RESEARCH AT SDSU and INSIGHTS FROM A GROWER
Richard Little
Copyright February 1999. Permission is granted
by the author to any individual person to make five or fewer hardcopies
of any information presented herein, provided that the author is conspicuously
credited on each copy. If more hardcopies are to be made or if electronic
transfers of copies are desired, requests should be made to the author
at one of the addresses listed at the bottom, and should
include contact information, business affiliation, and intended use of
the copies. Copying of more than one paragraph onto a web-site is forbidden.
This document provides a confident overview of basic agronomic research and a tentative view of the market situation of an American wildflower, Echinacea angustifolia var. angustifolia D.C, also known as black samson or narrow-leaved purple coneflower. Market and price information that follows is limited to my recent experience with only a small percentage of growers, wildcrafters, marketers and seedsmen who deal with Echinacea.
Most of my comments come from first-hand knowledge tempered by feedback, notably from my graduate advisor, Dr. Neil Reese. Where possible, I back up opinions with evidence from experiments conducted at SDSU. Major references are provided for supporting literature. Production ideas were developed 1992 through 1995 and were tested commercially in 1995 and 1996 during my tenure as a consultant and farm manager for Plantation Medicinals Inc (PMI). Many of the ideas were proven and some were discarded during this period. I am constrained from commenting on the commercial experience.
My sense of the resiliency of E. angustifolia in its native habitat was gained from periodic forays down country roads in South Dakota and surrounding states since 1991 and from conversations with wildcrafters and soil conservationists. A two-month seed collection trip from Texas to Montana in 1995 under the auspices of PMI solidified my sense of how E. angustifolia acts its native habitat.
This wildflower not only inspires research into its secrets and potentials; it provides a connection between the grower and all of nature. I hope to make this connection available to others by stimulating questions for further research at school or on the farm. From love of one native plant we can learn to appreciate other native plants and the diversity of nature.
I would like to acknowledge four authors: Steven Foster (Echinacea:
Nature's Immune Enhancer, 1991, Healing Arts Press, One Park Street,
Rochester VT 05767); Christopher Hobbs (The Echinacea Handbook,
1989, Eclectic Medical Publications, Portland OR); Richo Cech (Native
American Tonic Roots, Horizon Herbs, 541-846-6704); and Li, Thomas
S.C. (Echinacea: Cultivation and Medicinal Value, HortTechnology
8-2:122-129) whose writings complement the information presented in this
newsletter. Medicinal Herbs in the Garden, Field, and Marketplace ,
1999, by Lee Sturdevant and Tim Blakely, (San Juan Naturals, PO Box 642,
Friday Harbor, WA 98250) is also worth exploring.
"Am
I mad to consider growing E. angustifolia in 1999?"
If your seed source can be verified, you might be able to market seed to medicinal plant growers or to soil conservationists for roadside planting or prairie restoration. Seed can be harvested as early as the second year. As for harvesting this crop for drug content (at the end of the third or fourth year), the common wisdom is that the current root prices won't last more than a year, since there will likely be a glut of cultivated roots on the market by then. This prediction has been repeated for two or three years already. Either the expanded markets have outpaced cultivated production or the cultivated roots are taking longer than anticipated to reach the markets. Considering the common wisdom, based on production and pricing patterns for crops like ginseng and passion flowers, feasibility studies for growing E. angustifolia should be based on much lower than current root and seed prices. Seed prices are likely to plummet at the same time as root prices, since potential growers (seed buyers) will be discouraged from planting a first crop. The supply of seeds from cultivated plants will quickly exceed demand. A single medium-sized E. angustifolia farm (14 acres) can produce enough seed in two years to plant 1000 acres at 0.7 pounds pure live seed per acre. (From each acre, approximately 25 pounds of pure live seed can be harvested per year during the second and third years after planting). So I see only a short-lived opportunity to capitalize on the high-priced seed market. Any new plantings this coming year will probably miss the high-priced seed market.
Until recently, the demand for E. angustifolia roots was supplied exclusively from wild populations. Published estimates of historical wild root harvests (Steven Foster, Herbalgram, volumes 30 and 32, American Botanical Council, http://www.herbalgram.org) vary between 50 and 200 tons per year (45,000 and 180,000 kg yr-1). Demand for wild-harvested roots remained high in April 1998 with wholesale prices at $44 to $51 per kg dried root ($20 to $23 per pound) (personal communication with a ND collector), despite the anticipated arrival of cultivated produce on the market this year. Currently I estimate that over 40 ha (100 acres) are cultivated in the United States to supply fresh or dried roots for processing into medicinal extracts. Using a conservative estimate of 700 kg ha -1 dried root production (625 pounds per acre) after the third year of growth, at least 15% of the historic demand can be supplied from current cultivated acres.
Many individuals around the world have contacted me, voicing their intentions to research this crop or to grow it commercially. In 1994 and 1995 each of several growers who contacted me intended to plant one to five acres of this crop. A planting of 30 acres occurred in South Dakota in 1995, with equivalent plantings to occur in subsequent years (Brookings Register, June 3, 1995, pp. 1 and A9, Brookings SD 57007; Brookings Register, July 17, 1996, p. 1; SD Farm and Home Research 46-4:14-15, winter 1995). In 1996 a group of growers in ND calling itself Dakota Produce Inc. advertised on the internet that they could produce for contracts of up to 100 acres. They intended to plant 30 acres in 1996; but one member of this group informed me that, at least on her farm, only 7 acres were successfully planted. The Sahnish Farm Co-op in Newtown ND was seeking start-up funds for a ten-acre seed plot in 1996, with intentions to make three successive 150-acre plantings in subsequent years. Many farmers have been contacting extension agents in western ND in early 1999 for information on growing Echinacea, resulting in at least one extension meeting devoted to this subject.
Since growers of large acreages of botanicals tend to keep a low profile, I suspect that a much larger acreage is being cultivated than any company or individual knows. Harvey Clark (mentioned later in this section) claims that many acres have been planted in central Canada. Steven Foster told a group at Kansas on October 3, 1998 that he knows of approximately five hundred acres of E. angustifolia under cultivation. Harvey Clark estimates closer to a thousand acres.
The combination of production from a few large growers and many small-scale growers and unbridled harvesting from the wild points toward large root supplies in the next couple of years. When large cultivated quantities come on the market, prices might nosedive. Wholesalers who hoard and dump raw product can influence the market. Their activities will probably determine the time and extent of a price decline.
I project prices for E. angustifolia roots to fall and eventually level at about 30 % above the current prices for E. purpurea roots. Factors that have kept E. angustifolia root prices at a premium over E. purpurea include: 1) the common perception that E. angustifolia is difficult to grow; 2) lack of cultivation information; 3) limited seed availability; 4) lack of registered herbicides; 5) the common expectation that drug content will be highest if the crop is grown in its native subhumid and semi-arid regions with calcareous soils; 6) lack of income during first and second years of production due to poor markets for stems and leaves; 7) lower root yields than for E. purpurea or E. pallida because of slow plant growth; 8) and the perceived superiority of E. angustifolia for certain medicinal uses. The first three factors no longer operate. The fourth factor does not inhibit unscrupulous growers; one or two such growers can put a lot of roots on the market at a low cost of production. It is now possible for the fifth factor to be challenged by commercial labs.
The sixth and seventh factors have not been a strong enough barrier
to limit pioneering production, but will probably influence growers to
grow E. purpurea instead of E. angustifolia if its root prices
fall as low as for E. purpurea roots. [Unprecedented demand in 1995
for anything labeled E. angustifolia inflated the price of its "herb"
(stems and leaves) to $12.00 per kg, (personal communication with a North
Dakota grower) despite the very low drug content of these plant parts.
In contrast, the leaves and stems of E. purpurea have been documented
as having high contents of immune-stimulating polysaccharides. With plentiful
supplies of products from both plants, the price of stems and leaves of
E. angustifolia is destined to fall to next to nothing.] These two
factors will thus help to maintain a slight premium for E. angustifolia
roots, as long as customers continue to prefer E. angustifolia for
some applications.
Since the price of E. purpurea has been fairly steady due to its steady market and long-term production around the world, I would use current prices of E. purpurea as a baseline for projecting E. angustifolia prices. A good place to start a search on current prices would be Saskatchewan's classified ads for herbs, spices and fruit: http://www.aginfonet.sk.caunder "marketplace." Harvey Clark, a market analyst at the Saskatchewan Irrigation Diversification Centre (Box 700, Outlook, SK S0L 2N0, Phone: (306) 867-5402, FAX: (306) 867-9656, E-mail: pf22407@em.agr.ca) posted this marketplace address to the SAF on-line discussion group, and has contributed information on root prices and buyers. He also suggested using the predominantly North American herbal green pages:http://www.herbworld.com/classifieds.htm.
