We have reviewed seed dormancy and germination in the Rubiaceae, the fourth-largest angiosperm family (in terms of species richness), in relation to ecology, life form, biogeography and phylogeny (subfamily/tribe). Life forms include trees, shrubs, vines and herbs, and tropical rainforest trees have the greatest number of tribes and species. The family has five kinds of embryos: investing, linear-full, linear-underdeveloped, spatulate and spatulate-underdeveloped, and seeds are non-dormant (ND) or have morphological (MD), morphophysiological (MPD) or physiological (PD) dormancy. Except for the occurrence of the investing embryo only in dry fruits of Dialypetalanthoideae, each kind of embryo is found in dry and fleshy fruits of Dialypetalanthodies and of Rubioideae. In tropical and temperate regions, there are species with ND seeds and others whose seeds have MD, MPD or PD. A complete seed dormancy profile (i.e. some species with ND seeds and others whose seeds have MD, MPD or PD) was found for tropical rainforest trees and shrubs and semi-evergreen rainforest shrubs. Dormancy-break occurs during cold or warm stratification or dry-afterripening, depending on the species. Some tropical species have long periods of dormancy-break/germination extending for 4–5 to 30–40 weeks. Soil seed banks are found in 5 and 15 tribes of Rubiaceae in tropical and temperate regions, respectively. With increased distance from the Equator, diversity of life forms and seed dormancy decreases, resulting in only herbs with PD at high latitudes. We conclude that the low species richness of Rubiaceae in temperate regions is not related to low diversity of seed dormancy/germination.

‘Seed heteromorphism’ is a broadly- and loosely-defined term used to describe differences in size/mass, morphology, position on mother plants and ecological function (e.g. dispersal, dormancy/germination) of two or more seeds or other diaspores produced by an individual plant. The primary aim of this review paper was to characterize via an in-depth classification scheme the physical structural design (‘architecture’) of diaspore monomorphism and diaspore heteromorphism in angiosperms. The diaspore classification schemes of Mandák and Barker were expanded/modified, and in doing so some of the terminology that Zohary, Ellner and Shmida, and van der Pijl used for describing diaspore dispersal were incorporated into our system. Based on their (relative) size, morphology and position on the mother plant, diaspores of angiosperms were divided into two divisions and each of these into several successively lower hierarchical layers. Thus, our classification scheme, an earlier version of which was published in the second edition of ‘Seeds’ by Baskin and Baskin, includes not only heteromorphic but also monomorphic diaspores, the Division to which the diaspores of the vast majority of angiosperms belong. The scheme will be useful in describing the ecology, biogeography and evolution of seed heteromorphism in flowering plants.

Small isolated plant populations are one of the consequences of fragmentation of natural habitats by humans. We asked what effect does the creation of smaller populations from larger ones has on the plant fitness-related trait seed germination. Using information on 119 species (142 species entries) in 50 families, we found that seeds in only 35.2% of the species entries from larger populations germinated to higher percentages than those from smaller populations. In the other entries, seeds from large and small populations germinated equally well (57.7% of total entries) or seeds from small populations germinated better (7.0% of total entries) than those from large populations. These results indicate that population size is not a reliable predictor of seed germinability. Furthermore, there was little relationship between seed germination and either seed mass, genetic diversity or degree of population isolation, or between population size and genetic diversity.

In this review, we explore the origin of the rudimentary embryo, its relationship to other kinds of plant embryos and its role in the diversification of angiosperms. Rudimentary embryos have a length:width ratio of ≤2.0, and they have organs, including cotyledon(s) and a primary root. A literature survey failed to reveal rudimentary embryos in the pre-angiosperms, suggesting that this kind of embryo is an angiosperm invention. Although proembryos of some gymnosperms and angiosperms have a length:width ratio of ≤2.0, they have not formed meristems or organs. Thus, rudimentary embryos are not proembryos. During the development of rudimentary embryos in monocots and dicots (all non-monocots), the growth pattern of the epicotyledonary cells differs, resulting in differences in the placement of the shoot meristem and in one versus two cotyledons, respectively, but the embryo size is similar. Rudimentary embryos grow inside the seed prior to germination, which is true for linear-underdeveloped embryos, including those in some gymnosperms. Rudimentary embryos served as the starting point for the great diversification of embryos, and ultimately of seeds, in angiosperms, and they are still present in many families of extant angiosperms. The rudimentary embryo is part of the syndrome of changes, including increased speed of pollen germination and pollen tube growth, simplification of the female gametophyte, development of endosperm and elimination of multiple embryo production from each zygote, that distinguish angiosperm seed production from that of gymnosperms. We conclude that the rudimentary embryo was one of many new developments of angiosperms that contributed to their great success on earth.

