Introduction
In a previous paper (Adams et al., Reference Adams, Baskin and Baskin2005), we showed that seed lots of Aristolochia macrophylla and of A. tomentosa contain seeds with both morphological dormancy (MD) and non-deep morphophysiological dormancy (MPD), whereas seeds in seed lots of A. californica had intermediate or deep complex MPD (sensu Baskin and Baskin, Reference Baskin and Baskin2004). Further, it was demonstrated that: (1) seeds of A. macrophylla and of A. tomentosa exhibited a continuum of states with regard to depth of the physiological component of MPD; and (2) cumulative growth curves for embryos of all three species were sigmoidal. We hypothesized that these patterns of seed dormancy and of embryo growth indicated that the underdeveloped embryos (sensu Baskin and Baskin, Reference Baskin and Baskin1998) in cohorts of these three species grew at different rates, thus generating multiple size classes within the population at various time intervals between time-0 (start of incubation) and time-x, when all embryos had reached the critical (maximum) length for germination (see below). To investigate this further, we monitored sizes of embryos in seeds of A. macrophylla, A. tomentosa and A. californica, starting with fresh seeds and ending with those in which all embryos had attained the maximum length for germination. To our knowledge, this is the first report of using size-class distribution categories to monitor growth of underdeveloped embryos.
Materials and methods
Seeds
Seeds of A. macrophylla were collected in 1999 and in 2000 from the north slope of Pine Mountain, near Whitesburg, Letcher County, Kentucky; those of A. tomentosa in 2000 and in 2001 on the banks of the Cumberland River near Long Pond slough in Montgomery County, Tennessee; and those of A. californica in 2000 and in 2001 from bottomlands along Chico Creek in Butte County, California.
Embryo growth studies
Embryo growth studies were carried out using 12/12 h daily alternating temperature regimes of 15/6, 25/15 and 35/20°C (A. macrophylla, A. tomentosa) and constant temperatures of 5, 10 and 15°C (A. californica). The daily photoperiod was 14 h, and the light source for incubation and for cold-stratification was cool white fluorescent tubes, which produced a photon irradiance of approximately 40 μmol m− 2 s− 1, 400–700 nm. At the alternating temperature regimes, the photoperiod extended from 1 h before the beginning of the high-temperature period to 1 h after the beginning of the low-temperature period.
Seeds were placed in 5.5-cm-diameter Petri dishes on white quartz sand moistened with distilled water. Dishes were wrapped with plastic film to reduce water loss, and water was added to the dishes as needed to keep the sand moist. At intervals during the growth studies, 25 embryos were excised from seeds under a dissecting microscope, measured using a micrometer and placed into size classes (see below) designated for each of the three species based on mean length of embryo at seed maturity and maximum length of embryo required for radicle emergence.
To be able to assign a length (size) to an embryo (and not to the seedling) in a seed that had germinated between monitoring intervals, it was necessary to determine the maximum embryo length required for radicle emergence in each species. Fifty seeds were selected randomly from those of A. macrophylla collected in 1999, A. tomentosa in 2000 and A. californica in 2000. Embryos were excised and measured as soon as the radicle tip could be observed emerging from the seed, i.e. an embryo had attained its maximum length before germinating. The 50 measurements for embryos in seeds of each species were averaged to determine mean maximum length ( ± SE) that embryos attain prior to germination (radicle emergence). Therefore, if a seed had already germinated when embryo measurements were made, the mean maximum embryo length for a species was assigned to that embryo. For seeds to germinate, embryos must reach a mean ( ± SE) critical length of 4.15 ± 0.08 mm in A. macrophylla, 6.12 ± 0.11 mm in A. tomentosa and 5.69 ± 0.10 mm in A. californica (Adams, Reference Adams2003).
A. macrophylla
Embryo growth was monitored for 45 d at 25/15°C and for 60 d at 15/6 and at 35/20°C for seeds collected in both 1999 and 2000. At time 0 and at 5-d intervals for 45 or 60 d, embryos from a sample of 25 seeds were dissected out of the seeds, measured and each one of them placed into one of six size (length) classes [(1) 1.5–2.0 mm; (2) 2.1–2.5 mm; (3) 2.6–3.0 mm; (4) 3.1–3.5 mm; (5) 3.6–4.0 mm; and (6) 4.1–4.5 mm].
A. tomentosa
Embryo growth was monitored for 30 d at 35/20°C, for 40 d at 25/15°C and for 60 d at 15/6°C in seeds collected in 2000 and 2001. At time 0 and at 5-d intervals for 30, 40 and 60 d, embryos from a sample of 25 seeds were dissected out of the seeds, measured and placed into one of five size (length) classes [(1) 2.0–3.0 mm; (2) 3.1–4.0 mm; (3) 4.1–5.0 mm; (4) 5.1–6.0 mm; and (5) 6.1–6.5 mm].
