Seed Dormancy and Germination Vary Within and Among Species of Milkweed
Seed Dormancy and Germination Vary Within and Among Species of Milkweed
Seed Dormancy and Germination Vary Within and Among Species of Milkweed
Abstract. Pollinators in general and monarch butterflies in particular are in decline due to habitat loss. Efforts to
restore habitats for insects that rely on specific plant groups as larvae or adults depend on the ability of practition-
ers to grow and produce these plants. Monarch larvae feed exclusively on milkweed species, primarily in the genus
Asclepias, making propagation and restoration of these plants crucial for habitat restoration. Seed germination pro-
tocols for milkweeds are not well established, in part due to the large number of milkweed species and conflicting
reports of seed dormancy in the genus. We tested for seed dormancy and the optimum period of cold stratification
in 15 populations of A. speciosa and 1–2 populations of five additional species, including A. asperula, A. fascicularis,
A. subulata, A. subverticillata and A. syriaca. We exposed seeds to cold (5 °C) moist conditions for 0, 2, 4, 6 and 8
weeks and then moved them to 15 °C/25 °C alternating temperatures. In A. speciosa, dormancy was detected in
eight populations, and this dormancy was broken by 2–4 weeks of cold stratification. The remaining seven popula-
tions showed no dormancy. Seed dormancy was also detected in two populations of A. fascicularis (broken by 4–6
weeks of cold stratification) and a single population of A. syriaca (broken by 2 weeks of cold stratification). No dor-
mancy was detected in A. asperula, A. subulata or A. subverticillata. Seed dormancy appears to be widespread in the
genus (confirmed in 15 species) but can vary between populations even within the same species. Variation in seed
dormancy and cold stratification requirements within and among Asclepias species suggests local adaptation and
maternal environments may drive seedling ecology, and that growers should watch for low germination and use
cold stratification as needed to maximize seed germination and retain genetic variability in restored populations.
Keywords: Cold stratification; conservation biology; habitat restoration; monarch butterfly; plant propagation; pol-
linator conservation.
© The Author(s) 2018. Published by Oxford University Press on behalf of the Annals of Botany Company.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/
licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
15 years (Jepson et al. 2015), due in part to loss of milk- suggested germination protocols for Asclepias species
weed in their summer breeding habitat from increased are inconsistent or even contradictory (Borland 1987).
use of herbicides in intensive farming practices (Brower For example, cold stratification has been recommended
et al. 2012). to release seeds of A. speciosa from dormancy but some
Plants in the milkweed family (Asclepiadaceae) are practitioners report that no cold stratification is neces-
the exclusive food source for monarch butterfly larvae. sary for large-scale propagation (Stevens 2000).
Monarchs in North America feed on at least 27 species Here we present the results of research on seed dor-
in the genus Asclepias and a few closely related genera mancy and germination in six Asclepias species. In par-
(Malcolm and Brower 1986). Milkweed populations have ticular, we examine variation in dormancy and cold
suffered considerable declines in habitat in the central stratification requirements among populations of A. spe-
at 25 °C. We defined germination as emergence of the trials (not shown) we found that 95 % or more seeds
radicle at least 3 mm, and counted germinated seeds germinated within this period of time, and any addi-
after 10 days (Baskin and Baskin 2001). In preliminary tional germination took many weeks to complete. Seed
Table 1. Asclepias species and populations included in dormancy and germination tests, with seed sources, samples (50 seeds each) per
treatment and years in storage. †BSE indicates the US Forest Service, Bend Seed Extractory; GRIN indicates the US National Germplasm
Resources Information Network.
Species Population Seed source† (accession number) Samples per treatment Years in storage
viability was tested with tetrazolium (TZ) by the Oregon A. speciosa, with substantial improvement in germin-
State University Seed Lab using standard procedures ation after stratification for eight out of the 15 popu-
(Elias et al. 2012). The TZ test was used to estimate lations tested (grouped for ease of visual inspection in
the percentage of live and dead seeds in each seed Fig. 3A), but only slight or no dormancy in the remain-
source, regardless of dormancy level (Baskin and Baskin ing seven populations (Fig. 3B). Seed viability was over
2001). Seeds were cut longitudinally to expose interior 90 % in most A. speciosa populations as tested with TZ,
tissues and facilitate the entrance of TZ solution, and with the exception of two populations with lower viabil-
incubated in a 1 % TZ solution for 24 h at 35 °C. Seeds ity, Willamette Valley, OR (73 ± 6.0 %) and Malheur, OR
were inspected for TZ staining, specifically the pattern (28 ± 6.1 %). In populations with substantial dormancy,
and intensity of red colour in live tissues in seeds. Seed <60 % of seeds germinated without cold stratification,
with uniform stain colour were recorded as viable seeds. but with 2 or more weeks of cold stratification germin-
Tetrazolium tests were performed with samples of 157 ation increased to the levels of seed viability, or nearly
to 210 seeds per source population. so. For example, without cold stratification 32 ± 5 %
of seeds from Little Pend Oreille, WA, germinated, but
Analysis after 2 weeks of cold stratification germination rose to
We tested for effects of cold stratification and popula- 91 ± 5 %, close to the viability estimate of 96 ± 2.6 %
tion with a general linear model of analysis of variance, (Fig. 3A). Seed from Malheur, OR, required 4 weeks of
with cold stratification treatment as a fixed effect and cold stratification to fully break dormancy, while ger-
seed source as a random effect using NCSS statistical mination of seeds from Willamette Valley, OR, did not
software (Hintze 2008). Separate analyses were per- germinated to the level of their estimated viability.
