Wild-proso millet

Panicum miliaceum L.

Wild-proso millet immature inflorscence
Wild-proso millet inflorescence

Images above: Left: Wild-proso millet immature inflorescence (Joseph DiTomaso, University of California, Davis). Right: Wild-proso millet inflorescence (Joseph DiTomaso, University of California, Davis).

Identification

Other common names:  proso millet, proso, millet, hog millet, broom-corn millet, broom-corn, panic millet

Family:  grass family, Poaceae

Habit:  Tall, upright annual grass.

Description:  Seedlings are upright.  The leaves are rolled in the bud, lack auricles, and are densely covered in short, stiff hairs.  The ligule is a fringe of hairs 0.08-0.16” (0.2-0.4 cm) long that are fused at the base.  The seed remains attached to the seedling at the root.  Mature plants reach 1-4.25 ft (0.3-1.3 m) tall.  Stems are upright or occasionally nodding.  The collar is pale green or white and partially encloses the stem.  The sheath is open and densely hairy. The ligule is similar to that of the seedling.  Leaves are 4-12” (10-30 cm) long by 0.25-1” (0.6-2.5 cm) wide with a conspicuous, strongly ridged pale green or white midvein.  The leaves are sparsely to densely covered in long hairs (or occasionally hairless).  The roots are fibrous.  The inflorescence is a 3-12” (8-30 cm) long, terminal, upright to nodding, branching panicle which remains partially enclosed in the leaf.  Panicle branches are smooth to rough.  Spikelets occur singly at the end of branches, and are 0.18-0.21” (0.46-0.53 cm) long, smooth, and have prominent green veins.  Each spikelet produces a single seed, which is shiny, smooth, white to reddish-brown or black and 0.13” (0.33 cm) long by 0.1” (0.25 cm) wide.  The apparent seed includes a thin, tight covering of fruit tissue.
Similar species:  Witchgrass (Panicum capillare L.) and fall panicum (Panicum dichotomiflorum Michx.) have similar habits and inflorescences to wild-proso millet.  Witchgrass is shorter (8-36” = 20-90 cm tall) and has smaller spikelets (0.08-0.13” = 0.20-0.33 cm long) than wild-proso millet.  Fall panicum is hairless at maturity, and the seedling leaves have hair only on the underside.

Note:  Wild-proso millet populations vary greatly in the degree to which they depart from the domesticated crop and behave as weeds.  In general, black seeded populations tend to be weedier than populations with light colored seeds (Colosi et al. 1988).

Management

Crop rotation is an important component of wild-proso millet management because seed banks of this species are rapidly depleted without substantial annual input of seeds to the soil (Williams and Harvey 2002).  Any plants that emerge in alfalfa will be mowed before they can produce seeds, and most of the relatively short-lived seeds will die during the sod phase of the rotation.  For example, a Wisconsin study found that four years of alfalfa reduced seedlings emerging in a following corn crop by 80% (Harvey and McNevin 1990).  Winter wheat is very competitive against wild-proso millet and will usually be harvested before the weed can go to seed (Mickelson et al. 2002).  Spring cereals also suppress this weed since they become well established by the time it emerges (Wilson 1992), and, in principle, a large proportion of the seeds produced can be captured or destroyed during combine harvest.  In contrast, corn, soybeans and dry beans are poor competitors with wild-proso millet (Wilson 1992).  Late spring planting of summer crops, however, can reduce wild-proso millet density relative to earlier planting dates (Harvey and McNevin 1990).  Crop-like biotypes are more susceptible to competitive stress than dark-seeded biotypes (Warwick and Thompson 1987).

Avoid fall tillage if possible: the relatively large seeds are highly attractive to a wide range of seed predators (Colosi et al. 1988, Mickelson et al. 2002, Peachy and Mallory-Smith 2011), and are also more susceptible to other forms of mortality when near the soil surface (Colosi et al. 1988).  Even the soil disturbance associated with no-till drilling of cover crops may be sufficient to protect the seeds (Peachy and Mallory-Smith 2011).  Deep tillage in the spring, however, will reduce seedling emergence following a season with heavy seed rain, and since the annual death rate of seeds is high even deep in the soil, relatively few seeds will return to within emergence depth following tillage in subsequent years.

Rotary hoeing can kill a substantial proportion of the first flush of seedlings.  Time the hoeing for just before the seedlings emerge, about 7-10 days after last tillage (Mickelson 2002).  Some seedlings will emerge from below the depth of hoeing.  Configure tine weeders to bury these just after emergence.  The shoot will re-sprout if broken above the seed (Cavers and Kane 1990), so burying seedlings is preferable to breaking them.  Inter-row cultivation of row crops often catches the young wild-proso millet after most have emerged, and can provide up to 95% control (Wilson 1992).  Completely covering seedlings with 0.8” (2 cm) of soil kills them (Cavers and Kane 1990), so if possible, hill crop rows while seedlings are still small.

