1)  Flowers of Acianthus sinclairii (cover picture) remained open for six weeks in a vase in my study before shrivelling. Is A. sinclairii capable of selfing? If not it would be something of an exception among NZ native orchids.

Structurally it is believable: the pollinia in the anther cap are separated from the downward-facing stigma by a prominent double rostellar shelf, and appear to be attached to the back of the rostellum, which may be a viscidium. Bruce Irwin 's drawings show this clearly (see below). Pollination did not occur.

In my shrivelled flowers, the pollinia were held secure in a 3-lobed box made up of the two sides of the anther cap and the hardened stigma.

2)  Anther cap retention is said to promote cross-fertilisation.

Many flowers of Gastrodia "long column"  plants I saw in the southern Wairarapa had their anther cap and pollinia missing. In those flowers which still retained them, the anther cap was easily dislodged, containing its pollinia.

Peter and Johnson [1] examined anther cap retention in South Africa, and reviewed some of the adaptations orchids use to reduce the likelihood of self-pollination. Three of them use delay to reduce the availability of pollinia to the flower, or plant, they came from ....

(i) One adaptation is the change in orientation of  pollinaria following removal. Freshly removed pollinaria are oriented so that if the pollinator immediately revisited the flower, or other flowers on the inflorescence, the pollinia could not be deposited on the stigma. It is only after the caudicles or stipes change their shape after a specific interval that the pollinia are correctly oriented to strike the stigma.

Darwin observed that process and illustrated it for Orchis mascula.

The pollinarium of Orchis mascula.

A. when first removed;
B. after a period.

From Darwin C. Fertilisation of Orchids, 1904.

(ii) Pollinia of two species of Bulbophyllum have to dry and shrink over about 2 hours before they can be inserted into the stigmatic cavity.

(iii) A third mechanism involves the retention of the anther cap - anther tissue that covers the pollinarium while it is still in place on the column. In most species, the anther cap immediately drops off the pollinarium following its removal by a pollinator, but in a few species the anther cap clasps the pollinia, making them unavailable for a period, before the anther cap drops off. This delay may work to reduce the chance of insect-assisted selfing.

A number of researchers have measured anther cap retention times: eg, about 20 min in a number of species of Catasetum; 2 - 3 h in Cycnoches lehmanni; up to 40 min in Pleurothallis luteola; up to 30 min in Pleurothallis teres and P. ochreata. These last 3 are small South American orchids pollinated by small flies which spend a lot of  time visiting individual flowers. Anther cap retention is a mechanism to limit assisted selfing by insects that spend considerable time visiting an individual plant.

Catling and Catling [2] recorded anther cap retention times of between 8 and 110 min in Tipularia discolor, and noted that the two flaps of the anther cap that envelope the pollinia gradually open until the anther cap no longer grasps the pollinia and so falls off. The anther cap in T. discolor is small enough to fit into the stigmatic cavity. However, the differing microrelief of the pollinia compared with that of the anther cap effectively limits deposition in comparison with those pollinaria without anther caps.

Peter and Johnson observed that the eastern South African orchid Eulophia foliosa is pollinated almost exclusively by click-beetles, which remained on the flowers for on average 301 seconds. The anther cap was retained for an average of 512 sec after pollinarium removal by beetles. In this orchid the anther cap is bigger than the stigmatic cavity, thus preventing the deposition of pollinia until after the anther cap has dropped.

Thus retention of the anther cap for a duration longer than the beetle 's visit appears effectively to prevent insect-assisted self-pollination in E. foliosa.

What about Gastrodia "long column" this summer? I suspect insects (attracted to that freesia fragrance) spend a good deal of time at the plants, that the pollinia are removed with the anther caps intact (a membrane extends partly over the pollinia from the midline of the column - see Bruce Irwin 's drawing below), that the combined size of the insect and cap/pollinia bundle is too big to enter the tubular flowers (past a large rostellar shelf) and reach the stigma deep in the base, so the disappointed insect flies around until the cap comes off, by which time it is close to a different plant, and can effect cross-pollination.

Fall-back self pollination should be achievable by the flowers turning upward after a time, the pollinia loosening from the anther cap and falling onto the stigma. In populations I have seen fruit is very rarely set, bare stems being the norm, so these devices seem rather ineffective.

