Restoration in the Colorado Desert: Management Notes

Native Seed collection, processing, and storage for revegetation projects

Prepared for the California Department of Transportation
District 11, 2829 Juan Street, San Diego, CA, 92138
as part of the Desert Revegetation Project
October 1993 
Matthew W. Fidelibus and Robert T.F. Mac Aller
Biology Department
San Diego State University
San Diego, CA 92182 


State and federal laws and regulations are increasingly requiring sites disturbed by mines, highways, and other construction projects to be revegetated with native plants. Many state and local governments in the West are using native species for park and highway landscaping projects. In California, increased use of the deserts has resulted in plant and soil degradation which can best be minimized by revegetation with native shrubs (Kay et al., 1984). These revegetation projects can require considerable amounts of seed. Unfortunately, seed production in desert is erratic and often seeds of a particular species are not available from wild stocks for collection when needed. The establishment of a native seed bank is desirable for assured availability. The California Department of Transportation allows advance collection for this reason. If very large quantities are needed a growout phase of two to ten years may be required. Before a useful seed bank can be established, proper collection, storage and germination techniques for each species must be known.


Seed quality is highly variable from year to year and should be evaluated before collecting large quantities of seed. If the seed quality is very low it may not be worthwhile collecting seed, table 1. If the seed quality is very high it may be worth setting up a large scale collection program. Seed quality can be assessed by non-destructive X-ray analysis, dissection, and germination tests.

Table 1. Seed viability in Isomeris arborea collected at Red Rock Canyon State Park


% Good seed, dark seeds

% Good seed, light seeds










Lippit, 1992

For most revegetation projects, it is desirable to harvest seeds from a diverse population; at least 50 plants should be utilized for larger orders. Jones (1989) found that seeds of Juniperus excelsa from different maternal plants had significantly different germination rates. To further encourage genetic variation, plants should be selected from different stands in a range of comparable sites, as provenance may greatly affect germination and growth characteristics (Hartman; Kester, 1983). Within a stand, it is likely that most individuals are related. If seed is not collected across a broad genetic base, inbreeding will occur. Too much inbreeding can result in reduced diversity and inferior progeny. Stands are considered different if plants are separated by enough distance to prevent cross fertilization between populations. The United States Forest Service uses 200 yards as a minimum distance between stands of pine trees, but larger distances may be advisable for wind pollinated desert plants (Lippit, 1991).

Although major shrub diseases are not usually passed on from the diseased plant through the seed to the next generation, some shrub species have been shown to possess genetic strains which have some degree of resistance to certain diseases (Schopmeyer, 1974). To take advantage of this possibility, it is best to collect seed only from healthy shrubs. Shrubs with desirable phenotypes may be genetically superior (in a specific ecosystem) to neighboring plants with less desirable traits. The seeds of such plants would be more likely to successfully establish themselves in a revegetation project. Plants that are dominant or co-dominant (have a fast growth rate and high vigor), will often produce more viable seed (Lippitt, 1991). Diseased plants, or plants subjected to environmental stress such as drought, exhibit reduced vigor and usually generate fewer seeds. Some species may produce a stress crop with more seed. Seed vigor may or may not be compromised (Lippitt, 1992). Native stands of Jojoba (Simmondsia chinensis) produced many more seeds when provided with supplemental water from micro-catchment basins (Ehrler et al., 1978).

Revegetation efforts should use seed from local stands because local ecotypes are most likely to succeed and successfully reseed. Use of seeds from other areas, where climatic conditions are different, may lead to total failure (Dobbs et al., 1976). For this reason, the California Department of Forestry and Fire Protection has divided the state into numerous seed zones (Figure 1). Seed zones are established within broad biogeoclimatic regions. Careful consideration that might greatly affect tree or shrub growth (Schopmeyer, 1974).

Seed should generally be collected and planted within the same zone and within approximately 500ft (150 m) of its original elevation. Some shrubs, such as saltbush (Atriplex), appear to exhibit 'eco-speciation' (i.e. plants of the same species become so adapted to particular sites that their seeds exhibit poor survivorship when planted in a dissimilar site where local individuals of the same species flourish). For example, Atriblex seeds collected at low elevations in the southwest will not be winter-hardy when planted at northern locations (Young & Young, 1986).

Once a stand has been selected, the timing of seed collection can be crucial (Young & Young, 1986). For some species, ripe seed is available for several weeks or months, in others it may be for only a few days. Collection of immature seeds results in low seed viability or dormancy. Immature seeds are often incompletely developed, some proteins may not have reached a stable conformation that will allow for dehydration without denaturation (Priestly, 1986). Enzyme activity is subsequently lost in these seeds rendering them inviable. If seed collection is delayed the seeds are often consumed by various seed predators such as insects, birds, and rodents. Other seeds are likely to be lost since the seeds of many wild shrubs dehisce (fall from the seedhead) very rapidly.