Objectives
During my graduate studies my intention has been to use the information to develop a botanical farm that specializes in domesticating species native to America by selecting and growing superior cultivars from the broadest genetic base possible, with Echinacea angustifolia as my first crop of interest. This intention kept me focused on agronomic traits and cultural practices while I studied the genetic variation of five native populations of E. angustifolia in a half-acre common garden.
Garden Design
The environmental component of variation among plants is controlled by growing ecotypes or "clines" in a common garden. After the variation of traits has been observed and corrected for environmental influences, selection of desirable traits can be attempted.
It is useful for plant selection and breeding to learn whether there is as much variation within a population as there is between populations. If the variation within a population is greater than the variation between populations, the plant breeder can focus on selecting within a single population. If there is more variation between populations, many sources should be collected to enhance a breeding program.
It is also useful to know early in a breeding program whether plant density has a significant impact on a particular trait. Plant spacing is known to dramatically affect sunflowers, the agronomic crop of nearest relation to Echinacea. Experiments involving traits that are dramatically affected by plant spacing will be highly affected by plant mortality. The study of such traits will require a higher degree of management and expense than studies of traits that are not as affected by plant spacing.
Eighteen plants from each family were needed to provide sufficient data for a laboratory genetic study that linked with the common garden study. A family is all seeds from one seed-head. All plants derived from seeds of the same head are referred to as "half-sibs" because they only have the maternal parent in common.
90-cm spacing experiment
The 90-cm spacing study was designed to discern the amount of variation within and among five populations. Eight families for each of five populations, with 6 half-sib individuals of each family were represented in each of three blocks, for a total of 720 plants.
Variable-spacing Experiment at 30-, 60- and 90-cm spacing
The variable-spacing study was designed to discern differences among 3 plant densities and among populations at each density. Six families of three populations with 6 half-sib individuals of each family were represented in each of 3 blocks, for a total of 972 plants.
Garden Establishment
Javad Feghahati collected E. angustifolia seeds in five states for genetic studies. Seeds were planted February 15, 1992 using the method of germination developed by Dr. Feghahati and Dr. Reese.
I chose to transplant rather than to seed directly into the field. I was not interested in direct seeding because I wanted to keep track of individual plants from seeding to harvest. When seedlings began to develop the first true leaf (3 weeks after sowing) they were moved from blotter paper into 1" x 1" x 4" rootrainer cells in sterilized potting medium (3 parts silty clay loam soil: 2 parts coarse perlite: 2 parts sphagnum peat). This mix formed a marginally cohesive root-ball, but it didn't drain well. Saturated soil caused seedlings to quit growing after the 2-leaf stage. Leaves became cupped and sometimes brittle, classical symptoms of ethylene buildup around the roots.
Despite the effects of poor drainage, the plants performed well in the field. They were transplanted May 19-31. I watered them three times during the first month of dry 25 C weather. Only 15 % needed to be replaced through July 6. A late frost on May 25 killed some. Others may have died from damage to the root-ball, despite gentle handling during transplanting. Each winter about 25% of the remaining plants died from a root rot, possibly exacerbated by my practice of harvesting the foliage each fall by cutting to within 4 to 6 inches of the ground.
Extremely wet weather in late 1993 and throughout 1994 affected the plants noticeably. Plants toward the wet end of the plot had a 10-20% lower survival rate than those at the drier end of the field. Of the 1961 original plants, including replacements, 847 survived until root harvest in October 1994.
Cool weather in 1992 and 1993 might have skewed plant growth data from
what would be observed in a more normal year. Yet, "the semiarid climate
of South Dakota is always highly variable. While global temperatures for
1991-1995 were the warmest half-decade on record, South Dakota had its
own extremes. The precipitation totals from 1991-1995 were the greatest
in more than 100 years of South Dakota climate records, and the 1992 and
1993 summers were the coolest consecutive summer seasons in the climate
record beginning in 1890," according to Alan Bender, SD State Climatologist
(http://www.abs.sdstate.edu/ae/weather/weather.htm)
April May June July Aug Sept
Precipitation for Brookings normally 2.07 2.93 4.34 3.32 2.81 2.64
totaling 22.89 inches for 12 months
Precipitation for Brookings in 1992 1.79 1.19 7.98 3.69 4.67 1.87
totaling 26.72 inches for 12 months
Precipitation for Brookings in 1993 1.96 4.32 8.69 5.18 2.27 2.12
totaling 28.58 inches for 12 months
Precipitation for Brookings in 1994 2.99 1.52 10.21 2.08 3.57 2.75
totaling 26.52 inches for the year
Garden Design for Eliminating Environmental Variation
The environmental component of variation among plants is controlled
by growing ecotypes or "clines" in a common garden. The error is managed
by planting half-sibs in equally-spaced hills in randomized complete blocks.
After the variation of traits has been observed and corrected for environmental
influences, selection of desirable traits can be attempted. It is useful
for plant selection and breeding to learn whether there is as much variation
within a population as there is between populations.
I selected 17% of the surviving plants that had the following traits: minimal lodging or disease problems, compact growth habit, fairly uniform flowering and seed maturity, and moderately high seed yield. Dried roots were weighed after selections were made. The selected plants yielded 60% more dried root than the average.
A composite seed sample for each population was grown commercially.
Compared to progeny of wild plants, the selected plants were visually superior.
There were too many uncontrolled factors in the commercial field to make
further comparisons.
Genetic Laboratory Research
SDSU genetic studies
A genetic study using restriction fragment length polymorphic (RFLP) analysis is detailed in the doctoral thesis of Javad Feghahati (1995, Germination, Genetic Variation, and Phylogenetic Relationships within and among Populations of Echinacea angustifolia, SDSU, Brookings, SD).
DNA fragment (RFLP) frequencies revealed phenetic distinctions between
NE and KS populations (south group) and WY, ND and SD populations (north
group). DNA fragments (RFLPs) produced by enzymatic digestion of leaves
from maternal plants were used to create complementary DNA and genomic
DNA libraries. Fragments that were greater than 400 base pairs were used
as probes for 20 samples from each population. The frequency of each fragment
in each population revealed clusters of similarity among populations, using
Hedrick's model (1982, Genetics of Population). ND clustered near WY (.31
identity). The ND and WY group clustered with SD (.26 identity). NE clustered
near KS (.24 identity). The identity between the north group (ND, WY and
SD) and the south group (KS and NE) was .18.
Genetic Studies at other Universities
Genetic similarities with a north-south grouping were also revealed in isozyme analyses of ND, SD, KS and Texas populations by Baskauf (1993, Comparative Population Genetics and Ecophysiology of a Rare and a Widespread Species of Echinacea (Asteraceae). PhD Dissertation, Vanderbilt University). A phylogenetic tree resulting from Wagner distance analysis revealed that KS and SD populations were the least similar among the four populations.
Leuszler (1996, The Indirect Effects of Large-scale Insecticide Spraying on Populations of Echinacea angustifolia DC (Asteraceae) in the North Dakota Badlands, MS Thesis, Utah State University, Logan, Utah) examined isozyme loci of 23 populations populations of E. angustifolia in southwestern ND. Despite a high similarity index of 92.7% among populations, Leuszler discovered a significant association between genetic distance and the actual distance between sites. However the genetic distance between any two sites in 1993 or 1994 was no greater than 2%.
These three studies indicate that there is variation among populations, even within a small geographical distance that might be of value for plant breeding. To obtain the greatest genetic diversity, plant breeding programs should include populations of E. angustifolia from both northern and southern sources.
A hydroponic system using the ideas of William Skaife's U.S. patent #4,397,114 provided nearly optimal conditions for two seasons of growth and controlled pollinations of Echinacea angustifolia. The portable plants took up little space in the refrigerator thus enabling convenient forcing of flowering and ease in performing cross-pollinations. The system also provided a controlled root environment for comparison of root growth and morphology.
With this system and the germination protocol outlined later in this
newsletter, growth of 36 plants was controlled from germination through
harvest. Dormancy was broken and early growth was synchronized by imbibing
the seeds in 0.7 ml L-1 Ethrel (1mmol ethephon) under
continuous light at 2 C for 14 days. Germinated seeds were placed in 1-cm
diameter polyester "seed cells" (Cropking Inc., Medina OH) when radicals
were 3-7 mm long, with the polyester half-submerged in tap water adjusted
to pH 6.0. When roots had fully penetrated the 50 mm of polyester batting,
the "seed cells" were transferred to a very coarse, pH-adjusted (pH 5.5),
perlite-peat mix without disturbing the roots. The peat-lite mix was stuffed
into "socks" made from Delnet sunflower pollination bags and shaped with
a heat sealer. Each sock was 6 cm in diameter and 25 cm long. Socks were
placed into "sleeves" of flexible PVC (a product for covering asbestos
on hot water pipes) that allowed 1 cm of air space around the socks.