The Asteraceae with up to 30,000 species occurs on all continents except Antarctica and in all major vegetation zones on earth. Our primary aim was to consider cypselae dormancy-break and germination of Asteraceae in relation to ecology, vegetation zones and evolution. Cypselae are desiccation-tolerant and in various tribes, genera, species and life forms of Asteraceae are either non-dormant (ND) or have non-deep physiological dormancy (PD) at maturity. All six types of non-deep PD are found among the Asteraceae, and dormancy is broken by cold or warm stratification or by afterripening. Soil cypselae banks may be formed but mostly are short-lived. Much within-species variation in dormancy-break and germination has been found. Using data compiled for 1192 species in 373 genera and 35 tribes of Asteraceae, we considered ND and PD in relation to life form, vegetation zone and tribe. Senecioneae and Astereae had the best representation across the vegetation zones on earth. In evergreen and semi-evergreen rainforests, more species have ND than PD, but in all other vegetation zones, except alpine/high-latitude tundra (where ND and PD are equal), more species have PD than ND. Tribes in the basal and central grades and those in the Heliantheae Alliance have both ND and PD. The high diversity and lability of non-deep PD may have enhanced the rate of species diversification by promoting the survival of new species and/or species in new habitats that became available following globally disruptive events since the origin of the Asteraceae in the Late Cretaceous.

Dormancy-breaking requirements and level of morphophysiological dormancy (MPD) were determined for seeds of Ilex formosana and I. uraiensis from the subtropical region, and seeds of I. rotunda from both the subtropical and tropical regions of Taiwan. We hypothesized that some Ilex species would have deep simple MPD broken by warm stratification. Germination of seeds and embryo growth was monitored at 30/20, 25/15, 20/10, 15/5°C and at 25°C. Seeds were cold-stratified and then incubated at 25/15°C, and seeds treated with GA3 and GA4 were incubated at 25/15°C. Fresh seeds reached 50% germination after 11–45 weeks of warm stratification. Pre-treatment with GA increased germination percentages of I. formosana and I. rotunda (tropical) but not I. uraiensis and I. rotunda (subtropical), while cold stratification did not promote germination of either species but increased the germination rate of I. rotunda (tropical). Embryo length in seeds of all species increased ≥710% prior to root emergence, and growth occurred during warm stratification. The positive response to GA and relatively short time for beginning of germination and to reach 50% germination indicate non-deep simple MPD in seeds of I. formosana and I. rotunda (tropical). The negative response to GA and long time for beginning of germination and to reach 50% germination indicate deep simple MPD in seeds of I. uraiensis and I. rotunda (subtropical). Thus, in both the subtropical and tropical regions of Taiwan, the seeds of Ilex species have non-deep simple and deep simple MPD that are broken by warm stratification. Furthermore, GA treatment increases the germination rate and percentage of Ilex seeds with non-deep simple MPD, and cold stratification promotes the seed germination rate of Ilex species with non-deep simple MPD in tropical region.