Seeds that ripen and fall quickly can sometimes be collected by early placement of the seed head in a section of nylon stocking or netting. This has worked well for collecting ocotillo seeds. Canvas or plastic can be spread on the ground to collect seeds. This is most effective for heavy seeds that will drop to the tarp even in a mild wind (Lippitt, 1992). The time and difficulty required to spread the sheeting under the plant limits the value of this collection scheme, however. In some cases, it may be possible to collect seeds. Seed gathered from the soil surface, but many problems can be associated with ground collected seeds. Seed gathered from the soil surface us usually of low quality, requires excessive cleaning, (Young & Young, 1986), any may result in a fungal contamination of the seed (Lippitt, 1992). If left on the ground too long, fleshy seeds such as acorns may experience a reduction in moisture content to unacceptable levels. Some species, such as oaks, 'abort' or release problem seed early while the rest of the seed is retained on the tree for further ripening. Thus, it is often best to harvest seed directly from the plant. Bush collected seed of Larrea tridentata (creosote bush) had twice the viability of ground collected seed (Kay, 1977). Viability of blue oak (Quercus douglasii) acorns collected from the tree was more than double that of seed collected from the ground (Phillips, 1992).

Harvesting the seeds of wild plants usually requires manual labor since the desired species rarely grow in pure stands, and the topography often limits use of mechanical equipment (Young & Young, 1986). However, browse seed harvesters using vacuum suction for collection are being used more frequently. Commercially available leaf and yard vacuums may be of similar value in seed collection. Also available commercially is a box like attachment with a collection bag for rotary line trimmers (Environmental Survey Consulting, n.d.). Seed clipped from the selected plant is guided into the collection bag by an aluminum box surrounding the cutting head. Further improvement in such machines show promise in accelerating shrub seed production, particularly along roadsides, seeded plantations, and other readily accessible areas (Vallentine, 1971). Four-wing saltbush, (Atriplex canescens) a good candidate for native shrub revegetation in California deserts, has been grown on cultivated agriculutra lands; offering an alternative to sole reliance on harvesting native stands. The seeds of four-wing saltbush can be stripped by tractor drawn seed strippers (Young & Young, 1986).

Small seeds or fruits are generally harvested directly into a container carried or worn by a picker (Schopmeyer, 1974). A typical method of manual collection requires holding a tray or a box under the outstretched branches of a shrub while flailing the bushes with a stick or paddle or by sweeping the arms across the upper branches to loosen the seeds, which then fall into receptacle (Young & Young, 1986). A tough net or plastic bag with a hoop opening can also be used to collect seed. It can be placed over a branch being shaken or held under the branch like the tray.

The inyo tray was developed on the Inyo National Forest, on the east side of the Sierra Nevada Mountains in California, for collecting bitter brush seeds by hand (Young & Young, 1986). The aluminum tray is 20 in (51 cm) long by 30 in (76 cm) wide and rounded at the bottom to a depth of 8 in (20cm). A handle is inserted along the axis. A Hudson Bay Co. blueberry/cranberry harvesting device has been useful for collecting seed and pods from wild shrubs (Lippitt, 1992). The harvester is an open topped wooden box with tines along one edge. The device is held by a handle opposite the tines and is ranked across branches so that the seeds or pods are stripped and fall into the container.

For spiny shrubs, where the fruits must be physically stripped from the branches, a lightweight, 5-20 gallon barrel provides a ridged lip over which to bend shrub branches for removing fruits. Salad tongs are helpful in collecting cactus fruits and cholla buds.

Larger seeds and fruits such as palo verde pods can be collected by cleaning or tarping the area under the tree or shrub and then beating down the dry pods with a stick, bamboo pole, or 1/2"plastic pipe, or by shaking the branches. The pods are then raked or picked up by hand or concentrated by folding the tarp (Bainbridge & Clark, 1987).

Leaves, twigs, and other debris is collected with the fruits will fragment during drying and processing, increasing the difficulty of cleaning the seed (Shopmeyer, 1974). As much debris as possible should be separated during collection.


Seeds often require cleaning, dewinging, and debearding. Removing dirt, leaves, stems and chaff from the seeds will reduce the bulk for handling and storage, remove moist, material that may cause heating and mold formation in storage, and facilitate flow through seeding equipment (Vallentine, 1971). Higher percent purity also allows more accurate seeding, reducing seed use (Lippitt, 1992). Seeds can be cleaned in the field by hand screening if there are dry enough. The collected material is filtered through a screen with large enough openings to allow the seeds or fruit to pass through, leaving coarse trash and waste on top of the screen to be discarded. A second sieving is made with a smaller screen which prevents the desired seed from passing through. The seeds are retained while the fine waste passes through for discarding (Young & Young, 1986).

Meat grinders are commonly used to separate seeds from pods. Hand separating Prosopis seed from pods can produce several hundred clean seeds per hour, compared to about 7,000 seeds per hour using a meat grinder (Pasiecznik & Felker, 1992). The choice of plate hole size is important. Large holes (3/8 in; 9.5mm) facilitate material flow, but only liberate 20% of the seed. Smaller holes (1/4 in; 6.4mm) free the remaining seed, but reduce material flow. Mid-sized holes are not commercially available. This is also one of the seed cleaners at the California Department of Forestry Reforestation Center in Davis, CA.

