English (en) | Change Language
By: Tim Motis
Published: 2023-07-13

This article summarizes ECHO research published in Experimental Agriculture by Trail et al. (2022).



Farmers and those serving them require reliable and ready access to quality seeds. Where resources are scarce, farmers obtain as little as 10% of their seeds from commercial sources, relying on informal seed systems for most of their seeds (Coomes et al., 2015). Informal seed sources include seeds that farmers produce themselves, exchange with other farmers, or purchase from local markets. Small-scale community seed banks play an important role in supporting in-country seed networks and crop diversity. 

1Orthodox seeds tolerate desiccation and can be stored under dry and freezing conditions. Examples of crops that produce orthodox seeds include cereal grains, legumes, and most vegetables.

One way you can expand community access to seeds of diverse crops is to build capacity for small-scale seed saving entities to store orthodox 1 seeds for periods of time longer than annual growing cycles. The ability to store seeds longer than one growing season provides farmers with the option of planting a crop that they may not have grown and saved seeds of in the previous growing season. Additional crop options increase resilience to agricultural challenges such as drought and pests.

Seed storage requirements

2How dry should seeds be to prevent mold? The answer varies with crop and fungi species. Seed storage fungi will generally not be able to grow at seed moisture contents of less than 12% to 14% (Martin et al. 2022). Seed moisture is influenced by the humidity of the surrounding air. To prevent mold growth and stabilize seed moisture at acceptable levels, keep humidity levels below 65%.

Prior to storage, seeds must be dry enough 2to prevent the growth of fungi such as Aspergillus spp. (Reader and Motis, 2017). Reducing seed moisture is particularly important for newly harvested seeds, as they may still have moisture from the field. Drying seeds in preparation for storage can be done with a simple seed drying cabinet, described by Motis (2010) in an ECHO Technical Note entitled ‘Seed Saving: Steps and Technologies’. 

Here we focus on storing seeds that have already been processed to remove field moisture, debris, and pests. Factors that affect seed storage life are seed moisture, temperature, and oxygen levels. Store seeds dry and cool, and under low oxygen. These conditions slow the metabolism of seeds, thereby prolonging storage life. In tropical environments, the main constraints to seed storage are high humidity and heat.

Research purpose

Many smallholder farmers in the tropics do not have electricity or access to equipment for climate-controlled seed storage. There are, however, low-cost technologies that can be used instead. The main objective of this research, therefore, was to test the effectiveness of some of these technologies for keeping seeds dry in a hot and humid environment. Our research is relevant to situations in which you 1) have already processed harvested seeds in preparation for storage or 2) need to store seeds that you have acquired (e.g., from a seed store) that are already dry.

Technologies tested

EDN160 Figure 1

Figure 1. Technologies and ambient humidity and temperature under which seeds of okra, sorghum, and velvet bean were stored for a one-year period in Florida and Thailand. Source: Tim Motis

We explored technologies related to vacuum sealing and desiccants (Figure 1). Vacuum sealing removes air—along with water vapor and oxygen it contains—from a container. With less water vapor, humidity is reduced. Less oxygen not only slows respiration but reduces the activity of any insect pests that may be present. ECHO research in Asia showed that vacuum sealing, over a one-year period, prevented the population growth of bruchids (Callosobruchus maculatus) in cowpea seeds (Lawrence et al., 2017). 

3Some bicycle pumps are easier than others to modify. The Technical Note explains why this is so. Also, at the bottom of the section on bicycle pumps, note instructions for modifying a pump with fewer parts.

We compared two vacuum-sealing technologies: a machine-sealer and a modified bicycle pump. The machine sealer withdraws air from polyethylene (plastic) bags. Of the two technologies, the machine sealer is the most expensive (300 USD or more). It is used primarily for food processing. Bicycle pumps are inexpensive (15 USD or less) and available worldwide. We used a modified bicycle pump to remove air from small (118 ml [4 oz.]) glass jars with plastic-lined, metal lids. An ECHO Technical Note entitled Vacuum-Sealing Options for Storing Seeds (Motis, 2019) explains how to modify and operate a bicycle pump to draw vacuum.3

Desiccants absorb moisture from the air, thus lowering humidity in the air surrounding seeds. We explored the use of two desiccants: Drying Beads® and calcium oxide. Drying Beads® are a zeolite-based product that absorbs water, holding it in microscopic pores. They are manufactured in Thailand by Rhino Research and have been used primarily in Asia. Information on their use is available on the manufacturer’s website (http://www.dryingbeads.org/) which states that beads saturated with moisture can be reused by heating in an oven for two and-a-half hours at 250°C.