formed for Asclepias speciosa populations, and the five Populations with largely non-dormant seeds (Fig. 3B)
additional Asclepias species and their populations. Data germinated to very high rates at or near their viability
are reported with the mean ± 95 % confidence interval. estimate without any cold stratification. Seed germin-
Confidence intervals for the viability estimates were cal- ation was not higher after 8 weeks of cold stratification
culated from the normal approximation of the binomial than 6 weeks in any population tested, so this treatment
distribution using the proportion of seeds that were con- is not shown (Fig. 3).
sidered viable, the proportion of seeds that were consid-
ered non-viable and the total number of seeds tested. Interspecific differences in dormancy and
germination
Dormancy and germination varied among species of
Results
Asclepias examined here. Again there was a signifi-
cant interaction between cold stratification treatment
Intraspecific population differences in dormancy and population for the five taxa examined (F = 40.29,
and germination of A. speciosa df = 24;105, P < 0.00001), with two species showing
There was a significant stratification treatment by popu- positive effects on germination of cold stratification of
lation interaction (F = 17.62, df = 56;255, P < 0.0001) for 2 or more weeks, but the remaining taxa possessing
Table 2. Summary of seed dormancy in Asclepias species tested at the USDA/ARS National Laboratory for Genetic Resources Preservation,
Seed Quality Lab, with duration of cold stratification (if any), number of populations sampled and number of populations with and without
dormancy. Germination is relative to viability, and is shown as the amount or range determined for each period of stratification. Populations
were classified as possessing seed dormancy if <80 % of viable seeds germinated. Each population represents a separate seed collection.
Species Stratification No. of populations Viability Populations without dormancy Populations with dormancy
sampled
No. of populations Germination No. of populations Germination
germination requirements, and both vary widely in individual mother plants (Andersson and Milberg 1998).
cold stratification needs. Asclepias syriaca populations Seed dormancy can affect interactions within and
with seed dormancy may require 56 days or more of among species by determining the seasonal timing of
cold stratification (Green and Curtis 1950; Oegema and germination, seedbank dynamics, and exposure of seeds
Fletcher 1972), or as little as 7 days (Evetts and Burnside and seedlings to hazards and competition for resources
1972; Farmer et al. 1986). We found 2 weeks of cold strati- (Harper 1977; Baskin and Baskin 2001). Dormancy has
fication was sufficient to release dormancy in this spe- been shown to be under genetic control in some spe-
cies, but we did not try a shorter period. In A. tuberosa, cies, often in response to natural selection, such as in
as much as 90–120 days of cold stratification (Salac and Digitaria (Hacker 1984), Arabidopsis (Alonso-Blanco et al.
Hesse 1975; Cullina 2000; Blessman and Flood 2001) or 2003; Bentsink et al. 2010) and Oryza (Gu et al. 2004),
as little as 21 days (AOSA 2016) may be needed for opti- or it can result from conditions during seed maturation
mal germination of dormant seed lots. Similarly, 30 days in the environment of the maternal plant and zygote
(Vandevender and Lester 2014a) down to 1 week or less (Bodrone et al. 2017; Penfield and MacGregor 2017), or
(Lincoln 1983; Schultz et al. 2001a) of cold stratification both (Postma and Agren 2015). To our knowledge, nei-
may be needed to improve germination in A. incarnata. ther genetic nor environmental controls on dormancy
Among species tested by the National Laboratory for have been documented in Asclepias. Despite the wide-
Genetic Resources Preservation, most populations that spread presence of seed dormancy in Asclepias species,
received cold stratification of 7–21 days were released persistent seed banks have not been detected even
from dormancy (Table 2). when milkweeds are present in the above-ground vege-
tation (Smith and Kadlec 1983; Johnson and Anderson
Factors that affect dormancy and germination in 1986). Seedling emergence of A. syriaca exceeds 80 %
Asclepias for seeds buried 0.5–4 cm (Yenish et al. 1996) leaving
Seed dormancy can vary among species, popula- only a small proportion of seeds in the soil unaccounted
tions (Keith and Myerscough 2016; Siles et al. 2017), for and which could contribute to a seedbank or suc-
collections from the same population but different cumb to mortality. Germination of Asclepias seeds
years (Green and Curtis 1950; Kaye 1999) and among can also be affected by light, scarification, substrate,
Table 3. Duration of cold stratification needed to break dormancy in Asclepias species from published sources, with post stratification
temperatures used or recommended for germination, where reported.