On the other hand Eric Scanlen tells me in Invercargill it drops its seed capsules when the weather is too warm, whereas plants in deep shade set ample seed.

3)  More on pollination by fungus gnats: Goldblatt et al from St Louis [3], and Okuyama et al from Kyoto [4] studied saxifrages. A fungus gnat is the primary pollinator of Tolmiea species; pollen is transferred to ventral portions of the gnat's thorax while it probes for an unusually dilute nectar produced at the base of the floral tube. When it forages for nectar on a female phase flower, grains are transferred to the two receptive stigmas on each pistil. Fungus gnats were the principal pollinators of four Mitella species; they landed on the flowers, and pollen grains were attached to their bodies.

Ackerman and Mesler [5] looked at Listera cordata (the lesser twayblade orchid) in northern California. When an insect contacts the touch-sensitive trigger hairs of the rostellum, the pollinia are suddenly released and become cemented to its body. Visitors are attracted to the flowers by their foul odour and minute quantities of nectar presented on the surface of the labellum. The most abundant pollinators of L. cordata were fungus gnats.

Blanco and Barboza [6] studied the orchid Lepanthes glicensteinii. Sexually aroused males of a fungus gnat were the only visitors and pollinators. The initial long-distance attractant seemed to be olfactory. On finding a flower, the fly curled his abdomen under the labellum and grabbed the appendix with his genitalic claspers, then dismounted the flower and turned around to face away from it. The pollinarium became attached to his abdomen during this pivoting manoeuvre. Pollinia were deposited on the stigma during a subsequent flower visit. The flies appeared to ejaculate during pseudocopulation.

Thus the ubiquitous fungus gnat seeks to gratify his sweet tooth as well as his sexual appetite from different orchids.

4)  Molloybas cryptanthus is said to be predominantly self pollinating, and intuitively that would appear so - the flower is rarely seen above the leaf litter, and should be inaccessible to insects.

But that does not take into account the life cycle of the fungus gnat.  The adult fungus gnat lays its eggs singly or in clusters, in rotting leaf litter. In about 4 days, the eggs hatch into tiny, silvery white larvae. Within 2 weeks the larvae begin to spin tiny cocoons in the litter. The pupal stage lasts a week. Adult fungus gnats emerge and mate. The adult stage lasts only another week, but the female will lay around 100-150 eggs during that time. The entire life cycle lasts about 4 weeks.

When the adults hatch they are ready to mate, a traditional season for orchids to seduce the males into pseudocopulation.

Have a look at Bruce Irwin 's drawing of Molloybas cryptanthus below. The flower had been pollinated and he found the decomposing body of what appears to have been an adult fungus gnat stuck to the stigma.

Furthermore the stigma appears to be well-protected from self pollination by being deeply recessed and facing down and forward.

OK, now think about how you might prove insect pollination in a buried orchid flower ....

You can't say with any certainty that the insect was the pollinator, but what you can say is insect pollination by fungus gnats is possible (adult gnats hatch from cocoons in the leaf litter, and an adult gnat has been found stuck to a stigma), and that self pollination would be difficult (the structural relationship between anther and stigma creates a significant barrier).

5)  Some terms used by those who write about pollination may need explanation
      (after van der Pijl & Dodson [7]).

• The anther is a cap-like structure at the apex of the column: it contains the pollinia.
• The pollinia consist of bundles of pollen grains more or less bound together by viscin threads; a small amount of pollen with viscin strands is elongated to form a caudicle which attaches the pollinia to each other or to a viscidium (see below).
• The stigma is a shallow depression composed of the three stigmatic lobes of monocotyledons. It produces a sticky sweet nutrient solution for the pollen grains.
• In most orchids part of the third stigmatic lobe is sterile and specialised and forms the rostellum. In many orchids the rostellum forms a flap between the anther and the stigma proper. There it may form a barrier to self pollination, may deposit sticky liquid on a visitor, which then brushes the pollen which sticks to its body, or may be further specialised so the caudicles of the pollinia are already attached to the rostellum and a part of it comes off as a sticky pad called a viscidium.
• In some orchids a strap of nonsticky tissue (the stipe) connects the pollinia, via the caudicles, to the viscidium.
• The complex of pollinia, caudicles, stipe and viscidium is called the pollinarium.