The 100 hectare Yarling Nature Reserve  (36 kilometres east of Narrogin, 7 kilometres west of Wickepin, originally a mallet reserve), is a rarely visited patch of bush. Turn south from Williams Kondinin road on to Watt Road to access the reserve. The view from the road is mostly uninviting due to dense woodland, which hasn't been burnt for decades. This is not a place for the average tourist as there are no trails or facilities, and it is difficult to walk through some areas.
​However, it is an interesting spot with an attractive and wildflower-rich kwongan sand valley on the southern edge.

This area is part of a major watershed, which separates the Hotham, Arthur and Avon River catchments. It contains remnants of an ancient undulating lateritic plain, which has been eroded by river systems to the underlying granite. The millions of years old  plain can be likened to a layer-cake, with the top layer of gravel, sand, or ironstone overlying a pale/mottled clay (pallid zone) over decomposing granite or dolerite.
Like most reserves in the district, Yarling is mostly agriculturally poor laterite or rock. A line of breakaways separate a remnant of the ancient  plain from more recent soils on underlying layers to the north and west.

The northern breakaway roughly follows a dolerite dyke, which is is associated with the Binneringie Dyke further to the south. This is because the dyke was more resistant to erosion from the granite contact line being 'cooked' into more resistant rock, and stony laterite formed on dolerite being more erosion-resistant than that formed on granite.

Much of the sheoak forest is choked by dead saplings, which germinated after a wildfire over 50 years ago, and have slowly died off due to competition without any other fires to remove the dead litter. The shrub understorey has become impoverished because shrubs that depend on fire for germination have died.

A few patches of open Red Morrel woodland occur on dolerite brown loam.

Brrown Mallet whipstick woodland lines the breakaway slopes and edges and other areas where the pallid zone clay layer below the gravel is exposed. The mallets are mostly a uniform size due to mass regrowth after the fire over 50 years ago. Since then the stand has been very slowly thinning in a fairly monotonous landscape.

The ancient lateritic plain on the south eastern side contains lateritic stony and impoverished soils. There are a few stony dryandra ridges and an gentle slope of open Silver Mallet Eucalyptus falcata woodland interspersed with brown mallets. Silver Mallet is a beautiful tree, which can be confused with Salmon Gums. However, Silver Mallet only grows on lateritic gravel soils, and Salmon Gum occurs on clay slopes and valleys.
​
on the southern edge there is group of rare Brown Mallet / Red Morrel hybrids.

Species-rich winter-moist kwongan sandplain on a very broad sandy valley on the southern side contains an a wide range of unusual wildflowers, with varying flowering times from August to November. The lower end is mainly dense Eremaea pauciflora shrubland crossed by numerous kangaroo tracks. From September to November I noticed  plants flowering with the best in late October. The easiest way to reach this area is to walk east from the patch of Red Morrel trees on the western side, but it is not easy walking.

as one walks up the valley Eremaea pauciflora gives way to other species such as Calothamnus quadrifidus and Verticordias on yellow sand then Callitris pines.

A triangle on the south eastern edge was illegally cleared, which is very sad as most of it was wonderful kwongan sandy shrubland of little agricultural value. It was replanted some years later in 1996 to wandoos and understory shrubs. 
The wandoos and shrubs have grown well, apart from the deep white sand area. However after 27 years there are still extensive bare areas and very little recruitment of new species from the adjacent uncleared area.
​Our native bush sandy soils are very sensitive, and recover extremely slowly once organic matter, seeds and native microbes has been reduced. 

Comesperma is an Australian genus (mainly in WA), which belongs to the Polygalaceae  (Milkwort) family that is widespread in other countries. The name Comesperma means 'hairy seed", which  relates to the seeds bearing tufts of hair. They are climbers or small wiry shrubs (that are often grazed by kangaroos). I am intrigued by their small 'upside down pea' flowers: an evolutionary adaption to attract native bees that pollinate adjoining pea plants.

On closer inspection, it is obvious that Comesperma flowers are very different from peas. 

When a bee lands on a pea flower, its legs part the wing petals to reveal the underlying stamens and pistil.
​There is little internet information on Comesperma pollination apart from them being pollinated by native bees, and that the flowers use secondary pollen presentation.
Comesperma flowers are quite strange. Two of the five sepals comprise the coloured 'wings' (like an upside-down Donkey orchid - another pea flower mimic), and the petals form a pouch-like 'keel' enclosing the stamens and pistil) with two upright  petals at the end.

On close inspection the story gets even stranger. 
White UV fluorescent patterns lead the pollinator to a small hole between the keel and outer petals. What could be inside?

It was no mean effort cutting into the tiny keel without destroying its contents, so the following images are a bit ragged.
First Comesperma scoparia.
The style appears to grow up in a curve, pushing the cup-shaped stigma into anthers which are crowded at the top of the keel. This loads pollen into the cup shaped stigma ready for contact with a pollinator's tongue which presumably is poked into the entrance. The last image shows that the style had straightened out during dissection. Blowed if I know, but perhaps the style is held under tension inside the keel. Could it poke out after contact with the insect's tongue? The active part of the stigma is a patch on the side of the style underneath the stigma cup.

My attempt as dissecting Comesperma volubile shows that the stamens are attached as a band to the interior of the keel.
My next challenge is to see them being pollinated.

​Curious little plants!