Because of their high sugar content, fleshy pods must be thoroughly dried before being fed into a grinder. Pasoecznik & Felker (1992) recommend drying Prosopis pods at 125ェ (52イ) overnight. If the pods are not sufficiently dry, they can carmelize in the grinder and form a solid mass.

A hammer mill, cement mixer, or gem grinder can be used to remove seeds from chaff and pods. Hammer mills consist of many finger like hammers rotating inside a section of perforated metal cylinder. The seeds are forced through the holes, separating them from their appendages (Young & Young, 1986). Cement mixers and gem grinders have rotating drums with paddles that can de-wing seeds (Lippitt, 1992). Four-wing saltbush seed dewinged in a hammer mill was found easier to plant at proper depths and germinated more rapidly (Vallentine, 1971). Seeds extracted in a mill may already be sufficiently scarified from the action of the mill to readily germinate (Bainbridge & Clark, 1987), but mechanical injury can exert a depressing influence on storability (Priestly, 1986). As an alternative, seed is placed in a mesh bag (or a plastic bag with a few small holes in it) and compressed air can be used to blow the wings off seeds (Lippitt, 1992).

To improve seed purity and decrease the percent of empty or less viable seed, weed seeds, the seeds of other plants and empty seed must be removed. Seeds can be sorted and cleaned using an air separator, which utilizes the movement of air to divide materials according to their terminal velocities (Young; Young, 1986). A seed's size, shape, surface texture, and density are factors that contribute to it's terminal velocity. When fed into a rising airstream, seeds and debris of different terminal velocities will separate from each other. The velocity of the airstream can be manipulated to capitalize on the differences between the seeds or trash being sorted.

Disease Control

Insects and fungi are usually controlled by dry, near freezing, or subfreezing storage of seed. Many insects that attack stored seeds were originally from the tropics and have spread and adapted to colder climates by living in man-made seed storage shelters. Temperatures of greater than 50ェ(10イ) are usually needed to develop damaging populations (Young; Young, 1986). Brucids, a boring insect that may do considerable damage to mesquite seeds on the tree, the ground or in storage, can usually be controlled by drying pods immediately after collecting and then freezing them to kill the larvae (Bainbridge; Clark, 1987).

Sometimes the seed cannot be brought to a cold enough temperature to kill the insects without damaging the seed. A 20% solution of Malathion in water followed by a drying period and a subsequent dusting with 5% Sevin has proven effective in these cases (Desert Enterprises, 1992). Kay et al., (1984) used the insecticide Phostox (aluminum phosphide) to protect the seeds of Mojave desert shrubs.

In moist storage at cool temperatures, pretorage fumigation for fungi may be necessary (Shopmeyer, 1974). At one time, seed collectors routinely treated seeds with fungicides before storage or prior to sowing (Campbell; Landis, 1990). Because of possible toxicity to the seeds, as well as adverse effects on human health and the environment, less drastic measure to control seed pests are becoming increasingly popular. Powdered mustard and cinnamon have been used successfully to control mold on stored acorns and may have potential for other seeds. Scented baby powder seems to provide a modicum of bird and rodent control (Lippitt, 1992). The powder had been added to seeds to improve flow during sowing, and is now sometimes substituted for the fungicide/repellant Arasan 75.

As an alternative to fungicides, seeds may be surface sterilized before storage. Dumroese et al. (1988) reduced the levels of the pathogenic fungi Fusarium in Douglas-fir seeds to negligible levels after a 90 second microwave hot water soak. Similar heat treatments to sterilize seed have been successful using vegetable oils instead of water as the heating agent. Vegetable oils are not as easily imbibed as hot water and are thought to be less toxic to the embryo. Jones (1989) used a 1% sodium hypochlorite solution for 15 minutes and then washed the seeds five times with distilled water. Seeds can also be sterilized by soaking in a 40% solution of household bleach in tap water (2 parts bleach in 3 parts tap water) for ten minutes, then they are rinsed thoroughly in running water for at least 48 hours. A similar procedure can be followed using a 3% hydrogen peroxide solution (Campbell; Landis, 1990).

Unfortunately, both the bleach and hydrogen peroxide treatments are phytotoxic to many species. The 48 hour running water rinse was found to reduce levels of pathogenic fungi to a similar extent as chemical sterilizers without deleteriously affecting seed viability (Lippitt, 1992). For the running water rinse, a shower head is placed in the bottom of a bucket, and an aquarium bubbler is used to concentration of dissolved oxygen and promote water circulation.


