It's early spring, and a patch of endangered shrubs in the Blue Mountains of New South Wales has started to bloom.
Crouching among the plants is a group of scientists on a delicate mission.
Armed with tweezers, the scientists "emasculate" small pink and white flowers by removing the male reproductive organ called an anther.
They're after specks of pollen to hand-pollinate plants that are struggling to reproduce.
They're collecting sperm in the hope of performing the plant equivalent of artificial insemination.
Zieria covenyi blooms from September to December.
It has only two known populations in the world, both in the Blue Mountains National Park.
While the shrub is at risk from climate change, fertility issues are another threat to its survival.
Zieria covenyi flowers have both male and female parts: the orange-coloured male anther holds pollen, and the female stigma at the centre receives pollen carried from other flowers by pollinators.
If all goes to plan, the shrubs will produce fruit and seed that eventually becomes a plant with its own unique genetics.
While some plants at the national park's Breakfast Creek have fertile pollen, scientists suspect another population some 6 kilometres away at Narrow Neck Peninsula may be sterile.
"What I usually say is, these plants are shooting blanks," Angelina Ellis, an honours student at the University of New South Wales (UNSW), says.
"That's why we're looking for sperm donors."
The goal of UNSW's Threatened Rare Endemic Plant Research Group is to uncover the extent of the plant's fertility issues in both locations.
"How sterile is sterile?" Chantelle Doyle, UNSW conservation botanist and Ms Ellis's supervisor, says.
"If we find plants with viable pollen, can we cross-pollinate and create seed with plants that are sterile?"
The collected pollen is stored in a cool bag, and the scientists trek over to Narrow Neck.
The difference between the two locations is stark: while the Breakfast Creek shrubs stand up to 2 metres tall, they only reach knee height at Narrow Neck.
The team works fast to "inseminate" flowers on some plants, leaving others untouched as a control group.
Then the flowers are encased in "wedding party favour" bags, as Dr Doyle calls them, to protect against pollinators.
It will take two months to see if the pollination attempt has worked.
Most Zieria covenyi are thought to reproduce asexually from shoots growing from the roots of the parent plant in a cloning process called "root suckering".
The Narrow Neck shrubs are a good example of this cloning process. Genetically speaking, they are the same organism.
As a result, their genetic diversity is low. Even with a conservation plan, the shrub will remain at heightened risk if it cannot diversify its population.
"[Cloning] is a really good adaptation to persisting in an area when the habitat and the climate are optimal," Dr Doyle says.
"But when it shifts, [the shrubs] don't have any genetic diversity to adapt."
Genetic rescue programs, like the work the scientists are undertaking in the Blue Mountains, aim to increase genetic variation within a population of flora or fauna to make it more resilient to pests, diseases and environmental changes.
One of the first examples of genetic rescue in Australia was a program to translocate mountain pygmy possums in the Victorian Alps.
Two months after hand pollinating the Zieria covenyi shrubs, Ms Ellis and Dr Doyle return to Narrow Neck Peninsula to check the fruits of their labour.
One by one they sift through the organza "party favour" bags, keeping an eye out for small black seeds that Dr Doyle says look like "fly poo".
There are a few moments of false hope when they notice black dots in the bags.
But then those dots start to crawl. They are bugs scavenging the dried flowers.
As the hot day wears on, it becomes clear this hand pollination attempt has not been successful.
"The female part of the plant works. They produce a flower that is receptive and attracts pollinators," Dr Doyle says.
"But then it just drops off and never produces a fruit. The seed never develops because it doesn't get fertilised."
There may have also been issues with the process itself.
Although the team pollinated flowers within hours, the donor pollen might have degraded in transit.
"And just because the other population produces [seed] doesn't necessarily mean that it has 100 per cent pollen viability," Ms Ellis says.
"It's actually relatively low from what I can see."
There are a handful of flowers still intact that might produce seed.
They are re-bagged to be checked again in a few weeks.
"Just because it doesn't work this time, it doesn't mean it will never work," Dr Doyle says.
Other programs, such as the NSW government's Saving our Species program, successfully hand pollinated the endangered buttercup doubletail orchid.
But every plant is unique and can require a different approach.
The team heads back to Narrow Neck in the new year to check the remaining bags.
In one of a handful of bags, they find one black speck.
"In the entire Narrow Neck population, we have a single fully developed seed," Ms Ellis says.
This flower did not receive donor pollen at all.
Despite being bagged, it must have been pollinated by something in nature — a bee, a butterfly, or maybe a fly.
As Dr Doyle puts it after another long day in the sun: "Who's your daddy?"
"It's the great mystery, isn't it? One single seed in the entire population."
While this seed did not result from the team's handiwork, it is an encouraging sign.
The Narrow Neck shrub is not pollen sterile, which means scientists could one day find a fertility treatment that works.