Comparison of the seed germination of native andnon-native winter annual Apiaceae in North America,with particular focus on Cyclospermum leptophyllumnaturalized from South America

JEFFREY L. WALCK,*1 CAROL C. BASKIN,†‡ SITI N. HIDAYATI*1 and JERRY M. BASKIN†*Department of Biology, Middle Tennessee State University, Murfreesboro, Tennessee 37132, United States of America,†Department of Biology, University of Kentucky, Lexington, Kentucky 40506-0225, United States of America and ‡Departmentof Plant and Soil Science, University of Kentucky, Lexington, Kentucky 40546-0312, United States of America

Abstract

To increase our knowledge of the seed dormancy breaking and germination requirementsof native and non-native winter annual Apiaceae in temperate eastern North America, astudy was conducted on seeds of Cyclospermum (Apium) leptophyllum, which is native toSouth America, but naturalized in North America and elsewhere. A high percentage offresh seeds germinated over a narrow range of temperatures, but the range increasedwhen seeds were incubated continuously in light or buried in soil at high summertemperatures, indicating the presence of conditional physiological dormancy (PD).Embryos grew from approximately 0.3 to 1.25 mm (316% increase in length) inside theseeds before the radicle emerged; thus, seeds had morphological dormancy (MD) as wellas PD, that is, morphophysiological dormancy (MPD). Embryo growth did not occurduring summer in seeds exposed to natural temperatures, but embryos did grow inautumn. For seeds sown in the field in June, both embryo growth and germinationoccurred in autumn. Thus, seeds have non-deep simple MPD, with PD being broken insummer and MD in autumn. A comparison of C. leptophyllum with native and non-nativewinter annual Apiaceae in the region reveals that all species have MD, but the proportionof fresh seeds that also has PD (thus, MPD) varies between species. A combination offactors, including PD, an inability to germinate at high summer temperatures, a relativelylow temperature requirement for embryo growth and a lack of dispersal until autumn (intwo non-natives), delays germination of both natives and non-natives until autumn,when conditions are favorable for both germination and seedling establishment.

Keywords: Apiaceae, germination phenology, morphological seed dormancy, morphophysio-logical seed dormancy, winter annual.

Received 13 June 2007; accepted 25 October 2007

Introduction

The flora of temperate eastern North America includes 15genera and 24 species of winter annual Apiaceae (Radford

et al. 1968; Gleason & Cronquist 1991). Six genera (Chaero-phyllum, Cynosciadium, Daucus, Ptilimnium, Spermolepisand Trepocarpus) are native to the region, and 10 (Aethusa,Anethum, Amni, Anthriscus, Apium, Bupleurum, Cyclosper-mum, Coriandrum, Scandix and Torilis) are not native. Allspecies from the non-native genera, except Cyclospermumleptophyllum (Pers.) Sprague ex Britton & P. Wilson, arenative to Europe or Eurasia (Radford et al. 1968; Gleason& Cronquist 1991). Cyclospermum leptophyllum is native toSouth America, and it is widely naturalized as a weed in

Correspondence: Jeffrey L. WalckEmail: jwalck@mtsu.edu1Present address: Kings Park and Botanic Garden, Fraser Avenue,West Perth, Western Australia 6005, Australia and School of PlantBiology, The University of Western Australia, Crawley, WesternAustralia 6009, Australia.

Plant Species Biology (2008) 23, 33–42 doi: 10.1111/j.1442-1984.2008.00205.x

© 2008 The AuthorsJournal compilation © 2008 The Society for the Study of Species Biology

tropical and temperate regions (McIntyre et al. 2005;Menglan et al. 2005), including North America, where itoccurs from New York to Tennessee and Florida, west-ward across the southern portion of the United States toCalifornia, and northward to Oregon (NatureServe 2006).It grows in a variety of habitats: freshwater marshes,disturbed areas, roadside ditches, roadsides and lawns(Constance 1993; Weakley 2006).

As winter annual Apiaceae in temperate eastern NorthAmerica are represented by both native and non-nativespecies, we wished to determine if the two groupshave the same seed dormancy breaking and germinationrequirements. To date, seed dormancy and germinationhave been investigated in the natives Chaerophyllum tain-turieri Hook. (Baskin & Baskin 1990), Chaerophyllum proc-umbens (L) Crantz (Baskin et al. 2004), Ptilimnium nuttallii(DC.) Britt. (Baskin et al. 1999) and Trepocarpus aethusae(Nutt.) ex DC. (Baskin et al. 2003), and in the non-nativesBupleurum rotundifolium L. (Baskin & Baskin 1974) andTorilis arvensis (Hudson) Link. (Baskin & Baskin 1975).However, to make good comparisons between the germi-nation ecology of native and non-native winter annualApiaceae, more information is needed on non-nativespecies. As C. leptophyllum is the only non-native in thegroup that is not from Europe or Eurasia, we selected itfor study.

Previously C. leptophyllum was called Apium leptophyl-lum (Pers.) Benth., but Constance (1990; see also Burtt1989) transferred A. leptophyllum to the genus Ciclosper-mum (Cyclospermum, an orthographic variant; Weakley2006) based on life cycle (being annual), morphology(having very narrow leaf divisions, long cotyledons, sub-rhomboidal pollen) and chromosome number (n = 7). AsC. leptophyllum was previously placed in the genus Apium,we need to pay some attention to what is known aboutseed dormancy in Apium.