Although desert seeds are often long lived and may exhibit multiple dormancy, many seeds have their best germination potential at the moment they reach maturity on the plant (Harrington, 1988). Storage conditions are critical in order to maintain seed viability over an extended period of time. The two most important factors affecting seed longevity are seed moisture content and seed temperature (Evans, 1992). As a general rue, each 1% reduction in seed moisture doubles the life of the seeds (Young; Young, 1986). To protect from premature germination and seed pests, seeds should be dried as quickly as possible to <14% seed moisture and should be stored below this moisture content at all times. Post drying moisture content of seeds from a variety of Mojave desert shrub species range from 1.5%-9.4% (Kay et al., 1984). Seedlings do not emerge as quickly from dried seeds, and the seeds require more water and longer time to imbibe and germinate.

Temperatures can be controlled by storage location, refrigeration, or freezing. Refrigeration, though usually beneficial to seed life, is expensive and may not be cost effective for large quantities of seed. Seed moisture content is controlled by storing properly dried seed in tightly closed containers (or doubled 4 mil plastic bags sealed with barlok ties) or by regulating humidity in the storage area (Schopmeyer, 1974; Lippitt, 1992). Under long term storage seed viability is harmed by a high oxygen atmosphere and benefited by a high carbon-dioxide atmosphere (Frankel; Bennet, 1967). When seed is sealed in a container, respiration reduces oxygen concentrations, and increases carbon dioxide concentrations creating conditions conductive to long storage life. Moisture in tree and shrub seed has been largely controlled by putting dried seed in closed containers whereas much agricultural seed kept in dehumidified storage rooms.

Kay et al. (1984) performed a study to determine if agricultural seed warehouse storage conditions were as satisfactory as closed containers for storing seeds of Mojave desert shrubs. Seeds from the shrubs being stored in hermetically sealed containers were dried for six days at 95ェ(35イ), while seed for warehouse storage were not dried after processing. Seeds of most of the 22 species studied showed unchanged or increased germination rates after 9 years of storage in hermetically sealed glass containers. Germination for most species stored under warehouse conditions was significantly lower after nine years. Atriplex Canescnes, or four wing saltbush, was the only species of shrub which exhibited higher germination rates after being stored under warehouse conditions.

Seed storage must be in containers or facilities that protect the seed from rodents, birds and insects. Many seed collections have been destroyed by rodents. Metal tins or glass bottles are often best.

Processed seed lots often vary widely in quality. Stored seeds need to be identified with detailed label information about the seed. The label should indicate the species and variety (if known), the precise geographic location of the collection (including seed zone if known), the elevation, soil type, date of collection, and the signature of the collector (Vallentine, 1971; Stein et al., 1986). The number of stands and individual plants from which the seed was collected and notes regarding insect damage, seed/pd maturity, and possible poor pollination also contribute to better understanding and processing of particular seed lot (Stein et al., 1986). Seeds/pound, percent germination, percent moisture and seed treatment and storage conditions must also be included (Figure 2)(Lippitt, 1992).

Breaking seed dormancy

Seed dormancy is an ecological important device to optimize the spatial and temporal distribution of the species, but it is an obstacle to revegetation efforts where prompt, uniform, and complete germination is desirable to grow high quality planting stock (Rietveld, 1989). Nondormant seeds readily pass through three germination stages: 1)imbition of water, 2)activation of metabolic processes, and 3)growth of the embryo (Schopmeyer, 1974). If any of these stages are blocked, the seed remains in a state of dormancy. Impermeability of the seed coat to water or gases, hard sedededness, is the most common form of seed dormancy and is characteristic of certain families, including the legumes (Schopmeyer, 1974). If the seed imbibes moisture, but does not germinate, moist stratification may be needed (Young; Young, 1986). If the seed does not imbibe moisture, scarfication is necessary. Scarification involves the physical abrasion or removal of the seed coat to allow entry of water.

Scarification can be accomplished by soaking in acid, mechanical abrasion, or soaking in hot water. Concentrated sulfuric (H2SO4) or hydrochloric (HCl) acid is commonly used to scarify seeds, however many smaller nurseries and propagators are reluctant to use acid. N.T. Mirov designed a simple vessel for use in acid scarification (Young; Young, 1986). It consists of a dish made of a stainless steel screen that can be lowered into a glass container by a center pole. A small volume of acid is placed in the bottom of the container and the seeds are placed into the dish and lowered into the vessel. After the prescribed treatment interval, the basket is raised and the acid is allowed to drain from the seed.

Seeds that have been mechanically cleaned may have been broken seed coat, so milled seeds must be checked carefully before using acid on them to prevent destruction of the seeds (Bainbridge; Clark, 1987). Scarification causes a noticeable thinning of the seed coat; if seed is left too long in an acid bath, a hole will develop in the seed coat allowing acid to enter and kill the embryo (Jones, 1989). The length of time required for scarification must be determined by experimentation, and the treatment duration may vary significantly among seeds of the same species. Mesquite seeds, for example, require between 3 and 15 minutes in an acid bath (Bainbridge; Clark, 1987). Inter species treatment lengths vary from seconds to hours, depending upon the nature of the seed coat. Some particularly woody fruits may require a 24 hour acid treatment (Young; Young, 1986). Seeds should be thoroughly washed following an acid bath.

