Posted on

optimum temperature for cannabis seed germination

Seed germination ecology of Bidens pilosa and its implications for weed management

It is now widely recognized that Bidens pilosa has become a problematic broadleaf weed in many ecosystems across the world and, particularly in the light of recent climate change conditions, closer management strategies are required to curtail its impact on agricultural cropping. In this investigation, experiments were conducted to evaluate the effect of environmental factors on the germination and emergence of B. pilosa, and also on the response of this weed to commonly available post-emergence herbicides in Australia. The environmental factors of particular interest to this current work were the effect of light and temperature, salinity, burial depth and moisture on B. pilosa since these are key management issues in Australian agriculture. In addition, the effects of a number of commonly used herbicides were examined, because of concerns regarding emerging herbicide resistance. In the tested light/dark regimes, germination was found to be higher at fluctuating day/night temperatures of 25/15 °C and 30/20 °C (92–93%) than at 35/25 °C (79%), whilst across the different temperature ranges, germination was higher in the light/dark regime (79–93%) than in complete darkness (22–38%). The standard five-minute temperature pretreatment required for 50% inhibition of maximum germination was found to be 160 °C, and it was further shown that no seeds germinated at temperatures higher than 240 °C. With regard to salinity, some B. pilosa seeds germinated (3%) in 200 mM sodium chloride (NaCl) but all failed to germinate at 250 mM NaCl. Germination declined from 89% to 2% as the external osmotic potential decreased from 0 to −0.6 MPa, and germination ceased at −0.8 MPa. Seeding emergence of B. pilosa was maximum (71%) for seeds placed on the soil surface and it was found that no seedlings emerged from a depth of 8 cm or greater. A depth of 3.75 cm was required to inhibit the seeds to 50% of the maximum emergence. In this study, application of glufosinate, glyphosate and paraquat provided commercially acceptable control levels (generally accepted as >90%) when applied at the four-leaf stage of B. pilosa. However, none of the herbicide treatments involved in this study provided this level of control when applied at the six-leaf stage. In summary, B. pilosa germination has been clearly shown to be stimulated by light and thus its emergence was greatest from the soil surface. This suggests that infestation from this weed will remain as a problem in no-till conservation agriculture systems, the use of which is increasing now throughout the world. It is intended that information generated from this study be used to develop more effective integrated management programs for B. pilosa and similar weeds in commercial agricultural environments which are tending toward conservation approaches.

Introduction

Bidens pilosa, a member of the Asteraceae family, is a C3 annual broadleaf species with a life cycle of 150–360 days 1 . It is thought to have originated in tropical America, but it has now spread aggressively throughout similar climatic regions of the world 2 , 3 , consequently becoming a troublesome weed in both field and plantation crops. Whilst it is commonly found in gardens, along roadsides, fence rows and in open waste places, of particular interest to this study is that it has been reported to be now a common weed in no less than 31 different crops in more than 40 countries 2 . Some of the crops in which B. pilosa has been found as a principal weed are corn, sugarcane, sorghum, cotton, rice, vegetables, and pastures, all of which are critically important to many national economies 4 .

The fruits of B. pilosa are sparsely bristled with two to four barbed awns, which facilitate its unintended spread by workers, animals and prevailing winds 2 . As an indication of the serious nature of this problem, it is noted that a single isolated plant of this weed can produce over 30,000 seeds 1 . It also displays two flowering events during the growing season and, of these, the early flowering plants have been claimed to be crucial to its spread in dry seasons 5 by ensuring that at least some plants will always produce seeds for the next growing season. A previous study reported that B. pilosa plants produce two kinds of seeds: long seeds and short seeds 6 . Long seeds were found to germinate more readily than short seeds. The authors also observed that more short seeds are produced under long day conditions, suggesting that the fraction of long and short seeds changes with the season. Furthermore, it has been ascertained that B. pilosa commonly serves as a host for fungi, viruses and nematodes, which increases its damaging effect on those crops within which it has become established 2 .

Globally, there are a significant number of studies focusing on the effects of B. pilosa in cropping environments 7 , 8 , suggest that B. pilosa is becoming widely recognized as a major impediment to economic food production, but the foci on peculiar geographical situations and the recent effects of climate change issues have strongly suggested that specific Australian studies be carried out. It is therefore expected that, with this new information, a better understanding of the seed germination ecology of B. pilosa could be used to develop cultural management programs in a wide range of agricultural concerns, particularly in the light of management strategies which are eschewing the over-use of herbicides.

It is recognized that outcomes of previous studies have shown that seed germination of B. pilosa is affected by several environmental factors 9 . Light, for example, has been clearly established as being an important environmental factor for germination 10 , and it is well recognized, for management purposes, that weed seeds requiring light for germination do not emerge when buried deep in the soil. Conversely, for no-till cropping approaches, residual weed seeds are not significantly buried during crop preparation, and thus have ample opportunity to germinate. It has also been established that temperature is an important environmental factor affecting weed seed germination 9 , and it is noted that whilst some weed seeds germinate readily over a wide range of temperatures, others require a specific temperature range. In addition, some seeds are sensitive to pre-emergent temperatures, and knowledge of this also has important implications for weed management. As such, this information needs to be part of any containment process. For example, the narrow windrow burning technique is one of the common systems of harvest weed seed control (HWSC) in Australia 11 . In this technique, chaff and plant residues containing weed seeds are concentrated in a narrow windrow during crop harvest. Later, the chaff is burnt and the high temperature generated from burning destroys much of the weed seed bank on the soil surface. However, it is known that the temperature required to kill weed seeds differs considerably across common problematic weed species, and, currently, such information is not available on B. pilosa. Further, it is known that varying moisture conditions and salinity can also have important effects on weed seed germination, and across the Australian continent there is a wide range of conditions that would need to be carefully considered during crop management strategies.

Of increasing concern is that, in fallow situations, where cropping areas are being prepared for sowing, surface weeds, including B. pilosa, are commonly controlled using glyphosate and paraquat. However, it is known that herbicide efficacy is significantly reduced if applied to many varieties of large weedy plants, but such information is very limited in the case of B. pilosa, especially in ambient Australian conditions. In addition, it is becoming clear that reliance on one or two herbicides over a considerable time period may result in the unintended evolution of herbicide resistance in these weedy species. In this respect, biotypes of B. pilosa have already been reported as being resistant to several herbicides, including glyphosate and paraquat 12 , 13 . In Brazil, acetolactate synthase (ALS)-inhibiting herbicide-resistant biotypes have also been reported 14 . This is a concerning development and strongly suggests that there is a pressing need to evaluate the efficacy of a range of herbicides at different growth stages of B. pilosa in order that a more ecologically responsible approach to herbicide use can be developed.