During the first year of growth, in which plants developed taproots and a rosette of leaves without flowering, the plants were grown outdoors in 25-cm diameter PVC tubes half-buried with soil. In October, before frost occurred, the plants were removed from the PVC tube, packed in plastic bags, and transferred to a dark refrigeration unit at 2 C. In mid-March, the plants were transferred back to the PVC tubes, now erected in a heated greenhouse. All except two plants showed new leaf growth. Within two weeks all plants showed remarkable leaf growth, except for the two which proved to be dead.
Nutrient solution was continuously recycled through the PVC tubes from a 375-Liter tank. The solution level in the 25-cm diameter PVC tubes was maintained between 5 and 10 cm high. The system provided 11 Liters of a modified Hoagland's nutrient solution per plant.
Toward the end of May, 76 grams Na2SiO3 H2O was added to the solution in an attempt to curb the spread of powdery mildew. Otherwise the only additions, other than tap water, were HNO3 or KOH for pH adjustment. During the first week of operation in the greenhouse, it was necessary to adjust the pH down from above pH 7.0 until the plants were actively using the NH4 in the solution. Thereafter, NH4NO4 buffered the solution so that the pH stayed at 5.9 to 6.2.
Flowers were covered with Delnet pollination bags when rayflowers emerged. Bags were placed on the first flowers in early May. Bagging continued through early June. Self-pollinations were attempted for each plant. Cross-pollinations were made by temporarily removing the bags and rubbing heads of two plants. Crossing was begun 1 to 14 days after bagging, mostly within a week. Each flower combination was pollinated three or four times as the anthers proceeded from outer ring to top of the head over an average period of 4 to 8 days. By the end of June most heads had brown pedicels, an early sign of seed maturity. Harvesting was delayed until July 27, when entire stems were brown and dry.
Of the self-pollinated heads, only two heads developed any embryos. One of the two heads had seven embryos from 25 achenes. This result is similar to levels of self-compatibility that I found among plants in the common garden (less than 1%). These results are lower than self-crossing reported for wild populations. Leuszler et al. reported 9 % viable achenes from geitonogamous hand pollinations (among florets on the same head) and 7 % viable achenes from autogamous pollinations (with pollen and megaspore from the same floret) (June1996, Prairie Naturalist 28(2): 91-102, Reproductive Biology of Purple Coneflower in Southwestern North Dakota).
Roots were examined at harvest. Five plants appeared entirely dead, as the dried stems pulled easily away from the roots. Within four days after harvest, it was apparent that a rot was affecting the roots as nine of the roots had soft or rotted shoot meristems at the crown. Only two of the other plants had active apical meristems.
This hydroponic method was useful for observing plant growth for two
seasons and flowering for one season. The portable plants made cross-pollination
easy. Despite fast growth of the entire plant, root production using this
method was a failure because of the root rot problems.
A potting soil mix containing 50 % clay loam soil formed good root balls but drained poorly. I also experimented with a peat-lite soil mix containing 12% Styrofoam beads and 12% sand. The drainage was much improved over the previous mix but the root balls were loose. A better option may be to use a peat-lite mix with a wetting agent. Placing sand underneath the plug trays also improved drainage by pulling water out of the plugs by capillary tension. However, when trays were left too long on sand, the roots grew into the sand. The dangling roots were vulnerable to desiccation when transplanted.
Seedling growth in trays seemed to be limited by early root pruning when trays were placed on a hard surface. Seedlings that were allowed to extend their roots into wet sand underneath the trays continued to grow new larger leaves; whereas seedlings in trays on a hard surface quit growing at the 2-leaf stage.
Planting and Spacing Recommendations
It seems sensible to do as much field preparation in the fall as possible
so that efforts can be focused on transplanting the following spring. Building
beds or ridges, laying landscape fabric and seeding a perennial cover crop
in the wheel tracks can all be accomplished in August or September after
a small grain or pea crop.
Advantages for transplanting over direct-seeding:
1. Seed is valuable ($200- $300/lb.) and is used more efficiently by
transplanting.
2. Solid-seeded beds may be more difficult to weed since mechanical cultivation would be impossible. The plants grow slowly and do not compete well with weeds, because they are slow to produce a canopy. In my plot, by the end of the first growing season, plants from seeds of northern sources (SD, WY, ND) had formed a rosette of leaves of only 6 to 10 inches diameter. Plants from more southerly sources grew more quickly and some flowered the first year, but the canopy diameter was approximately the same as for plants of northern source.
If mulch is used for early-season weed-control, it is easier to mulch
a space-planted crop than a direct-seeded crop. Straw mulch is easier to
weave around uniformly-emerged, evenly-spaced plants than around randomly-scattered,
direct-seeded plants.
3. Cone transplanters are available that can plant into 3-mil plastic
mulch, to limit early-season weed problems. This option is limited to the
use of plugs (rather than bare-root transplants). However, as indicated
under "Soil Choice and Management", the lack of aeration under the plastic
may contribute to root rot problems.
Many commercial efforts to direct seed E. angustifolia have been limited to fall seeding due to stratification requirements. Spring seeding would allow more flexibility in early season weed control.
Leroy Ballard of Nature's Cathedral near Norway, Iowa described to me a method of fall seeding. He scattered seeds on raised beds in the fall, gently raking them in, and covered them with an inch of loose straw to keep them moist. Unlike some seed sources of E. purpurea, seeds of E. angustifolia require light for germination, so it is necessary to keep the mulch thin and loose. A method that would require the least labor input would be to sow into wheat stubble in the fall and to flail-chop the stubble to provide mulch.
There are several advantages for direct seeding in fall, winter or spring
rather than transplanting:
1. Direct seeding is less labor intensive than transplanting.
2. It is less labor-intensive to thin direct-seeded plants to obtain the appropriate density after plants are established than to replace transplants that have died. Thinning can be done by hoe or by machine.
3. No mechanical transplanters are currently capable of transplanting
into heavy landscape fabric. Also, most mechanical transplanters can plant
no closer than 12 inches within or between rows, thus setting an upper
limit on plant population that is lower than the optimal spacing.
4. The use of a precision drill seeder would place the seeds into tillable rows, thus eliminating the drawbacks of solid seeding. Precision seeders can more closely approach the desired plant population than can mechanical transplanters.
In general, I recommend a plant spacing of nine inches each direction.
The recommendation would vary with the seed source. At a nine-inch spacing,
there will be no competition between plants during the first season, since
the radius of the rosette of leaves in my plot ranged from 6 ± .5
inches for plants of northern source (WY, SD, ND) to 9 ± 1.5 inches
for plants of southern source (KS, NE). Plants would compete for sunshine
if they are spaced closer than this. A wider spacing would not help against
spread of the most commonly observed disease, a phytoplasma similar to
aster yellows, since leafhoppers spread this disease.
Soil in my plot was a clay loam. This fine texture was suitable at the dry end of the plot, but was obviously detrimental towards the poorly drained end of the plot. For this reason, I consider choice of soil type to be of utmost importance. My choice is a moderately coarse-textured soil (sandy loam to fine silt loam) overlying gravel. Such a soil requires irrigation, at least during the first month of establishment of transplanted seedlings.
If the soil texture is coarse enough to provide good drainage, I would still recommend planting on raised beds or ridges to keep soil from washing over the slow-growing seedlings. However, salt buildup on ridge tops might be a problem for some soils. Therefore it is generally wise to choose soil that has a low electrical conductivity (EC), although the ability of Echinacea angustifolia to tolerate salt is unknown. An EC of over 3000 mmhos/cm can inhibit water uptake by roots of salt-intolerant plants.
A coarse-textured soil will require a cover crop or mulch to keep sand from blowing in the wheel tracks. An ideal cover crop would establish quickly to smother weeds, would not become a weed itself, would not contribute to disease or insect problems, would be perennial and would tolerate wheel traffic. Spring-seeded winter wheat would provide cover for one season without producing seed, but it will not survive the winter. Most winter wheat must be vernalized (exposed to cold) for 7 weeks in order to produce seed. Therefore winter wheat will not go to seed even if it is planted a month before the coneflowers are transplanted. There is a potential perennial cover crop on the market from Hybritech, (a subsidiary of Monsanto, 316-755-1249) called "Regreen," which is a sterile hybrid between winter wheat and slender wheatgrass. Legumes that are not overly prolific can be another option. Sava snail medic, one of the less-prolific medics, is available from Timeless Seeds in Idaho (406-278-5720). Black medic has been touted as a potential cover crop for interplanting with corn. But the wet conditions of my research plot turned volunteer black medic into a choking weed. The black medic will set seed unless mowed.