Some seeds are very sensitive to acid scarification. Stidham et al. (1980) found that chemical scarification reduced germination in many species of woody shrubs including four-wing saltbush. In addition to removing the seed coat, the process generates considerable heat which drives moisture from the seed (Young; Young, 1986). The reaction temperature can be lowered by pre-chilling the acid before treatment and cooling the acid seed mixture during treatment by immersing the reaction vessel in a cool water bath.

Mechanical scarification machines are commercially available. Sharp gravel in a cement mixer may also work or a feed chute can be used to move seeds onto a sanding disc or drum (Bainbridge; Clark, 1987). Small batches of seeds can be scarified by hand, using a file or knife to make a nick or slice in the seed coat. Some care must be taken to avoid injuring the radicle. Seeds that were scarified by sanding showed increase susceptibility to fungus and mold, presumably from the presence of the small particles of seed coat.

A common and effective substitute to acid scarification is the boiling water dip or soak. Most seed processors allow the seeds to soak in boiling water that is allowed to cool. Lippitt (1992) prefers briefly dipping the seeds to soak in boiling water. The boiling water dip is safer than acid scarification, and seed coat thickness is less of a factor in damage. The same length of treatment can be used from year to year which is risky with acid.

If seeds still fail to germinate following scarification, stratification can be used to overcome dormancy. Many seeds with physiological/physical dormancy require exposure to either high or low temperatures before being placed in conditions favorable for germination. Cold-moist stratification is the type most commonly used. In most cases, the seed must be fully imbibed before temperature can be effective in breaking dormancy. Stratification works by stimulating embryo growth, removing inhibitors or both.

The embryo of many seeds fail to germinate because oxygen does not diffuse through the seed coat. Oxygen is more soluble in cold water, so the oxygen requirements of the embryo can be better satisfied during cold moist stratification (Young; Young, 1986). Sometimes oxygen is kept near saturation levels by forcing compressed air through the water. The correct temperature and duration for stratification varies according to the species and must be determined by experimentation. Temperatures generally range from 34-40ェ (1-4イ); with duration varying from weeks to months. Any seed lot to be stratified for more than 30 days should be surface dried after imbition or periodically surface dried and then put back into stratification. This procedure can reduce or prevent mold development which is more likely to develop in the high humidity conditions that exist during long term stratification (Campbell et al., 199); Lippitt, 1992).

Plastic bags are good containers for stratification. The seeds can be placed in bags with a variety of substances, or stratification can occur in bags without substrate; a "naked" stratification (Lippitt, 1992). Common stratification substrates include moist sand, activated charcoal, varmiculite, or kitty litter and stored at low temperatures (33-50ェ, 0.5-10イ) until the stratification requirement is satisfied. Substrates help maintain moisture levels in the bag and some, such as activated charcoal, absorb soluble germination inhibitors (Young; Young, 1986). Many desert seeds have water soluble growth inhibitors that prevent the seed from germinating unless there is sufficient moisture to establish a seedling. If all of the inhibitor is not washed away at once, the seed produces more inhibitor (Bryant, 1985). Graves et al. (1975) found that Ambrosia dumosa (Mojave desert seed sources) seed germination was improved by stratification in activated charcoal or moist sand at 36ェ (2イ) for 30 days.

Seeds from the desert may require high temperature 120ェ(50イ) stratification rather than the low temperature/moist stratification commonly used (Capon; Van Asdall, 1966). High temperature stratification appears to promote seed maturation in desert species (Table 2). Seeds show vastly improved germination percentages during the first five weeks of high temperature storage (compared to seed stored at room temperature), but continued storage at high temperatures results in a loss in viability. Seeds stored at room temperature, 68ェ(20イ), do not achieve similar germination success until after five months of storage. Thus, if a seed lot is to be planted during the same season that it is collected, it may be beneficial to try high temperature stratification. If the seed needs to be stored for longer periods of time, lower temperatures may be desirable.

Table 2. Percentage germination of seeds of several species of desert annuals in response to temperature pre-treatment.

Storage in weeks


@ 20イ


@ 50イ












&127;P. insularis










&127;S. altissimum










&127; G. cansecens










S. arizonicus










If sufficient information on the seed system is available, appropriate scarification treatments can often be deduced. Seed dispersed by animals or birds if often amenable to acid scarification. Flood dispersed seeds (i.e. Palo Verde) often require mechanical chipping or sanding. Stratification times can sometimes be approximated by winter snow duration or temperature cycling. These estimates can reduce the time required to develop an effective treatment, but they are not always reliable (Lippitt, 1992). Some species that don't have long winters have required longer stratification, for example. Thickness of seed coat seems to be as big a factor as winter length. The quality of the seed and the effectiveness of treatments on germination should be recorded on a permanent record, figure 3.