4CAUTION This reaction produces heat. This is not a problem when absorbing water vapor in air, but do not allow calcium oxide to mix with water as a liquid. Also, avoid breathing in the dust of calcium oxide powder. In using calcium oxide for keeping seeds dry, place the calcium oxide in an envelope or cloth to avoid direct contact of the seeds with the calcium oxide.

Calcium oxide, also known as burnt lime, is a powder made by subjecting calcium-carbonate-containing materials–like limestone, seashells, or wood ash–to high heat (500 to 1000°C [Teker Ercan, 2003; Suwannasingha et al., 2022]). Combustion temperature influences the extent to which calcium carbonate will convert to calcium oxide. We have been able to achieve temperatures of 600°C and higher with a small forge used by blacksmiths. As calcium oxide absorbs water it converts to calcium hydroxide 4. It is possible to convert calcium hydroxide back to calcium oxide, for reuse, by heating at 450 to (Materic and Smedley, 2011) to 600°C (Powers and Calvo, 2003).

To be effective, vacuum-sealing and desiccant approaches require hermetic (airtight) containers. The entry of air into a poorly sealed container causes loss of vacuum and faster desiccant saturation than with an airtight lid. Thus, we used glass jars with screw-on lids in conjunction with vacuum sealing and desiccants. In seed banks with climate-controlled storage space, desiccants may be removed from seeds once the seeds are dry enough for storage. We kept the desiccant with the seeds being stored, as this would be preferable in situations where ambient humidity cannot be controlled and could otherwise rehydrate the seeds.


When and where

We conducted the trial over a one-year period between 11 July 2017 and 16 July 2018. The study was implemented at two sites: the ECHO Asia Regional Impact Center in Chiang Mai, Thailand and at ECHO in Florida, USA. At each location, seed containers were kept in cardboard boxes outside under the ambient conditions (see Figure 1) of screened-in porches. This allowed us to approximate the context in which many smallholder farmers in the world store their seeds.


EDN160 Figure 2

Figure 2. Containers used in evaluating seed storage treatments. Source: Nate Flood    

Treatments involved the previously mentioned technologies aimed at controlling humidity in seed storage containers. Between temperature and humidity, humidity in containers is easier to control, and we wanted to see if doing so would improve seed viability under ambient temperatures. The treatments we compared are listed in Table 1.

Table 1. Treatments corresponding to seed storage technologies tested in Thailand and Florida.
Treatment Container Type
1. Vacuum-sealing with modified bicycle pump Glass jars (118 ml [4 oz])
2. Vacuum-sealing with machine sealer Polyethylene (plastic) bags
3. Desiccant, using zeolite Drying Beads® Glass jars (207 ml [7 oz])*
4. Desiccant, using calcium oxide powder  Glass jars (207 ml [7 oz])*
5. Nontreated seeds Small paper bags 
*Desiccants were placed inside the glass jars, in small breathable paper bags on top of the seeds to avoid direct contact of desiccants with seeds. They were kept at a seed-to desiccant ratio, by weight, of two to one (60 g of seed with 30 g of desiccant). To compensate for the volume taken up by desiccant, the glass jars for treatments 2 and 3 were larger than those for vacuum sealing.    

Each treatment was tested with seeds of okra (Abelmoschus esculentus), sorghum (Sorghum bicolor), and velvet bean (Mucuna pruriens). For each crop, 60 g of seeds were stored per container. We randomly assigned each container to 1 of the 5 treatments, 1 of 3 seed types, 1 of 5 sampling times, and 1 of 4 replicates, for a total of 300 containers per location (Figure 2).


EDN160 Figure 3

Figure 3. Germination testing of okra seeds in a petri dish. Source: Nate Flood

For each treatment, we sampled seeds at 0 (initial baseline), 1, 3, 6, 9, and 12 months of storage. At each sampling time, we opened the containers assigned to that sampling time for data collection, leaving the remaining containers closed. By taking data from a different batch of seeds each time, we did not have to open and reseal containers over the course of the trial. At each sampling time we measured:

5This meter works by heating the seeds (at a high temperature [at 120°C on the automatic mode]) until they stop losing weight. The machine calculates seed moisture by taking the difference between initial and final weight (initial weight minus final constant weight) and dividing that number by the initial weight. You can do this in an oven if you do not have a moisture meter. Crushing the seeds helps you achieve a constant weight more quickly than if you leave the seeds intact.
  • Seed moisture content: We measured seed moisture content with a 5 g subsample of seeds for each container. Moisture content of crushed seeds was measured with a DSH-50-1 Precision Halogen Moisture Meter.5
  • Seed germination percentage: We tested a subsample of 50 seeds from each container for germination using the Petri dish method (Rao et al., 2006; Figure 3). Petri dishes were placed in a germination cabinet (Seedburo® Model 548 [Florida]; a custom-built cabinet, like the one in Florida, was used in Thailand) that maintained temperature at 29 ± 2°C and humidity at 60 ± 5%.
  • EDN160 Figure 4

