The proposal is made that seed scientists need an internationally acceptable hierarchical system of classification for seed dormancy. Further, we suggest that a modified version of the scheme of the Russian seed physiologist Marianna G. Nikolaeva be adopted. The modified system includes three hierarchical layers – class, level and type; thus, a class may contain levels and types, and a level may contain only types. The system includes five classes of dormancy: physiological dormancy (PD), morphological dormancy (MD), morphophysiological dormancy (MPD), physical dormancy (PY) and combinational dormancy (PY + PD). The most extensive classification schemes are for PD, which contains three levels and five types (in the non-deep level), and MPD, which contains eight levels but no types. PY is not subdivided at all but probably should be, for reasons given. Justifications are presented for not including mechanical dormancy or chemical dormancy in the modified scheme. PD (non-deep level) is the most common kind of dormancy, and occurs in gymnosperms (Coniferales, Gnetales) and in all major clades of angiosperms. Since, first, this is the class and level of dormancy in seeds of wild populations of Arabidopsis thaliana and, secondly, Type 1 (to which seeds of A. thaliana belong) is also common, and geographically and phylogenetically widespread, it seems that biochemical, molecular and genetic studies on seed dormancy in this model species might have rather broad application in explaining the basic mechanism(s) of physiological dormancy in seeds.

A novel technique allows long-term monitoring of real-time oxygen consumption during seed germination in an open system. Most current techniques used to detect oxygen consumption by seeds measure the decrease in oxygen concentration in a closed chamber. This is not ideal for long-term experiments because the chamber must be replenished with air periodically, subjecting the seeds to abrupt changes in oxygen concentration. The current technique employs an open system, in which seeds are submerged in a continuously aerated aqueous environment. Oxygen electrodes are used to measure the steady-state concentration of oxygen in the solution, which is a function of both the rate of oxygen consumption by the seed and the rate of aeration from the atmosphere. The rate of aeration is directly dependent on the oxygen concentration of the bathing solution; therefore, previous calibration of the system allows the direct conversion of steady-state oxygen concentrations into oxygen consumption rates. Because oxygen is not limiting, the experimental design described here can monitor the same sample non-intrusively every minute for more than 24 h, allowing for greater precision than hourly readings often reported with current techniques. Multiple treatments and/or replicates can be run simultaneously, allowing sensitive comparison of various seed treatments or seed types. To illustrate its potential application, the technique was used to follow the rehydration and pre-emergence phases of germination of cucumber (Cucumis sativum), pea (Pisum sativum) and mustard (Brassica juncea) seeds, detect the inhibitory effects of surface sterilization techniques on seed respiration of cucumber, and follow the interaction of a bacterial biocontrol agent with germinating cucumber and pea seeds.

Dormancy and desiccation tolerance were studied in Fagus sylvatica seeds subsequent to 0–26 weeks of either moist prechilling (5°C) or warm (15°C) pretreatment at a restricted water content of 34% (fresh weight basis), i.e. c. 5% below full imbibition. After the pretreatments, seeds were either fully imbibed at 5 or 15°C, or dried for 7 d to around 13% moisture content and then fully imbibed at 5°C. Most of the dormancy was released after 5 weeks of prechilling, with nearly 70% of the seeds germinating at 15°C, and dormancy was fully released after 11 weeks of prechilling. Most of the seed desiccation tolerance, measured as the final germination percentage for dried seeds, was lost between 11 and 13 weeks of prechilling. A further decline in desiccation tolerance was recorded from 13 to 19 weeks, with a moderate recovery after 26 weeks. Germination was compared with the expression of a previously described abscisic-acid-responsive GRPF1 mRNA for a glycine-rich protein in F. sylvatica. There was a clear decline in accumulation of the transcript in seeds during prechilling, but not during the warm pretreatment. The amount of GRPF1 expression correlated significantly (P < 0.001) with dormancy status.

The temperate grass species, red fescue (Festuca rubra ssp. litoralis), perennial ryegrass (Lolium perenne) and Kentucky bluegrass (Poa pratensis), are often sown in mixtures for turfgrass. Differences in germination response to environmental conditions often result in different establishment success. The three species were compared in a study of imbibition at reduced water potential (Ψ) and in a study of the effects of Ψ and temperature on germination. In all three species, imbibition rate and extent were reduced with decreasing Ψ, causing a prolonged lag-phase before germination, particularly in Kentucky bluegrass. In perennial ryegrass and Kentucky bluegrass, water content at radicle emergence decreased non-linearly with decreasing Ψ, with a lower asymptote expressing the critical water content for germination. Seeds were germinated in factorial combinations of two temperatures (10 and 25°C) and nine Ψ levels (–2 to 0 MPa). Kentucky bluegrass required a longer thermal time to germinate than the other species and exhibited conditional dormancy at 25°C, which affected the use of the thermal time model. The effects of Ψ were described by a hydrotime model. Kentucky bluegrass had higher base water potential (Ψb) and required a longer hydrotime for germination than the other species. However, Ψb increased with temperature in Kentucky bluegrass and red fescue, but decreased in perennial ryegrass, and the base temperatures also changed with Ψ. The interacting effects of Ψ and temperature affected the ability of the hydrothermal time model to predict germination performance across all temperature and Ψ conditions.