It is noted that some useful information about the germination requirements of B. pilosa is available for populations in the USA 7 , but no similar information is available for specifically Australian populations. The concern here is that the previous study in the USA was conducted more than 20 years ago and there is a significant possibility that germination requirements of B. pilosa in Australia are different from those of B. pilosa in the USA. In addition, very limited information is currently available in the literature globally on the efficacy of a range of herbicides on B. pilosa, leaving a gap in the possibility of varying treatments in a specific area to avoid herbicide-resistant evolution.

Therefore, in this study, experiments were conducted in order to build a more detailed knowledge of the biology of B. pilosa. In particular, work has been done to determine the effects of temperature, light, salt stress, osmotic stress, and seed burial depth on the germination and emergence of seeds of this species, and to evaluate the relative responses of this weed to a range of commonly available post-emergence herbicides in Australia.

Results and Discussion

Effect of temperature and light on germination rate

The role of temperature in the germination of weed species is a complex issue. This observation that fluctuating temperature is a significant factor affecting weed seed germination has been previously recorded. For example, Eclipta prostrata (L.) L. has been shown to germinate over a range of temperatures 15 , whereas other species require critical levels of temperatures before they germinate. In the present study, seeds germinated at all the tested temperature ranges, which normally occur during the spring-summer seasons in Queensland, Australia. Our experiments indicated that, in the light/dark regime, germination of B. pilosa was higher at alternating day/night temperatures of 25/15 °C and 30/20 °C (92–93%) than at 35/25 °C (79%). These results suggest that B. pilosa can emerge throughout the spring and summer seasons in Australia and may also germinate in the early autumn. As an indication of the complexity of this issue, in a previous study carried out in the USA, B. pilosa incubated at temperatures ranging from 25/20 to 35/30 °C had maximum germination 7 , and in that study, only a 5 °C difference was maintained between the fluctuating temperatures. However, in the northern regions of Australia, temperatures can fluctuate more than 10 °C between day and night, therefore, this difference of 10 °C was selected between the fluctuating temperatures in the present study. A previous study from South Africa reported that the optimum temperature for germination was 25 °C, which is similar to the temperature (30/20 °C) found in our study 6 .

Whilst it was found that light was not a prerequisite for germination, it did significantly stimulate germination. This was shown by results which indicated across different temperatures, germination was higher in the light/dark regime (79–93%) than in complete darkness (22–38%) (Fig.  1 ). It is generally known that light plays an important role in affecting weed seed germination, but germination responses to light vary from species to species. Seeds of some species (e.g., E. prostrata) have an absolute requirement of light for germination 15 and others can germinate in both light and dark conditions 16 . Whilst in this study, B. pilosa germinated in both light regimes, seed germination under a 12 h photoperiod was higher than under continuous darkness. In previous studies, it was found that light had a stimulatory effect on B. pilosa germination 7 , 8 .

Effect of alternating day/night temperatures (25/15 °C, 30/20 °C, and 35/25 °C) and light (light/dark and dark regimes) on seed germination of Bidens pilosa. The vertical bars represent the standard errors of the mean.

The results of the present study suggest that B. pilosa emergence in the field may be encouraged by reduced or no-till systems, in which seeds are exposed to light and thus a large proportion of weed seeds remain on or close to the soil surface 17 . However, it was observed that a proportion of seeds germinated in darkness, thus B. pilosa has the potential of (somewhat reduced) germination even when buried in the soil.

Effect of pretreatment high temperature on germination

A sigmoid response was observed in the germination of B. pilosa seeds with increases in pretreatment (for 5 min) high temperature from 20 to 240 °C (Fig.  2 ). Germination was >80% when exposed to temperatures up to 120 °C for 5 min, but decreased when exposed to higher temperature. The critical temperature (for 5 min) to inhibit 50% germination was 160 °C and no seeds germinated at 240 °C.

Effect of oven temperature (pre-treatment) for 5 min on seed germination of Bidens pilosa when incubated at 30/20 °C in the light/dark regime for 14 days. The line shows a three-parameter sigmoid model fitted to the germination values and the arrow shows the temperature required to inhibit 50% of the maximum germination.

It has been claimed that vegetation burning can increase temperature on the soil surface to greater than 500 °C for more than 5 min 18 . These results suggest that, depending on stubble amount, B. pilosa seeds on the soil surface should be easily destroyed by narrow windrow burning. These harvest weed seed control (HWSC) systems have now become an essential part of integrated weed management programs in Australia. For example, a previous study on another Asteraceae weed, E. prostrata, reported that it required a temperature of 200 °C for 5 min to completely kill the seeds 15 .

Effect of salt stress on germination

Salinity is a major natural resource issue throughout Australia and is particularly so in southern Australia. Soils with 40 to 100 mM NaCl are considered to have ‘moderate’ salt content and soils with higher levels of NaCl are referred to as ‘high’ salt content 19 . In these investigations, a sigmoid response was observed in the germination of B. pilosa seeds with increases in NaCl concentrations from 0 to 250 mM (Fig.  3 ). The germination was at a maximum (92%) when seeds were incubated in no-stress conditions. Whilst seeds did not germinate at 250 mM NaCl, more than 3% seeds still germinated at 200 mM NaCl. A concentration of 120 mM NaCl inhibited 50% germination.

Effect of sodium chloride concentrations on seed germination of Bidens pilosa. The line shows a three-parameter sigmoid model fitted to the germination values and the arrow shows the concentration required to inhibit 50% of the maximum germination.

Similar to B. pilosa, seeds of another Asteraceae weed (E. prostrata) germinated at high salt concentrations 15 . In contrast, the previous study in the USA reported no germination of B. pilosa at 200 mM NaCl, and that study suggested that the weed germination was sensitive to salt stress. Of particular concern in this regard is that in Queensland, while the area of saline land was 48,000 hectares in 2001, it had increased to 107,000 ha by 2002 20 . The relevance of the results of the present study is that they suggest that B. pilosa can aggressively germinate in such soil conditions, therefore crop production may be affected in the future by both soil salinity increases and weed competition.

