additional long period of cold stratification in order to germinate. Thus, without any pretreatment, seeds of F. caroliniensis will germinate at 5°C after a long period of time at this temperature (Threadgill et al., Reference Threadgill, Baskin and Baskin1981).
Concluding remarks
There are several advantages in adopting a symbol/formula system into seed dormancy classification. Thus, assignment of a symbol or formula to each class, level and type of dormancy would: (1) be a step in the right direction toward standardizing a seed dormancy classification system for use by the international seed biology community; (2) improve accuracy in describing dormancy, thus allowing the reader to understand more precisely what kind of dormancy is being discussed and how to break it; (3) more clearly indicate the interrelationships among the various kinds of dormancy; (4) help organize information on seed dormancy and communicate it in a standard way (e.g. Pendry et al., Reference Pendry, Dick, Pullan, Knees, Miller, Neale and Watson2007), especially among those who speak different languages; and (5) allow ready access to the kinds of dormancy in seeds of hundreds of plant taxa, compiled by Nikolaeva and her colleagues, to those of us who do not read the Russian language. Certainly a detailed dormancy classification system acceptable to the international seed biology community is essential in describing ecological/biogeographical patterns and phylogenetic relationships of the diverse kinds of seed dormancy (Baskin and Baskin, Reference Baskin and Baskin1998, Reference Baskin, Baskin, Smith, Dickie, Linnington, Pritchard and Probert2003; Nikolaeva, Reference Nikolaeva1999, Reference Nikolaeva2001). Without doubt, a set of dormancy formulae would greatly facilitate description of these patterns and relationships, especially at the level and type layers in the hierarchy of a seed dormancy classification system (sensu Baskin and Baskin, Reference Baskin and Baskin2004). Thus for now, we suggest that the appropriate symbols and formulae of Nikolaeva (Reference Nikolaeva2001) be used with the corresponding kinds (classes, levels and types) of dormancy suggested by Baskin and Baskin (Reference Baskin and Baskin2004) (Table 1).
Undoubtedly, however, this set of formulae will require revision as seed scientists learn more about the various kinds of dormancy via experimentation and logic, and this is especially true for MPD. For example, using two of the cases discussed above, i.e. non-deep simple MPD in C. tainturieri and deep complex MPD in F. caroliniensis, it seems that justification can be made that dormancy could be more clearly represented by modifying the Nikolaeva (Reference Nikolaeva2001) formulae for these types (sensu Nikolaeva). In both examples, Nikolaeva's stages 2 and 3 of dormancy-break in seeds with MPD (Nikolaeva, Reference Nikolaeva2001, see above) appear to represent a continuum of events that occur without a change in the temperature regime. Thus, perhaps the hyphen between C1bB-C1b (C. tainturieri) and that between C3B-C3 (F. caroliniensis) should be replaced by an arrow, becoming C1bB → C1b and C3B → C3, respectively. Another option might be to drop the second part of the Nikolaeva formulae. In which case, the dormancy formula for C. tainturieri would be C1bB and that for F. caroliniensis C3B. C1bB and C3B would indicate that all three stages of dormancy-break in seeds with these levels of MPD take place during continuous warm stratification (C. tainturieri) or during continuous cold stratification (F. caroliniensis).
These one-part formulae would also show the interrelationship between PD and MPD. Thus, for example, the dormancy formula for the winter annual race of Arabidopsis thaliana, with a fully developed embryo and non-deep PD, is C1b (sensu Nikolaeva, Reference Nikolaeva2001, Table 1), indicating that seeds need only warm stratification [or warm-dry storage (afterripening)] to come out of dormancy (Baskin and Baskin, Reference Baskin and Baskin1983). The C1bB formula for dormancy in seeds of C. tainturieri, a winter annual with an underdeveloped embryo (B) and PD (C1b), also indicates that warm stratification only is required for seed dormancy-break. Likewise, the one-part formula for non-deep simple MPD shows the relationship between morphological dormancy (MD) and non-deep simple MPD. Thus, the formula for MD is B (Table 1), indicating that the seed requires only warm stratification to grow, and that for non-deep MPD is C1bB, indicating that the seed has non-deep PD (C1b) in addition to MD (B).
However, neither the two-part (C1bB-C1b) nor the one-part (C1bB) formula for seed dormancy in C. tainturieri shows that following dormancy (stage 1 of MPD, sensu Nikolaeva, Reference Nikolaeva2001), the (now) morphologically dormant embryo (B) requires a light stimulus to grow, i.e. to undergo postdevelopment (stage 2). Otherwise, the non-dormant underdeveloped embryo re-enters dormancy [i.e. non-deep PD (C1b)], and thus the seed goes from MD back to MPD. In other words, as long as the seed remains viable the underdeveloped embryo can cycle between non-deep PD and non-dormancy until the seed is exposed to light at the proper time of the year. Then, postdevelopment of the embryo and germination occur (Baskin and Baskin, Reference Baskin and Baskin1990). Using the one-part formula for seed dormancy, the dynamics of dormancy and germination in C. tainturieri are summarized in Fig. 1. Obviously then, further thought needs to be given to representation of the light requirement for embryo postdevelopment in a formula for those MPD seeds that need this stimulus to become fully non-dormant.
Summary of the dynamics of dormancy and germination in Chaerophyllum tainturieri using the one-part formula for seed dormancy. MPD, Morphophysiological dormancy; MD, morphological dormancy; ND, non-dormant; hν, light.
Finally, although seed biologists have been slow in adopting the Nikolaeva dormancy classification scheme, which was first published in English in 1969 (Nikolaeva, Reference Nikolaeva1967 [1969]), there is quite a bit of recent evidence that they are beginning to use her terminology (e.g. Forbis and Diggle, Reference Forbis and Diggle2001; Forbis et al., Reference Forbis, Floyd and de Querioz2002; Walck et al., Reference Walck, Hidayati and Okagami2002; Wen et al., Reference Wen, Lowry, Walck and Yoo2002; Thompson et al., Reference Thompson, Ceriani, Bakker and Bekker2003; Tweedle et al., Reference Tweedle, Dickie, Baskin and Baskin2003; Walck and Hidayati, Reference Walck and Hidayati2004; Finch-Savage and Leubner-Metzger, Reference Finch-Savage and Leubner-Metzger2006; Kondo et al., Reference Kondo, Sato, Baskin and Baskin2006). We hope that our proposed refinements to the terminology and its notation will enhance the utility and widespread adoption of the Nikolaeva formula system.
