The possibility of lactating dinosaurs sat for 15 years in an ideas folder on Professor Paul Else's computer.
There it stayed, year after year, niggling away at him.
He had hoped, that in time, someone working in the land of dinosaurs would come up with the same idea so he could hit the delete button. But, when no-one did, there was only one thing left to do - put it out into the world himself.
So, during a study leave period, he sent his hypothesis to the editor of The Journal of Experimental Biology. Keen on the idea of dinosaurs feeding milk to their nest-bound offspring, they commissioned 3000 words.
Away from his day job as a physiology expert and molecular biologist researching membrane lipids in humans at the University of Wollongong, Else began to nut out his theory that dinosaur milk, laced with additives and growth hormones, may have been responsible for the enormous size of the dinosaur.
Published last month, his controversial proposition has attracted comments in scientific journals and from science bloggers worldwide.
Lactation is physiologically associated almost always with mammals so when it comes to dinosaurs there's no breastfeeding involved. Instead Else suggests a milk-like substance may have been secreted from the upper digestive tract which dinosaurs would then feed to their young through their mouths.
Modern birds are descendants of dinosaurs and some of them, including pigeons, emperor penguins and flamingos, feed their hatchlings in this way, producing a type of milk from crop glands or glands of the oesophagus. So, if some birds do it, thought Else, is it such a stretch to think dinosaurs lactated?
"The greatest potential advantage of dinosaur lactation is that milk fed to the young can be 'spiked' with additives, such as antibodies, antioxidants, calcium and minerals and growth hormones," says Else.
"These are all examples of additives, some of which [are] found in the milk of pigeons that allow their young to grow at phenomenal rates."
Some birds produce an oily, yellowish cheese-like substance, while pigeons produce a tiny rice pellet of milk, like a seed, in their crop. They then add value to it, including epidermal growth factor, which allows their hatchlings to grow to 85 per cent of their adult weight within four weeks.
The milk produced by some birds is similar to mammals' milk and contains fat and protein, along with carotenoids and antibodies which aid immune-enhancing factors. This bird milk allows chicks to grow much faster than they would if they received only regurgitated food.
Palaeontologists know that dinosaur babies also grew quickly and the only way that was possible, argues Else, is if they were fed growth hormones through the lactation process.
There is, of course, no evidence of dinosaur milk in fossil records. That would require investigation of soft tissues, which are not preserved as fossils.
But to make his point Else used herbivorous duckbills (hadrosaurs) as a case study. Hadrosaurs were herd breeders with nest-bound young fed by parents.
There is also strong evidence that hadrosaurs raised their hatchlings in vast nesting colonies where they cared for them for some time, similar to the nesting colonies of seabirds.
Else says baby hadrosaurs would not have been able to feed themselves enough to sustain life, yet somehow they managed to grow rapidly - from a 300 gram hatchling to a 20 kilogram juvenile in just eight weeks.
"One thing that always struck me as unresolved about dinosaurs was how could a dinosaur parent weighing several tonnes feed its young of only a few kilograms? It seems obvious to me, a form of lactation was involved, similar to that present in modern birds.
"I propose that some dinosaurs likely used a feeding strategy in which they used specialised lactation then progressed to plant regurgitation."
Dinosaur lactation could also have facilitated immune responses as well as extending parental protection as a result of feeding newly hatched young in nest environments, says Else.
Birds are great parents, raising their chicks from birth until they can literally fly the nest. One of the mysteries of paleontology is the question of how good dinosaurs were as parents and did they hand down their nurturing instincts to modern birds?
"It seems plausible though that, along with other methods of parental care, adult dinosaurs of all kinds nursed their young with secretions of some sort," says Else.
Professor Frank Seebacher, a biologist at the University of Sydney, has told Australian Geographic that the controversial idea would be impossible to prove.
"I think, in my opinion, it will remain an idea, but that's what science is made out of," Seebacher says.
Dinosaur expert Dr Steve Salisbury, of the University of Queensland, says the use of lactation is misleading because the process is "so different to mammals".
He explains Else's theory as a logical progression in our understanding of the links between birds and dinosaurs.
"Rather than adding some amazing new insight, it's just altering what we are already coming to accept about dinosaurs," he told Australian Geographic. ■