Because of the economic importance of Apium graveo-lens L. (celery), its seeds have received considerableresearch attention (Thompson 1974a,b; Thomas et al. 1975;Jacobsen et al. 1976; Pressman et al. 1977, 1988; Biddington& Thomas 1979; Jacobsen & Pressman 1979; Coolbear et al.1991; Pérez-García et al. 1995). Seeds of this species have asmall, linear-shaped embryo that is surrounded byendosperm. If seeds are incubated in light at 20°C, theembryo grows to two-thirds the length of the seed in only5 days, after which time the seeds germinate (Jacobsen &Pressman 1979). As embryo growth and germinationoccur in <1 month, seeds of A. graveolens have morpho-logical dormancy (MD) (sensu Baskin & Baskin 2004). Ifthe underdeveloped embryos also had physiologicaldormancy (PD) and, therefore, required a dormancy-breaking treatment before or after the embryos grewinside the seeds, seeds would have morphophysiologicaldormancy (MPD) (Baskin & Baskin 2004).

Although much is known about the germination ofA. graveolens, little is known about the seeds of non-cultivated members of this genus. Cochrane and Probert(2006) reported that seeds of Apium prostratum ssp.phillipii, an herbaceous perennial endemic to WesternAustralia, have small embryos relative to the length ofthe endosperm. Seeds of this species germinated to�92% in light during 29 days of incubation at constanttemperatures of 15 and 20°C and at an alternating tem-perature of 25/10°C, with and without surgical treat-ment and potassium nitrate. Although some after-ripening (breaking of PD) may have occurred betweencollection (late February) and the initiation of the experi-ments (mid-March) (A. Cochrane, pers. comm., 2007), itseems probable that freshly matured seeds of A. prostra-tum have MD.

The primary focus of our research on the germinationecophysiology of C. leptophyllum was to determine ifseeds have MD or MPD, and if they have MPD, whichone of the eight levels of MPD do they have (sensuBaskin & Baskin 2004). We also wished to determinehow the timing of germination is controlled in nature.Specifically, the following aspects were investigated: (i)the light and temperature requirements for dormancybreak and embryo growth; (ii) the germination andembryo growth responses of buried seeds; and (iii) thephenology of germination and embryo growth. Finally,we compared our results with those for other winterannual Apiaceae in temperate eastern North Americaand with the germination requirements of Apiumgraveolens.

Materials and methods

Plant material

Mericarps (seeds) were collected from plants of C. lepto-phyllum growing at three locations: a cedar glade in Fran-klin County, Alabama, USA, on 7 July 1991; a cedar gladein Bedford County, Tennessee, USA, on 24 May 2003 anda grassy/weedy area alongside a paved trail and theedge of woods in a floodplain in Rutherford County,Tennessee, USA, on 5 June 2005. Seeds from the Alabamacollections were received via mail at the University ofKentucky (Lexington, Kentucky) on 26 July 1991 andstudies initiated on the same day; all studies conductedon the Alabama collection were done at the University ofKentucky. Studies conducted on 2003 and 2005 seedswere initiated within 1 and 9 days, respectively, follow-ing collection, and they were done at Middle TennesseeState University (Murfreesboro, Tennessee) or in agarden in the same town. The seeds were stored dry atroom temperature between the collection and initiationof experiments.

34 J . L . WA L C K ET AL.

© 2008 The Authors Plant Species Biology 23, 33–42Journal compilation © 2008 The Society for the Study of Species Biology

General procedures

Laboratory studies were done in temperature-controlledand light-controlled incubators set at 12/12 h with dailytemperature regimes of 15/6, 20/10, 25/15, 30/15 and35/20°C. These regimes approximated the mean dailymaximum and minimum monthly air temperatures inTennessee and northern Alabama, where the seeds werecollected, during March and November (15/6°C), Apriland October (20/10°C), May (25/15°C), June and Septem-ber (30/15°C) and July and August (35/20°C) (NationalOceanic and Atmospheric Administration 2003). The dailyphotoperiod was 14 h, extending from 1 h before thebeginning of the high-temperature period to 1 h after thebeginning of the low-temperature period. Another incu-bator was set at a constant 5°C and a 14-h photoperiod.The light source was 20 W cool white fluorescent tubes,with a photon flux density (400–700 nm) at seed levelfrom the top to the bottom shelf ranging from 48 to72 mmol m-2 s-1.

Seeds were placed on white quartz sand moistenedwith distilled water in 6.0 cm ¥ 1.5 cm (diameter ¥ depth)plastic Petri dishes. Water was added to the dishes asneeded to keep the substrate moist. Three replicates of 50(1991 collection) or 25 (2003 and 2005 collections) seedsper dish were used in each treatment, unless otherwisestated. Dishes were wrapped with plastic film to reducewater loss, and the dishes in the dark treatments wereadditionally wrapped in aluminum foil. Protrusion of theradicle was the criterion for germination. Ungerminatedseeds were pinched with forceps to determine if embryoswere firm and white, indicating they were viable butdormant, or soft and gray, indicating they were non-viable. Germination percentages were determined on thebasis of the number of viable seeds and were arcsinesquare root transformed for statistical analyses.

For embryo growth studies, one set of 50 (1991 collec-tion) or 25 (2005 collection) seeds was used each timemeasurements were made. Fresh seeds were allowed toimbibe for 24 h in the dark at room temperature. Embryoswere excised from seeds with a razor blade and theirlengths measured under a dissecting microscopeequipped with a micrometer. If a seed had germinated theembryo length was recorded as 1.25 mm, which is thecritical length for germination (i.e. the approximate lengthof the embryo when it had grown enough to start splittingthe seed coat).

Germination of freshly matured seeds

Three dishes of seeds collected in 2003 were incubated inlight at each of the following temperatures: 15/6, 20/10,25/15, 30/15 and 35/20°C for 14 weeks and were checkedfor germination at 2-week intervals. Another set of three

dishes of seeds was tested for germination in darkness ateach of the five temperature regimes, and final germina-tion percentages were determined after only 2 weeksbecause of limited seed availability.

Three dishes of seeds collected in 2005 were incubatedin light at each of the five temperature regimes for16 weeks with the same set of dishes checked for germi-nation at 2-week intervals. In addition, 24 dishes of seedswere tested for germination in darkness at each of the fivetemperature regimes with one set of three dishes beingexamined for germination at 2-week intervals for a total of16 weeks.