    Figure 4. Using a vacuum gauge to measure vacuum pressure. Source: Tim Motis

    Vacuum pressure: Vacuum pressure was collected for treatment-two (bicycle sealing) and treatment-three (machine sealing) containers, just prior to opening the containers. To measure the level of vacuum in each container we used a simple pressure gauge attached to a hypodermic needle (Figure 4). The needle was pushed directly into treatment-one polyethylene bags or through the tape covering the hole in each glass jar lid (through which vacuum was drawn). The aforementioned ECHO Technical Note explains the use of the pressure gauge and gaskets on jar lids. 

For this article we will focus mainly on findings for initial (month 0 baseline) and final (month 12) seed moisture and germination.


Seed moisture

The moisture content of seeds is influenced by the humidity of the surrounding air and the composition of the seed. Oil and water do not mix, but water is attracted to protein and starch (McDonald, 2007). We did not have the seeds analyzed for these components, but values in the literature are shown in Table 2. Initial values shown in Figure 5 indicate baseline seed moisture, just before treatments were set up. As seen in our initial moisture content values, the moisture content of seeds in a seed bank—under similar storage conditions—will vary with crop, which is due at least in part to differences in their composition. Perhaps the combination of high protein and starch explains why initial seed moisture content was highest in velvet bean seeds.

Table 2. Reported values for oil, protein, and starch in seeds of okra, sorghum, and velvet bean
Crop Oil  Protein Starch Source
Okra 14% 19-41% 7-37% Ofori et al. (2020); Omoniyi et al. (2018)
Sorghum 5-8% 4-21% 56-75% Mehmood et al. (2008); Khalid et al. (2022)
Velvet bean 6% 25-29% 39-41% Baby et al. (2022; Omeh et al. (2014)
EDN160 Figure 5

Figure 5. Moisture content of okra, sorghum, and velvet bean seeds as influenced by storage treatments. Data are shown for values measured before treatment (month 0) and after 12 months of storage under ambient conditions under outdoor screened porches. The value shown for each bar is an average of eight observations (four from Florida and four from Thailand).

Table 3 shows general trends for the effect on seed moisture of the various treatments over time from month 1 to 12. Of the two vacuum sealing treatments, only bicycle vacuum kept moisture content from increasing in seeds of all three crops. Both desiccants prevented increases in seed moisture over time. Drying Beads® was the only treatment that decreased seed moisture over time; it did so for all three crops. With no treatment, seed moisture fluctuated over time, with lowest and highest values coinciding with humidity levels during dry and rainy seasons, respectively. The highest moisture content recorded over the trial was 14%, at month 3 with nontreated velvet bean seed.

Table 3. Effect of time on seed moisture and germination for each treatment, as indicated by the statistical significance of time and overall trends over time.
    Statistical significance of time and overall trend over time*
Seed parameter Treatment Okra Sorghum Velvet bean
Moisture Machine vacuum *** (increase) *** (increase) NS (no change)
Moisture Bicycle vacuum NS (no change) NS (no change) NS (no change)
Moisture Zeolite *** (decrease) * (decrease) *** (decrease)
Moisture Calcium oxide NS (no change) NS (no change) NS (no change)
Moisture No treatment *** (fluctuate) *** (fluctuate) *** (fluctuate)
Germination Machine vacuum NS (no change) *** (decrease) NS (no change)
Germination Bicycle vacuum NS (no change) ** (decrease) NS (no change)
Germination Zeolite NS (no change) NS (no change) *** (decrease)
Germination Calcium oxide NS (no change) ** (decrease) NS (no change)
Germination No treatment *** (decrease) *** (decrease) NS (no change)
*The effect of time was statistically nonsignificant (NS) or significant at a P level of 0.05 (*), 0.01 (**) or, 0.001 (***).

Final (month 12) seed moisture values are shown in Figure 5. By the end of the trial, all storage technologies evaluated kept seed moisture content below that with no treatment. This was true for all crops. Storage technologies differed, however, in the extent to which they kept seeds dry. Only zeolite Drying Beads® reduced seed moisture below initial values. It brought seed moisture content levels down to less than 5% for all crops. Calcium oxide and bicycle vacuum sealing kept seed moisture values close to baseline values. Among the storage technologies tested, final seed moisture values were highest with machine sealing.