Retention of moisture by mulches might cause insect and disease problems. In my research plot I used wheat straw mulch, which led to an aphid problem where volunteer wheat sprang up. A grower in British Columbia (see http://www.forthrt.com/~roland/herbfarm.html) noted a high incidence of cutworms only where newspapers were used as a mulch. A grower in Iowa using black plastic mulch claimed that it caused roots to rot. Air circulation under woven landscape fabric should prevent such a problem.
Several resources have aided my search for an ideal cover crop or mulch. Dr. Joanna Fraser of The Agriculture and Agri-food Research Centre (P.O. Box 3000, Lethbridge, Alberta, Canada T1J 4B1, 403-327-4591 Ext. 450) has evaluated several annual legumes at several sites for establishment, weed content, plant height, reseeding or regeneration, and percentage ground cover. The long-term goal for this program has been "to identify and register more annual legumes suitable for forage, seed or green manure, or for use as cover crops in rotations." Also, a newsletter for organic market gardeners has recorded growers' experiences with cover crops and landscape fabric. (Growing for Market, P. O. Box 3747, Lawrence, KS 66046). The Minnesota Department of Agriculture annually publishes the free "Greenbook" which has chronicled the experiences of farmers who have used various mulching systems while co-operating with university researchers. (Energy and Sustainable Agriculture Program, MN Dept. of Agriculture, 90 West Plato Boulevard, St. Paul MN 55107, 612-296-7673, prescott.berg@State.mn.US) I have gained ideas locally, too, such as planting milo, corn or oats in the wheel tracks in late August to catch snow to protect plants from thawing and re-freezing. By default, perhaps weeds might work equally well for catching snow.
Herbicides are not labeled for use on E. angustifolia, nor are they ever likely to be labeled in the United States without a strong University proponent who is willing to do the exacting lab and field work and can convince Interregional Research Project (IRP) personnel to provide funding. The process for registering an herbicide for use on food crops (including medicinal plants) is more involved than the process for ornamental plants. Even though some formulations of two herbicides, DCPA and oryzalin, have been labeled in the U.S. for use on ornamental E. purpurea, they should not be used on any Echinacea plants intended for medicinal use. Effects of the herbicides on plant metabolism must be studied in order to register the product for use on specific plants. Since it takes three years to produce roots that are large enough to harvest, it will take at least that long to provide specimens for residue and metabolite analysis and perhaps another year to complete analyses and complete the registration process. The production of the specimens as well as the analyses must be done according to scientific procedure by a qualified technician. Governmental funding is provided only for crops and projects near the top of IRP priority lists.
Weeds in the SDSU plot were controlled by mulches, by hand pulling and by hoe. When I attempted to cultivate with a tractor and row-crop cultivator, several plants were destroyed because stems too easily broke off at the base of the plant. It was difficult to avoid this damage, since many plants had stems that were lodged and sprawled to up to two feet from the base of the plant after the first year. This lodging problem occurred during excessively wet years. Mechanical cultivation might be less destructive during drier years since lodging might not be as much of a problem.
Seed source might have an effect on stem spread and lodging. Seed from
South Dakota consistently produced plants with a broader average stem spread
than from the other sources (22 inches across for SD vs. 16 inches for
the other states in the driest environment). The broader spread for SD
reflected a higher rate of lodging than for plants of other sources.
SDSU Lab Research on Seed Germination
Research was conducted at SDSU to maximize the germination rate of Echinacea angustifolia seeds and to minimize the germination time. To accomplish this goal, the effects of scarification and chemical treatments in conjunction with short periods of prechilling were examined. The most successful chemical treatment was ethephon, which releases the growth hormone, ethylene, when imbibed by the seed.
The recommended protocol from this research is as follows: Place seeds on blotter paper moistened with 1 millimolar Ethephon (3 grams Ethrel(1)
per gallon of distilled water) in covered transparent plastic seed boxes(2). Place under continuous light (75 to 200 footcandles, cool-white fluorescent lamps) in a cold environment (2 to 6 C) for 14 days followed by moving the seed boxes to 25 C under continuous light (400 to 700 footcandles, warm and cool fluorescent lamps).
This protocol allowed transplanting of seedlings to the field within 9 weeks after placing the seeds on blotter paper. Most seeds germinated within 18 days after placing on blotter paper. Over 95% of viable seeds germinated. Viability was determined by tetrazolium tests on sub-samples of the same seed source. This protocol not only increased total germination; it also tended to synchronize germination so that all seedlings were the same size when transplanted to the field. The protocol also shortened the period of warm temperatures needed after the prechill treatment.
Results are detailed in the publication: Feghahati, S.M. Javad and R.N. Reese. 1994. Ethylene-, Light-, and Prechill-enhanced Germination of Echinacea angustifolia Seeds, J. Amer. Soc. Hort. Sci. 119(4): 853-858. Other important discoveries noted in this publication were that: 1. Seed maturity, as measured by the ease with which achenes were removed from the head, was an important factor in germination. 2. No water-soluble germination inhibitors were discovered. 3. Mechanical scarification was ineffective. 4. Chemical scarification with sulfuric acid was ineffective at up to 40% acid and damaged the seeds at greater concentrations. 5. Prechilling for two weeks in the dark was ineffective. 6. Two weeks prechill under continuous light increased germination from 15% to 30%. 7. Addition of ethephon to the germination boxes further increased germination from 30% to 80%. 8. The 1-week prechill treatment with ethephon did not increase germination above controls or above the germination induced by one week prechill in the light.
For a recent review of stratification requirements without the use of ethylene (of seed obtained from Otto Richter and Sons, Canada), see Parmenter et al., 1996, Chilling Requirement of Commercial Echinacea Seed, New Zealand Journal of Crop and Horticultural Science 24:109-114.
Update on Fungicide Seed Treatment
Although a Vitavax-200 seed treatment was used in most of the experiments in the above publication, commercial use of this fungicide can't be recommended for the following reasons: 1. It is not labeled for use on Echinacea seeds. 2. Tests were not done with other anti-fungal treatments that might work better, such as soaking in a 10 % hypochlorite solution for 15 minutes or using a preliminary hot water treatment. 3. The best germination percentages reported above were accomplished without the use of a fungicide. In follow-up experiments, the Vitavax-200 treatment appeared to inhibit the germination of viable seedlings. Seedlings from treated seeds had elongated roots with few root hairs. Without the fungicide, the seedlings had whiter, shorter roots with many more root hairs. 4. Fungicidal treatment might not be necessary when only the heavy seeds are used. Fungicide was used in most of the original experiments because the lightest achenes in preliminary tests were known to support fungal growth that would contaminate the whole seed box. In the experiments with heavy seed, it was noted that there was almost no fungal contamination despite the lack of fungicide. On the other hand, most experiments that involved empty achenes had some fungal hyphae growth even with the Vitavax-200 treatment. We suspect that the fungi grow primarily on empty achenes and non-viable seeds, and will not grow on viable seeds.
Seed Source and Response to Treatment
The cause of the low percentage of seed germination in the Kansas and Nebraska populations was examined with seeds from several samples not used in the original germination study. Tetrazolium viability tests revealed that the low percentage of germination was matched by low seed viability. The seeds from these two locations were difficult to dislodge from the heads, indicating that they were apparently harvested too soon and were immature. On the other hand, there is evidence that some populations have inherently low germination rates: Germination of seeds from one SD location has been consistently 30% lower than seeds from a second SD location.
Germination requirements and viability also vary by species. For example the percentage and speed of germination of six purchased lots of Echinacea purpurea were unaffected by seed weight or light (Wartidiningsih and R.L. Geneve, 1994, HortScience 29(12): 1443-1448. Two articles: Seed Source and Quality Influence Germination in Purple Coneflower; and Osmotic Priming or Chilling Stratification Improves Seed Germination of Purple Coneflower.) Growers will benefit from doing their own experimentation with specific seed lots.
Distributors of native seeds rarely provide the information on purity and viability of seed that buyers would desire. However, seed distributors might be willing to provide another easily obtainable indirect measure of seed viability such as test-weight. Low test-weight seed might indicate that seeds within the achenes are undeveloped or missing altogether. In early experiments it was observed that heavier seed germinated at a greater rate than lighter seed. Empty achenes were an obvious cause of the low germination rate of lighter seed. Damaged or undeveloped embryos could also contribute to the lower germination rates of lighter seed.