Repeated-measure analyses of variance (rmanovas)were used to test the effects of length of incubation(within subject) and temperature regime (betweensubject) on the germination responses of 2003 seeds, andthe effects of length of incubation (within subject) andlight and temperature regimes (between subject) on 2005seeds (SAS 2003). For the 2003 experiment, light regimewas not included as a factor because the length of incuba-tion differed between the tests conducted in darkness(2 weeks) and in light (up to 14 weeks). For the 2005experiment, germination responses in light and darknesswere treated similarly (as ‘repeated’) even though a dif-ferent set of seeds incubated in darkness was examinedevery 2 weeks. Greenhouse–Geisser corrected probabili-ties are reported for the rmanovas.

Effects of dry storage and cold stratification onseed dormancy

Seeds collected in 2003 and 2005 were air dried beforebeing placed into closed glass jars at ambient laboratoryconditions (22°C) for 12 and 16 weeks, respectively. Fol-lowing dry storage, the seeds were incubated in light andin darkness for 4 weeks at 15/6, 20/10, 25/15, 30/15 and35/20°C. Seeds collected in 2005 were cold stratified at5°C for 4, 8 and 12 weeks in darkness and then incubatedin darkness at 15/6, 20/10, 25/15, 30/15 and 35/20°C for4 weeks. Seeds could not be stratified in light because ahigh percentage of the seeds germinated in light at 5°C.Seeds both stratified and incubated in darkness were notchecked until the end of the incubation period.

Temperature requirements for loss of dormancy

Approximately 1500 seeds collected in 1991 were placedinto 16 fine-mesh nylon bags and the bags were buriedindividually at a depth of 7 cm in soil (3:1 v/v mixture oflimestone-derived topsoil and river sand) in 15-cm-diameter plastic pots with drainage holes. One pot ofburied seeds each was placed at 5, 15/6, 20/10, 25/15 and30/15°C. Soil moisture was kept near field capacity, and

CYCLOSPERMUM LEPTOPHYLLUM S E E D G E R M I N AT I O N 35

Plant Species Biology 23, 33–42 © 2008 The AuthorsJournal compilation © 2008 The Society for the Study of Species Biology

the pots were covered with aluminum foil. After12 weeks, seeds were removed from one pot at each tem-perature regime, placed into Petri dishes and incubatedin light at 15/6, 20/10, 25/15, 30/15 and 35/20°C for2 weeks.

Germination and embryo growth of buried seeds

Approximately 2000 seeds collected in 1991 were placedinto six nylon bags and the bags were buried individuallyto a depth of 7 cm in soil (as described above). Pots wereplaced under a bench in a non-heated greenhouse inLexington, Kentucky. The greenhouse received no heatingor air-conditioning, and its windows were kept openthroughout the year; thus, exposing the seeds to near-natural seasonal temperatures (Baskin & Baskin 1985). Soilwas watered to field capacity once per week from 26 Julyto 31 August 1991, and then daily during the remainder ofthe study, except on winter days when the soil was frozen.

Freshly matured seeds and those exhumed on the firstday of each month for the dates shown in Figure 4 weretested for germination in light and darkness at 15/6,20/10, 25/15, 30/15 and 35/20°C. When buried seedswere exhumed, seeds that were to be incubated in dark-ness were not exposed to any light. In darkness, the bag ofseeds was removed from the soil and a ‘pinch’ of 50–75seeds was placed into each of three Petri dishes to beincubated in darkness at each temperature. Then, disheswere wrapped with plastic film and two layers of alumi-num foil. Seeds to be incubated in light were countedunder fluorescent room light, and three replicates of 50seeds each were placed in light at each temperature. Allgermination tests were terminated after 2 weeks.

To determine if embryos grow in buried seeds exposedto natural temperatures, embryos were excised from 50freshly matured seeds and from 50 seeds each time theywere exhumed between September 1991 and January1992. Embryos were not measured in March 1992 becauseapproximately 75% of the exhumed seeds had alreadygerminated.

A three-way anova was used to examine the effectsand interactions of time of exhumation, temperatureregime and light regime on germination responses and aone-way anova was used to examine the effect of time ofexhumation on embryo length (SAS 2003).

Phenology of seed germination and embryo growth

To monitor germination phenology, three replicates of 300seeds each collected in 2005 were sown on limestone-derived topsoil in plastic trays (16 cm [width] ¥ 24 cm[length] ¥ 8 cm [depth]) on 18 June 2005. The trays wereplaced outside under a wooden bench for protection fromdirect precipitation, covered with a metal screen and

watered under two moisture regimes (see below). Atweekly intervals until 31 December 2006, all germinatedseeds were counted and removed from the trays.

To monitor embryo growth phenology, embryos wereexcised from fresh 2005 seeds and at regular intervalsfrom seeds exposed to natural temperatures. In addition,the length of fresh seeds was measured prior to excisingthe embryo. Twenty-five seeds were placed in each of 22nylon bags that were placed outside on soil (either in traysor directly in a garden), covered with a metal screen andwatered under two moisture regimes (see below) startingon 18 June 2005. On 1 and 15 July and on the first day ofeach month between August 2005 and April 2006, one bagwas removed from each moisture treatment and placed onwet paper towels in plastic bags in darkness for 24 h. Thenumber of germinated seeds in the bag was counted, andembryos were excised from ungerminated seeds andmeasured.

Two moisture treatments were used in the phenologyof seed germination and embryo growth studies: (i) thesoil on which seeds were sown and on which bags wereplaced was kept continuously moist throughout the studyperiod; and (ii) the soil was subjected to a natural precipi-tation regime. Bags containing seeds were placed in agarden, where they received direct precipitation. Follow-ing each rainfall event, the soil in the trays was wetted tofield capacity and kept moist as long as the soil surface inthe garden was moist. On the day the study was initiatedthe soil in both treatments was watered to field capacity.Then, the two moisture treatments were maintained until1 April 2006. Following 1 April 2006, the soil in all flatswas watered once each week until 31 August, after whichit was watered daily, unless frozen, until 31 December2006. On 4 November 2006 the soil in all flats was dis-turbed by stirring.

