[Front Page] [Features] [Departments] [SGAP Home Page] [Subscribe]

Australian Plants online

A Study of Eremophila Seed Germination

Paul Rezl

Eighteen species of Eremophila from two different sources were sown on 14 April 1996. Most of them (11) were collected from a property near Mullewa, Western Australia in September/October 1995. These were:

E.clarkeiE.fraseri subsp. galeataE platycalx

Another seven species were obtained from a commercial seed service in February 1996 but the age of the fruits was unknown. These were:

E.cuneifoliaE.laaniiE racemosa

Seed Extraction

It is known that Eremophila seeds are locked in woody fruits and for germination to take place they need to be released. This is, in most cases, quite a laborious process, but nevertheless the most important one. Eremophila fruits exhibit a remarkable diversity, not only in the shape and size (ranging from as little as 2 mm in E.densifolia to 15 mm in E.maculata), but also in the internal structure. Some fruits are corky inside and the seed is quite loose but in many species the inside is more or less woody.

Eremophila racemosa is a species with a very restricted occurrence in nature but which is reasonably well established in cultivation. Select the thumbnail image or plant name for a higher resolution image (26k).

Fruits consist of four chambers and these contain up to 12 seeds according to Richmond and Osborne ("Australian Plants" No. 134) but in practice I have never found more than six fully developed seeds per fruit. Seeds are elongated, approximately 2-6 mm long. They are covered with a thin skin-like testa. To release the seeds from fruits it is necessary to open each of the four seed chambers and that means basically splitting the fruits in halves and quarters. I use a board made of soft wood (pine - any offcuts would do), a light hammer and a Swiss army knife (or similar). Fruits are placed in small dents on the board for better stability and, while holding the edge of a knife somewhere close to the central seam of the fruits, they are split longitudinally. This needs to be done very carefully at first as different species require different force. Some seeds are inevitably damaged during the process.

Sometimes only a part of the seed is exposed and it is good to loosen it up by scraping off the surrounding shell using whatever is handy. Details on seed extraction are summarised in Table 1.

Table 1 - Seed Extraction
SpeciesAvg. No. of Seeds in the FruitStructure of the FruitEase of Seed Extraction
E.densifolia2-4softvery difficult, tiny fruit
E.divaricata0-2semi-hardmedium, seed very small
E.eriocalyx1-4very soft, corkyvery easy
E.fraseri subsp."galeata"1-2hardmedium
E.longifolia1very hardvery difficult
E.maculata4very hardmedium
E.oldfieldii0-1semi hardeasy, low seed count
E.platycalyx4-6soft, brittlemedium
E."spuria"1-5soft, brittleeasy

Seed Sowing

Soil less propagating mixture was used for sowing. It is based on Lignocel (milled coconut shells which are compressed into brick-like blocks) and perlite in the rate of 3:1. I use this universally for all sowings. It is similar to peat/perlite mix but it has better water holding capacity and the structure is more open, doesn't compact easily and unlike peat, absorbs water readily even when dry. As a result, young seedlings root very well in this mixture.

".....the box was placed in a fridge every night and removed in the morning for the period of first two weeks from the date of sowing."

I do not sterilise the mix; it is generally treated with fungicide. In this case a new product, PolyversumTM, was used for the first time. It is a new generation biofungicide that works on the wide spectrum of soil-borne diseases. The active agent is the fungus Pythium oligandrum which is naturally present in small quantities in the soil. This "friendly fungus" lives in symbiosis with the root system of the plants. As stated by the manufacturer, the product is non-chemical and therefore completely harmless. Remarkable is the long storage capacity of 1O years (provided it is kept dry). When preparing the propagating mixture a Lignocel block must be water saturated first and this was done by adding 5 litres of water to the compressed block. In this case 0.05% Polyversum was used in order to inoculate the substrate. Seeds were treated with Polyversum prior to sowing and sown immediately in small containers only lightly covered. Containers were watered by soaking in 0.05% Polyversum and placed in the propagating box with the temperature regime of 25 C day/5 C night in order to simulate the oscillating temperatures inland where most eremophilas grow. As I do not have a growth room or a thermostat, the box was placed in a fridge every night and removed in the morning for the period of first two weeks from the date of sowing. The propagating box was then left on 25 C permanently. Results are summarized in Table 2.