See also  pie 95 seeds

Effect of water stress on germination

It is widely recognized that, in Australian agricultural areas, water variability, causing levels of water stress, is an increasing phenomenon. An increase in drought conditions characteristic of climate change is beginning to make areas of traditional agriculture unprofitable because the response to water stress conditions, and its management, are not yet well understood.

In this investigations, a sigmoid response was observed in the germination of B. pilosa seeds with decreases in osmotic potential from 0 to −0.8 MPa (Fig.  4 ). Germination declined from 89% to 2% as osmotic potential decreased from 0 to −0.6 MPa. Germination was completely inhibited at an osmotic potential of −0.8 MPa and an osmotic potential of −0.44 MPa inhibited 50% germination. These results suggest that B. pilosa is a weed favored by a moist environment, which resonates with results from the previous study in the USA which indicated that B. pilosa was sensitive to low water potential 7 . This finding concurs with the positive association between this species and moist environments which was reported previously 2 . An outcome of this work suggests that a lack of moisture may be the overriding control factor for seed germination in dry soil conditions 21 since seeds germinate when the water content of the embryo is sufficient to support biochemical processes that lead to cell expansion. Although the effect of flooding (excess soil moisture) was not evaluated in our study, a previous study suggested that seedling emergence of B. pilosa is greatly reduced when flooding is maintained 7 . These results suggest that this species may not persist in flooded areas.

Effect of osmotic potential on seed germination of Bidens pilosa. The line shows a three-parameter sigmoid model fitted to the germination values and the arrow shows the osmotic potential required to inhibit 50% of the maximum germination.

Effect of seed burial depth on emergence

It was clear that Bidens pilosa emergence is greatly affected by seed burial depth which is consistent with the positive effect of light on germination. From previous studies, it is known that increased seed burial depth is known to reduce weed seedling emergence 22 because light can only penetrate minimally into the top soil layer 23 . Our results showed that seedling emergence declined linearly with increased burial depth (Fig.  5 ), with germination being greatest (71%) for seeds placed on the soil surface. As burial depth increased, seedling emergence declined by 10% for every centimeter increase in immersion in the soil. No seedlings emerged from a depth of 8 cm or greater, and the burial depth required for 50% inhibition of the maximum emergence was 3.75 cm.

Effect of seed burial depth on seedling emergence of Bidens pilosa. The line shows a linear model fitted to the emergence values and the arrow shows the depth required to inhibit 50% of the maximum emergence.

Although Bidens pilosa germination was maximum on the soil surface, up to 31% of seeds germinated in darkness. It is thought, therefore, that additional factors might be responsible for the (albeit limited) emergence of deeply buried seeds. While light stimulation clearly has an important role in seed germination, and its absence will have a noticeable retarding effect, it is possible that insufficient seed reserves (due to the small size of the seeds) to support seedling emergence could be another factor for low emergence from increased soil depths.

Notwithstanding the incomplete picture of seed emergence parameters, the greater level of surface germination strongly suggests that no-till or conservation agriculture practices would enhance B. pilosa seedling emergence. This is because, in such farming systems, a large fraction of weed seeds dropped by previous infestations remain on or close to the soil surface 17 . These results are entirely consistent with the previous study conducted in the USA 7 , in which results were related with observed severe infestations of B. pilosa in open undisturbed areas such as citrus row middles.

With regard to possible management strategies, two issues should be noted. First, weed seeds which are present on the soil surface are prone to environmentally facilitated death and from predation by insects. Thus there may be a useful mitigating effect on B. pilosa weed infestation where environmental interaction is severe enough or where natural seed predators are common. Second, the physical buildup of B. pilosa seeds on the soil surface can be managed by a deep tillage operation that buries B. pilosa and other weed seeds below their maximum depths of emergence (8 cm or deeper in the case of B. pilosa).

Effect of herbicides on survival and biomass

Currently, and in the recent past, selective and non-selective post-emergence herbicides have been a major tool to control B. pilosa both in productive crops and in fallow conditions, However, it is being recognised that overuse of a particular herbicide can lead to developed resistance by weed species, and also to unanticipated environmental effects in surrounding areas. In this work, a number of herbicides was evaluated in order to ascertain whether a more sensitive and efficient herbicide management approach might be possible to control B. pilosa.

It was found that, at the four-leaf stage of emergence, no plants of B. pilosa survived following the application of paraquat, and only 3% plants survived following the application of glufosinate (Table  1 ). Subsequently, application of these treatments at the six-leaf stage resulted in 18% and 38% surviving plants, respectively. Less than 20% plants survived following the application of 2,4-D or 2,4-D plus picloram at the four-leaf stage; however, the number of surviving plants increased to greater than 50% when these herbicides were applied at the six-leaf stage. In these experiments, fluroxypyr and saflufenacil were the poorest herbicide treatments, with 97–100% of plants surviving the treatment, irrespective of their leaf stage.

Table 1

Effect of post-emergence herbicides on seedling survival, shoot biomass, and control of Bidens pilosa when applied at the four- and six-leaf developmental stages.

Herbicide Rate (g ha −1 ) Four-leaf stage Six-leaf stage
Survival (%) Biomass (g/plant) Control (%) Survival (%) Biomass (g/plant) Control (%)
Control 0.155 100 0.244
2,4-D 1050 17 0.031 80 64 0.116 52
2,4-D + picloram 375 15 0.016 90 58 0.143 41
Fluroxypyr 100 97 0.070 55 100 0.160 34
Glufosinate 750 3 0.002 99 18 0.050 80
Glyphosate 675 48 0.011 93 100 0.104 57
Paraquat 500 00 0.000 100 38 0.155 36
Saflufenacil* 12 100 0.057 63 100 0.122 50
LSD 19.5 0.030 16.8 0.100

* Hasten at 1% was added to saflufenacil.

Our shoot biomass data suggested that glufosinate, glyphosate and paraquat provided greater than 90% control when applied at the four-leaf stage. However, none of these herbicide treatments provided greater than 80% control when applied at the six-leaf stage, and, regardless of growth stage, other herbicide treatments did not provide commercially acceptable control (>90%) of B. pilosa.