Maximum and minimum daily air temperatures foreach week of the study (18 June 2005–1 April 2006) wererecorded at the Murfreesboro, Tennessee, weather station,approximately 5.5 km from the study site.

Results

Germination of freshly matured seeds

Ninety-three to 100% of fresh 2003 seeds germinated at15/6–25/15°C and 3–39% germinated at 30/15–35/20°Cduring 4 weeks of incubation in light; no seeds germi-nated during 2 weeks of incubation in darkness (Fig. 1).Germination in light significantly increased to 99–100% atall temperatures during 14 weeks of incubation (timefactor, P < 0.0001; interaction, P < 0.0001).

Sixty to 85% of fresh 2005 seeds germinated at 15/6–30/15°C and 0–2% germinated at 5 and 35/20°C during4 weeks of incubation in light, and 0–3% seeds germinated

36 J . L . WA L C K ET AL.

© 2008 The Authors Plant Species Biology 23, 33–42Journal compilation © 2008 The Society for the Study of Species Biology

in darkness over the range of temperatures (Fig. 2). Ger-mination significantly increased to 93–100% at all tem-peratures during 16 weeks of incubation in light and to18–52% only at 20/10–30/15°C in darkness (time factor,P < 0.0001; all 2-way and 3–way interactions, P < 0.0001).

Effects of dry storage and of cold stratification onseed dormancy

Seeds collected in 2003 and 2005 and stored dry for 12 or16 weeks, respectively, germinated to 73–95% at 15/6–30/15°C and to 0–13% at 35/20°C during 4 weeks of incuba-tion in light and to 0–1% at all temperatures in darkness.Seeds cold stratified in darkness germinated to 1–23% indarkness, regardless of the length of stratification or theincubation temperature.

Temperature requirements for loss of dormancy

Regardless of the burial temperature, �91% of seedsgerminated at 15/6–25/15°C and �1% germinated at35/20°C (Fig. 3). In contrast, �22% seeds germinated at30/15°C when buried at 5–15/6°C and �44% germinatedwhen buried at 20/10–30/15°C.

Germination and embryo growth responses ofburied seeds

Germination responses of seeds to temperature and lightdiffered significantly among exhumation dates (date,temperature regime and light factors and all 2-way and3–way interactions, P < 0.0001). At maturity in July,71–97% of seeds germinated at 15/6–25/15°C and 0–1%

germinated at 30/15–35/20°C in light, and no seeds ger-minated in darkness (Fig. 4). Regardless of the exhuma-tion date, �75 and �5% of seeds germinated at 15/6–25/15 and 35/20°C, respectively, in light; germination at30/15°C varied from 2%, when seeds were exhumed inSeptember 1991, to 91%, when exhumed in November1991. Seeds exhumed between September 1991 andDecember 1991 germinated to 0–37% in darkness overthe range of temperatures and 21–91% of those exhumedbetween January 1992 and March 1992 germinated onlyat 15/6–25/15°C.

The length of embryos varied significantly with exhu-mation date (P < 0.0001). Mean embryo length in freshseeds was 0.34 mm (standard error [SE] = 0.01). Embryoswere 0.34–0.36 mm long in seeds exhumed in September1991, November 1991 and December 1991 and 0.43–

Fig. 1 Cumulative germination percentages (mean � 1 standarderror [SE], SE shown if �5%) of Cyclospermum leptophyllum seedscollected in 2003 and incubated for 14 weeks in light at five tem-perature regimes.

(b)

(a)

Fig. 2 Germination percentages (mean � 1 standard error [SE],SE shown if �5%) of Cyclospermum leptophyllum seeds collectedin 2005 and incubated for 16 weeks in (a) light or (b) darkness atsix temperature regimes. The same set of seeds was checkedevery 2 weeks for seeds incubated in the light and the percent-ages in (a) are cumulative; a different set of seeds was checked(and then destroyed) every time germination was examinedfor seeds incubated in darkness and percentages in (b) are notcumulative.

CYCLOSPERMUM LEPTOPHYLLUM S E E D G E R M I N AT I O N 37

Plant Species Biology 23, 33–42 © 2008 The AuthorsJournal compilation © 2008 The Society for the Study of Species Biology

0.45 mm long in those exhumed in October 1991 andJanuary 1992.

Phenology of seed germination and embryo growth

Seeds sown on 18 June 2005 and subjected to continuousmoisture began germinating between 11 and 17 Septem-ber 2005, when mean weekly maximum and minimumtemperatures were 32.2 and 16.6°C, respectively, and peakgermination occurred between 6 and 12 November 2005,when temperatures were 23.6 and 6.5°C (Fig. 5). Seedlingswith cotyledons and a tiny (approximately 1-mm long)leaf that emerged in November 2005 survived over-winter(until April 2006), even though the soil froze on someoccasions. Two seedlings emerged following soil distur-bance in November 2006.

Precipitation increased in November 2005, and the soilremained moist for a longer period of time because oflower air temperatures and less evaporation. Thus, seedssubjected to a natural precipitation regime began germi-nating between 13 and 19 November 2005, when meanweekly maximum and minimum temperatures were 15.8and 0.9°C, respectively, and peak germination occurredbetween 20 and 26 November 2005, when temperatureswere 13.0 and -0.6°C (Fig. 5). Ten seedlings emerged fol-lowing soil disturbance in November 2006.