To test this hypothesis, I removed most of the empty achenes (as determined by periodically crushing a sub-sample of the achenes in the heavy seed fraction) and separated seed of each of four sources (progeny of ND, SD, NE and KS populations) into two fractions with a Carter-Day seed blower. Then I used the SDSU germination protocol as a test of seed viability. Test weights were measured on standardized volumes of seeds prior to germinating. Test weights for the heavy and light seed fractions were: ND 215 and 200 mgcm-3 ; SD 224 and 205 mgcm-3 ; NE 208 and 188 mgcm-3 ; and KS 232 and 219 mgcm-3. Mean germination rates for the heavy and light seed fractions were: ND 58.0 ± 2.6% vs. 53.7 ± 2.8%; SD 82.3 ± 2.3% vs. 72.3 ± 3.1%: NE 28.7 ± 2.3% vs. 28.0 ± 2.3%; KS 80.0 ± 3.4% vs. 75.0 ± 3.0%.
The parameters of test-weight and seed fraction explained 68% of the variation in germination rates. Test-weight values accounted for 49% of the variation. Seed fraction accounted for an additional 19% of variation. (Surprisingly, the germination rate was not affected by seed density for these samples in which density ranged from .39 to .57 g per 100 seeds).
Therefore, even though seed fraction accounted for a fair amount of variation in viability, the test-weight was more important, but still did not explain some of the variation. Although low test-weights were correlated with low germination rates, the degree of difference in germination rate between high and low test-weight samples could not be explained by test weight. Neither parameter adequately predicted the low (28% and 58%) germination of two of the sources. The low viability of theses two sources was confirmed by tetrazolium tests that revealed many full-sized turgid embryos that were not viable. I suspect that seed from these sources was affected by moisture and heat damage during storage. It is reasonable then, for the buyer to request measurements of both the test weight and seed viability, with an emphasis on the latter. Test weights above 200 mgcm-3 should be fairly free of empty achenes. A germination or tetrazolium test must still be conducted to determine whether the seeds are viable.
The result of the above experiment contrasts with previous tests of a large quantity of seeds that were stored in a hot attic during the summer before the seeds were systematically separated by density. Only the heaviest fraction of seed was retained. The germination rate on these attic-stored seeds was above 90%. I suspect that any damaged seeds were dehydrated while in storage, and were subsequently discarded with the light seed fraction.
The resilience of E. angustifolia populations after grazing or harvesting is a subject of speculation and research. There is evidence that plants in Kansas have been somewhat resilient despite abundant harvests over a number of years. Populations in the historical range of E. angustifolia north of Kansas appear to be fairly abundant. I observed a greater abundance in northwestern Texas, the Dakotas and Montana than in Kansas. Yet this flower is rare enough wherever I travel that I get excited every time I see several of these plants in one location.
In Kansas, the dynamics of some E. angustifolia populations are being studied. Dana Price Hurlburt of the Kansas Biological Survey estimates that 100,000 pounds of E. angustifolia are harvested each year in Kansas, mostly in the west central portion of the state. Locations that she is systematically surveying have been quite resilient in total numbers of plants, although mature plants seem to be less prevalent in grazed pastures than in pristine prairies. Roots routinely grow back from the bases of roots previously harvested at six to eight inches below the ground.
In the Central Plains, pasture management practices might have more of an impact on E. angustifolia abundance than wild-harvesting practices. Range scientists refer to E. angustifolia as a "decreaser" in that this plant becomes less abundant as grazing pressure intensifies. Young coneflower leaves appear quite palatable, judging by observations of plants in pastures that have been chewed by cattle and in the research plot by gophers. During 1996, I observed large coneflower populations in pastures in which the rancher consistently, over a number of years, kept cattle out until late summer and avoided overgrazing. The predominant cause for reduced E. angustifolia stands on rangeland appears to be cattle chewing on flowers so that seeds are not replenished (personal communication with Dana Price Hurlburt, 1998).
There is no information on herbicide impact on coneflower populations. It is possible that the thick cuticles and long dense hairs on coneflower plants repel some herbicides. Also, coneflowers are likely to be dormant before herbicides are applied in the fall to control some perennial weeds. The main threat from herbicides may come from attempts to control new perennial pests such as red cedar and musk thistles, which are sprayed throughout the year with powerful herbicides not previously used so extensively on rangeland.
On the sub-humid prairies during wet years, it has become difficult to find once-thriving populations among the tall dense grasses east of the Missouri River, because E. angustifolia cannot compete for sunshine with grasses when water is abundant. I observed large conspicuous populations during the second and third wet years following several normal (dry) years on the Plains. By the fourth wet year in central SD (1997), I was unable to find as many large populations in bloom as in previous years. In 1998 dense populations were more likely to be found west of the Missouri River in SD.
Apparently the roots can survive in either the moist or dry environments for several years without needing much foliage to replenish them. Then when there is adequate moisture and light the plants will grow and bloom.
The Problem of Over-harvesting
At an Echinacea conference in Kansas in October 1998, (Proceedings in process), it was generally agreed that over-harvesting of E. angustifolia in Kansas is not the reason why conservationists are discouraging wild-crafting. Rather, buying practices have led to damaging and illegal harvests of related species, some of which are endangered. Buyers will take any species of Echinacea on the market and will re-sell it as "Kansas Snakeroot", "Missouri Snakeroot" or even as "Echinacea angustifolia." Thus, there is a persistent outcry against the wildcrafting of Echinacea.
In Montana and North Dakota and areas near the Black Hills, the concern of over-harvesting of E. angustifolia, (the only Echinacea species in that region), is different. Root harvest in this region is a relatively new phenomenon resulting in the rapid decimation of populations. High unemployment on Indian reservations and surrounding communities, lack of cultural norms for wild harvesting, and the premium price for Northern Plains root (above the currently high price), contribute to the rapacity of root collection in this region (Monique Kolster, U of Montana, personal communication, 1998).
Echinacea germplasm is being collected systematically by public and private institutions. There is a USDA project to collect germplasm of all Echinacea species, for which I am short on details. Native American Seed, (Junction, Texas, 800-728-4043), searches for seeds in areas that are slated for development in order to preserve wild populations.
Although there are at least two seed companies which are growing E.
angustifolia for seed, the preservation of threatened plant populations
in areas impacted by urban development, overgrazing, herbicides, or over-harvesting
will require the efforts of growers and collectors who are more interested
in genetic preservation than in commercial production. Collectors should
look for sources that represent the extent of genetic variation. Only one
E.
angustifolia population should be maintained by each preservationist
to avoid crossing between populations. Growers should keep track
of the source of their seeds and should store the seeds at low humidity
(11% R.H.) and low temperature. Without such efforts, it is possible only
to save mixed populations, as I have done at SDSU.
In return for the currently exorbitant prices, the buyer should demand about 90% pure live seed or better with a precise identification of seed source and certified verification of species. Seed companies must have expertise in working with wild species and have adequate seed cleaning, conditioning and controlled-environment storage equipment in order to guarantee high quality seeds of the correct species.
Seed should be tested not only for purity (freedom from debris and seed of other species), but also for viability. It is easy to test for the presence of empty achenes by crushing seeds of a representative sample with a fingernail. Underdeveloped or damaged embryos can be detected by incubating halved embryos at 35 C. in tetrazolium solution for two hours. A dark pink color throughout the tissue indicates that the seed is respiring and hence is viable. State seed testing labs can perform tetrazolium and germination tests. The SDSU germination protocol, with references, should be specified since it has not been published as an official protocol.
A 1995 British Columbia BMAFF publication (Agdex 253-810) listed seed companies that supply E. angustifolia a, E. pallidab, or E. purpurea c. For brevity I will simply name the company and location: Abundant Life Seed Foundation, Fort Townsend WA a,b,c; Elixir Farm Botanicals, Brixey MO a,b,c; Prairie Moon Nursery, Winona, MN a,b,c; Prairie Nursery, Westfield WI b,c; Johnny's Selected Seeds a,c; Otto Richter & Sons, Goodwood ON a,b,c; Missouri Wildflowers Nursery, Jefferson City MO a,c; Prairie Ridge Nursery, Mt. Horeb WI b,c; and Westcan Horticulture, Calgary AB b. For other potential seed sources try http://www.hort.purdue.edu/newcrop/med-aro/seedsources.html or use the search utilities at http://eru.usask.ca/safonline/.