This study found that none of the selective herbicides (for example, 2,4-D) provided commercially accepted control of this weed, which suggests that growers will face significant challenges for controlling B. pilosa in herbicide-resistant cereals. To achieve any reasonable economic level of control, it appears that they would need to integrate herbicides such as 2,4-D with other cultural tools to control B. pilosa.

Non-selective post-emergence herbicides (glufosinate, glyphosate and paraquat) provided >90% control but only when applied at the four-leaf stage, suggesting that B. pilosa could be controlled effectively in fallows or before crop planting provided that these herbicides are applied at or before the four-leaf stage. Generally, herbicide efficacy is reduced when applied on large plants 24 and therefore, their rates may need to be increased to achieve effective control of large plants. In the absence of effective herbicides for controlling large B. pilosa plants, protoporphyrinogen oxidase (PPO) and photosystem II inhibitors should be evaluated for B. pilosa control. PPO herbicides inhibit the PPO enzyme which disrupts photosynthesis and causes membrane disruption and plant death 25 . Photosystem II inhibitors interfere with the photosynthesis process by which plants fix energy from sunlight 25 .

Saflufenacil is a relatively new herbicide in Australia and it has been found effective on other Asteraceae species, such as Conyza canadensis 26 . In our study, this herbicide provided poor control of B. pilosa; however, saflufenacil is not registered to control B. pilosa in Australia. This tolerance might be due to its ability to rapidly metabolize and detoxify the herbicide.

Plants in our study were grown in pots in a screen house, where they received conducive growing conditions. Plants grown in the field condition may be stressed and of variable sizes, which may reduce herbicide efficacy further. Therefore, there is a need to evaluate herbicides in the field conditions.

Summary

As these results have shown that B. pilosa germinates over a wide range of spring and summer temperatures, this suggests that the weed will germinate throughout the spring and summer seasons in the northern region of Australia. The robust nature of this species was also demonstrated by the observation that whilst light greatly stimulated seed germination of B. pilosa, a significant number of seeds also germinated in the dark. In addition, B. pilosa was found to germinate at high salt levels, but, in common with many other species, preferred a moist environment.

Based on available evidence, it is suspected that tillage operations that bury seeds deeper than 8 cm would limit B. pilosa seedling emergence, and that no-till or reduced-till systems, which do not cover seeds, would be most likely to promote seedling emergence. These results are consistent with observed B. pilosa infestations in the northern region of Australia, where no-till is the dominant farming system. The temperature investigations suggest that to support the no-till farming system, windrow burning after crop harvest may help in reducing weed infestation by destroying the seed bank on the surface layers of the soil. In addition, with no-till fallow periods, application of the non-selective post-emergence herbicides at or before the four-leaf stage of B. pilosa can significantly reduce its density and prevent reseeding in the following season. Knowledge gained from these studies in Australian conditions should be helpful for devising regional control measures in cropping situations and for limiting the future spread of B. pilosa. Inferences from the results of this study should be carefully related to the field conditions as this study was conducted in the laboratory and screenhouse, where conditions are different from the field. In the field, results may differ; therefore, there is a need to conduct field trials.

Material and Methods

Seed collection and germination test

Seeds of B. pilosa were collected during May 2015 from the edges of established soybean and sorghum fields at the research farm of the University of Queensland, Gatton (27°32′S, 152°20′E), Queensland, Australia. As the first author work at the university, no permission was needed to collect the weed seeds, as well as that the studies did not involve endangered or protected species. Seeds were collected from at least 100 plants, dried in a glasshouse for two days, cleaned, and stored in a labelled airtight container at room temperature (20 °C) until used in the experiments.

Aspects of Bidens pilosa germination were evaluated by placing experimental lots of 25 seeds in a 9 cm diameter Petri dish containing two layers of Whatman No.1 filter paper moistened with either 5 ml of distilled water or a relevant experimental solution. The dishes were placed in sealed plastic bags (to avoid moisture loss due to evaporation) and then placed in an incubator at fluctuating day/light temperatures of 30/20 °C in light/dark conditions (optimum conditions), unless otherwise indicated. The photoperiod in the incubator was set at 12 h light to coincide with the high-temperature period. In this investigation, seed germination was evaluated up to a maximum of 14 days after the start of the experiment, as no seeds germinated in a preliminary control experiment in the incubator after this period. For the purposes of common reportage, when there was visible protrusion of the radicle, the seed was considered ‘germinated’.

Effect of temperature and light on germination

The effect of variable temperature and light conditions on the germination of B. pilosa was determined by placing seeds in incubators set at three different day/night temperature regimes of 25/15 °C, 30/20 °C, and 35/25 °C in both light/dark and dark regimes. These temperature regimes were selected to reflect the temperature fluctuation during the spring to summer period in Queensland. In the dark regime, the Petri dishes were wrapped in double layers of aluminum foil.

Effect of ‘pretreatment’ high temperature on germination

The effect of pretreatment high temperature on germination of B. pilosa was determined by placing seeds in an oven for 5 min at 40, 80, 120, 160, 200 and 240 °C. The treated seeds were then incubated at the optimum conditions described earlier. Seeds stored at room temperature (20 °C) were used in the control treatment.

Effect of salt stress on germination

The effect of salt stress 13 on germination of B. pilosa was determined by placing seeds on filter paper in Petri dishes with 5 ml solutions of 0, 25, 50, 100, 150, 200, and 250 mM of sodium chloride (NaCl). These dishes were incubated at the optimum conditions described earlier.

Effect of water stress on germination

The effect of water stress on germination of B. pilosa was determined by placing seeds on filter papers in Petri dishes with 5 ml solution of 0, −0.2, −0.4, −0.6, −0.8, and −1.0 MPa. The solutions were prepared by dissolving polyethylene glycol 8000 in distilled water 27 .

Effect of seed burial depth on emergence

The effect of seed burial depth on seedling emergence of B. pilosa was determined in the screen house weed science facility of the University of Queensland, Gatton, Queensland. Fifty weed seeds were placed on the soil surface in 15-cm diameter plastic pots or covered with the same soil to achieve burial depths of 1, 2, 4, 6, 8, and 10 cm. The soil used in this experiment was red clay and had a pH of 5.5 and an electrical conductivity of 0.6 dS m −1 . There was no background seedbank of B. pilosa in the soil. Initially, an overhead mist sprinkler was used to irrigate pots, but during the experiment the pots were sub-irrigated. Emergence was defined as the first appearance of cotyledons, and the experiment was run until 35 d after seed burial.