False Boronia, Lysiandra calycina (fomerly Phyllanthus calycinus) is an unobtrusive little plant with an interesting story. Found on sandy gravel soils in Narrogin, it belongs to the family Phyllanthaceae, which  mostly occurs in subtropical areas. The plant has soft green leaves and white flowers with pinkish green shades, which dangle down from the stem on long  flower stalks (pedicels). Flowers have no petals. The apparent petals are sepals, hence the name Leaf Flower. 

​Now for the interesting part

Each plant has male and female flowers (monoecious). Females flowers are larger and intermixed with more numerous male flowers.

​I have never seen any pollinators approach the flowers. This is probably because the flowers have no scent, but they do have large nectar glands at the base of each flower.

Other Phyllanthaceae species are pollinated by insects which partially parasitise them.
One pollinator is a moth which fed on nectar, transmitted pollen to the stigma and laid an egg. When the grub hatched it burrowed down the style to the ovary, ate ovules and pupated to complete its life cycle. As the moth left some ovules, to develop seeds, This is a win win for plant and moth. It also occurs in Boronias.
For other species a tiny gall midge fed on nectar at the spongy floral disc of male flowers and laid eggs in male flower buds, picking up pollen and contacting the styles of nearby female flowers in the process. Infested buds developed into sterile galls, within which midge larvae completed their development. 

​Nature is amazing!

Donkey Orchids (Orchidaceae Diuris species) are only distantly related to pea flowers (Fabiaceae), do not produce nectar reward, and at first glance look quite dissimilar. However, bees' vision differs from humans', and Diuris  flowers have evolved to be great pea flower mimics.
There are a number of Diuris species In this district, but several are difficult for average person to separate. For simplicity I put them into three groups.

1. Donkey orchids (most common)  D. brachyscapa Western Wheatbelt Donkey Orchid, D. corymbosa Common Donkey Orchid, Diuris porrifolia, D. setaceae Bristly Donkey Orchid (at Newman Block).
2. Bee Orchids D. decrementum Bee Orchid, D. laxiflora Common Bee orchid.
3. Purple Pansy Orchid (Tarwonga Road) D. longifolia.

Donkey orchids are less common than pea flowers in the bush, so to attract bees to them from a distance (up to 8 metres), they developed larger flowers and waving bright and reflective 'donkey-ear' petals.
​The researchers measured equal numbers of Tricocolletes and honey bees visiting Diuris brumalis in the Perth hills, and departing with the same number of orchid pollinea on their head. However honey bees were much less effective in transferring pollinea to the orchids to pollinate them This is another example of possible negative effects of honeybees on biodiversity.

​Further Reading: Footprints in the Pollen

Williams Nature Reserve is a high hill on 230 hectares of bushland vested in the DBCA approximately 6km by road north of Williams. To get there turn off Albany Highway at Mckenzie Road and follow signs to the parking area. Facilities include picnic tables, an information bay, a 1.5km walk trail, and a lookout. There is no toilet. Fires and dogs are not permitted.
The Grade 3 walk trail suits moderately fit walkers as it has steep sections, and low ironstone outcrops adjoining the breakaway. The best time to visit is from June and November with peak flowering from August to October.

The Echidna walk trail leads to a breakaway lookout then loops down from the ironstone plateau through underlying rock/soil layers, which can be recognised by characteristic vegetation.

• Jarrah (Eucalyptus marginata) / Powderbark (E. accedens) woodland on ironstone ridges and gravelly upper slopes.
• Powderbark open woodland with almost no understorey on underlying 'pallid zone' pale to hard setting reddish sand / loam soil.
• Wandoo - Rock Sheoak (Allocasuarina huegeliana) woodland on weathered bedrock soil.
• Rock Sheoak, Wandoo (sandy surfaced), York gum (Eucalyptus loxophleba, loam) on flatter surrounding farmland soils.​

Unlike nearby reserves, there is little Brown Mallet (E. astringens - breakaway slopes) or Marri (Corymbia calophylla - deep sand, sandy gravel soil) here on the upland.

Echidna Walk Trail 1.5km
​( approximately 45 minutes to 1 hour)
​
1. Walk upslope on the Jarrah  ironstone mesa to a lookout at a breakaway.

​The trail continues north near the breakaway edge. This section is very uneven with ironstone outcrops on the trail.

2. The trail turns downhill  to Powderbark woodland on reddish pallid zone soil below the ironstone gravel layer.

3. Further downhill  the soil changes to to open Wandoo woodland on reddish soil overlying bedrock

4. Further downhill the soil becomes rockier as you enter Rock Sheoak woodland with winter ferns and low scrub and  annual herbs. Many annuals are weeds, but I found a Blue Lady Orchid Thelymitra crinita flower one October.

5. The trail becomes rocky and weaves through small wandoos and Calothamnus quadrifidus bushes as it crosses a small gully and ascends the other side.

6. The trail gently climbs an upland saddle/valley with vegetation that varies from Powderbark and Wandoo overlooking the Calothamnus gully, and then below the lookout plateau.


7. The trail becomes sandy as it weaves through granite rocks under rock sheoaks.
Check for orchids here in August- September.

8. Vegetation becomes sparse as you ascend through  powderbarks on red brown 'pallid zone' soil to the gravelly plateau.
​
9. The last stage of the trail is through Jarrah / Powderbark woodland with a dense understorey of Pingle Banksia/Dryandra squarrosa.