Seed evaluation/germination testing

To determine the value of a seed lot or the rate of seeding needed for a successful planting, every seed batch should be evaluated for purity and percentage of sound seed prior to storage, periodically during storage, and again just before the entire seed lot. A seed lot is generally defined as a quantity of seed collected from a particular location and elevation during one season (Stein, 1986). If the seed lot is stored in several containers, a sample should be drawn from each (Goor, 1963;Stein, 1986). Since laboratory test results can only indicate the quality of the sample submitted, it is critically that the sample taken be representative of the entire lot (Stein, 1986).

Seed lots from wild shrubs are rarely homogenous, so the sampling method must produce a representative sample. Equal subsamples should be taken from each storage container. If the containers are of unequal size or fullness, the amount taken should be proportional to the volume of seed in the container. If the seed lot is divide into small packets, individual packets may be used as subsamples (Stein, 1986). Seed lot subsamples are combined into one composite sample.

Bags or containers of free flowing seed can be accurately sample with a partitioned probe or trier (Stein, 1986). The length of the trier to be used is determined by the depth of the container or bag being sampled. The probe is inserted closed, then opened so seeds from different positions in the container are simultaneously admitted into the slots. Successive thrusts into a container should be along different paths.

Seeds that are not free flowing, such as Larrea, and some other shrub species are usually sampled by hand. An open hand with the fingers held closely together , is inserted into the seeds, closed, and withdrawn holding a representative subsample. As with trier sampling, seeds are drawn from well-separated points within each container sampled (Stein, 1986).

Purity can be expressed as the weight of clean seeds of a species divided by the weight of the total sample (which includes impurities). Multiplying this value by 100 give the percent purity of the sample (Goor, 1963). Purity is determined by inspection of the sample, pure or clean seeds being put aside and weighed separately.

Percentage of sound seeds can be determined by different methods depending upon the species. X-ray evaluation is a non-destructive method of assessing seed fill and potential viability. By combining X-rays with cutting tests it is possible to relatively quickly determine seed quality. A cutting test can be a quick and easy way of determining the number of full seeds, but requires some experience with the species being tested (Schopmeyer, 1974;Lippitt, 1992). Seeds are cut open, and those with firm, undamaged, healthy looking tissue are determined to be viable. Small seeds can be cut by placing them on tape with the sticky side up to prevent the seed from sliding (Goor, 1963). In some cases, seeds can be crushed instead of cut. Determination of desert seed quality can be challenging because even dry seeds may be viable.

Sometimes, the viability of a seed can be predicted by external color. Dark seed of Isomeris aborea, for example is more likely to be viable than the white form. Immature, inviable seed is often light colored (Hanscom, 1984). Color may be an effective indicator for determining viability in the field, but it can be less effective after storage. The soundness of many seeds can also be determined by their weight or density. Most sound seeds will sink in water, while empty ones will float. Heavy, well filled seeds can be separated from e empty, immature ones with an air separator (Stidham et al., 1980).

Biochemical stains are available that stain only living tissue. Of these, tetrazolium salts have proved to be the most successful (Schopmeyer, 1974). In solution, tetrazolium salts are colorless, but in living tissue they are reduced by dehydrigenase enzymes to form a stable red pigment which is insoluble in water. The localization and proportion of dead unstained tissue is the key to classification of seed as potentially viable or inviable. The tetrazolium test is very fast, but lack of uniformity of staining, failure to detect seeds that will germinate abnormally, and difficulty in interpreting different degrees of staining are important drawbacks. Testers need a lot of experience with a species before much useful information can be obtained from the test. Tetrazolium stains have proven to be uncertain on seeds with very effective dormancy. The may work better following stratification.

The most reliable method of determining potential germination is to germinate a representative sample of the seed lot (Schopmeyer, 1974), but this is only effective when the stratification/scarification requirements are relatively well understood. Seeds to be used in germination tests should be separated from debris and tested,. The results are then combined with seed purity information (Lippitt, 1992). The groups of seed are spread on trays of suitable substrate and placed into humid germinators. Alternatively, seed can be placed in enclosed germination dishes or containers where the dishes maintain humidity. Enclosed containers can be advantageous as moisture loss is not a problem as it may be with open trays of substrate. Seed should be spaced properly (2-5 times the normal seed width) to prevent the spread of fungi, especially in large seeds (Schopmeyer, 1974). Theproper germination temperature is species specific and needs to be determined by experimentation. An overview of stratification and scarification requirements is included as Table 2.

The substrate for germination tests must meet the following requirements: 1) nontoxic to the germinating seedling, 2) relatively free of molds, other microorganisms, and their spores, and 3)provide adequate aeration and moisture for germinating seeds (Schopmeyer, 1974). Now that small cabinet germinators are common, natural substrates such as sand or peat are less often used. Paper substrate, including germination blotters, paper towels, and laboratory filter paper, have become more popular. Paper towels sometimes contain germicides which may inhibit germination. Some nurseries use perlite as a substrate because root development is less restricted than on paper and seed can be more easily checked and replaced since roots don't attach to perlite as they do with paper (Lippitt, 1992). Published germination standards for seeds should specify the substrate used (Young; Young, 1986). Results should be reported on the seed lot tracking form, figure 3.