(Continued)
temperatures, after-ripening and other factors. Light Baskin and Baskin (1977) suggest that with after-ripen-
may be required for germination in some Asclepias spe- ing A. syriaca seeds can germinate at lower tempera-
cies, such as A. incarnata (Lincoln 1976; Deno 1993), or tures and after shorter periods of cold stratification, and
have no effect on others, including A. tuberosa (Mitchell Bhowmik (1978) found that germination increased grad-
1926; Deno 1993). Light has little or no effect on ger- ually with time in storage from 1 to 11 months. Seed
mination of A. syriaca once adequate (≥2 weeks) cold dormancy in A. speciosa populations included in our
stratification has been provided (Baskin and Baskin experiments may have been affected by the amount of
1977; Lincoln 1983). Mechanical scarification of the time seeds were in storage as well as local environmen-
seed coat improves germination in A. syriaca (Oegema tal conditions. We used fresh seeds that had been col-
and Fletcher 1972; Evetts and Burnside 1972). Smoke lected in the same year of the experiment and had been
treatments that mimic wildfire smoke can improve stored for ~3 months, as well as seeds that had been in
germination of A. syriaca as well (Mojzes and Kalapos dry, cold (<0 °C) storage for up to 6 years. Fresh seeds
2015), but burning and soil disturbance reduced seed- from six populations had >40 % dormancy, while seeds
ling emergence in A. meadii (Roels 2013). Experiments stored for 2, 5 and 6 years had <15 % dormancy. But
with temperature and substrate have found that A. syri- seeds stored for 4 years from two populations had dor-
aca germinates best under alternating temperatures of mancy levels similar to fresh seeds, suggesting that dor-
20 °C/30 °C on clay or clay mixed with peat (Farmer et al. mancy can persist even in stored seeds and may involve
1986), or 21 °C/26 °C on sand or loam (Radivojevic et al. other factors beyond storage. Further confounding the
2016). Temperatures that promote successful germin- effects of storage time was that most of our fresh seeds
ation after cold stratification of Asclepias also vary con- came from latitudes north of 40° and all of the stored
siderably, and most reports recommend an alternating seeds came from farther south, and no seed dormancy
temperature regime. For example, a temperature cycle was found in the southern populations; in other words,
of 10 °C/30 °C is effective for A. perennis (Edwards et al. most non-dormant seeds had been stored and were
1994) and the standard for seed testing in A. tuberosa from southern latitudes. Therefore, it is not possible for
(AOSA 2016). Alternating 16 °C/27 °C (Lincoln 1983) has us to separate the effects of storage time from local
been successfully used for A. incarnata, and cycles of environments, but we speculate that both seed storage
15 °C/30 °C (Baskin and Baskin 1977) and 20 °C/30 °C and local conditions (Seglias et al. 2018) could influ-
(Evetts and Burnside 1972; Lincoln 1976; Farmer et al. ence seed dormancy in A. speciosa and possibly other
1986) have been recommended for A. syriaca. Bhowmik Asclepias species.
found germination in A. syriaca increased with increas-
ing temperature from 10 °C to 27 °C. We used alternat-
ing 15 °C/25 °C in our trials and this cycle was generally Conclusions
effective for germination of the species and populations Growers of Asclepias species should consider cold strati-
we examined. In the case of the A. speciosa population fication to break seed dormancy when it is encountered.