A. californica
There were two different sets of monitoring events for embryo growth in A. californica. In the first set, embryo growth was monitored for 12 weeks at 10°C in seeds collected in 2000 and in 2001. At time 0 and after 4, 7 and 12 weeks, 25 embryos of seeds collected in 2000 and 2001 were dissected out of the seeds, measured and placed into one of five size (length) classes [(1) 1.5–2.0 mm; (2) 2.1–3.0 mm; (3) 3.1–4.0 mm; (4) 4.1–5.0 mm; and (5) 5.1–6.0 mm].
In the second set, embryo length was measured in fresh seeds, in seeds cold-stratified for 12 weeks at 5°C, in seeds cold-stratified at 5°C for 12 weeks and then incubated at 15°C for 2 weeks and in seeds cold-stratified at 5°C for 12 weeks and then incubated at 15°C for 3 weeks. Thus, the embryo growth monitoring sequence was: fresh seeds → 12 weeks at 5°C → 2 weeks at 15°C → 3 weeks at 15°C. Embryos were placed into the appropriate size class following measurement.
Results
A. macrophylla
Embryo size-class structure for 1999 and 2000 cohorts is shown only for days 0, 25, 35 and 45. Mean ( ± SE) embryo length of fresh seeds was 1.89 ± 0.06 mm, and all embryos were in size classes 1 and 2 (Fig. 1). At 25/15°C, the optimal regime for germination (Adams, Reference Adams2003), embryos grew faster than at either of the other two temperature regimes. Within 25 d of incubation, most embryos in the 1999 and 2000 colletions had moved out of size class 1, and 14 (1999+2000 cohorts) had even reached size class 6. By day 40 in 2000 (data not shown) and day 45 in 1999 (Fig. 1), all embryos had moved into size class 6.
Figure 1 Embryo growth in Aristolochia macrophylla seeds collected in 1999 and in 2000 and incubated at 25/15°C for 0, 25, 35 and 45 d. Embryos from 25 seeds were excised and measured on each of the four days and then placed into one of six size classes.
At 35/20°C, embryos grew relatively slowly in both years (data not shown). By day 60 in 1999 and 2000, only 17 and 9 embryos, respectively, had reached size class 6. After 60 d, there were embryos in each of the six size classes (except class 3 for 1999 embryos), and 12 of the 50 (1999+2000 cohorts) still remained in the lowest two size classes.
Embryos in seeds incubated at 15/6°C grew slower than those at 25/15 or at 35/20°C (data not shown). It took 25 d for any of the embryos in the 1999 and 2000 collections to reach size class 6, and by day 60 only 18 (1999+2000 cohorts) of the embryos had reached this size class. Twelve embryos in the 1999 cohort and nine in the 2000 cohort remained in the lowest two size classes at the end of the study.
A. tomentosa
Embryo size-class structure for 2000 and 2001 cohorts is shown only for days 0, 15, 25 and 30. Mean ( ± SE) embryo length of fresh seeds was 2.49 ± 0.05 mm, and all embryos were in size classes 1 and 2 (Fig. 2). At 35/20°C, embryos grew relatively rapidly in both years. For example, by day 15 in 2000 only three embryos each remained in size classes 1 and 2, while 14 had moved into size class 5. By day 30, all embryos in both 2000- and 2001-collected seeds had moved into size class 5 (Fig. 2).
Figure 2 Embryo growth in Aristolochia tomentosa seeds collected in 2000 and in 2001 and incubated at 35/20°C for 0, 15, 25 and 30 d. Embryos from 25 seeds were excised and measured on each of the four days and then placed into one of five size classes.
Embryos in seeds incubated at 25/15°C grew relatively rapidly, but not quite as fast as those incubated at 35/20°C (data not shown). There were embryos in all five size classes by day 15 for both seed cohorts, with 11 (2000) and 18 (2001) in size class 1 and seven (2000) and three (2001) in size class 5. All embryos in seeds in the 2000 cohort had moved into size class 5 by day 35, and all of those in the 2001 cohort had done so by day 40.
Embryos in seeds incubated at 15/6°C grew very slowly (data not shown). It took 20 d for embryos in seeds of the 2000 cohort and 25 d for those in the 2001 cohort to move into size class 3. The first embryo of the 2000 cohort did not reach size class 5 until day 30, and the first embryo of the 2001 cohort did not reach size class 5 until day 35. By day 60, 21 embryos in seeds from the 2000 cohort had reached size class 5; however, embryos of the remaining four seeds remained in size class 1. Overall, embryos in seeds from the 2001 cohort grew more slowly than those in seeds from the 2000 cohort. By day 60, 12 embryos from the 2001 cohort still remained in size classes 1 and 2, while 13 embryos had reached size class 5.
A. californica
Embryo size-class structure for 2000 and 2001 cohorts is shown only for weeks 0, 4, 7 and 12. Mean ( ± SE) embryo length of fresh seeds was 2.44 ± 0.05 mm, and all embryos were in size classes 1 and 2 (Fig. 3). Embryos grew slowly at 10°C and had not moved out of the first two size classes until week 4 in both 2000 and 2001. It took 7 weeks for any embryos to move into size class 5; by this time, only two embryos (both in 2001 cohort) remained in size class 1. All embryos had moved into size class 5 after 11 weeks in 2000 and after 12 weeks in 2001.