Effect of herbicides on survival and biomass

Fifteen seeds of B. pilosa were planted at 0.5 cm depth in 11-cm diameter plastic pots, using a weed-free commercial potting mix. Seedlings were thinned to five plants per pot three days after emergence and sprayed with different herbicide treatments at the four- and six-leaf stages. There were control treatments for each leaf stage in which herbicides were not sprayed. The herbicides and their application rates are given in Table  1 . Herbicides were applied using a research track sprayer that delivered 108 L ha −1 water at a spray pressure of 200 kPa. Flat nozzles (Teejet 110015) were used in the sprayer. Seedling survival was determined 28 days after herbicide treatment with the criterion of at least one green leaf on the plant indicating survival. Survived plants were cut at the soil surface and dried in an oven at 70 °C for 72 h to measure dry biomass, and the results were expressed as percent control.

Statistical analyses

All experiments were conducted in a randomized complete block design and each treatment was replicated three times. All experiments were conducted twice and the second run was started within a month of termination of the first experiment. The data variance was visually examined by plotting the residuals to check homogeneity of variance before statistical analysis. ANOVA was performed on the original values, and showed that there were no differences between the two experimental periods. Henceforth, the data from the repeated experiments were combined for analysis.

The data on the temperature and light and the herbicide experiments were subjected to ANOVA, and means were separated using the least significant difference (LSD) test at the 5% level of significance (Gen Stat 18 th Edition). For other experiments, regression analysis was carried out using SigmaPlot 13.0 (Systat Software, Inc). Germination data from high temperature, salt and water potential treatments were modelled using a functional three-parameter sigmoid model:

In this model, Y is the total germination (%) at the NaCl concentration or osmotic potential x, Ymax is the maximum germination (%), x is the NaCl concentration or osmotic potential required for 50% inhibition of maximum germination, and b indicates the slope. Seedling emergence resulting from different burial depths was modelled using a linear model:

In this model, Y is the predicted emergence (%) as a function of burial depth x, a is the Y intercept, and b describes the slope of the curve.

Author contributions

Conceptualization: Bhagirath Singh Chauhan. Data curation: Bhagirath Singh Chauhan. Formal analysis: Bhagirath Singh Chauhan. Funding acquisition: Bhagirath Singh Chauhan. Methodology: Bhagirath Singh Chauhan. Resources: Bhagirath Singh Chauhan. Writing ± original draft: Bhagirath Singh Chauhan. Writing ± review & editing: Hafiz Haider Ali, Singarayer Florentine.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Cardinal Temperatures and Thermal Time for Seed Germination of Brunonia australis (Goodeniaceae) and Calandrinia sp. (Portulacaceae)

Seed germination of Brunonia australis Sm. ex R.Br. and Calandrinia sp. (Mt. Clere: not yet fully classified) was investigated using a thermogradient plate set at different constant temperatures to determine seed propagation requirements of these potential floriculture species. Germination responses were tested at 3, 7, 11, 15, 18, 22, 25, 29, 34, and 38 °C. Germination data were modeled using the cumulative distribution function of the inverse normal, which provides information on lag, rate, and maximum seed germination for each temperature regime. To determine cardinal temperatures, the reciprocal time to median germination (1/t50) and percentage germination per day were calculated and regressed against temperature. Base temperature estimates for B. australis were 4.9 and 5.5 °C and optimum temperatures were 21.4 and 21.9 °C, whereas maximum temperatures were 35.9 and 103.5 °C, with the latter being clearly overestimated using the 1/t50 index. Base temperatures for Calandrinia sp. were 5.8 and 7.9 °C, whereas optimum and maximum temperature estimates of 22.5 and 42.7 °C, respectively, were reported using the percentage germination per day index. Maximum seed germination of 0.8 to 0.9, expressed as the probability of a seed germinating, occurred at 11 to 25 °C for B. australis, whereas maximum germination for Calandrinia sp. was 0.5 to 0.7 at 18 to 25 °C. Thermal time, the accumulation of daily mean temperate above a base temperature, was calculated for different germination percentages. Estimates of thermal time (°Cd) for 50% seed germination were 54 and 90 °Cd for B. australis and Calandrinia sp., respectively.

Abstract

Seed germination of Brunonia australis Sm. ex R.Br. and Calandrinia sp. (Mt. Clere: not yet fully classified) was investigated using a thermogradient plate set at different constant temperatures to determine seed propagation requirements of these potential floriculture species. Germination responses were tested at 3, 7, 11, 15, 18, 22, 25, 29, 34, and 38 °C. Germination data were modeled using the cumulative distribution function of the inverse normal, which provides information on lag, rate, and maximum seed germination for each temperature regime. To determine cardinal temperatures, the reciprocal time to median germination (1/t50) and percentage germination per day were calculated and regressed against temperature. Base temperature estimates for B. australis were 4.9 and 5.5 °C and optimum temperatures were 21.4 and 21.9 °C, whereas maximum temperatures were 35.9 and 103.5 °C, with the latter being clearly overestimated using the 1/t50 index. Base temperatures for Calandrinia sp. were 5.8 and 7.9 °C, whereas optimum and maximum temperature estimates of 22.5 and 42.7 °C, respectively, were reported using the percentage germination per day index. Maximum seed germination of 0.8 to 0.9, expressed as the probability of a seed germinating, occurred at 11 to 25 °C for B. australis, whereas maximum germination for Calandrinia sp. was 0.5 to 0.7 at 18 to 25 °C. Thermal time, the accumulation of daily mean temperate above a base temperature, was calculated for different germination percentages. Estimates of thermal time (°Cd) for 50% seed germination were 54 and 90 °Cd for B. australis and Calandrinia sp., respectively.

Brunonia australis and Calandrinia sp. are Australian native herbs with potential for commercialization as flowering potted plants or bedding plants. Both species are ideally grown as annuals and, like many traditional short-lived garden plants, are most economically propagated by seed. Nursery managers often schedule plant production for peak market periods outside the natural growing season. Therefore, it is necessary to understand factors that impact propagation as well as growth and development to facilitate production protocols for year-round cultivation. Although Roche et al. (1997) investigated the effects of smoke on seed dormancy of B. australis, this is the first study to investigate the effect of different constant temperatures on seed germination of this species and of Calandrinia sp.