Wholesale seed prices since 1992 for kilogram quantities or greater have ranged from $340 to $1500 per kg. Prairie Moon Nursery (PMN)(507-452-1362) has been a steady source of E. angustifolia seeds collected in South Dakota and North Dakota. PMN's wholesale seed prices for 1992 to 1996 have ranged from $400 to $660 per kg ($180 to $300 per pound), with higher purity in recent years (as low as 5% foreign matter with the seed) providing a higher value product for the same price. PMN sells seed from collectors on consignment and retains about one-third of the money for cleaning, storage, marketing and handling.
Brenda Reese, a seed collector from Oacoma, South Dakota, has sold through PMN and also to Western Native Seed (719-539-1071) of Salida, Colorado. She indicated that she will sell directly to the consumer, and has used the services of Heyne Custom Seed Services to clean her seed.
In addition to custom cleaning, Heyne sells seeds from native plants. Heyne's April 1997 price list included E. pallida (at $13.50 per ounce), E. purpurea (at $24.00 per pound), and E. angustifolia (at $18.00 per ounce). See http:///www.netins.net/showcase/bluestem/.
Seed from Wild vs. Cultivated Sources
Commercial seed has historically been collected in the wild. With the large number of commercial growers in recent years, seed is now available from cultivated sources, which introduces questions about the quality of seed from the wild versus cultivated sources. Cultivated sources are likely to have adequate moisture and nutrients to fill seed out. In the wild, moisture is often short during the seed development period of July through August.
Pollination of dense stands of this crop may require the use of honeybees. More importantly, however, cultivated sources have potential problems of hybridization and contamination with other seeds that are less likely to occur with wild sources. One seed grower has addressed the problem of hybridization by arranging to have E. angustifolia grown in an isolated location in New Mexico in 1996.
Be wary of seeds from cultivated E. angustifolia if grown in areas with native or cultivated populations of E. pallida, E. atrorubens, E. angustifolia var. strigosa or E. simulata. Native American Seed (Company) has collected seed in the Dallas/Fort Worth environs from plants that fit the description of E. angustifolia var. strigosa, which is thought to be a hybrid between E. angustifolia and E. atrorubens. The achenes were larger and rougher than seeds of E. angustifolia var. angustifolia. (The isolation distance to avoid hybridization has not been determined. It is likely that wild bees, honeybees and other pollinators can transport pollen several miles between plants). The geographic range of Echinacea species is illustrated in McGregor, R.L., 1968, "The taxonomy of the genus Echinacea (Compositae)," Univ. Kansas Sci. Bul. 48, and in Barker, W.T. et al., 1977, Atlas of the Flora of the Great Plains, Iowa State University Press.
An experience with trying to market roots from hybrid plants is reported
at http://www.geocities.com/HotSprings/5760/ourfarm.html.
Each locality has about a four- to six-week window of opportunity for collecting seeds, made shorter by high winds after the seeds have dried out. In moist years, tall grass will protect against seed-shattering winds. For the same reason, coneflowers in areas protected by trees will retain seeds longer.
The ideal time for collecting seeds in northern states in 1993 and 1994 was a week or two before and after the first hard frost. The same held true in 1995 even though the summer was much hotter and drier. In 1993, early heads were harvested September 25. By October 3 some seeds were shattering. While harvesting heads on October 9, one day after the first hard frost, I noted that less than 5% of seeds were shattering. The heads that I harvested on October 30 had lost about 15% of seeds from shattering. In 1994, harvest on the early heads began September 10, two weeks earlier than in 1993. By September 18, the number of mature heads was at a peak, and seeds were just beginning to shatter. Half of the plants had ten or more mature heads. Two days later, I noted about 4% shattering of seeds.
Maturation of seed heads in Nebraska and Kansas was a few weeks later than in the northern states in the year that our seeds were collected for research (1989,1990). In contrast, harvesting farther south (in Texas) has been possible as early as the first week in August.
I am very cautious not to collect too early. Immature seeds have very low or no germination. Seed samples from Texas collected in mid-July of 1995 had no germination. Seeds collected August 1990 in Nebraska had only 37.3% ±4.5% viable seed. Signs of immaturity were evident on some of the Nebraska seed heads: seeds were very difficult to remove and were very pungent.
Another reason to delay harvest is to collect less debris. For some seed sources, the brown florets that are attached to each seed will wither and fall off the head if harvest is delayed. If collected with the seed, the florets are very difficult to remove since they are the same size and weight as the seeds.
It is possible to direct combine E. angustifolia (with a cutter bar). But the seeds will be lost if any wind is used. Therefore cleaning must be done with a stationary cleaner.
Prairie Habitats Inc. (Box 1, Argyle MB, Canada ROC OBO, 204-467-9371) lists a hand-held machine which was developed to strip seeds from wild plants. The machine has recently been equipped to efficiently strip E. angustifolia heads from the stems in the field. (http://www.prairiehabitats.com.)
Harvesting and Cleaning Seed by Hand
Clip brown, dry heads free of stems. At the ideal stage, seeds should be loose enough to remove with only 20 pounds of pressure. If your timing is perfect, it is sometimes possible to strip the seeds without clipping off the head, but the risk of shattering is too great. Bracts ("spines") can chafe the skin. I handle the spiny heads with "paint-prep" synthetic gloves manufactured by Magla Edmont.
Rub seed heads between two boards surfaced with corrugated rubber (the backside of heavy carpets). Anchor the bottom board and rub the heads with about 30 pounds pressure. If more pressure is required, the seeds are probably immature. Seeds that will not dislodge easily are not worth the extra effort, as they will probably not germinate.
If florets were collected with the seed, it will be necessary to apply
enough pressure to crush the florets to allow them to pass through the
#5 screen. The extra pressure might destroy the pericarp (shell) of individual
seeds. Without the pericarp, the naked seeds that remain are still viable
unless the tough tan translucent seedcoat is ruptured. The main problem
with naked seeds is that they pass through all of the screens into the
discarded fraction.
Stack screens (available from Crippen Manufacturing Co., Inc., Alma
MI 48801, 517-463-2119) as follows:
1. 8 x 3/4 rectangular-slot Retains cone pieces
2. 9 1/2 x 64 round-hole Retains large bracts
3. 5 x 1/2 rectangular-slot Retains large seed that would clog the next screen
4. 8 1/2 x 64 round-hole Retains medium bracts
5. 4 x 1/2 rectangular-slot Retains desired fraction
6. Solid pan Contains small bracts and florets
For samples with large seed, remove #4. For samples of small seeds, remove #2 and #3. Often, #5 will retain many bracts. The bracts will "float" on top of the seeds as you shake the screen, and can be skimmed off the top.
Though percent seed yield varies greatly by source and quality of the crop, a common yield by weight is: 18-23% achenes (shells with or without seeds inside), 35-40 % cones, 27-33 % bracts and 10-14 % very small achenes and florets. For five seed sources, the ratio of the weight of heavy seeds to the weight of heads (clipped free from stems) ranged from .22 to .48.
Often many of the achenes have no seeds inside. These shells must be removed with a blower to obtain 100% viable seeds. Very small samples can be cleaned with an inexpensive model 757 seed blower available from Seedburo in Chicago, IL. However, I recommend using a mechanical cleaner with a fan rather than using hand screens and seed blower. Grain elevators are usually equipped with a Carter Dockage Tester, which is suitable for mechanical cleaning of coneflower seeds.
Seed Cleaning with Small-scale Equipment
The following sequence for threshing and cleaning seeds was developed over a two-year period, in which I experimented with several machines used for seed research at SDSU.
Heads are easily threshed by passing through concave screens in a threshing
mill that has rubber-slatted threshing bars. If this method doesn't work
(heads roll with neither bracts nor seeds being dislodged) then I would
say that the heads are not mature enough or are too wet.
The most efficient method for cleaning samples larger than 100 grams
was to combine the use of a Carter Dockage Tester with the use of hand
screens (for screen sizes that we didn't have for the machine). On the
Dockage Tester, I removed bracts with no. 00 and no. 1 "riddle" screens
that resemble straw-walkers on a combine. Other particles except for "disc
florets" were removed with a series of rectangular slot screens, similar
in size to the #3 and #5 hand screens, on the Carter Dockage Tester. The
large seed fraction above the largest rectangular slot screen was 99.5%
pure unless cone pieces of this size were present in the sample, (which
were impossible to remove except by hand). Florets that remained in the
small seed fraction were removed by using the 4 X 1/2 hand screen, resulting
in 90 to 98 % purity for this fraction. Small particles passed through
a small round-holed screen on the bottom. Very small seeds that were retained
on top of this small screen were examined for presence of viable seeds.