Figure 3 Embryo growth in Aristolochia californica seeds collected in 2000 and in 2001 and incubated at 10°C for 0, 4, 7 and 12 weeks. Embryos from 25 seeds were excised and measured on each of the four dates and then placed into one of five size classes.
All embryos in fresh seeds were in size classes 1 and 2, and they grew very slowly at 5°C (Fig. 4, two upper graphs). Only one embryo in the 2000 cohort and two in the 2001 cohort had moved into size class 5 after 12 weeks at 5°C (12 w 5). However, after embryos were then moved to 15°C they grew rapidly. Thus, 38 embryos (2000+2001 cohorts) had moved into size class 5 by week 2 (2 w 15), and all 50 (2000+2001 cohorts) had done so by week 3 (3 w 15) (Fig. 4, bottom graphs).
Figure 4 Embryo growth in Aristolochia californica seeds collected in 2000 and in 2001, cold-stratified at 5°C for 12 weeks and then moved to 15°C. In fresh seeds (week 0), after 12 weeks at 5°C (12 w 5) and after 2 (2 w 15) and 3 (3 w 15) weeks at 15°C following 12 weeks at 5°C, 25 embryos were excised from seeds and measured and then placed into one of five size classes.
Discussion
Data collected on change in embryo size over time can be presented either as a cumulative growth curve, size-structure diagrams or both. However, studies that monitor growth of underdeveloped embryos in seeds typically present data (mean ± SE) as cumulative curves (e.g. Baskin and Baskin, Reference Baskin and Baskin1998; Finneseth et al., Reference Finneseth, Layne and Geneve1998; Forbis and Diggle, Reference Forbis and Diggle2001; Kondo et al., Reference Kondo, Okuba, Miura, Honda and Ishikawa2002, Reference Kondo, Miura, Okubo, Shimada, Baskin and Baskin2004, Reference Kondo, Sato, Baskin and Baskin2006; Walck et al., Reference Walck, Hidayati and Okagami2002; Phartyal et al., Reference Phartyal, Kondo, Baskin and Baskin2009). We are not aware of any previous studies in seed ecology that have monitored growth of underdeveloped embryos using the size-class distribution method.
Sorting embryos into size classes during growth adds a level of detail to understanding dormancy-break and germination in seeds with MD and MPD that cannot be obtained from cumulative growth curves, by allowing the investigator to identify subpopulations of embryos that are growing at different rates or not at all. This information, in turn, can be used to help explain, at the sub-whole-seed level, aspects of a species' life history, such as germination phenology and seed persistence in soil of species with MD or MPD (Baskin et al., Reference Baskin, Baskin and Chester2003; Thompson et al., Reference Thompson, Ceriani, Bakker and Bekker2003). Thus, for example, data on size-class distribution of underdeveloped embryos over time can be used to explain at the embryo level why a portion of seeds in a population germinates early in the season (embryos advance relatively rapidly through the size classes), others germinate later in the season (embryos advance relatively slowly through the size classes) and still others do not germinate at all (embryos do not advance to the size required for radicle emergence). Regarding the latter point, the mechanism of differential rates of embryo growth detected by this method may sometimes allow essentially transient seeds to persist in the soil for a further year, i.e. germinate in the second germination season after dispersal rather than in the first.
In the case of A. macrophylla and A. tomentosa, size-class distribution of embryos over time gives an embryo-level explanation of why some seeds in a cohort have MD and others non-deep MPD. Thus, for example, at 25/15°C embryos in seeds with MD move more rapidly through the size classes (no physiological component to dormancy) than do those with non-deep MPD (some physiological dormancy). Further, slow growth of embryos in seeds of these two species at 15/6°C accounts for their lack of germination after dispersal in autumn, whereas failure of embryo growth at 5°C but relatively rapid growth at 15/6 and 20/10°C following a 12-week cold-stratification period explains why they germinate in spring (Adams, Reference Adams2003). In A. californica, movement of embryos through the size classes at 10°C and lack of growth of embryos in fresh seeds at 15°C but rapid growth at 15°C following cold-stratification of the seeds at 5°C categorizes this species as having intermediate or deep complex MPD and explains, at the embryo level, why germination of seeds of this species, which are dispersed in autumn, is delayed until early spring.
Thus, in conclusion, the size-class structure method of presenting data on growth of underdeveloped embryos in seeds over time allows the seed ecologist working primarily at the whole-seed or descriptive (N) level of study (sensu Passioura, Reference Passioura1979; Thornley, Reference Thornley1980), to explain dormancy and germination at the embryo-growth response (N-1) level more precisely than does presenting embryo growth data in the form of a cumulative curve.
Acknowledgements
We thank Dr Wayne Chester (Department of Biology, Austin Peay State University, Clarksville, Tennessee, USA) and Dr Rob Schlising (Department of Biology, Chico State University, Chico, California, USA) for collecting seeds of Aristolochia. Funding for this project was provided by the University of Kentucky departments of Biology and of Plant and Soil Sciences, the Southern Appalachian Botanical Society and the Kentucky Natural History Society.