Information in the literature and unpublished results of the authors on Dobinea were used to determine the kind [class(es)] of seed (true seed + endocarp) dormancy and of non-dormancy of genera in all five tribes of Anacardiaceae, and the results were examined in relation to the taxonomic position and endocarp anatomy within the family. Reports of both seed germination and endocarp anatomy were found for 15 genera in tribe Spondiadeae, 6 in tribe Anacardieae, 30 in tribe Rhoeae, 3 in tribe Semecarpeae and 1 in tribe Dobineeae. In Spondiadeae (Spondias-type endocarp), Anacardieae, Semecarpeae and Dobineeae (Anacardium-type endocarp), seeds are either non-dormant (ND) or have physiological dormancy (PD). In Rhoeae (Anacardium-type Rhoeae Groups A, B, C and D endocarps), on the other hand, seeds are ND or have physical dormancy (PY), PD or PY + PD. PY/PY + PD in this tribe seems to be restricted (or nearly so) to Rhus s.s. and closely related genera (e.g. Cotinus, Malosma and Toxicodendron) with an Anacardium-type Rhoeae Group A endocarp. However, seeds of other genera (e.g. Astronium and Schinus) with this type of endocarp and those with Rhoeae Group B (e.g. Pistacia), Group C (e.g. Pentaspadon) and Group D (e.g. Heeria) endocarps are either ND or have PD. The fossil fruit record strongly suggests that present-day relationships between diaspore dormancy (or non-dormancy), endocarp structure and taxonomic position within Anacardiaceae extend back to at least the Palaeogene.

Germination of seeds of some summer annuals in Kentucky (eastern USA) in late-winter lead to the hypothesis that under present climate conditions the whole length of the winter cold stratification (CS) period is not required for dormancy-break of seeds of summer annuals with physiological dormancy (PD). We evaluated our data from germination phenology studies of 45 species (69 datasets) and buried-seed studies of 33 species (44 datasets). We determined time and temperature of germination after CS and percentage of the total number of hours of CS during winter (% of winter CS) seeds received prior to start of germination. In the phenology studies, mean temperature during the week of first germination for C3 and C4 species was 11.1 and 14.4°C, respectively, and % of winter CS was 80.8 and 87.4, respectively. In the buried-seed studies, % of CS for C3 and C4 species was 40.8 and 48.1, respectively, when they germinated to 25% at 20/10°C. For 32 of 33 species in the buried-seed studies, the minimum temperature at which seeds germinated decreased with increased CS; thus, seeds had Type 2 non-deep PD. The time of germination is controlled by a number of hours of CS, a decrease in minimum temperature at which seeds can germinate and a temperature increase in early spring. Seeds can germinate at relatively high temperatures as early as December and January, but they continue to be CS until spring. Temperature increases in eastern North America due to global warming are not likely to inhibit the germination of summer annuals with PD in spring.

To persist (without immigration) in habitats with unpredictable environmental conditions, annuals must produce seeds each year or have a seed bank. Thus, we predicted that compared to perennials, annuals have a wider germination temperature range (GTR, the difference in temperature between the week with the highest and the week with the lowest germination during the natural germination season). We determined the GTR via germination phenology data for 350 herbaceous species in 59 families from the eastern USA: summer annuals (SA), 63; winter annuals (WA), 83; monocarpic perennials (MP), 28; and polycarpic perennials (PP), 176. There was no significant phylogenetic signal for the GTR. The width of the GTR during the first spring germination season was 9.6, 8.7 and 8.8°C for MP, PP and SA, respectively, and during the first autumn germination season 12.8, 11.8 and 12.4°C for MP, PP and WA, respectively. Annuals did not have a wider GTR than perennials in either the spring or the autumn germination season. Our data suggest that selection for early germination in either spring or autumn has resulted in only small differences in the GTR. We predict that global warming will have little or no effect on reshaping the germination phenology of herbaceous species of temperate eastern North America.

This review provides a revised and expanded word-formula system of whole-seed primary dormancy classification that integrates the scheme of Nikolaeva with that of Baskin and Baskin. Notable changes include the following. (1) The number of named tiers (layers) in the classification hierarchy is increased from three to seven. (2) Formulae are provided for the known kinds of dormancy. (3) Seven subclasses of class morphological dormancy are designated: ‘dust seeds’ of mycoheterotrophs, holoparasites and autotrophs; diaspores of palms; and seeds with cryptogeal germination are new to the system. (4) Level non-deep physiological dormancy (PD) has been divided into two sublevels, each containing three types, and Type 6 is new to the system. (5) Subclass epicotyl PD with two levels, each with three types, has been added to class PD. (6) Level deep (regular) PD is divided into two types. (7) The simple and complex levels of class morphophysiological dormancy (MPD) have been expanded to 12 subclasses, 24 levels and 16 types. (8) Level non-deep simple epicotyl MPD with four types is added to the system. (9) Level deep simple regular epicotyl MPD is divided into four types. (10) Level deep simple double MPD is divided into two types. (11) Seeds with a water-impermeable seed coat in which the embryo-haustorium grows after germination (Canna) has been added to the class combinational dormancy. The hierarchical division of primary seed dormancy into many distinct categories highlights its great diversity and complexity at the whole-seed level, which can be expressed most accurately by dormancy formulae.