Wild-proso millet is still spreading to new fields in most of its range and preventing new infestations, particularly of the black seeded biotypes, is essential.  Since tillage equipment, combines and forage harvesters are major means of spread, till and harvest infested fields last (Mickelson et al. 2002).  Similarly, clean machinery before moving from an infested to a clean field.  Pull up new outbreaks and if the plants have flowered, remove them from the field (Mickelson et al. 2002).

Ecology

Origin and distribution:  The species is native to Eurasia, and was one of the first domesticated grain crops.  The first black seeded biotype in North America was found in Minnesota and Wisconsin in 1970 (Colosi and Schaal 1997) and was probably introduced from Europe or Asia (Bough et al. 1986).  Crop-like weedy biotypes appear to have developed spontaneously by mutation from domesticated proso millet.  Wild proso millet now occurs in most of the U.S.A. and southern Canada (USDA Plants), but is most problematic from southern Ontario through the Midwest to the Northwest.  In addition to temperate North America, it occurs in Mediterranean Europe and the Middle East, South Asia, Japan, Australia, New Zealand, and South Africa (Holm et al. 1979)

Seed weight:  3.87-7.24 mg (Warwick and Thompson 1987), 3.8-5.9 mg (Eberlein et al. 1990).

Dormancy and germination:  Seed dormancy at maturity varies with biotype.  In contrast with domesticated proso millet which had 100% germination at maturity, three wild biotypes had 51-87% germination even though seeds had greater than 99% viability (Eberlein et al. 1990).  In another study, only 22% of seeds were capable of germination 10 weeks after appearance of the inflorescence (Carpenter and Hopen 1985).  Biotypes with dark colored seeds take up water more slowly and germinate more slowly than those with white or golden colored seeds (Khan et al. 1997).  Domesticated proso millet germinates at temperatures from 50-108 °F (10-45 °C) (Baltensperger 1996), with the optimum from 68-86 °F (20-30 °C) (Oelke et al. 1990).  The lower threshold for wild-proso millet germination is 45 °F (7 °C) (Wiese and Binning 1987).  Under field conditions, seedlings begin emerging the week after day/night temperatures exceed 77/49 °F (25/9.5 °C) (Anderson and Nielsen 1996).  Dormancy is overcome by moist, cold temperatures of 41 °F (5 °C) and the absence of light during germination (Carpenter and Hopen 1985).

Seed longevity:  Generally, seeds from light seeded, crop-like populations have very high mortality over the winter, and none survive a second winter.  In contrast, black seeded populations have a relatively low overwinter mortality rate.  In Canada, survival of black seeded populations buried at 8” (20 cm) for 11 months averaged 40%, and very few seeds survived for 42 months (Colosi et al. 1988).  Another study found an average annual seed mortality of 39% over a 54-month period (Harvey and McNevin 1990).  This figure is close to the 42% mortality rate from a 1-year study (Davis et al. 2005).  Seeds survived better in a moderately drained soil than in a well-drained sand or a poorly drained clay loam (Colosi et al. 1988).  Seed predators rapidly remove seeds from the soil surface (Peachy and Mallory-Smith 2011).  Although deep burial at 6-8” (15-20 cm) promotes seed survival relative to more shallow burial at 2” (5 cm) (Colosi et al. 1988, Peachy and Mallory-Smith 2011), the increase in survival is small relative to many other weed species.

Season of emergence:  In Ontario, Wisconsin and Colorado, most seedlings emerge in late May and June with a few continuing to emerge later in the summer (Colosi et al. 1988, Mickelson et al. 2002, Anderson and Nielsen 1996). 

Emergence depth:  The average depth of emergence is 1-2” (2.5-5 cm) (Bough and Cavers 1987, Colosi et al. 1988), and the maximum depth from which seedlings can emerge is 5.4” (13.5 cm) (Colosi et al. 1988).  Emergence from the soil surface is minimal (Colosi et al. 1988).

Photosynthetic pathway:  C4 (Elmore and Paul 1983)

Sensitivity to frost:  The species tolerates cold better than most C4 plants, but will not tolerate frost (Baltensperger 1996).

Drought tolerance:  Domesticated proso millet is possibly the most drought tolerant of all cereals, despite a shallow root system (Baltensperger 1996), and wild proso millet is similarly drought tolerant.  Young wild-proso millet can tolerate at least 2 weeks of drought, and many plants survive such droughts even when raked half out of the soil (Cavers and Kane 1990).

