The collection of high quality, fully ripened seeds from wild plant populations requires careful, on-the-spot decision making. Seed collectors need good judgment, training and experience, and in many cases an intimate knowledge of the plant species.
A seed collection made for conservation purposes is far more than just a packet of dry seeds in a freezer. The value of any seed collection relies on information regarding which plant species they came from, where the collection was made and at what time of year. This kind of information is vital for any potential use of the seeds in future years, therefore careful data recording in the field, at the time of collection is extremely important.
Once it is decided that a plant population is suitable to make a conservation seed collection from, a set of herbarium specimens are required. This is crucial for the species to be positively identified. A herbarium specimen is usually a pressed and dried adult plant specimen. In seed collecting this is known as a voucher specimen and is forever associated with that seed collection. A voucher specimen ensures that if any future taxonomic revision takes place, the voucher specimen can be re-examined to determine the exact taxon of the seed collection.
For example consider a collection of a poorly understood species held in a seed bank. Several years after collection the “species” is recognised as being two very similar species. At this point the voucher specimen can be reviewed to determine whether a name change for the seed collection is necessary.
All material collected by the ‘SeedSafe’ project is verified by the Tasmanian Herbarium. Currently ‘SeedSafe’ collects three voucher specimens for every seed collection made; one for the Tasmanian Herbarium, one for the Kew Herbarium (the MSBP) in the UK and the third is sent to an institution working on the taxonomic group.
Whenever a seed collection is made, important field data is collected regarding the plant species, the location of collection, the habitat, the plant population size and the number of plants collected from. This meta-data is recorded on a collecting sheet (fig. 2), which is kept along with the seed collection. All this information is also stored on a database. Without this meta-data the seed collection is of very little use.
To collect or not to collect; that is the question.
One of the biggest logistical challenges for a seed collector is ensuring that they are in the right place at the right time to collect seeds. It is well understood that the seeds of most plants reach their peak of maturity, viability and potential longevity at the point of natural dispersal. Collecting too early or too late can result in collections with poor long-term storage potential. Therefore collecting seeds at the peak of ripeness is one of the most fundamental aspects of seed collecting.
Indications of seed maturation include:
- Seeds become loose in capsules and capsules start to open.
- Berries and other ‘wet’ fruits take on distinct colours and become easier to detach from the plant.
- The seed contents are usually firm and uniform in colour.
- Seeds start naturally dispersing from the plant.
These indications, and many others, are important for a seed collector. Not only do they help to determine when to collect but they are also useful in determining a method for collection.
Maximising seed longevity
Figure 3 details the developmental phases of seeds leading up to dispersal. You can see that the potential longevity (the ability of seeds to survive for a long time) is at its peak for only a short period of time coinciding with the dispersal stage. This is the optimum time to collect. After dispersal, natural fluctuations in humidity and temperature will begin to age seeds, reducing their longevity.
Under certain circumstances it is possible to collect seeds a few weeks prematurely and allow post-harvest maturation to take place ex situ in a humidity controlled chamber. This technique however is far from perfect and in most cases some reduction in seed viability and longevity should be expected. Nevertheless this technique can be hugely valuable when access to fruiting material is extremely difficult and collecting opportunities are rare.
How many to collect?
For the purposes of long-term seed conservation, seed collections must be relatively large in order to adequately represent the genetic diversity of the local population.
A good conservation seed collection should contain between 10,000 to 20,000 viable seeds. Collections should also be sampled evenly from 50 or more individuals to ensure a thorough genetic representation of the population.
However, it is very important that harvesting does not impact heavily on the natural breeding strategy of the plant population. This is particularly important if the species is rare or threatened.
Therefore as a general rule it is recommended that no more than 20%of the available ripe seed be collected from any one population in any one season.
When confronted with a plant in fruit it is very important to ascertain how much seed is available:
- How big is the population?
- How many seeds per fruit?
- How much full seed is being produced?
- Is the population pure?
In many cases when these questions are asked it will be determined that there is little use in collecting the seed.
For example, consider seed collection of the grassland paper daisy Leucochrysum albicans (Asteraceae, fig. 4). This species is endangered within Tasmania and an ‘obligate seeder’ i.e. it relies heavily on seed production to rejuvenate its population for the following year. It was therefore vital when collecting L. albicans seeds that at least 80 to 90% of the available seed was left in situ.
Other plants, such as Acacia species (Mimosaceae), work up large soil seed banks containing physically dormant seeds that can survive for decades under the soil surface before germinating. In such cases the seed crop of any one year is not nearly so important to the population.
Checking seed viability in the field
Before making a collection it is necessary and relatively easy to assess the viability of the seeds by performing a simple cut-test in the field. Viability is determined by cutting the seed open and examining the contents. In particular, the seed embryo must appear intact and healthy and the endosperm must appear clean and firm (fig. 5), which can be confirmed through seed viability testing.
How to collect
The best collecting method should result in a collection with the highest quality seed available and the minimum amount of debris. Seeds of many species ripen over a period of time and there are often both mature and immature seeds present on the same plant at the same time. Collecting the ripe seeds whilst avoiding the immature ones will greatly enhance the final quality of the collection.
Some practical seed collection methods include:
- Striping – A technique used for grasses, but also useful for plants with small fleshy fruits. When grass seeds mature they separate more readily from the parent plant. By running the infrutescence through the hand or fingers whilst applying gentle pressure, mature seeds can be harvested and immature seeds left behind.
- Pruning – For many species this is the most practical way to collect seeds. Secateurs or in some cases pole pruners can be used to prune off small branches which have many fruit attached. This technique is commonly used for species with dry capsules. On drying out, the seeds are shed and the branches can just be shaken and removed leaving the seed and associated structures. This approach can also be used for tough stemmed species with dry dehiscent and/or fluffy fruits.
- Picking – Similar to pruning but without the secateurs, this technique is useful for larger fleshy or dry fruit and the larger herbaceous daisy flowers.
- Vacuuming – a small hand held rechargeable vacuum cleaner can be incredibly useful for collecting loose seeds from small prostrate species, reducing collecting time from many hours to a few minutes (fig. 7). It also however delivers a fine soil collection and an interesting entomological collection; as do many other methods.