Table 2 - Seed Germination
SpeciesNo. of Fruits UsedNo. of Seeds SownFirst Seedling
No. of Seeds Germinated
E.fraseri subsp."galeata"66212

Results and Observations

0f the total of 132 seeds sown, 49 have germinated and that means 37% germination in total. I have deliberately not counted the germination for each species because the number of seeds sown varied, in some cases being as little as 1.

Generally all species from Mullewa have germinated. They showed much better germination rate (48%) than the seed from the commercial source (16%). Three species did not germinate and they were all from the commercial source.

Germination was rapid. Most of the germination occurred in the second and third week from the date of sowing and after one month I considered it to be complete.

With two seeds of E.maculata which refused to germinate in spite of all efforts, I have decided to remove the testa. It was carefully peeled off and both seeds germinated immediately.

The most commonly grown Eremophila is E.maculata, the "Spotted emu bush". It occurs in a range of colours from deep red through to yellow. Select the thumbnail image or plant name for a higher resolution image (29k).

Pots with seedlings were placed under fluorescent lights with a 14 hour day/l0 hour night cycle. They were growing vigorously as expected and quickly showed some signs of nutrient deficiency. Because the propagating mixture is very low in nutrients, I started to apply regular doses of complete fertiliser and the seedlings responded immediately by strong healthy growth.

At this stage two applications of 0.05% Polyversum were applied to the seedlings (watering) as recommended by the manufacturer.

Seven seedlings were lost before the first repotting, possibly due to an airborne fungal infection, drying out or natural weakness. First repotting was done approximately one month from the date of sowing and most of the seedlings had 2-4 pairs of leaves. Potting mix was based on loam/peat 1:1 which I am using for all of my plants. I have noticed a very healthy, well developed root system in all seedlings.

There were no losses after the first repotting and all of the plants (42) are healthy and growing well at the time of writing.


Seed extraction is the single most effective method of sowing Eremophila seed. Other methods mostly use fruits and not the seed for sowing and therefore are much less effective. Many fruits contain no viable seeds and these are eliminated by the extraction method.

Several germination delay mechanisms other than the physical one (which, in all cases, is the impervious woody fruit) have been observed. These are the effect of temperature (oscillating temperatures promote germination) and the presence of inhibitor in the seed testa. These two mechanisms are not developed in all species and many germinate under the constant temperature. Most seeds will germinate without the removal of testa but in some cases as demonstrated on E.maculata this procedure is very helpful.

Contrary to what has been previously published, fresh seeds germinate better than old.

Treatment of seed, soil and seedlings with Polyversum has proved to be very beneficial. Root systems on the first repotting were well developed. Plants appear to be very compact with short internodes and many side shoots. Polyversum seems to stimulate the growth through the root tissue. I sincerely recommend it for testing with many other Australian native plants.

Sowing of Eremophila seed as described above can be a good alternative to vegetative propagation and can be used for breeding new cultivars and propagation of the rare species.

Reprinted from the August 1996 issue of the newsletter of the Society's Eremophila Study Group.
Note that comparison of the viability between wild and commercial sources of seed was not an objective of this study. Accordingly, the identity of the commercial seed supplier has been deleted from the original article because a) there is no suggestion that seed from any other commercial source would have achieved better results and b) there was no duplication of species between the commercial and Mullewa sources.

[Front Page] [Features] [Departments] [SGAP Home Page] [Subscribe]

Australian Plants online - June 1997
The Society for Growing Australian Plants