Martin placed the lateral embryo, which occurs only in grasses, adjacent to the broad embryo at the base of his family tree of seed phylogeny. Since Poales and Poaceae are derived monocots, we questioned the evolutionary relationship between the lateral embryo and other kinds of monocot embryos. Information was compiled on embryo and seed characteristics for the various families of monocots, kind of embryogenesis for families in Poales and germination morphology of families with lateral (only Poaceae) and broad embryos. The kinds of monocot embryos are broad, capitate, lateral, linear fully developed, linear underdeveloped and undifferentiated, but only broad and lateral embryos are restricted to Poales. Asterad embryogenesis occurs in Poaceae with a lateral embryo and in Eriocaulaceae, Rapataceae and Xyridaceae with a broad embryo. In developing grass seeds, the growing scutellum (cotyledon) pushes the coleoptile, mesocotyl and coleorhiza to the side. In the organless broad embryo, the cotyledonary sector is larger than the epicotyledonary sector. During germination of grass seeds, the coleorhiza and then the coleoptile emerge, while in a seed with a broad embryo the elongating cotyledon pushes the epicotyledonary sector outside the seed, after which a root–shoot axis is differentiated at a right angle to the cotyledon inside the seed. Broad and lateral embryos are closely related; however, the lateral embryo is more advanced in seed/embryo traits and germination morphology than the other kinds of monocot embryos, suggesting that its position on the family tree of seed phylogeny should be higher than of the other monocot embryos.

Viburnum is a temperate-zone genus that also occurs in mountains of South America and Malesia, and seeds of many species have morphophysiological dormancy (MPD). Information on the level of MPD in seeds of species in various clades of Viburnum potentially would increase our understanding of the evolutionary relationships between the nine levels of MPD. Our aim was to determine the level of MPD in seeds of Viburnum plicatum var. formosanum that is endemic to mountains (1800–3000 m a.s.l.) in Taiwan and a member of the Lutescentia clade. The temperature requirements for embryo growth and root and shoot emergence and response of seeds to gibberellic acid (GA) were determined. No fresh seeds germinated during 16 weeks of incubation at 15/5, 20/10, 25/15, 30/20 or 25°C. Embryo growth and root emergence occurred during moist cold stratification at 5°C or at a temperature sequence of 15/5 to 5°C. During cold stratification, embryos length increased from 0.76 ± 0.06 to 3.40 ± 0.26 mm and the embryo length:seed length ratio from 0.20 ± 0.02 to 0.68 ± 0.07. In a temperature sequence simulating field conditions, embryos grew inside seeds at 5°C, roots emerged at 15/5°C and shoots emerged at 20/10°C. The optimum temperature for embryo growth was 5°C. Neither GA3 nor GA4 was effective in promoting root emergence. We conclude that seeds of V. plicatum var. formosanum have deep complex MPD, which is a first report for Viburnum. Dormancy release during the cool season at high elevations helps to ensure that seeds germinate at the beginning of the warm season.

Although the level of seed desiccation sensitivity (LSDS) may have an impact on plant species conservation, information is available for <10% of tropical angiosperms. A study was conducted to assess the LSDS of 28 tropical montane species in Sri Lanka. Seeds were extracted from freshly collected fruits. Initial weight was recorded, and thousand seed weight (TSW) was calculated. Seed moisture content (MC) was determined. LSDS was determined using seed desiccation experiments and predicted using the TSW–MC criterion. Seed storage behaviour was predicted using LSDS and storage data and using a model based on phylogenetic affiliation. The relationship between LSDS and seed dormancy, life form and forest strata was evaluated. Fresh seeds of only 12 species germinated to >80%. Although seeds of the other species had >80% viability, only 0–70% germinated due to dormancy. Seeds of five species had MC <15%, indicating desiccation tolerance (DT). Seeds of 12 species lost viability after desiccation, indicating desiccation sensitivity (DS). Seeds of Ardisia missionis, Psychotria gartneri and Psychotria nigra remained viable after desiccation, showing DT. Seeds of 17 species were DS and those of 11 species DT. The TSW of four species was >500 g. Thus, seeds of other species were predicted to be DT by the TSW–MC criterion. A relationship was identified between LSDS and the forest strata of the species. More canopy species produced DS than DT seeds. Since seeds of most of the studied species were DS, these species may be threatened due to prolonged droughts predicted for the region due to climate change.