The mean length of freshly matured seeds collected in2005 was 1.25 mm (SE = 0.02), and embryos were 0.31 mm(0.01) (Fig. 5). In seeds subjected to continuous soil mois-ture, maximum embryo elongation (187%) occurredbetween September 2005 and November 2005. Althoughgerminated seeds were present in bags between October2005 and April 2006, seedlings in the bag exhumed inApril 2006 were rotten and no ungerminated (viable)seeds remained. In contrast, in seeds subjected to naturalprecipitation, maximum embryo elongation (89%)occurred between October 2005 and December 2005. Ger-minated seeds were present in bags exhumed betweenDecember 2005 and April 2006, but many seedlings in theApril 2006 sample were rotten and ungerminated (viable)seeds were present.

Discussion

Freshly matured seeds of C. leptophyllum are water-permeable, as evidenced by their softness after 24 h on amoist substrate at room temperature; thus, they do nothave physical dormancy (Baskin & Baskin 2004).However, before radicles emerged, embryos elongatedfrom approximately 0.3 to 1.25 mm, a 316% increase.Clearly, embryos were underdeveloped (sensu Grush-vitzky 1967), and seeds had MD. In addition, whereas�80% of fresh seeds germinated over a relatively narrowrange of temperatures (Figs 1,2a,4a), 90–100% eventually

Fig. 3 Germination percentages (mean � 1 standard error [SE],SE shown if �5%) of Cyclospermum leptophyllum seeds collectedin 1991 and incubated for 2 weeks in light at 15/6, 20/10, 25/15,30/15 and 35/20°C following 12 weeks of burial at 5, 15/6,20/10, 25/15 and 30/15°C. No seeds buried at 15/6, 20/10 or30/15°C germinated at the test temperature of 35/20°C.

(a)

(b)

Fig. 4 Germination percentages (mean � 1 standard error [SE],SE shown if �5%) of Cyclospermum leptophyllum seeds collectedin 1991 and incubated for 2 weeks in (a) light or (b) darkness atfive temperature regimes following approximately 0–8 monthsburial in a non-heated greenhouse.

38 J . L . WA L C K ET AL.

© 2008 The Authors Plant Species Biology 23, 33–42Journal compilation © 2008 The Society for the Study of Species Biology

germinated over a broader range of temperatures, indicat-ing the presence of PD. Thus, seeds of C. leptophyllum haveboth MD and PD, that is, MPD, unlike seeds of Apiumspecies, but similar to most native winter annual Apiaceaein eastern North America (Table 1).

Seeds of C. leptophyllum can not germinate until both PDand MD have been broken. The level of PD in seeds ofC. leptophyllum is non-deep, that is, quickly broken, andmany seeds are in a state of conditional PD at maturity. Thismeans fresh seeds can germinate to high percentages at oneor more temperature regimes, but after PD is broken seedsgerminate over an increased range of temperatures. When

seeds germinate over the widest range of temperaturespossible for the species, they are non-dormant (Baskin &Baskin 2004). The optimal burial temperature for increas-ing the temperature range for germination of C. leptophyl-lum seeds was 30/15°C (Fig. 3). Likewise, the optimumtemperature to break PD in seeds of other winter annualApiaceae with MPD was high (Table 1). High summertemperatures also promote the breaking of PD in seeds ofwinter annuals with fully developed embryos (e.g. Baskin& Baskin 1986). The optimum temperatures for germina-tion of C. leptophyllum seeds were 15/6–30/15°C, similar tomost other Apiaceae (Table 1).

(a)

(b)

(c)

Fig. 5 Phenology of germination (a) (mean � 1 standard error [SE], SE shown if �5%) and of embryo growth (b) (mean � 1 SE, SE shownif �0.03 mm) for Cyclospermum leptophyllum seeds collected on 5 June 2005 and sown outside on 18 June 2005. Mean weekly maximumand minimum temperatures and total weekly precipitation (c) are shown for the study period (12 June 2005–1 April 2006). Soil was kepteither continuously moist or was subjected to a moisture regime that simulated natural precipitation and soil moisture levels.

CYCLOSPERMUM LEPTOPHYLLUM S E E D G E R M I N AT I O N 39

Plant Species Biology 23, 33–42 © 2008 The AuthorsJournal compilation © 2008 The Society for the Study of Species Biology

In laboratory studies, C. leptophyllum seeds germinatedto high percentages at 30/15 and at 35/20°C when incu-bated for up to 16 weeks (Figs 1,2). Of these two tempera-ture regimes, seeds buried under natural temperatures innorth-central Kentucky gained the ability to germinate tohigh percentages only at 30/15°C (Fig. 4). Apparently, thediscrepancy between the laboratory and greenhouseresults can be explained by the fact that seeds do not gainthe ability to germinate at 35/20°C unless they have beenexposed to this temperature for several weeks. In thegreenhouse, buried seeds were never exposed to temper-atures higher than 30°C (mean monthly maximum/minimum temperatures were 30.1/23.2°C in July and28.8/21.0°C in August; see Baskin et al. 1993).

In the greenhouse in north-central Kentucky and in thefield in middle Tennessee, little or no embryo growthoccurred during summer. When seeds buried in thegreenhouse during mid-summer 1991 were exhumed inSeptember, embryos were still tiny (0.35 mm). However,the September–exhumed seeds germinated to 93–100% inlight at 15/6–25/15°C after 2 weeks (Fig. 4). Apparently,in C. leptophyllum seeds PD must be broken during thesummer before embryo growth (breaking of MD) takesplace in autumn. Likewise, MD is not broken in seeds ofChaerophyllum tainturieri (Baskin & Baskin 1990), C. proc-umbens (Baskin et al. 2004) and Ptilimnium nuttallii (Baskinet al. 1999) during summer. As in seeds of several otherwinter annual Apiaceae (Table 1), embryos grow inmost seeds of C. leptophyllum during autumn (Fig. 5).Consequently, embryos grow at temperatures above thosesuitable for cold stratification, that is, above 0–10°C; thus,we conclude that seeds of C. leptophyllum have non-deepsimple MPD (Baskin & Baskin 2004).