B. australis flowers naturally in spring to early summer from September to December, producing blue inflorescences on long slender stalks (Carolin, 1992; Stanley and Ross, 1986). This species is widespread throughout Australia, including southern Queensland, where average minimum winter temperatures are ≈5 to 7 °C and maximum temperatures are 19 to 21 °C (Bureau of Meteorology, 2010a). In spring and early summer, minimum temperatures are ≈10 to 20 °C with daytime temperatures reaching 25 to 34 °C. Calandrinia sp. is a drought-tolerant succulent (Harrison et al., 2009) with large pink flowers. There are only two known collections of this species, and both occur in the Gascoyne region of Western Australia (F. Obbens, personal communication). There are no published reports on growth and flowering of this species in its natural habitat, but other species within this genus usually flower in spring (September to November). The Gascoyne region has an arid climate where the average minimum temperature during winter is ≈9 °C and the maximum temperature is 21 to 23 °C (Bureau of Meteorology, 2010b). In spring, minimum temperatures are ≈12 to 19 °C, and maximum temperatures range from 27 to 34 °C.

Temperature is an important regulator of seed germination and is thought to be perceived within seed membranes (Mayer and Marbach, 1981). Temperatures that promote seed germination of many species typically correspond to natural environmental conditions, such as adequate soil moisture, that are conducive to seedling survival (Bell, 1999; Bell et al., 1993). Most seeds will germinate at ≈20 to 24 °C (Mayer and Poljakoff-Mayber, 1989). Imbibed seeds may not germinate at low temperatures, but in some species, cold stratification at ≈0 to 10 °C is required to break dormancy (Baskin and Baskin, 2004; Mayer and Poljakoff-Mayber, 1989). High temperatures usually induce or maintain dormancy of imbibed seeds, but seed death may occur as observed in several Australian species (Bell et al., 1993; Bellairs and Bell, 1990).

Germination rate indices are typically used to quantify cardinal temperatures for seed germination and include reciprocal time to median germination and percentage germination per day. Linear regression of a germination rate function versus sub- and supraoptimal temperatures and extrapolation of the same lines to the x-axis (intersect) are usually used to estimate minimum and maximum temperatures (Holt and Orcutt, 1996). The point where the two lines intersect defines the estimated optimum temperature at which germination is typically more rapid and the percentage germination greater. Usually, more than one germination rate index is used, and the adequacy of fit statistically compared to provide the most accurate method for estimating cardinal temperatures.

A robust estimate of the base (minimum) temperature for germination, at which development stops, is an essential requirement for calculating thermal time (Slafer and Savin, 1991). Thermal time, defined as the accumulation of daily mean temperature above a base temperature, is widely used in scheduling ornamental crop production. The approach can be more useful for predicting plant development stages than calendar date when temperatures for germination, or other development stages, are outside the experimental data range (Slafer and Savin, 1991).

The objectives of the study were to investigate seed germination responses of B. australis and Calandrinia sp. at different constant substrate temperatures to establish base, optimum, and maximum temperatures and to determine thermal time requirements. Two germination rate indices, used to provide estimates of cardinal temperatures, were compared. This information will be valuable for scheduling year-round production of these seed-propagated species and for providing insight into germination responses in the plant’s natural habitat.

Materials and Methods

Seed source and experimental procedures.

Seeds of B. australis were collected from a wild population in southern Queensland (lat. 27°54.905′ S, long. 149°51.139′ E) in Oct. 2003. Calandrinia sp. seeds were collected from plants grown at The University of Queensland, Gatton Campus (UQG) in 2004. Both species were stored at 10 °C in a specialized seed store at UQG. Undamaged seeds were selected and surface-sterilized in 2 g·L −1 chlorine with a drop of non-ionic wetting agent (Agral; ICI Crop Care, Australia) for 10 min and then triple-rinsed with deionized water. Seeds were then transferred to 65-mm diameter, 10-mm plastic petri dishes lined with two sheets of filter paper (Whatman No. 1) moistened with deionized water.

Germination responses were measured at 3, 7, 11, 15, 18, 22, 25, 29, 34, and 38 °C on a one-way thermogradient plate with 10 insulated chambers (Lindner and May Pty. Ltd., Windsor, Brisbane, Australia). The table surface temperature for each chamber was logged every 10 min using a standard thermistor probe (PB-5001) coupled to a data logger (Tinytag; Hastings data loggers, Port Macquarie, New South Wales, Australia). Variation from the set point was ± 0.3 to 0.9 °C. The temperature at the table surface was not considered different from the temperature of the substrate (moist filter paper) based on a preliminary study. Three petri dishes, each containing 20 seeds, were placed across the width of the table at each temperature division for both species (total of 30 dishes per species). Seeds of B. australis were germinated in the dark by wrapping the petri dishes in aluminium foil and seeds of Calandrinia sp. were germinated under a 16-h daylength as per requirements identified in a preliminary study. Light intensity of less than 10 μmol·m −2 ·s −1 was provided by 40-W wide-spectrum florescent lamps (Gro-Lux; Osram Sylvania, Danvers, MA).

Seeds were observed up to four times daily at the beginning of the experiment when germination was rapid and then gradually reduced to once a day until all seeds had germinated or no additional germination had occurred for 5 d. Germinated seeds were removed from the petri dish when the radicle or cotyledons (Calandrinia sp.) were 2 mm or greater. Seeds that failed to germinate within 31 ± 0.5 d after sowing were transferred to a temperature-controlled room at 18 ± 1.0 °C. Seeds of Calandrinia sp. were illuminated with a light intensity of 16 μmol·m −2 ·s −1 for 16 h·d −1 and seeds of B. australis were germinated under darkness. Seeds were assessed periodically for germination. After ≈25 d, when no further germination was observed, seeds were cut using a scalpel to assess viability and soft or discolored seeds were considered unviable. The experiment was repeated once over time during 2008 (Run 1 and Run 2).

Data analysis.

Results

Adequacy of fit of the inverse normal distribution.

The adequacy of fit (R 2 ) of the inverse normal distribution at different substrate temperatures for cumulative seed germination of B. australis and Calandrinia sp. is demonstrated in Figures 1A and B, respectively. Values of R 2 were greater than 0.90 for all temperature regimes, except at 7 and 11 °C (lowest temperature that germination occurred), where values of ≈0.55 and 0.80 were reported for B. australis and Calandrinia sp., respectively.

