The successful establishment of a plant species in a location is closely related to the rapidity of germination. Different genotype and/or environmental factors can affect this process by increasing or decreasing this rate. Amongst the environmental factors, light is one that does not prevent the germination of seeds, if it acts as a signal (25) to cause a change in the germination rate and in final germination (31) and, therefore, in thermal time parameters. This factor is one of the main determinants of the accumulation of a persistent seed bank of numerous weeds in the soil (48), and it is necessary for the germination of many species (31) mainly of plants with small seeds (31,48) because large seeds can emerge from a much greater depth than light can penetrate (49). Exposure time to light may be short, less than a minute, or long. Short exposure time is more commonly effective with small weed seeds, such as cattail seeds, than with large weeds (31).
Light exposure influences the germination of different Typha species (5,50). In this work, no germination was obtained in DT20d treatments and a delay in the germination was observed in treatments with a 24h dark photoperiod (DT3d to DT10d). This effect may be explained by the development of a secondary dormancy related to phytochrome activation/deactivation processes which occur through the stimulus of light on cattail seeds. Phytochromes are the principal mechanism triggering germination of Typha because they participate in breaking the dormancy (22,25). These pigments have two mutually photoconvertible forms: Pfr (considered the active form for seed germination) and Pr (considered the inactive form) (25,49). Pfr is established during the formation of the seed in the mother plant; however, this phytochrome form can reconvert to Pr in darkness (18,31). In these circumstances, the secondary dormancy does not break, and a period is needed to reconvert the phytochrome to its active form (Pfr) (51,52). This secondary dormancy can explain the results in darkness treatments. For example, in the darkness treatments (DT3d to DT10d), thermal time increases as the number of days in darkness increases (Table 4). In the case of DT20d treatments, no germination was measured after 20 days in darkness. These cattail seeds, although they absorbed water and began to swell, did not break their coatings to allow germination. This may explain the death of every seed after 20 days in darkness or the delay produced by secondary dormancy. We support this second option but, since no subsequent germination data was collected, a further study would be necessary to determine it.
Treatments of the same population had an increase in θT(50) as the number of days in darkness increased. There was a relationship between log(θT(50)) and darkness treatments (R2>0.90) (Table 4). Initially, a linear increase in thermal time was expected as the number of days in darkness increased. Indeed, there was an increase, but it was not proportional; for example, in the case of the population of Cu with constant temperatures, θT(50) at PT0d was 267 °h, and the value corresponding to DT10d was 411 °h. This means a 50% increase in thermal time, not a 100% increase as expected. This modification would indicate that T. domingensis seeds accumulate hours of temperature and that when receiving light, the dormancy is broken by the activation of Prf and the germination response occurs more quickly than expected. Darkness treatments, such as DT3d, DT5d, DT7d and DT10d, had lower seed germination than treatments without 24h dark photoperiod (DT0d). These results indicate that longer days of darkness may decrease the light sensitivity of T. domingensis. Dormancy broken in the presence of light and the influence of phytochrome has been studied and is common in small seeds, such as cattails (22,52). A buried environment is associated with darkness and cattail seeds do not germinate in darkness at any temperature; hence, buried seeds of T. domingensis could be a control method for the establishment in aquatic ecosystems. Darkness is also related to the depth of water(25); so the establishment of cattails in the GFF system may be successful if seeds are sowed above the soil submerged in water or on a floating structure of this system.
Although water depth was not a factor in this study, it is another factor that is related to the amount of light and the ease of germination of cattail seeds. The depth used was enough to saturate the paper and seed (<0.4 cm) due to the fact that germination in cattail seeds is greater and faster in aquatic conditions (2,24,33). Some authors have stated that flooded areas increase the germination of Typha species, and this increase in germination has a direct relation to depth (17,53). This feature may be caused by a decrease in the level of oxygen, rather than by the lower intensity of light in these situations (33). Other studies, however, show no relationship between the germination rate and depth (34,54). The limit of the depth of germination in Typha species in clear water is around 40cm (2) or 1 cm in sediment (55). There is an extreme case where cattail seeds germinated under 80 cm water (and survived 8 weeks) (56).