Some seed sources had many viable seeds in this fraction. For other sources
this fraction was discarded. Here is the preceding process in a nutshell:
Stack the Carter Dockage Tester as follows, setting blower at 2 1/2:
#00, #1 or #2 riddle Retains bracts, stems, cone pieces
#5 rectangular slot Retains large seed and some cone pieces
#4 rounded rectangular slot Retains small seed, bracts and disc florets which lodge in the screen and must be periodically removed by hand, then cleaned by hand in a Crippen 4 X 1/2 rectangular slot screen.
#2 round hole Retains very small seed and florets, which can be run
through a Carter-Day blower if achenes have seeds inside them.
Any further separation was accomplished in a time-consuming process of using a Carter-Day blower that separates by density, thus removing empty fruits. Much of this chaff can also be removed by setting the fan on the Carter dockage-tester machine between "2" and "3". But the dockage-tester fan can not be controlled as precisely as the Carter-Day blower and might blow away too much viable seed.
A gravity seed separator is another option for very quickly grading samples of 5 pounds or more. The heavier fraction of seeds is about 95% pure. Debris is concentrated with the lighter fraction of seeds, which must be cleaned by another method.
Seed-cleaning Research at other Universities
A study of "Echinacea [pallida] threshing and seed cleaning" was presented at the May 28-30, 1997 CSAE (Canadian Society for Agricultural Engineering) Annual Conference in Sherbrooke, Quebec, Canada (CSAE-SCGR-Paper, 1997, No. 97-318:1-7.) An in-house threshing device operated best at 2600 r.p.m. After the heads were threshed, three devices were evaluated for seed cleaning. "The Carter Dockage Tester, equipped with a 001 riddle, yielded the highest seed purity to seed loss ratio." They also experimented with the Carter Day Superior Fractionating Aspirator and the Forsberg Gravity Table. The combined use of these machines was apparently not evaluated. The average mass of the E. pallida seeds, (4.5 to 5.1 mg per seed), matched the heaviest fraction of E. angustifolia seeds that I worked with. Therefore the screens for cleaning the E. pallida would probably be different than those for E. angustifolia.
The need is great among small growers for understanding how to produce a high-quality crop that can find a market niche when Echinacea becomes a commodity. The current involvement of large companies in E. angustifolia production, in particular, can quickly turn the wholesale market for this species into a buyer's market. It is imperative that the best information on production methods be kept in the public domain, in order to maintain opportunities for small growers in markets where quality is of primary importance.
Groups in Canada, the United States, Europe, Australia, New Zealand and the Mid-East have been attempting to answer questions about the effects of genetics and environment on drug production of Echinacea species without the benefit of a common vision. Research funds for high value crops tend to come as "soft money" from corporations, provincial sources or private individuals. Because of the helter-skelter funding of new-crops research, concerted efforts to develop these new crops in ways that benefit the general public are rare. If there is a concerted effort, established corporations rather owner-operated businesses are often the beneficiaries.
I anticipate an initial difficulty in getting the various researchers to work together. That is where this proposal comes in. By providing a list and overview of common questions and hypotheses that several research groups are addressing, I am providing each researcher with the opportunity to learn the strengths of other research groups and to recognize the possibility of benefiting from networking and collaboration. By providing the rationale and framework for a multi-institution research project, I hope to unite researchers around a common vision to serve the commercial clientele of owner-operated farms and the academic clientele of free-thinking students.
Recent and Current Research on Chemical Analysis
v Chemical Analysis Researchers
A research group at the Biology Dept., University of Ottawa (in conjunction with Trout Lake Farms and Amway Corp.) is doing an extensive study of chemical content of several Echinacea species from wild and greenhouse environments. This is one component of a larger study to revise the taxonomy of Echinacea. The university researchers are John Livesey, Chantal Bergeron, Dennis Awang and John Arnason. Their innovative extraction procedure will soon be published, and was presented to industry workshop participants. Shannon Binns (sbinns@science.uottowa.ca) is a graduate student in ethnobotany working with the group to characterize species in the wild, phytochemically and morphologically.
Dean Gray (c648324@showme.missouri.edu), a graduate student at the U of Missouri, is studying the "Effects of water stress and time of harvest on chemical concentrations in Echinacea purpurea and Hypericum perforatum."
Kim Bauer (ka-bauer@uiuc.edu), a graduate student at the U of Illinois, is comparing the chemical components of the three commercial Echinacea species under nitrogen stress in sand culture.
My graduate advisor, R. Neil Reese (reesen@mg.sdstate.edu), is currently investigating the effects of a few environments on secondary metabolites of E. angustifolia and E. purpurea. The metabolites that he is working with have been shown to have allelopathic effects. Several compounds have been identified.
v Research questions concerning chemical analysis
The first question confronted by these researchers is, "Which chemicals should we analyze for?" Echinacoside, which is used by some companies for standardization of Echinacea products, is not a historically medicinally valuable component of Echinacea. It has been used as a marker for certain Echinacea species. The alkylamide components may have been over-rated by the German group (Bauer and Wagner) according to Steve Moring (smoring@hbc.ukans.edu), a plant biochemist at the U of Kansas who wrote an early scientific review of Echinacea. On the other hand, high molecular-weight polysaccharides are difficult to identify and quantify. So, which chemicals do you focus on? The above researchers have all skirted the difficulty of quantifying the polysaccharides by focusing on less valuable, but easier to analyze compounds. The Ottawa group has focused on the most abundant alkylamide (tetraene) and the low molecular-weight hydrophylic compounds (chlorogenic acid, cichoric acid and echinacoside). Dean Gray is focusing on cichoric acid and tetraene ("the predominant phenolic and isobutylamide in Echinacea roots"). Neil Reese is focusing on compounds that were identified in his own lab as having allelopathic effects using biological assays.
None of these approaches addresses the dictum that a host of chemicals work together in producing biological effects on humans. None of the metabolites by itself has a pronounced biological effect. Dr. Reese's approach (of using a biological assay to determine potency of the compounds) can reflect the effect of all compounds, but cannot give a picture of the effect on humans. There are two approaches to take to determine the potency of Echinacea products on humans: 1) Develop an immunological assay (Elisa test), which is an interest of Steve Moring; or 2) Do double-blind clinical trials, which are extremely expensive, especially if multiple products are to be tested.
Recent Research on Plant Spacing and Yield
The question of plant spacing will have to go beyond speculation to empirical studies. The hypotheses of interest are as follows: 1) Echinacea will respond like sunflowers to high-density plantings: individual plants will yield less, but total yield will be the same or higher. 2) Yield responses to plant spacing will vary among species according to growth habit, especially the herbage growth rate, root growth rate and the root morphology (rhizome vs. taproot). 3) The relationship between plant yield and drug yield can be altered by plant density through the induction of moisture, nutrient, disease or insect stresses or phytochrome responses that favor the growth of one plant organ over another. 4) Biomass yield responses to plant spacing will be environment-specific as affected by disease incidence and over-all growth rates, (as functions of light-levels, temperature, soil moisture, relative humidity in the microenvironment, presence of disease vectors and proximity to disease sources). 5) Slower growth due to competition for light at higher plant-spacings will make roots too small in diameter for efficient harvesting at the end of two or three years of production.
v Two recent reports address the first hypothesis:
An Egyptian study(3) supported the first hypothesis for E. purpurea in the Egyptian environment in 1992 and 1993. At the spacings of 20, 40 or 60 cm within rows and 50 cm between rows, total herbage was 1662, 1020, and 578 g/sq.m. (LSD = 28) for the first season, and 1822, 1184 and 822 g/sq.m. (LSD=36) for the second season, respectively. Root yields were 253, 196 and 88 g/sq.m. (LSD=10) for the first season and 236, 193 and 119 g/sq.m. for the second season, respectively.
A New Zealand study of E. purpurea (4) also supported the first hypothesis. Total yields leveled off above 10-14 plants per meter of .9-meter-wide raised bed with a .3-meter wheel track (7-9 plants/sq.m.) Plants at this optimum density were 22 cm apart within rows, 30 cm between rows (3 rows) and 90 cm between outside rows of adjacent beds. Plant survival over a two-year period was not affected by densities of up to 63.6 plants per meter of bed (42 plants/sq.m.), except that survival at the lowest density of 1 plant/sq.m. actually was lower (80% vs. 98% overall). Yield per plant for lowest to highest density dropped from 33 grams to 4 grams per plant.
v The New Zealand study also touches upon three of the other hypotheses:
An unpublished study at another New Zealand location, cited by Parmenter et al., showed that yield per unit area leveled off at 20 plants per meter of bed. This optimal spacing is a bit higher than Parmenter's and was possibly an effect of the overall higher yields at the second location (hypothesis #4).
Parmenter et al. also provided initial evidence that plant spacing alters the proportion of the plant that is involved in drug production (hypothesis #3). They reported that the percentage of root as rhizomatous tissue decreased linearly with the log of plant density.