from Malheur, OR, we also used a cycle of 20 °C/30 °C but We found widespread evidence of seed dormancy in
found germination to be reduced by about 30 % com- the genus through germination tests reported here as
pared to the 15 °C/25 °C cycle (data not shown). well as published and unpublished reports. Seed dor-
After-ripening, a period of time after seed dispersal in mancy has been detected in at least some populations
which changes in the seeds affect their ability to germin- of 15 species in the genus, including A. amplexicaulis,
ate, can affect some species of Asclepias. For example, A. exaltata, A. fascicularis, A. floridiana, A. incarnata,
A. hirtella, A. meadii, A. ovalifolia, A. purpurascens, Management Seeds of Success Program; A. Pilmanis
A. speciosa, A. sullivantii, A. syriaca, A. tuberosa, A. ver- and R. Hosna from the Bureau of Land Management;
ticillata and A. viridiflora (Table 3). Cold stratification D. Perkins from College of Idaho; L. Boyer from Heritage
is a relatively simple method to break this dormancy Seedlings, Inc.; K. See from the Uncompahgre Native
and increase seed germination substantially, and can Plant Program; and K. Herriman from the US Forest
shorten germination time (Salac and Hesse 1975) and Service Bend Seed Extractory. M. Cashman of the US
even out differences in germination across populations Department of Agriculture, Agricultural Research Service
(Milberg and Andersson 1998). For native plant special- Western Regional Plant Introduction Station in Pullman,
ists, commercial growers and managers who want to WA, assisted with making seeds available through the
grow milkweed to support habitat restoration for mon- US National Germplasm Resources Information Network
pathways. Proceedings of the National Academy of Sciences Deno NC. 1993. Seed germination theory and practice, 2nd edn.
107:4264–4269. State College, PA: Pennsylvania State University.
Betz RF. 1989. Ecology of Mead’s milkweed (Asclepias meadii Dumroese RK, Luna T, Pinto JR, Landis TD. 2016. Forbs: foundation
Torrey). Proceedings of the North American Prairie Conferences for restoration of monarch butterflies, other pollinators, and
13:187–191. greater sage-grouse in the western United States. Natural Areas
Bhowmik PC. 1978. Germination, growth, and development of com- Journal 36:499–511.
mon milkweed (Asclepias syriaca). Canadian Journal of Plant Ecker R, Barzilay A. 1993. Propagation of Asclepias tuberosa from
Science 58:493–498. short root segments. Scientia Horticulturae 56:171–174.
Biesmeijer JC, Roberts SP, Reemer M, Ohlemüller R, Edwards M, Edwards AL, Wyatt R, Sharitz RR. 1994. Seed buoyancy and viabil-
Peeters T, Schaffers AP, Potts SG, Kleukers R, Thomas CD, ity of the wetland milkweed Asclepias perennis and an upland
Settele J, Kunin WE. 2006. Parallel declines in pollinators and milkweed, Asclepias exaltata. Bulletin of the Torrey Botanical
Johnson RG, Anderson RC. 1986. The seed bank of a tallgrass prairie Postma FM, Ågren J. 2015. Maternal environment affects the gen-
in Illinois. The American Midland Naturalist 115:123–130. etic basis of seed dormancy in Arabidopsis thaliana. Molecular
Kaye TN. 1999. Propagation of endangered species: variable ger- Ecology 24:785–797.
mination of pink sandverbena from Pacific Coast beaches. Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, Kunin
Combined Proceedings of the International Plant Propagators WE. 2010. Global pollinator declines: trends, impacts and driv-
Society 49:617–621. ers. Trends in Ecology & Evolution 25:345–353.
Keith DA, Myerscough PJ. 2016. Population variation in germination Radivojevic L, Saric-Krsmanovic M, Umiljendic JG, Bozic D, Santric L.
traits and its implications for responses to climate change in 2016. The impacts of temperature, soil type and soil herbicides
a fire-prone plant species complex. Plant Ecology 217:781–788. on seed germination and early establishment of common milk-
Klein AM, Vaissiere BE, Cane JH, Steffan-Dewenter I, Cunningham weed (Asclepias syriaca L.). Notulae Botanicae Horti Agrobotanici
SA, Kremen C, Tscharntke T. 2007. Importance of pollinators in Cluj-Napoca 44:291–295.
Vandevender J, Lester R. 2014a. Propagation protocol for produc- for Reforestation, Nurseries, and Genetic Resources. http://
tion of Container (plug) Asclepias incarnata L. plants USDA NRCS - NativePlantNetwork.org (16 June 2016).
Appalachian Plant Materials Center Alderson, West Virginia. Williams SL. 2001. Reduced genetic diversity in eelgrass transplan-
US Department of Agriculture, Forest Service, National Center tations affects both population growth and individual fitness.
for Reforestation, Nurseries, and Genetic Resources. http:// Ecological Applications 11:1472–1488.
NativePlantNetwork.org (16 June 2016). Williams PH, Osborne JL. 2009. Bumblebee vulnerability and con-
Vandevender J, Lester R. 2014b. Propagation protocol for produc- servation world-wide. Apidologie 40:367–387.
tion of Container (plug) Asclepias tuberosa L. plants USDA NRCS - Yenish JP, Fry TA, Durgan BR, Wyse DL. 1996. Tillage effects on seed
Appalachian Plant Materials Center Alderson, West Virginia. distribution and common milkweed (Asclepias syriaca) estab-
US Department of Agriculture, Forest Service, National Center lishment. Weed Science 44:815–820.