The germination response in plants of different origins could also be different (49). Differences related to the origin of a population are frequent in numerous species of plants, whether crops (57) or weeds (27,58). Successfully colonizing a new location is related to the greater adaptive capacity of these populations to harsher environmental conditions compared to other populations (59), thus allowing these populations to have greater flexibility and adaptability to different locations or future climate change scenarios (17).
Cattail seeds were grouped into northern (Cu and Ma) and southern (Ba and Se) populations (Figure 5). Mean temperatures of germination within each group were similar, but there were differences between the groups. The northern populations have lower values than the southern populations (Table 5). The results of the thermal time study also show differences between northern and southern populations. In treatments with the same temperature and darkness periods, the northern population presented lower values of thermal time and a higher germination response than the southern populations (Tables 2 and 3).
These differences among populations are consistent with the results of other studies carried out with Typha latifolia L. in fifteen European populations(17) and in USA populations(60); in both studies, in comparison to northern populations, southern populations germinated at a lower temperature. However, in our study, the opposite scenario occurred. Before providing conclusions, some points concerning these studies must be clarified. For example, T. domingensis is a species more adapted to warmer areas compared to T. latifolia. In the European study, only two Mediterranean populations were used, and both populations germinated more rapidly than northern populations; the distances between the origins of the populations were greater than those in our study. Some authors mention that other factors, such as temperature or nutrient supply, are more important than the origin of the seeds in the case of neighbouring populations(17).
In this study, the estimated mean Tb was 16.4°C and no differences greater than 0.6°C were observed regardless of origin, darkness treatments, or level or amplitude of temperatures. We could have considered that Tb was constant; however, other studies with crops (46) or weeds (61) estimated different Tb values for the different amplitudes of temperatures. There were significant differences in the germination responses both in terms of the level and amplitude of temperatures (Table 2). In comparison to treatments with other Tm, treatments with Tm close to Tb achieved a lower germination response in all treatments (Table 2). No data were found for the calculated Tb for Typha species, but the estimated values of Tb for cattail seeds in this study were very similar to those obtained in other studies with summer weeds (29,62). Steinmaus (2000) established a relation between the slope of the line used to estimate Tb and germination rate; this rate will be greater with a higher slope. In our study, higher slopes occurred in Cu in thermal regimes with both constant and alternating temperatures and coincided with the lower θT(50) of all populations studied (Figures 1 and 2).
Differences in To were obtained in the results of the multifactor analysis, mainly between the northern (Cu and Ma) and southern populations (Ba and Se) (Table 4). This difference in To is comparable with the results of other studies with different populations of weeds or with T. latifolia (17,22,23). The To for the Swedish populations of T. latifolia was approximately 20°C (23) or 10/30°C with alternating temperatures in Italian populations (22). Australian populations of the Typha genus germinate readily at high temperatures and decline when the Tm is lower than 20°C (63).
Table 7 shows the results from different studies of the seed germination of T. latifolia and T. domingensis. There are few studies on the seed germination of T. domingensis. Lorenzen et al. (2000) stated that a To of 30°C and 25/10°C occurred in southeastern American populations of T. domingensis at constant and alternating temperatures, respectively. These To values are distinct from those obtained in our study (22.5-25°C), but there are other studies with To values very similar to those obtained in this work (Table 7). These results showed different To values according to the places of origin of the seed and were closely related to climaticc conditions at each location (17). Some conditions, such as the temperature of the mother plants (38,64), may determine the germination of populations, regardless of whether the seeds were of the same species (24,37).
Table 7. Optimal temperature in T. domingensis and T. latifolia in different populations from various studies.
Plant species
|
Reference
|
C
|
A
|
Seed location
|
Typha domingensis
|
This study
|
22.5; 25 ºC
|
|
Spain
|
Lorenzen et al. (2000)
|
30 ºC
|
25/10 ºC
|
Florida, U.S.
|
Royal Botanic Gardens (2002)(65)
|
20 ºC
|
|
Wakehurst, England
|
Typha latifolia
|
Sifton H.B (1959)
|
30 ºC
|
20/30 ºC
|
Ontario, Canada
|
Bonnewell, V. et al (1983)
|
35 ºC
|
|
Minnesota, U.S.