The test substrate must be kept moist enough to provide sufficient moisture for seed germination, though excessive moisture restricts aeration, favors damping off, and inhibits germination (Schopmeyer, 1974). For most seeds, if a film of water forms around the seed, or a film of water forms on the finger when the substrate is pressed, then the paper is too moist. Placing the paper on a bed of sand can improve aeration. Small amounts of water should be added periodically if the germinator does not maintain high humidity.

The time required for germination tests vary among species. Germination tests typically last between 1 week and 1 month. In official testing, germination is defined as "the emergence and development from the seed embryo of those essential structures which, for the kind of seed in question, are indicative of the ability to produce a normal plant under favorable conditions" (Schopmeyer, 1974).

Pre-sowing treatments

Seed can be pelletized to improve handling and to reduce predation. In may provide methods of placing chipped or scarified seed in the field to improve germination. Coated pellets can be made by moistening the seed then shaking it in a clay or powdered soil in a tray. Repeated moistening and shaking can develop a thick covering, but a minimum coating is usually desired (Lippitt, 1992). Alternatively seed can be pelletized by spraying the seed with glue while is shaken in a recipro-rotating tray and then sprinkling it with fine dry silica sand (Walters; Geary, 1989).

Extruded pellets can be made by pressing a pasty seed and soil mixture through holes, and compressed pellets can be made by running a seed and soil mixture through pressure disks.

Some pelleting processes have greatly reduced seed germination (Vallentine, 1979). Thick coatings can inhibit root emergence, especially in delicate-rooted plants (Lippitt, 1992). Often, insufficient water in the filed prevents the coating from breaking down and thus keeps the seed from imbibing the water (Lippitt, 1992). Broadcasting pelleted seed from air or by hand has resulted in poor germination and establishment, and increase seeding rates have not overcome this limitation. Furthermore, the cost associated with the pelleting and handling the extra bulk is high. For these reasons, pelleting may not be a cost effective way of preventing seed predation. Considerable work may be needed to develop and effective pelleting process.

Abnormally colored seed can reduce predation by birds (Vallentine, 1979). Unpublished observations by Kathy Truman of the University of California, Riverside's Anthropology department suggested reduce predation of unusually colored (purple) seeds an seedlings of ancient European wheat strains.

It is sometimes necessary to used chemical repellents or poisons. A combination of the fungicide/repellent Arasan 75 and the insecticide/rodenticide edrin plus an adhesive has greatly reduced rodent depredation when applied to seed prior to planting. Baby powder may be a good substitute to more toxic repellents (Lippitt, 1992).

Many Legumes, such as mesquite, depend upon effective nodulation and nitrogen fixation to become successfully established. In order to assure nodulation, legume seed should be treated with a good commercial inoculant prepared from a strain of Rhizobia bacteria specific for the legume being planted (Vallentine, 1979; Green et al., 1984; Phillips ;Williams, 1987). Large quantities of seed can be coated in a cement mixer.


Native seeds are essential for successful revegetation projects. Too insure high quality, mature seed should be collected from healthy, local stands using a sufficiently broad genetic base. Care identification of the site characteristics and tracking seed viability overtime is essential. Cleaning, dewinging, debearding, and upgrading seed before storage can: 1) reduce the weight and bulk of seed, 2)improve the storage life, 3) increase the percent germination, and 4)make production and planting easier.

The seeds of many native plants quickly lost their viability if they are not stored under controlled conditions. These can be very species specific. Proper storage conditions will usually control seed pests, eliminating the need for chemical pesticides. Seeds should be protected from rodents and other pests in storage.

Multiple dormancy is common in the seeds of many desert species and experimentation is often necessary to determine the best way to break seed dormancy. This can be complicated by year to year and plant to plant variation. Seed lots should be tested for viability before storage, periodically during storage, and prior to sowing. For sites with limited infection potential, leguminous plant seed should be inoculated with Rhizobia prior to sowing. Dyeing seeds unnatural colors is a non-toxic way to protect them for predation.

The foundation of a successful revegetation or restoration program is quality seed. This requires careful collection, processing and storage.


Bainbridge, D.A. and P.C. Clark. 1987. Planting mesquite, carob, and similar seeds. Dry Lands Research Institute, University of California, Riverside. 3 p.

Bryant, J.A. 1985. Seed Physiology. Edward Arnold, Baltimore. 76 p.

Campbell, S.J. and T.D. Landis. 1990. Mananging seedborne diseases in western forest nurseries. Tree Planters Notes. 41(4):3-7.

Capon, B., and W.Van Asdall. 1966. Heat pre-treatment as a means of increasing germination of desert annual seeds. Ecology. 48 (2):305-306.

Desert Enterprises. 1992. Personal Communication. P.O. Box 23, Morristown, Arizona 85342.