After PD was broken in summer, seeds of C. leptophyl-lum incubated at low autumn temperatures (15/6, 20/10,25/15°C) germinated to high percentages in light but notin darkness (Fig. 4). However, by March 1992 approxi-mately 75% of the buried seeds had germinated. Thus,embryo growth and germination are faster in light than indarkness, but light is not absolutely required. After PD isbroken, seeds of Chaerophyllum tainturieri require light forembryo growth and germination (Baskin & Baskin 1990),and the percentage of seeds with embryo growth andgermination is higher in light than in darkness for seeds ofC. procumbens (Baskin et al. 2004), Ptilimnium nuttallii(Baskin et al. 1999) and Trepocarpus aethusae (Baskin et al.2003). Seeds of Apium graveolens required light for germi-nation, but this requirement was highly dependent ontemperature, with germination in darkness decreasing astemperatures increased (Thomas et al. 1975; Pressmanet al. 1977). Moreover, A. graveolens seeds exhibit skoto-dormancy, with effective temperatures for its inductionbetween 28 and 40°C (Biddington & Thomas 1979; Bid-dington et al. 1980b). However, seeds of C. leptophyllumT

able

1D

orm

ancy

,ger

min

atio

nan

dfl

ower

ing

char

acte

rist

ics

oftw

oA

pium

taxa

and

for

win

ter

annu

alA

piac

eae

inN

orth

Am

eric

a

Spec

ies

Stat

usin

Nor

thA

mer

ica

Type

ofd

orm

ancy

Opt

imum

tem

p.(°

C)

tobr

eak

phys

iolo

gica

ld

orm

ancy

Opt

imum

alte

rnat

ing

tem

p.(°

C)

for

germ

inat

ion

Seas

onof

embr

yogr

owth

Tim

eof

flow

erin

g†R

efer

ence

(s)

Api

umgr

aveo

lens

Non

-nat

ive

MD

—‡

22–2

5/12

–15

—‡

May

–Jul

yTh

omps

on(1

974a

,b);

Jaco

bsen

&Pr

essm

an(1

979)

;Bid

din

gton

etal

.(1

980a

);Pé

rez-

Gar

cía

etal

.(19

95)

A.p

rost

ratu

mss

p.ph

illip

ii—

§M

D—

‡25

/10

—‡

—‡

Coc

hran

e&

Prob

ert

(200

6)B

uple

urum

rotu

ndifo

lium

Non

-nat

ive

MPD

Sum

mer

15/

6–30

/15

Aut

umn

June

Bask

in&

Bask

in(1

974)

Cha

erop

hyllu

mpr

ocum

bens

Nat

ive

MPD

Sum

mer

20/

10–2

5/15

Aut

umn

Apr

il–M

ayBa

skin

etal

.(20

04)

Cha

erop

hyllu

mta

intu

rier

iN

ativ

eM

PD30

/15

,35/

2025

/15

Aut

umn

Mar

ch–M

ayBa

skin

&Ba

skin

(199

0)C

yclo

sper

mum

lept

ophy

llum

Non

-nat

ive

MPD

30/

1515

/6–

30/

15A

utum

nA

pril–

earl

yJu

nePr

esen

tst

udy

Pti

limni

umnu

ttal

liiN

ativ

eM

PD25

/15

,30/

1520

/10

–25/

15A

utum

nJu

ne–J

uly

Bask

inet

al.(

1999

)T

orili

sar

vens

isN

on-n

ativ

eM

D—

‡20

/10

–30/

15A

utum

nJu

ne–J

uly

Bask

in&

Bask

in(1

975)

Tre

poca

rpus

aeth

usae

Nat

ive

MPD

30/

1515

/6–

25/

15A

utum

nM

ay–J

une

Bask

inet

al.(

2003

)

† Gle

ason

and

Cro

nqui

st(1

991)

,Wea

kley

(200

6).‡ N

otap

plic

able

(or

not

obse

rved

/te

sted

).§ Su

bspe

cies

end

emic

toA

ustr

alia

.Api

umpr

ostr

atum

isin

trod

uced

into

Was

hing

ton

stat

e(U

SA)

(Nat

ureS

erve

2006

).M

D,m

orph

olog

ical

dor

man

cy;M

PD,m

orph

ophy

siol

ogic

ald

orm

ancy

.

40 J . L . WA L C K ET AL.

© 2008 The Authors Plant Species Biology 23, 33–42Journal compilation © 2008 The Society for the Study of Species Biology

buried in moist soil at 30/15°C for 12 weeks germinatedfrom moderate to high percentages at 15/6–30/15°Cduring 2 weeks of incubation in light (Fig. 3). Germinationat the highest test temperature (35/20°C) was inhibiteddespite pretreatment temperature regimes (5–30/15°C).Apparently, skotodormancy is absent from seeds ofC. leptophyllum.

Some seeds of C. leptophyllum in Tennessee had alreadybeen dispersed in late May and June (J. L. Walck & S. N.Hidayati, pers. obs., 2003, 2005). At simulated summerhabitat temperatures, germination is low in fresh seeds,but it gradually increases over 3 months (Figs 1,2). Thebeginning of germination for seeds sown outside in mid-June 2005 was dependent on soil moisture (Fig. 5). Germi-nation began 2 months earlier under optimum conditionsof continuous moisture than on alternating wet/dry sub-strate simulating natural precipitation. Under both mois-ture regimes, the phenology of embryo growth closelymatched that of germination. Thus, the germination phe-nology of the species is controlled by: (i) a slow germina-tion rate at high (summer) temperatures; and (ii) soilmoisture. Seedlings that emerged in autumn survivedoverwinter, as expected for a winter annual.