|
Lombardi, T et al. (1997)
|
|
20/30 ºC
|
Pisa, Italy
|
Ekstam and Forseby (1999)
|
20 ºC
|
|
Linköping, Sweden
|
Heinz, S (2011)
|
25 ºC
|
10/25 ºC
|
Germany
|
Meng, H. et al. (2016)
|
|
25/15 ºC
|
Northeast of China.
|
C: constant temperature. A: alternating temperature
In the Typha genus, temperature and amplitude were shown to be factors related to germination (23). The favourable effect of alternating temperatures on the germination response is well known in different weeds (22) because the effect enables a seed to understand when it is buried and to inhibit germination. In nature, seeds of the cattail are usually submerged. In this situation, fluctuations in the ambient temperature are rare; therefore, an increase in this fluctuation could indicate that seeds have reached land and germination could be stimulated. In this study, both thermal factors (level and fluctuation in temperatures) influenced the final germination of cattail seeds. In the treatments within the same population and in the darkness treatment, there was a greater germination response as the temperature approached To from values close to Tb, causing the existence of significant differences depending on the temperature level (Tables 1 and 2). An increase in the germination response is obtained with higher temperatures up to To; above this value, germination begins to decrease. The same results occur in other studies with Typha (17,22,23,33) and weeds (27,29).
The use of different amplitudes of temperature is related to the loss of dormancy in weeds (29,66) or crops such as lentil (30). In the case of cattail seeds, the loss of dormancy is related to changes in germination responses. Treatments with ΔT=0°C and 15°C had a higher germination response than those with ΔT= 5°C and 10°C (Table 1), so these last two amplitudes of temperature negatively affect germination. However, in studies with T. latifolia, treatments with constant temperature regimes (ΔT=0°C) achieved a lower germination response than alternating regimes (ΔT>0°C) (17). On the other hand, θT(50) corresponds to treatments of the same population, and ΔT=0°C is lower than treatments with ΔT≥0°C (Figure 4), in contrast to Solanum physalifolium (29) whose thermal time is considerably reduced in an alternating regime (Table 3). These data are consistent with the germination rate (Figure 3), in which treatments with alternating temperatures reach lower values than those corresponding to constant temperatures. According to these results, the best season to germinate T. domingensis would be spring or autumn because these seasons have a temperature regime of approximately ΔT=15°C under natural conditions.
The thermal time value of different populations of cattail seeds (Table 3) was substantially lower than that of other weeds such as different species of Solanum (20,22) or tropical species such as Pennisetum typhoydes (45,67). This indicates a rapid germination response compared with those of other plant species. There were also differences between populations, with Cu being the one with the lowest thermal time, both in ΔT=0 and ΔT>0 treatments. Although Cu and Ma obtained similar germination values (Table 2), θT(50) was the highest in Ma.
Therefore, Cu could be the population that presents the most vigour during this process because this population had the fastest germination under the conditions tested. The final germination percentages were very similar in all populations. It would be necessary to carry out new tests to determine whether the development in other stages of plant growth would also be fast in this population.
In comparison to other species of the genus, such as T. angustifolia, T. domingensis is a plant species more adapted to warm temperatures. In Spain, it has been observed that T. domingensis has been colonizing places where T. angustifolia once stood (3). If this capacity occurs with an increase in temperatures due to climate change, then it is possible to consider that T. domingensis might increase its expansion to the detriment of other Typha species such as T. angustifolia.
According to the evaluation of the models developed in this work (Table 6), there were some differences between the results that were related to thermal regimes and darkness treatments. The greatest coincidences are in the models developed with constant and alternating thermal regimes and DT0d treatments with R2 mean values ≥ 0.9, except for the population of Ma that present values slightly below 0.9 in both thermal regimes. The best coincidence was between expected and observed values in constant regimes and treatments without 24h dark photoperiod (DT0d). Darkness treatments affect the coincidence between expected and observed values, the R2 decreased to mean values between 0.84 to 0.78. In the DT3d treatments, R2 values were greater than the other darkness treatments. In the remaining darkness treatments (DT5d, DT7d and DT10d), R2 values were similar (Table 6). Ba and Ma populations show both the most and least coincidental values in all treatments, respectively, but the differences were small among all populations.
This work shows that environmental factors and the origin of populations affect the germination responses of cattail seed and shows how they could be used to develop models for improving the establishment of this species in wastewater systems such as GFF.