Adequacy of fit (R 2 ) of the inverse normal distribution to observed data for cumulative seed germination of B. australis (A) and Calandrinia sp. (B) at three contrasting substrate temperatures. A total of 60 seeds was used for each temperature regime and species. The experiment was repeated once over time and data from the first run (Run 1) are shown.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Adequacy of fit (R 2 ) of the inverse normal distribution to observed data for cumulative seed germination of B. australis (A) and Calandrinia sp. (B) at three contrasting substrate temperatures. A total of 60 seeds was used for each temperature regime and species. The experiment was repeated once over time and data from the first run (Run 1) are shown.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Adequacy of fit (R 2 ) of the inverse normal distribution to observed data for cumulative seed germination of B. australis (A) and Calandrinia sp. (B) at three contrasting substrate temperatures. A total of 60 seeds was used for each temperature regime and species. The experiment was repeated once over time and data from the first run (Run 1) are shown.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Probability of a seed germinating.

B. australis showed a higher (greater than 0.75) probability of seed germination at ≈11 to 25 °C, whereas Calandrinia sp. showed germination of greater than 0.5 at ≈18 to 25 °C (Fig. 2A–B). While no seeds germinated at the lowest temperature tested (3 °C) for either species, some germination was observed at 7 and 11 °C for B. australis and Calandrinia sp., respectively. At the highest temperature regimes of 34 and 38 °C, germination of Calandrinia sp. was reduced to ≈0.14 and inhibited for B. australis.

Probability of a seed germinating after 31 ± 0.5 d of incubation at substrate temperatures of 3 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B).

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Probability of a seed germinating after 31 ± 0.5 d of incubation at substrate temperatures of 3 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B).

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Probability of a seed germinating after 31 ± 0.5 d of incubation at substrate temperatures of 3 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B).

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

When non-germinated seeds of B. australis were transferred to 18 °C, 58% germination was observed for seeds removed from 3 °C, whereas 2% and 7% germination were obtained for seeds previously incubated at 34 and 38 °C, respectively. Similarly, seed of Calandrinia sp. transferred to 18 °C after incubation at 3 °C showed higher germination of 40%, whereas only 12% of seeds germinated after incubation at 7 °C and 2% or less of seeds germinated when removed from 34 and 38 °C. Seeds of both species that were incubated at 11 to 25 °C showed germination levels of 1% or less after being transferred to 18 °C. Most seeds that failed to germinate appeared to be viable, i.e., firm and not discolored.

Rate of seed germination.

Plotting the reciprocal of time (days) taken to reach a given germination percentage (G) against temperature showed mostly linear relationships for both species (Fig. 3A–B). For all values of G, the germination rate of B. australis increased markedly with rising temperature from a common base temperature of 5.1 ± 0.3 °C and then peaked at an optimum temperature of 22.3 ± 0.4 °C before decreasing linearly as temperatures rose in the supraoptimal range. The germination rate of Calandrinia sp. increased linearly for 1/t60 or less from a common base temperature of 6.3 ± 0.4 °C. For higher values of G, the rate began to plateau after reaching an optimum temperature, except for values at 38 °C that were omitted from analyses because they were obviously above the optimum for seed germination (Fig. 3B).

Rate of seed germination, expressed as the reciprocal response time (t, days) to reach a given percentage of total seed germination (G) at substrate temperatures of 3 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B). Values for 38 °C were omitted from analyses because they were clearly above the optimum for seed germination.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Rate of seed germination, expressed as the reciprocal response time (t, days) to reach a given percentage of total seed germination (G) at substrate temperatures of 3 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B). Values for 38 °C were omitted from analyses because they were clearly above the optimum for seed germination.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Rate of seed germination, expressed as the reciprocal response time (t, days) to reach a given percentage of total seed germination (G) at substrate temperatures of 3 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B). Values for 38 °C were omitted from analyses because they were clearly above the optimum for seed germination.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

The time (days) taken for 50% of seeds to germinate and the delay between sowing and the onset of germination (lag) generally decreased as temperature rose (Table 1). B. australis showed a minimum requirement of ≈1.5 d for radicle emergence, whereas Calandrinia sp. required at least 1 d for germination to start.

Median response time (t50, days) and delay (lag, days) in the onset of seed germination at substrate temperatures of 7 to 38 °C for B. australis and Calandrinia sp.

Cardinal temperatures and thermal time.

A broken linear model (Eq. [3]) was used to describe the effect of temperature on reciprocal time to median germination (1/t50) and percentage germination per day for B. australis (Figs. 3A and 4A). Both germination rate indices gave similar estimated values for base (Tb) and optimum (To) temperatures (Table 2). However, the maximum temperature (Tm) for seed germination was clearly overestimated using the 1/t50 index. Calandrinia sp. showed a linear response for temperature versus 1/t50 (rate = –0.059 + 0.010 × temperature), whereas temperature versus percent germination per day was described using a broken linear response (Figs. 3B and 4B). Estimates of Tb were similar for both germination rate indices (Table 2). Optimum and maximum temperatures for seed germination were estimated using the percent germination per day index. Values of 38 °C were omitted from analyses because they were clearly above the optimum for seed germination of Calandrinia sp. and inclusion would result in extrapolations for Tm that were outside biological limits for seed germination of most species.

Estimates of base (Tb), optimum (To) and maximum temperature (Tm), and optimum germination rate (Ropt) for seeds of B. australis and Calandrinia sp. calculated using two germination rate indices: reciprocal time to median germination (1/t50) and percentage germination per day (% germination/day). z

Rate of seed germination, expressed as percentage seed germination per day, at substrate temperatures of 7 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B). The relationship between temperature and germination rate was described using a broken linear function (Eq. [3]). Values for 38 °C (open circle) were omitted from analyses because they were clearly above the optimum for seed germination.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Rate of seed germination, expressed as percentage seed germination per day, at substrate temperatures of 7 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B). The relationship between temperature and germination rate was described using a broken linear function (Eq. [3]). Values for 38 °C (open circle) were omitted from analyses because they were clearly above the optimum for seed germination.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Rate of seed germination, expressed as percentage seed germination per day, at substrate temperatures of 7 to 38 °C for Brunonia australis (A) and Calandrinia sp. (B). The relationship between temperature and germination rate was described using a broken linear function (Eq. [3]). Values for 38 °C (open circle) were omitted from analyses because they were clearly above the optimum for seed germination.