Unlike seeds of Chaerophyllum tainturieri (Baskin &Baskin 1990), Ptilimnium nuttallii (Baskin et al. 1999) andTorilis arvensis (Baskin & Baskin 1975), which re-enter PDin autumn/winter, seeds of C. leptophyllum remained non-dormant overwinter and were not induced into dormancyby low temperatures (Fig. 4). In Trepocarpus aethusae, only1–3% of the seeds re-entered PD (Baskin et al. 2003). Seedsof C. leptophyllum exhumed from soil between July 1991and March 1992 germinated up to 100% in light. Incontrast, germination percentages gradually increasedbetween July (0%) and January/March (91%) for seedsexhumed from soil and incubated in darkness. The capac-ity of C. leptophyllum to form a persistent soil seed bank(sensu Walck et al. 2005) appears to be very low because: (i)moderate to high percentages of seeds germinated indarkness (Figs 2b,4b), (ii) high percentages of seeds ger-minated in bags while buried in soil; and (iii) only 0.7%emergence occurred during the second autumn germina-tion period from soil in which seeds were sown, that is,approximately 1.5 years following sowing.

According to information on the flowering season ofwinter annual Apiaceae, natives flower from March to Julyand non-natives flower from April to July, with most flow-ering in May and June, respectively (Table 1). Thus, thereare no clear differences in the flowering phenology ofnatives and non-natives. With regard to seed dormancy,all natives and non-natives studied thus far have MD, andall but Apium graveolens and Torilis arvensis, have somePD, that is, MPD (Table 1). The proportion of seeds ofnative species with PD (and thus MPD) generally varieswith the time that the seeds mature, with seeds maturing

early having more PD than those maturing relatively late.None of the fresh seeds of the natives Chaerophyllum tain-turieri (Baskin & Baskin 1990) or C. procumbens (Baskinet al. 2004), which mature in late May or early June, ger-minated (all had MPD). In contrast, 30% of fresh seeds ofthe native Ptilimnium nuttallii (Baskin et al. 1999), whichmature in mid-July, germinated, and 13–36% of freshseeds of Trepocarpus aethusae (Baskin et al. 2003), whichmatures in early September, germinated. Fresh seeds ofC. leptophyllum, which matures in middle Tennessee at thesame time as C. tainturieri and C. procumbens, germinatedto 47–70%, depending on year and temperature, and thoseof the non-native Torilis arvensis (Baskin & Baskin 1975),which matures approximately 1 month before T. aethusae,germinated to 74–95%. However, this pattern of seeds ofnon-native species having less PD than natives is not con-sistent because fresh seeds of the non-native Bupleurumrotundifolium, which matures in early July, germinated toonly 5–18% (Baskin & Baskin 1974). Furthermore, seeds ofB. rotundifolium remain attached to the dead, uprightmother plants during summer, during which time PD isbroken, and then seeds germinate when dispersed inautumn (Baskin & Baskin 1974).

In Torilis arvensis, whose seeds have only MD, seeddispersal is delayed until autumn, when conditions aresuitable for seed germination and seedling survival(Baskin & Baskin 1975). Germination of both native andnon-native species with MPD is prevented in summer,largely because seeds have conditional PD, whichprevents them from germinating at high (35/20°C)summer temperatures. With regard to the portion of theseed population with PD (and MPD), C. leptophyllum ismost similar to the late-maturing native Trepocarpus aethu-sae than to other native or non-native species. The PD isbroken in seeds of both native and non-native speciesduring summer, allowing MD to be broken as soon as thesoil moisture increases and temperatures decrease inautumn. Thus, mechanisms (e.g. lack of dispersal, pres-ence of PD, low temperature requirement to break MD)that prevent germination of seeds of non-native winterannual Apiaceae during summer, allow seeds of thesespecies to germinate in autumn, along with those of nativespecies, when conditions are favorable for seedling estab-lishment and growth.

Acknowledgments

We thank Dr David H. Webb for collecting seeds inAlabama in 1991 and Marta Rolig and Shea Cofer forassisting with the study.

References

Baskin C. C., Baskin J. M. & Chester E. W. (1993) Germinationecology of Leptochloa panicoides, a summer annual grass ofseasonally dewatered mudflats. Acta Oecologica 14: 693–704.

CYCLOSPERMUM LEPTOPHYLLUM S E E D G E R M I N AT I O N 41

Plant Species Biology 23, 33–42 © 2008 The AuthorsJournal compilation © 2008 The Society for the Study of Species Biology

Baskin C. C., Baskin J. M. & Chester E. W. (1999) Seed dormancyin the wetland winter annual Ptilimnium nuttallii (Apiaceae).Wetlands 19: 359–364.

Baskin C. C., Baskin J. M. & Chester E. W. (2003) Ecological lifecycle of Trepocarpus aethusae Nutt ex DC. and comparisonswith two other winter annual Apiaceae native to easternUnited States. Castanea 68: 43–55.

Baskin C. C., Hawkins T. S. & Baskin J. M. (2004) Ecological lifecycle of Chaerophyllum procumbens variety shortii (Apiaceae), awinter annual of the North American eastern deciduousforest. Journal of the Torrey Botanical Society 131: 126–139.

Baskin J. M. & Baskin C. C. (1974) Some aspects of the autecologyof Bupleurum rotundifolium in Tennessee cedar glades. Journalof the Tennessee Academy of Science 49: 21–24.

Baskin J. M. & Baskin C. C. (1975) Ecophysiology of seed dor-mancy and germination in Torilis japonica [arvensis] in relationto its life cycle strategy. Bulletin of the Torrey Botanical Club 102:67–72.

Baskin J. M. & Baskin C. C. (1985) Does seed dormancy play a rolein the germination ecology of Rumex crispus? Weed Science 33:340–343.

Baskin J. M. & Baskin C. C. (1986) Temperature requirements forafter-ripening in seeds of nine winter annuals. Weed Research26: 375–380.

Baskin J. M. & Baskin C. C. (1990) Germination ecophysiology ofseeds of the winter annual Chaerophyllum tainturieri: a newtype of morphophysiological dormancy. Journal of Ecology 78:993–1004.