Interception by plant canopies during wind dispersal can affect the final destination of diaspores. However, how the interaction of wind speed, canopy type and diaspore attributes affects interception of diaspores by the plant canopy has rarely been studied. We investigated canopy interception for 29 species with different diaspore attributes, six canopy types and six wind speeds in controlled experiments in a wind tunnel. Shrub canopy interception of diaspores were controlled by wind speed and diaspore attributes, but the latter had a greater influence on canopy interception than the former. At low wind speed, diaspore wing loading had a large influence on canopy interception, whereas at high wind speed, diaspore projection area had a large influence. The chance of canopy interception at a particular wind speed was additionally affected by the type of canopy. This study increases our knowledge of the dispersal process, corrects the previous understanding of diaspore dispersal potential and improves the theoretical basis for predicting spatial pattern and dynamics of plant populations.

‘Dust seeds’ with an undifferentiated (organless) embryo are known to be produced by mycoheterotrophic species (MH) in nine families of angiosperms. However, aside from the numerous studies on seed germination of orchids, relatively little is known about germination in MH families. In the Ericaceae, some degree of mycoheterotrophy (full, partial or initial) and dust seeds with an undifferentiated embryo occur in all species in the three tribes of Monotropoideae, the only subfamily of Ericaceae with this combination of characters. In most species, the seed is <0.90 mm in the greatest dimension, the endosperm is absent (Pityopus) or consists of few to many (30–40) cells, and the embryo is minute, consisting of as few as two cells in Monotropa. Germination in Monotropoideae is monopolar, with only the radicular pole of the embryo participating in germination. Thus, germination polarity differs from that of the dust seeds of orchids in which only the plumular pole of the embryo (protocorm) participates in germination. The dust seeds in Monotropoideae require the presence of fungi, either direct contact with a fungus or the presence of a diffusible substance therefrom, to germinate (symbiotic germination). Recently, representatives of the four genera of tribe Pyroleae have been successfully germinated asymbiotically in vitro. We present a broad overview of dust-size seeds in angiosperms and conclude that they should be subdivided into at least two major categories.

Astragalus is the largest genus of seed plants with 3000 or more species that occurs naturally on several continents. The genus has some use as a forage and medicine and in industry, and many of the species are rare endemics threatened with extinction. The seeds are reported to be dormant at maturity, and various treatments have been used in an attempt to germinate them. Our primary aim was to determine via a meta-analysis the most effective way(s) to break dormancy in seeds of this species-rich genus. Mechanical and chemical (conc. sulphuric acid) scarification were by far the best of 12 treatments for breaking seed dormancy of the 40 species included in our meta-analysis, whereas prechilling, gibberellin and smoke were ineffective. These results along with those of imbibition tests confirm that seeds of the examined Astragalus species have physical dormancy (PY). Further, PY in these 40 species and (its documented occurrence) in 118 species that could not be included in our meta-analysis transcends climatic and geographic boundaries, edaphic conditions, life cycle/life form types and infrageneric phylogeny. Thus, it seems likely that most species of Astragalus have PY. However, in addition to PY, physiological dormancy (PD), that is, combinational dormancy (PY + PD), has been reported in a few species of Astragalus. This study should be useful to both basic and applied scientists who want to germinate seeds of Astragalus.