Citation: HortScience horts 46, 5; 10.21273/HORTSCI.46.5.753

Thermal times (°Cd) for different percentages of seed germination (G) are shown in Table 3. Values for B. australis were calculated using suboptimal temperatures (linear increase phase) and a Tb of 5.1 °C, whereas °Cd for Calandrinia sp. were obtained using a Tb of 6.3 °C and regimes of 11 to 25 °C (temperatures above this range were excluded as a result of non-linearity). Both species showed °Cd to increase linearly with increasing percentage germination. Seeds of B. australis required 54 to 67 °Cd for 50% to 90% germination and Calandrinia sp. accumulated 90 to 117 °Cd for the same percentage germination.

Estimates of thermal time (°Cd) for different germination percentages (G) of B. australis and Calandrinia sp. z

Discussion

Cardinal temperatures were obtained using seed germination rate indices that are typically used to determine germination responses of agricultural crops and weed species. This study appears to be the first published report that uses seed germination rate indices to investigate responses of Australian flora and could be suitable for developing other native plants. Both indices, reciprocal time to median germination (Eq. [1]) and percent germination per day (Eq. [2]), gave consistent estimates of base temperature that were biologically relevant and statistically robust (Table 2). The reciprocal time to median germination index was unsuitable for quantifying maximum temperature, whereas the percent germination per day index provided good estimates for all cardinal temperatures.

Uniform estimates for base temperature, and general linear responses for temperature versus reciprocal time taken to reach a given percentage germination, supported the use of thermal time to describe germination responses of B. australis and Calandrinia sp. Some studies suggest that species adapted to cooler temperatures have a lower base temperature and a higher thermal time requirement for seed germination (Trudgill et al., 2000). However, this relationship was not evident in the present study, because B. australis showed a lower base temperature and thermal time requirement, whereas Calandrinia sp. had a higher base temperature and greater thermal time requirement.

Both species appear well adapted for germination in cooler temperatures because they had relatively low base temperature estimates that were below or similar to the average minimum temperature for winter in their natural habitat (Table 2). Temperatures for optimum rate of germination and maximum seed germination of B. australis (Table 2; Fig. 2A) corresponded with cooler temperatures typically experienced in late fall to early spring for inland southern Queensland. Although rainfall in this region is generally lower during these seasons, mild temperatures reduce the rate of evaporation and soil water is more likely to be retained long enough for seedlings to become established, whereas moisture from summer rainfall can quickly evaporate. Similarly, temperatures for optimum germination rate and maximum germination of Calandrinia sp. seeds (Table 2; Fig. 2B) coincided with cooler temperatures in late fall and early spring in the plant’s natural habitat: the Gascoyne region of Western Australia. However, the average monthly rainfall in early spring is ≈5 mm, whereas rainfall for late fall is 15 to 25 mm, suggesting that natural seed germination would take place during late fall when the soil moisture content was higher. Interestingly, some seeds showed rapid germination at 29 to 38 °C and this is comparable to temperatures experienced during summer in the Gascoyne region. It is possible that some seeds may germinate under hot conditions, particularly in microhabitats where soil moisture is retained for longer periods. Flowering of these plants most likely would be delayed, because studies have shown that temperatures below 20 °C are required to hasten flower bud production (Cave et al., 2010).

Generally, high temperatures of 34 and 38 °C were unfavorable for seed germination because maximum germination percentage was low. This response is common for many Australian native plants with the exception of some species within the family Casuarinaceae and Poaceae that germinate readily at 35 °C (Bell, 1999). High temperatures are thought to reinforce or induce seed dormancy to prevent germination during conditions unfavorable for seeding survival by strengthening the endosperm or pericarp and creating a physical barrier that is impervious to the radicle (Bell, 1999; Bell et al., 1993). On return to cooler temperatures, seeds of some species may germinate, but germination percentages are likely to be reduced as a result of seed death caused by exposure to high temperatures (Bellairs and Bell, 1990). Likewise, few seeds of B. australis and Calandrinia sp. germinated when transferred to 18 °C after 31 d of moist incubation at 34 and 38 °C.

Seeds that failed to germinate at high temperatures appeared viable. However, it is likely these seeds were dead given that ample time was allowed for germination to occur after transfer to favorable conditions and the inherent subjective interpretation of seed viability when using the cut method on small seeds (Gosling, 2003). Seed death caused by moist incubation at 35 °C has been reported in a number of Australian native species, including Velleia trinervis that belongs to the family Goodeniaceae (Bellairs and Bell, 1990). Loss of viability may be the result of increases in cell membrane permeability and the ensuing loss of K + and total electrolytes (Corbineau et al., 2002). In contrast, approximately half of the seeds of B. australis and Calandrinia sp. that were incubated for 31 d below the estimated base temperature germinated after being transferred to 18 °C, which suggested that cold temperatures can cause seed dormancy in these species.

Although both species germinated readily in the present study under favorable conditions, reports indicate that physiological dormancy may be present in fresh seeds (Roche et al., 1997; R.L. Cave, unpublished data). Physiological dormancy, a condition mainly controlled by factors within the embryo, can be alleviated using pre-treatments such as cold stratification, mechanical and chemical scarification, and dry after-ripening (Baskin and Baskin, 1998, 2004). Seeds used in this study were stored dry at 10 °C for ≈4.5 years and may explain why seed dormancy was not observed. Given this information, it is reasonable to infer that pre-treatment may be necessary to break dormancy.

Conclusion

Both germination rate indices, reciprocal time to median germination and percent germination per day, gave consistent estimates of base temperature for both species, whereas the percent germination per day index provided biologically relevant estimates for base, optimum, and maximum temperatures. Seed germination of B. australis and Calandrinia sp. appeared well adapted to cooler temperatures with base temperature estimates of 4.9 and 5.5 °C and 5.8 and 7.9 °C, respectively. Thermal time (°Cd) increased with increasing percentage germination. B. australis required 54 °Cd and Calandrinia sp. 90 °Cd for 50% of the seed population to germinate.

See also  dubble tropicana seeds