Exposure to smoke is required for the germination of seeds from dormant genotypes of Nicotiana attenuata, a post-fire annual of the Great Basin Desert. Germination can be elicited by GA1,3,4,7 treatments and inhibited by the GA biosynthesis inhibitor, paclobutrazol (PAC), abscisic acid (ABA) and terpenes leached from unburned litter of the plant’s natural habitat. We analysed the endogenous GA and ABA dynamics during the 22 h after imbibition, when smoke-treated dormant seeds commit to germination. Extractable GA1+3 pools decreased in all seeds, but the decrease was more dramatic within 2 h of smoke exposure, which was followed by an increase between hours 2 and 4. Extractable ABA pools increased shortly after imbibition and remained stable in control, water-treated seeds, but decreased sharply in smoke-treated seeds. PAC completely inhibited smoke-induced germination when seeds were treated for 12 h after smoke exposure, consistent with the requirement of de novo GA synthesis for germination. Smoke treatment in the dark did not result in germination, whereas GA3 treatment did, a result consistent with phytochrome-mediated GA biosynthesis. Smoke exposure dramatically increased the sensitivity of seeds to exogenous GA3 treatments in both the light and dark, and light exposure increased this sensitivity an additional tenfold. Taken together, these results suggest that while germination requires endogenous GA synthesis, the effects of smoke treatment increase GA sensitivity, which is correlated with a decrease in endogenous ABA pools.

Expansins are a multi-gene family of proteins involved in changes in cell wall properties. In seeds where the embryo is completely surrounded by the endosperm, dormancy breakage requires weakening of the micropylar endosperm, increased embryo growth potential or both. Cell wall alterations are fundamental components of these processes, and expansins are thought to participate in them. Here, we explore the possible involvement of expansins in the control of germination by phytochrome in Datura ferox L. seeds. Based on the conserved sequences of known expansin genes, corresponding primers were designed to investigate the expression of expansin mRNAs by reverse transcriptase polymerase chain reaction (RT-PCR). One expansin mRNA was detected in micropylar endosperm of D. ferox, while two mRNAs were present in the embryo prior to radicle emergence. Expansin transcript content is promoted by red (R) light, both in the micropylar endosperm and the embryo; these effects of R are far-red light (FR) reversible, displaying a typical low-fluence response (LFR) in a way fully consistent with the photocontrol of germination. On the other hand, when the promotion of germination by the LFR is antagonized by exposing the seeds to continuous FR through a high-irradiance response (HIR) or by ABA, the inhibitory treatments do not affect expansin mRNA abundance. The results support the participation of expansins in the promotion of germination by LFR of phytochromes, and suggest that inhibition of germination by HIR or ABA does not include a reduction in the transcription of expansin genes.

Both the California chaparral species, Emmenanthe penduliflora Benth. (Hydrophyllaceae), and a tobacco native to the Great Basin Desert of south-western Utah, Nicotiana attenuata Torr. ex Wats. (Solanaceae), germinate in response to component(s) of wood smoke. Nitrogen oxides (NO and NO2), in amounts produced by a fire, have been proposed to be germination signals for E. penduliflora. We examined the germination response of dormant seeds of E. penduliflora and N. attenuata to aqueous solutions of smoke adjusted to different pHs, and two NO donors [sodium nitroprusside (SNP) and S-nitroso-N-acetylpenicillamine (SNAP)]. The smoke solutions, at pH 4 or 5, induced the maximum germination response. Aqueous solutions of SNP and SNAP, releasing NOx as high as 42 μM, had no effect on germination. Additionally, NO2– could not be detected in aqueous smoke extracts derived from combusted cellulose or wood. Therefore, unidentified cellulose combustion factors, rather than NOx, are likely to be the ecologically relevant germination signals.

Dormancy in seeds of Dodonaea viscosa is due to a water-impermeable seed coat (physical dormancy, PY). Thus, mechanically scarified seeds imbibed water (c. 95% increase in mass) and germinated to high percentages over a wide range of temperature regimes in both white light and darkness, whereas non-scarified seeds did not take up water. Dry heat at 80–160°C and dipping in boiling water for 1–60 s also broke dormancy in a high percentage of the seeds, and continuous far-red light was not inhibitory to germination. However, dry storage in the laboratory for >1 year did not overcome dormancy. Seeds made water-permeable by boiling imbibed water, and thus germinated, at a much slower rate than those made water-permeable by mechanical scarification. We suggest that boiling opened the ‘water gap’ in the seed coat (not yet described in Sapindaceae but present in other taxa with PY) and that water entered the seed only through this small opening, thereby accounting for the slow rate of imbibition and subsequent germination. Physical dormancy has now been shown to occur in seeds of this polymorphic, worldwide species from Australia, Brazil, Hawaii, Mexico and New Zealand. The low level of dormancy reported for seed lots of D. viscosa in China, India and Pakistan is probably due to collection of seeds before they dried to the critical moisture content for development of water-impermeability of the seed coat. Germination of non-dormant seeds over a wide range of temperatures and in white light, far-red (leaf-canopy shade) light and darkness are part of the germination strategy of D. viscosa and of other taxa whose seeds have PY at maturity.