Seed germination

All seeds tested germinated well at the beginning (month 0) of the trial, with germination percentages of 88-96% (Figure 6). Between months 1 and 12, the germination percentage of okra and velvet bean remained stable with most of the treatments (Table 3). Sorghum germination declined over time with all treatments except zeolite Drying Beads®. Zeolite Drying Beads®, however, led to a rapid decline of velvet bean seed germination over time. With no treatment, we saw a significant decline in the germination of okra and sorghum but not velvet bean seeds. 

At the end of the trial, germination with no treatment was 34% for okra and 0% for sorghum (Figure 6). Interestingly, velvet bean seed proved very tolerant of ambient humidity and temperature, with 96% of seeds germinating at month 12. Vacuum sealing (both machine and bicycle) and calcium oxide resulted in final seed germination percentages of 78-83% for okra and 96-100% for velvet bean. With zeolite Drying Beads®, final germination percentages were highest with sorghum. Only zeolite Drying Beads® kept final sorghum germination above 70%.

EDN160 Figure 6

Figure 6. Germination percentage of okra, sorghum, and velvet bean seeds as influenced by storage treatments. Data are shown for values measured before treatment (month 0) and after 12 months of storage under ambient conditions under screened porches. The value shown for each bar is an average of eight observations (four from Florida and four from Thailand).

Vacuum pressure

Vacuum pressure (not shown) was steadily lost over time with the polyethylene bags used with the machine sealer. Conversely, glass jars used in conjunction with bicycle pump sealing prevented a decline in vacuum. Neither of the vacuum sealing technologies evacuated all of the air from respective containers.


Tolerance to ambient conditions varies with crop. Our findings showed that velvet bean seeds tolerated high heat and humidity much better than sorghum seeds. This shows that crops can vary significantly in their tolerance of unfavorable storage conditions.
Of the two vacuum sealing treatments, vacuum drawn on glass jars with a modified bicycle pump proved to be as or more effective in maintaining seed moisture content in storage than a much more expensive machine sealer. Our findings with vacuum sealing, in combination with previous findings from research at ECHO’s Regional Impact Center in Asia, have the following implications:

  • It is possible to store seeds well with modest levels of vacuum. The modified bicycle pump withdrew about 50% of the air from glass containers. Though the machine sealer was capable of a stronger initial vacuum, the polyethylene bags commonly used with it were not ideal for maintaining vacuum over a long period of time. 
  • Vacuum sealing can be implemented with very inexpensive technology. The cost of the pump is the main expense. If bicycle pumps in your area are difficult to modify for drawing vacuum, consider other approaches like brake bleeder pumps and syringes, as explained in ECHO Technical Note #93
  • Success with vacuum sealing requires a container that keeps air from diffusing in and out and that is well-sealed. If using glass jars, make sure the lids provide a good seal.

Both desiccants proved effective in maintaining baseline seed moisture content, but only the zeolite Drying Beads® reduced seed moisture over time. Calcium oxide is more widely available (as burnt lime) than zeolite Drying Beads® but takes more heat to reuse—after saturation with moisture—than zeolite Drying Beads®. Zeolite Drying Beads®, at the ratio used in the trial, dried seeds to the point where final germination percentages were adversely affected. Findings with desiccants have the following implications:

  • Seeds can be kept dry with both calcium oxide and zeolite Drying Beads®.
  • It is possible to achieve ultra-dry (below 5%) seed moisture with zeolite Drying Beads®. With respect to germination of seeds exposed to water immediately after storage, as would occur with seeds planted soon after storage, tolerance to ultra-dry conditions varied between crops. Ultra-dry storage can extend the storage life of sorghum, but seed banking personnel may want to experiment before implementing it with a wide variety of crops.
  • Seed savers and seed banking personnel may want to experiment with the ratio of seeds to desiccant to come up with best practices for the crops they work with. With respect to zeolite Drying Beads®, the manufacturers website (www.dryingbeads.org) explains how to tailor the seed-to-bead ratio for specific crops.


You can keep seeds dry for at least a year with low-cost technologies. Vacuum sealing with a modified bicycle pump and calcium oxide are the two least expensive technologies tested in this trial. Zeolite Drying Beads® are also an inexpensive option that is extremely effective in keeping seeds dry. 

Results of this trial are relevant to situations in which temperature is not controlled. Under high and fluctuating temperatures, it is possible to store seeds of some crops—without significant reductions in germination—by excluding humidity. Some seeds (e.g., sorghum in this trial) lose viability faster than others and may require drier conditions to slow the rate of loss in viability.