Only a few studies have been performed on seed germination of perennial ephemeral species native to the cold deserts of central Asia. We hypothesized that seeds of the cold desert perennial ephemeral Leontice incerta exhibit versatility in the timing of germination, that is, having the capacity to germinate at any time in summer, autumn and next spring. At dispersal in late May, only about 30% of the seeds could germinate; thus, a high percentage of the seeds was dormant. Seeds had a fully developed embryo, and dry storage, cold stratification, warm stratification and gibberellin promoted germination; we concluded that they have non-deep physiological dormancy. Seeds buried under natural conditions during summer germinated to 57–86% in autumn (late October) when exhumed and incubated at 5/2–25/15°C. However, seeds were sown in soil exposed to natural temperature and (low) precipitation did not germinate until next spring when the soil was moist. Thus, like various cold desert annuals, seeds of the perennial L. incerta can germinate in summer, autumn and next spring, depending on the availability of soil moisture (rainfall). Rainfall in cold deserts can play an important role in shaping seed germination traits of desert plants.

Hermaphroditic angiosperms, especially outcrossers, generally produce many more flowers and ovules than they can mature into fruits and seeds. One of the several hypotheses to account for the production of ‘excess’ flowers is selective abortion, which has been shown to increase offspring quality in plants. Our primary aim was to review the literature on the effects of selective abortion on seed germination and post-germination offspring vigour. Of 14 case studies (11 species in 10 genera and four families of flowering plants), germination percentage or rate (speed) increased in six and did not increase in eight, whereas post-germination offspring performance increased in 11 and did not increase in three. In six of the eight cases in which germination was not increased, seedling/juvenile vigour was increased. Seed mass was less likely to influence seed germination than seedling/juvenile vigour. Although selective abortion has been shown to increase progeny vigour of the early life history stages of plants, neither its demographic nor evolutionary importance has been demonstrated.

As a tribute to A.C. Martin's classic work on embryos in seeds, we have attempted to gain a better understanding of the peripheral embryo, which puzzled Martin. The peripheral embryo is strongly curved and in contact with the inner surface of the seed coat, and Martin placed it at the base of his family tree of seed phylogeny and called it a ‘phylogenetic orphan’. We evaluated ovule/seed development, kind of embryo and occurrence of perisperm in families with and without a peripheral embryo. All families with a peripheral embryo occur in the Caryophyllales. Seeds with a peripheral embryo have a low cotyledon width:radicle width ratio that coincides with Martin's (full-sized) linear embryo. The peripheral embryo develops in campylotropous and/or amphitropous ovules and is pushed to the side of the seed as the perisperm develops. Linear-full embryos and perisperm are widely distributed across extant angiosperms but are rarely found together, except in core Caryophyllales. The non-core Caryophyllales with endosperm and various kinds of embryos, including the linear-full, diverged before the core Caryophyllales. Thus, the ancestral linear-full embryo appears to have been retained when the core lineage developed campylotropous and/or amphitropous ovules and perisperm. Seeds with a peripheral embryo merit a position on Martin's family tree; however, the position should be a side branch (‘orphan’) slightly above (more advanced than) his linear embryo and not at the base. We conclude that Martin had great insight into the relationships between the kinds of embryos and rightly questioned the position of the peripheral embryo.

Maternal effects on offspring seeds are mainly caused by seed position on, and the abiotic environment of, the mother plant. Seed heteromorphism, a special form of position effect, is the production by an individual plant of morphologically distinct seed types, usually with different ecological behaviours. Seed heteromorphism is assumed to be a form of bet hedging and provides an ideal biological model to test theoretical predictions. Most studies of maternal effects on seeds have focused on abiotic environmental factors and changes in mean seed traits of offspring. Suaeda salsa is an annual halophyte that produces dimorphic seeds within the same inflorescence. We tested the hypothesis that plants grown from brown seeds of S. salsa have a higher offspring brown seed:black seed morph ratio and variance in seed size than plants from black seeds. Results from a pot experiment showed that plants grown from brown seeds had a higher brown seed:black seed ratio than plants grown from black seeds. This is the first layer of dynamic bet hedging. Brown seeds had higher size variation than black seeds, and seeds produced by plants from brown seeds also had higher seed size variation than plants grown from black seeds. This is the second layer of dynamic bet hedging. Thus, the maternal effect of seed heteromorphism is dual dynamic bet hedging. Furthermore, for seed traits we verified for the first time the theoretical prediction that an increase in offspring size variability induces an increase in the mean size of offspring.