Baskin J. M. & Baskin C. C. (2004) A classification system for seeddormancy. Seed Science Research 14: 1–16.

Biddington N. L. & Thomas T. H. (1979) Residual effects of hightemperature pre-treatments on germination of celery seeds(Apium graveolens). Physiologia Plantarum 47: 211–214.

Biddington N. L., Thomas T. H. & Dearman A. S. (1980a) Thepromotive effect on subsequent germination of treatingimbibed celery seeds with high temperature before or duringdrying. Plant, Cell and Environment 3: 461–465.

Biddington N. L., Thomas T. H. & Dearman A. S. (1980b) Theeffect of temperature on the germination-promoting activitiesof cytokinin and gibberellin applied to celery seeds (Apiumgraveolens). Physiologia Plantarum 49: 68–70.

Burtt B. L. (1989) Proposal to conserve 6004a CiclospermumLagasca (Umbelliferae) with Ciclospermum leptophyllum (Pers.)Sprague as typ. cons. Taxon 38: 507–509.

Cochrane A. & Probert R. (2006) Temperature and dormancy-breaking treatments: germination of endemic and geographi-cally restricted herbaceous perennials. Australian Journal ofBotany 54: 349–356.

Constance L. (1990) Tardy transfers from Apium to Ciclospermum(Umbelliferae). Brittonia 42: 276–278.

Constance L. (1993) Apiaceae [Umbelliferae]. In: Hickman J. C.(ed.). The Jepson Manual: Higher Plants of California. Universityof California Press, Berkeley, pp. 136–166.

Coolbear P., Toledo P. E. & Setagoses U. (1991) Effects of tem-perature of pre-sowing hydration treatment and subsequentdrying rates on the germination performance of celery seed.New Zealand Journal of Crop and Horticultural Science 19: 9–14.

Gleason H. A. & Cronquist A. (1991) Manual of Vascular Plants ofNortheastern United States and Adjacent Canada, 2nd edn. TheNew York Botanical Garden, Bronx.

Grushvitzky I. V. (1967) After-ripening of seeds of primitivetribes of angiosperms, conditions and peculiarities. In: BorrissH. (ed.). Physiologie, Ökologie und Biochemie der Keimung. Ernst-Moritz-Arndt-Universität, Griefswald, pp. 329–336 + 8figures.

Jacobsen J. V. & Pressman E. (1979) A structural study of germi-nation in celery (Apium graveolens L.) seed with emphasis onendosperm breakdown. Planta 144: 241–248.

Jacobsen J. V., Pressman E. & Pyliotis N. A. (1976) Gibberellin-induced separation of cells in isolated endosperm of celeryseed. Planta 129: 113–122.

McIntyre S., Martin T. G., Heard K. M. & Kinloch J. (2005) Planttraits predict impact of invading species: an analysis of herba-ceous vegetation in the subtropics. Australian Journal of Botany53: 757–770.

Menglan S., Fading P., Zehui P. et al. (2005) Apiaceae (Umbel-liferae). In: Flora of China Editorial Committee (ed.). Flora ofChina, Vol. 14 Apiaceae Through Ericaceae. Science Press, Beijingand Missouri. Botanical Garden Press, St. Louis, pp. 1–205.

NatureServe (2006) Natureserve Explorer: An Online Encyclopedia ofLife, Version 6.1. NatureServe, Arlington. [cited 7 December2006]. Available from URL: http://www.natureserve.org/explorer

National Oceanic and Atmospheric Administration (2003) Com-parative Climatic Data for the United States Through 2002.National Climatic Data Center, National Oceanic and Atmo-spheric Administration, US Department of Commerce,Asheville.

Pérez-García P., Pita J. M., Gonzáles-Benito M. E. & Iriondo J. M.(1995) Effects of light, temperature and seed primingon germination of celery seeds (Apium graveolens L.). SeedScience and Technology 23: 377–383.

Pressman E., Negbi M., Sachs M. & Jacobsen J. V. (1977) Varietaldifferences in light requirements for germination of celery(Apium graveolens L.) seeds and the effects of thermal andsolute stress. Australian Journal of Plant Physiology 4: 821–831.

Pressman E., Shaked R. & Negbi M. (1988) Germination of seedsof annual and biennial celery (Apium graveolens). PhysiologiaPlantarum 72: 65–69.

Radford A. E., Ahles H. E. & Bell C. R. (1968) Manual of theVascular Flora of the Carolinas. University of North CarolinaPress, Chapel Hill.

SAS (2003) The SAS System for Windows, Version 9.1. SAS Insti-tute, Cary.

Thomas T. H., Palevitch D., Biddington N. L. & Austin R. B. (1975)Growth regulators and the phytochrome-mediated dormancyof celery seeds. Physiologia Plantarum 35: 101–106.

Thompson P. A. (1974a) Germination of celery (Apium graveolensL.) in response to fluctuating temperatures. Journal of Experi-mental Botany 25: 156–163.

Thompson P. A. (1974b) Effects of fluctuating temperatures ongermination. Journal of Experimental Botany 25: 164–175.

Walck J. L., Baskin J. M., Baskin C. C. & Hidayati S. N. (2005)Defining transient and persistent seed banks in species withpronounced seasonal dormancy and germination patterns.Seed Science Research 15: 189–196.

Weakley A. S. (2006) Flora of the Carolinas, Virginia, and Georgia,and Surrounding Areas, working draft of 9 August 2006. Uni-versity of North Carolina Herbarium, North Carolina Botani-cal Garden, Chapel Hill. [cited 7 December 2006]. Availablefrom URL: http://www.herbarium.unc.edu/flora.htm

42 J . L . WA L C K ET AL.

© 2008 The Authors Plant Species Biology 23, 33–42Journal compilation © 2008 The Society for the Study of Species Biology