The authors suggest that future researchers test seed viability over time with varying levels of initial vacuum pressure. Such knowledge would be useful in documenting the effectiveness of a broad range of vacuum devices, including modified syringes less likely to remove as much air as a modified bicycle pump. Combining technologies for keeping seeds dry with those that keep seeds cool and/or stable is another promising area of study. An example of a technology that minimizes temperature fluctuation is earthbag construction, written about in an ECHO Technical Note entitled “Earthbag Seed Banks.“ If you have experience storing seeds under challenging conditions, let us know what you have learned.


Baby C., S. Kaur, J. Singh, and R. Prasad. 2022. Velvet bean (Mucuna pruriens): a sustainable protein source for tomorrow. Legume Science e178. 

Coomes O.T., McGuire S.J., Garine E., Caillon S., McKey D., Demeulenaere E., Jarvis D., Aistara G., Barnaud A., Clouvel P., Emperaire L., Loufi S., Martin P., Massol F., Pautasso M., Violon C. and Wencelius J. 2015. Farmer seed networks make a limited contribution to agriculture? Four common misconceptions. Food Policy 56, 41–50. doi: 10.1016/j.foodpol.2015.07.008

Feed the Future Innovation Lab (FFIL). 2017. Drying beads save high quality seeds. https://horticulture.ucdavis.edu/information/drying-beads-save-high-quality-seeds.

Khalid, W. A. Ali, M.S. Arshad, F. Afzal, R. Akram, A. Siddeeg, S. Kousar, M.A. Rahim, A. Aziz, Z. Maqbool, and A. Saeed. 2022. Nutrients and bioactive compounds of Sorghum bicolor L. used to prepare functional foods: a review on the efficacy against different chronic disorders. International Journal of Food Properties 25(1):1045-1062.

Lawrence, B., A. Bicksler, and K. Duncan. 2017. Local treatments and vacuum sealing as novel control strategies for stored seed pests in the tropics. Agronomy for Sustainable Development 37:6. DOI 10.1007/s13593-017-0415-0

Martin, I., L. Gálvez, L. Guasch, and D. Palmero. 2022. Fungal pathogens and seed storage in the dry state. Plants 11(22): 3167, https://doi.org/10.3390/plants11223167

Materic, V. and S.I. Smedley. 2011. High temperature carbonation of Ca(OH)2. Industrial & Engineering Chemistry Research 50(10):5927-5932.

McDonald, M. B. 2007. Seed moisture and the equilibrium seed moisture content curve. Seed Technology 29(1):7-18. 

Mehmood, S., I. Orhan, Z. Ahsan, S. Aslan, and M. Gulfraz. 2008. Fatty acid composition of seed oil of different Sorghum bicolor varieties. Food Chemistry 109(4):855-859.

Motis, T. 2010. Seed Saving. ECHO Technical Note no. 63.

Motis T. 2019. Vacuum-sealing options for storing seeds. ECHO Technical Note no. 93.

Ofori, J., C. Tortoe, and J.K. Agbenorhevi. 2020. Physiochemical and functional properties of dried okra (Abelmoschus esculentus L.) seed flour. Food Science and Nutrition 8:4291-4296.

Omeh, Y.N., D. Akachukwu, and O.U. Njoku. 2014. Physiochemical properties of Mucuna pruriens seed oil (MPSO), and the toxicological effects of a MPSO-based diet. Biokemistri 26(3):88-93. 

Omoniyi, S.A., M. A. Idowu, and A.A. Adeola. 2018. Potential domestic and industrial utilizations of okra seed: a review. Annals. Food Science and Technology. 19(4):722-730.

Powers, T.H. and W.J. Calvo. 2003. Moisture regulation. In Ahvenainen R. (ed), Woodland Publishing Series in Novel Food Packaging Techniques. Cambridge: Woodhead Publishing, pp. 172-185.

Reader, S. and T. Motis. 2017. Are my seeds dry enough to store? ECHO Development Notes no. 36.

Rao N.K., J. Hanson, M.E. Dulloo, K. Ghosh, D. Nowell, and M. Larinde. 2006. Manual of seed handling in genebanks. In: Handbooks for Genebanks No. 8. Rome: Bioversity International, pp. 28-29. 

Suwannasingha, N., A. Kantavong, S. Turnkiijjanukij, C. Aenglong, H-B. Liu, and W. Klaypradit. Effect of calcination temperature on structure and characteristics of calcium oxide powder derived from marine shell waste. Journal of Saudi Chemical Society 26(2):101441. 

Trail, P., T. Motis, S. Swartz, and A. Bicksler. 2022. Low-cost seed storage technologies for development impact of small-scale seed saving entities in tropical climates. Experimental Agriculture 57(5-6):324-337.