Dobbs, R.C., D.G.W. Edwards, J. Konishi and D. Wallinger. 1976. Guideline to collecintes of B.C. Conifers. British Columbia Forest Service/ CanadianForest Service, Joint Report No. #. Victoria, B.C. 98 p.

Dumroese, R.K., R.L> James, D.L. Weeny and C.J. Gilligan. 1988. Douglas-fir seed treatments:effects on seed germination and seed-borne organisms. pp. 155IN T.D. Landis, coord. Proceedings, Combined Meeting of the Western Forest Nursery Council and Intermountain Nurseryman's Association. USDA Forest Service Forest and Range Experiment Station RM -167. Fort Collins, CO.

Ehreler, W.L., D.H. Fink, and S.T. Mitchell. 1978. Growth and yield of jojoba plants in native stands using runoff-collecting microcatchments. Agronomy Journal. 70:1005-1009

Environmental Survey Consulting. n.d. The Grin Reaper. 4062 Placid Place. Austin TX 78731.

Evans, M. 192. Seed storage. SER News 5(2):6-7

Frankel, O.H. and E. Bennett. 1967. Genetic Resources in Plants: Their Exploration and Conservation. F.A. Davis & Co. Rome, 554 p.

Goor, Y.A. 1963. Tree Planting Pratices For Arid Zones. Food and Agriculture Organization of the United Nations, Rome, 233 p.

Graves, W.L., B.L. Kay and W.A. Williams. 1975. Seed treatment of Mojave Desert shrubs. Agronomy Journal. 67:773-777.

Green,J.T., Mueller, and D.J. Chambles. 1984. Inoculation of Forage Legumes. North Carolina Exten. Service, AG226, NCSU Raleigh, 15 p.

Hanscom, P.R. 1984. Seed Size in Isomeris arborea Nutt. MS Thesis. Dept, of Biology, San Diego State University, 46 p.

Harrington, B.F. 1988. Storing Seeds. Small Farm News. March/April:1.

Hartman, H.T. and D.E. Kester. 1983. Plant Propagation:Principles and Practices. Prentice Hall, Englewood Cliffs, NJ 727 p.

Jones, S. 1989. The influence of stratification, scarification, hot water and maternal plant on the germination of Juniiperus exclesa seeds from Eritrea. The International Tree Crops Journal. 5: 221-230.

Kay, B.L., C.M. Ross and W.L. Graves. 1977. Creosote Bush. Mojave Restoration Notes 9:10 p.

Kay, B.L. , C.C. Pergler and W.L. Graves. 1984. Storage of seed of Mojave Desert shrubs. Journal of Seed Technology. 9(1):20-23.

Lippitt, L. 1991. CDF Cone Collection Notes. L.A. Moran Reforestation Center, California Department of Forestry and Fire Protection, Davis, CA 16 p>,/p> Lippitt, L. 1992. Manage, L.A. Moran Reforestation Center. Personal Communication

Lippitt, LX. 1992. X-ray analysis of seed from Red Rock Canyon State Park, notes for SDSU/CDP&R 8/92.

Pasiecznik, N. and P. Felker. 1992. Mechanical cleaning of Prosopis seed. Nitrogen Fixing Tree Research Reports. 10:186-188.

Phillips, D.A. and W.A. Williams. 1987. Range Legume Inoculation and Nitrogen Fixation by Root Nodule Bacteria. U.C. Agrig. Exp. Station, Berkeley, CA. 13. p.

Phillips, R.L. 1992. Elevation, on -or off- tree collection influence the viability of blue oak acorns. California Agriculture 46(5):30-32.

Priestly, D.A. 1986. Seed Aging: Implicatons For Seed Storage and Persistence in the Soil. Comstock Publishing Associates, London. 288 p.

Rietveld, W.J. 1989. Variable seed dormancy in Rocky Mountain Juniper. pp. 60-64. In T. Landis, coord. Proceedings, Intermountain Forest Nursery Association, USDA- Forest Service Forest and Range Station, RM-184. Fort Collins, CO.

Schopmeyer, C.S. 1974. Seeds of Woody Plants in the United States. Forest Service, U.S. Department of Agriculture, Washington, D.C. 883 p.

Stein, W.I., R. Danielson, N. Shaw, S. Wolff, & D. Gerdes. 1986. Users Guide for Seeds of Western Trees and Shrubs. Forest Service, U.S. Department of Agriculture, Pacific Northwest Research Station. PNW-193. 45p.

Stidham, N.D. , R.M. Ahring, J. Powell and P.L. Claypool. 1980. Chemical scarification, moist prechilling, and Thiourea effects on germination of 18 shrub Species. Journal of Range Management. 32(2):115-118.

Vallentine, J.F. 1979. Range Development and Improvements. 2nd ed. Brigham Young University Press. Provo, UT. p. 281-289.

Walters, G.A. and T.F. Geary. 1989. Cleaning, pelletizing and sowing Eucalyptus seed. Tree Planters Notes 40(1):27-30.

Young, J.A. and C.G. Young. 1986. Seeds of Wildland Plants. Timber Press, Oregon 236 p.