Root size at harvest (hypothesis #5) may be the most important finding of the New Zealand study. Before evaluating their data, it would be useful to know the smallest root that can be harvested efficiently. An assumed minimum of 8-10 g per plant corresponds with the optimum density of 10-14 plants in Parmenter's study.
The only hypothesis not touched upon by these studies is hypothesis #2 regarding the expected differences among species. None of the hypotheses has extensive support from published studies to date.
Recent Research on the Effects of Cultural Practices on Yield
v Bud Harvest
A German study by Frank et al. (5) of the three commercial echinacea species revealed that if buds were removed in the second year of cultivation, "up to 10-20% higher root yield resulted."
v Nutrition
An Egyptian study (cited above, Shalaby et al., 1997) showed a root yield response to either nitrogen or potassium the first season, but not the second season. Nitrogen antagonized the effect of potassium on root yield both seasons except at the application rate of 100kg/ha each during the second season. Thomas Li has reviewed the use of several organic amendments and green manures(6) . As expected, the amendments with higher nitrogen and phosphorous levels increased fresh and dry weight of leaves. The highest nitrogen application (16 mM) also increased the yield of flowers and roots.
Cultivar Effect on Drug and Biomass Yield
v E. purpurea
In German literature there are two studies(7), (8) by the Institute for Pharmaceutical Biology at the University of Marburg comparing drug and biomass yield of ten E. purpurea cultivars grown for three years at two locations. The studies recommended the culitvars "Schleissheim" "Hybrida" and "Verbessere Leuchstern" for high first-year herbage yield. For second year herb harvesting, the latter two cultivars were recommended. "Rubinstern" was recommended for a good root yield in the second year. "Rubinstern" and "Vergesserte Leuchtstern" were found to be suitable for combined herb and root yield. Other cultivars were recommended with reservation. Drug content, presented in graphs, (so that I can understand enough to be dangerous), showed that drug percentage by weight was lowest for "Schleissheim" at one location, with "Hybrida" and "Verb. Leuchstern" in the high range and "Rubinstern" in the middle. Standard errors were not given, but it appears that there was no significant difference among cultivars for drug percentage at the other location.
v E. angustifolia
Preliminarily, I have found a north-south clinal variation among five populations of E. angustifolia grown in a common garden for survival, growth habits and biomass yield. This variation matches the findings of Javad Feghahati, who conducted an RFLP analysis(9)of the same populations that were grown in my common garden. The Ontario group will soon analyze the metabolite content of composite root samples of the five populations.
Research at the University of Newcastle in Australia has been focused on alkamide content of Echinacea species(10). Current research is concerned with metabolite content under various production systems (http://www.uq.oz.au/~gagrego/newslett/ncn18113.htm.)
Recommendations for Further Research
An immunological assay should first be developed for the effectiveness
of Echinacea extracts using the extraction procedure of the Ottawa group.
If such an assay is beyond the scope of the research community, then other
analytical procedures should be investigated, such as the use of nitrogen-phophorous
gas chromatography as recommended by Steve Moring. Efficient analyses for
high-molecular-weight polysaccharides should be developed.
A study of chemical components from the three commercial Echinacea species
grown under different production systems would be of interest to medicinal
plant growers and to the scientific community. Taking the lead from Kim
Bauer and Dean Gray, treatments for the production systems should include
with and without water and nitrogen stresses, or more subtly, with water-soluble
fertilizer vs. time-released fertilizer (such as compost or specially-formulated
fertilizer). Two or three levels of potassium fertilization should be included.
The effects of plant spacing and time of harvest on drug quality (or
quantity) in floral buds, other herbage and roots, and on the size of roots
of E.angustifolia,
E. pallida and E. purpurea should
be investigated at multiple locations. I suggest locating field trials
in British Columbia, Saskatchewan, South Dakota and Kansas to represent
the range of conditions where E. angustifolia is currently grown
and where researchers of this crop are located. By testing at several locations,
the need for replications at each site will be reduced(11),
thus increasing the number of cultivars that could be tested at each location
without increasing costs. This scheme would efficiently use the resources
that are presently dedicated to Echinacea research.
Source of seed or choice of cultivar is necessary for the success of
such a study. I am willing to supply second generation seed selected from
these populations if such a study is conducted. For seeds of E. pallida,
sources should be sought within the research community. E. purpurea
varieties should be chosen on the basis of the German literature cited
earlier plus input from current research on E. purpurea as an ornamental
in horticultural publications and as a medicinal crop in S.A.R.E.- funded
projects. Perhaps seed could be obtained from growers who have selected
for various traits. An Herbpharm representative claimed in a letter-to-the
editor of a recent Herbalgram magazine that they have developed a "buzzier"
strain of E. purpurea.
Production methods will have to be standardized among the various locations.
I am willing to help with this process. To remove genetic variation from
the equation, I would use tissue culture techniques developed by Dr. Jim
Harbage of SDSU. Clones would be sent to the various locations at the appropriate
time for starting in a greenhouse. Potting medium, containers, early fertilization
and photoperiod length would have to be agreed upon. Questions of field
preparation and the use of raised beds or flat ground would have to be
resolved.
This proposal is timely from the grower's perspective. Is it timely for researchers?
1. 1Ethrel varies from other sources of ethephon (tradenames Cepha, Florel, or Ethepha) by percent of ethephon. Ethrel is 21.3 % ethephon, (2 lbs. ethephon per gallon). It is available in gallon quantities in states where labeled, and is manufactured by Rhone-Poulenc Ag Company, 509-697-3236.
However, a more accessible source of ethephon is Florel, which is available to commercial growers in quarts, product code 03-0400-1 from Hummert Intern., phone: 314-739-4500, orders: 1-800-325-3055
2. 2Clear plastic boxes, 5" x 5-1/4" x 1-3/8" ID, with transparent friction tops are available from:
Hoffman Manufacturing Company
325 11th Ave. SE
P.O. Box 547
Albany, Oregon 97321 503-926-2920 FAX 1-800-343-6724
3. 3 Shalagy A.S. El-Gengaihi, S.E. Agina, E.A. E.L.-Khayat and S.F. Hendawy, 1997. "Growth and Yield of Echinacea purpurea L. as Influenced by Planting Density and Fertilization," J. Herbs, Spices and Medicinal Plants, 5(1): 69-77.
4 Parmenter, G. and R. Littlejohn, 1997. "Planting density effects on root yield of purple coneflower," New Zealand J. of Crop and Hort. Sci. 25: 169-175.
5. 5 Franke R., Schenk R., Schaser J., Nagell A., 1997. Influence of Different Methods of Cultivation on Yield and Active Principles of Echinaea pallida (Nutt.) Nutt. Zeitschrifr fur Arznei und Gewurzpflanzen. 2(4):173-185.
6. 6 Li, T., 1998. "Echinacea: Cultivation and Medicinal Value," HortTechnology 8(2): 122-129.
7. 7 Von U.Bomme, J. Holzl, Chr. Hessler and Th. Stahn, 1992, February. "Wie Beeinflusst die Sorte wirkstoffgehalt und Ertrag von Echinacea purpurea (L.) Moench im Hinblick auf die pharmazeutische Nutzung?" Landwirtschaftliches Jahrbuch. 69(2): 149-164.
8. 8 Von U.Bomme, J. Holzl, Chr. Hessler and Th. Stahn, 1992, March. "Wie Beeinflusst die Sorte wirkstoffgehalt und Ertrag von Echinacea purpurea (L.) Moench im Hinblick auf die pharmazeutische Nutzung?" Landwirtschaftliches Jahrbuch. 69(3): 323-342.
9. 9 Feghahati, J., 1995. Germination, Genetic Variation and Phylogenetic Relationships within and among Populations of Echinacea angustifolia. PhD Thesis, South Dakota State University.
10 Rogers, K.L. Grice, I.D. Mitchell, C.J. and L.R. Griffiths, 1998. "High-Perfor,ance Liquid-Chromatography Determined Alkamide Levels in australian-Grown Echinacea spp.," Australian Journal of Experimental Agriculture, 38(4): 403-408.
11 Johnson, J.J. Ulrich, S.E. and J.R. Alldredge,
1990. "On-farm Testing of Additional Sorghum Cultivars by Replacement of
Replications with Additional Locations," Poster Paper Abstracts: Genotype-by-Environment
Interaction and Plant Breeding Symposium, p.11, Louisiana Ag. Expt.
Station).
P.O. Box 512 E-mail: rlittle@choicetech.net or Richard_Little@sdstate.edu
Volga SD 57071 See also http://biomicro.sdstate.edul/Reesen/Echinaca.htm