Mycorrhiza:  No reports are available, but wild-proso millet probably is mycorrhizal based on the mycorrhizal status of other Panicum species (Dhillion and Friese 1994).

Response to fertility:  No information is available on the response of weed biotypes to nutrient applications.  Domestic proso millet is less responsive to fertility than most crops (Baltensperger 1996).  The recommended N application of 36 lb/A (40 kg/ha) only resulted in a 6 to 40% increase in dry weight (Anderson 2000).  The species does best on soils with a pH between 5.8 and 6.8 (USDA Plants).

Soil physical requirements:  The species is well adapted to medium and fine textured soils but does poorly on course textured soils.  It does not tolerate anaerobic or saline soils (USDA Plants).

Response to shade:  Experiments in Washington and Illinois found a negative correlation between wild-proso millet seed production and the percentage of light intercepted by sweet corn canopies (Williams et al. 2007).  Nevertheless, wild-proso millet is moderately shade tolerant.  Dry weight was reduced only about 25% with 47% shade.  With 90% shade, however, dry weight was reduced about 80%, but this effect is lower than that for many other weeds.  Plants grew substantially taller, but thinner, when shaded, which potentially allows them to grow through crop canopies.  Even with 90% shade, plants still produced seeds (Carpenter and Hopen 1985).

Sensitivity to disturbance:  Wild-proso millet is highly resistant to physical damage.  Crushing young plants with tractor tires or repeated mowing had little effect on survival or subsequent growth.  Many plants even survived rotary tilling or disking.  In controlled experiments, even complete removal of either roots (below the seed) or shoots had little effect on survival of seedlings at the 3 or 6 leaf stage, and even at the 1 and 2 leaf stages, many plants recovered from severe damage.  Only a very few plants, however, recovered from complete burial under 0.8” (2 cm) of soil.  (Cavers and Kane 1990)

Time from emergence to reproduction:  Maturation is more rapid as day length shortens.  Flowering occurs in 2.5-4 weeks at a 10-12 hr day length, whereas flowering takes 7.5 weeks at a 16 hr day length (Carpenter and Hopen 1985, Patterson et al. 1986).  Some seeds will germinate as early as 4 weeks after appearance of the inflorescence, but even after 10 weeks nearly 80% of seeds were still dormant (Carpenter and Hopen 1985).  Dark seeded biotypes mature in 6-7 weeks after emergence whereas many light seeded biotypes take 9 weeks (Bough et al. 1986).

Pollination:  Wild-proso millet is primarily self-pollinating, with out-crossing rates of less than 10% (Colosi and Schaal 1997).

Reproduction:  Wild proso millet plants grown without competition produced 69,000-94,000 seeds per plant (Eberlin et al. 1990).  Under competitive conditions, seed production of 420 to 620 seeds per plant were observed (Williams and Harvey 2002). 

Dispersal:  Seeds are spread by birds, mammals, and farm machinery (Mickelson et al. 2002, Khan et al. 1997).   Combine harvesters spread the seeds substantial distances within fields and probably between fields as well (McCanny and Cavers 1988).

Common natural enemies:  Sphacelotheca destruens (head smut) and Ustilago crameri (kernel smut) can substantially damage plants (Oelke et al. 1990).  Birds, rodents and insects eat many seeds (Colosi et al. 1988, Mickelson et al. 2002).

Palatability:  Cultivated proso millet is harvested as grain for human and animal consumption or as fodder (Oelke et al. 1990, Baltensperger 1996), and at least the crop-like, light-seed-color, weedy biotypes are similarly palatable.  Wild proso millet can, however, cause poisoning of young sheep and goats (Burrows and Tyrl 2006).

References:

  • Anderson, R. L.  2000.  A cultural system approach can eliminate herbicide need in semiarid proso millet (Panicum miliaceum).  Weed Technology 14:602-607.
  • Anderson, R. L., and D. C. Nielson.  1996.  Emergence pattern of five weed species in the central Great Plains.  Weed Technology 10:744-749.
  • Baltensperger, D. D.  1996.  Foxtail and proso millet.  In, Progress in New Crops, J. Janick (ed.), p. 182-190.  ASHS Press: Alexandria, VA.
  • Bough, M., J. C. Colosi, and P. B. Cavers.  1986.  The major weedy biotypes of proso millet (Panicum miliaceum) in Canada.  Canadian Journal of Botany 64:1188-1198.
  • Bough, N., and P. B. Cavers.  1987.  Proso Millet.  Fact Sheet, Ontario Ministry of Agriculture Food and Rural Affairs.  http://www.omafra.gov.on.ca/english/crops/facts/87-025.htm
  • Burrows, G. E., and D. J. Tyrl.  2006.  Handbook of Toxic Plants of North America. Blackwell:Ames, IA.
  • Carpenter, J. L., and H. L. Hopen.  1985.  A comparison of the biology of wild and cultivated proso millet (Panicum miliaceum).  Weed Science 33:795-799.
  • Cavers, P. B., and M. Kane.  1990.  Responses of proso millet (Panicum miliaceum) seedlings to mechanical damage and/or drought treatments.  Weed Technology 4:425-432.
  • Colosi, J. C., P.B. Cavers, M. A. Bough.  1988.  Dormancy and survival in buried seeds of proso millet (Panicum miliaceum).  Canadian Journal of Botany 66:161-168.
  • Colosi, J. C., and B. A. Schaal.  1997.  Wild proso millet (Panicum miliaceum) is genetically variable and distinct from crop varieties of proso millet.  Weed Science 45:509-518.
  • Davis, A. S., J. Cardina, F. Forcella, G. A. Johnson, G. Kegode, J. L. Lindquist, E. C. Luschei, K. A. Renner, C. L. Sprague, and M. M. Williams II.  2005.  Environmental factors affecting seed persistence of annual weeds across the U. S. corn belt.  Weed Science 53:860-868.
  • Dhillion, S. S., and C. F. Friese.  1994.  The occurrence of mycorrhizas in prairies: Applications to ecological restoration.  Thirteenth North American Prairie Conference 13:103-114.
  • Eberlein, C. V., E. L. Lurvey, T. L. Miller, and J. L. Michael.  1990.  Growth and development of wild-proso millet (Panicum miliaceum) biotypes.  Weed Technology 4:415-419.
  • Elmore, C. D., and R. N. Paul.  1983.  Composite list of C4 weeds.  Weed Science 31:686-692.
  • Harvey, R. G., and G. R. McNevin.  1990.  Combining cultural practices and herbicides to control wild-proso millet (Panicum miliaceum).  Weed Technology 4:4333-439.
  • Holm, L., J. V. Pancho, J. P. Herberger, and D. Plucknett.  1979.  A Geographical Atlas of World Weeds.  Wiley: New York.
  • Khan, M., P. B. Cavers, M. Kane, and K. Thompson.  1997.  Role of the pigmented seed coat of proso millet (Panicum miliaceum L.) in inbibition, germination and seed persistence.  Seed Science Research 7:21-25.
  • McCanny, S. J., and P. B. Cavers.  1988.  Spread of proso millet (Panicum miliaceum L.) in Ontario, Canada. II. Dispersal by combines.  Weed Research 28:67-72.
  • Mickelson, J. A., C. M. Boerboom, and R. G. Harvey.  2002.  Wooly cupgrass and wild proso-millet management.  Weed Science, University of Wisconsin. http://128.104.239.6/uw_weeds/extensions/articles/woolwildpro.htm.
  • Oelke, E. A., E. S. Oplinger, D. H. Putnam, B. R. Durgan, J. D. Doll, and D. J. Undersander.  1990.  Millets.  Alternative Field Crops Manual. February 22, 2007. http://www.hort.purdue.edu/newcrop/afcm/millet.html.
  • Patterson, D. T., A. E. Russell, D. A. Mortensen, R. D. Coffin, and E. P. Flint.  1986.  Effects of temperature and photoperiod on Texas panicum (Panicum texanum) and wild proso millet (Panicum miliaceum).  Weed Science 34:876-882.
  • Peachy, R. E., and C. Mallory-Smith.  2011.  Effect of fall tillage and cover crop strategies on wild-proso millet (Panicum miliaceum) emergence and interference in snap beans.  Weed Technology 25:119-126.
  • USDA Plants. USDA, Natural Resources Conservation Service.  Plants Database. http://plants.usda.gov/
  • Warwick, S. I., and B. K. Thompson.  1987.  Differential response to competition in weedy biotypes of proso millet.  Canadian Journal of Botany 65:1403-1409.
  • Wiese, A. M., and L. K. Binning.  1987.  Calculating the threshold temperature of development for weeds.  Weed Science 35:177-179.
  • Williams, B. J., and R. G. Harvey.  2002.  Influence of simulated seed rain on the seed bank of wild-proso millet.  Weed Science 50:340-343.
  • Williams II, M. M., R. A. Boydston, and A. S. Davis.  2007.  Wild proso millet (Panicum milliaceum) suppressive ability among three sweet corn hybrids.  Weed Science 55:245-251.
  • Wilson, R. G.  1992.  Wild proso millet.  NebGuide G83-648-A.  Cooperative Extension, Institute of Agriculture and Natural Resources, University of Nebraska, Lincoln, NB.