Nearly 12 years ago evolutionary biologist Geoff McFadden of the University of Melbourne stunned malaria researchers by announcing he had found a degenerate chloroplast in Plasmodium falciparum, the deadly agent of human malaria. The deadliest of five Plasmodium species that infect humans, P. falciparum kills more than a million children under the age of two every year.

McFadden's team later confirmed that the apicoplast has retained some of the biosynthetic machinery of the original chloroplast by showing that the broad-spectrum herbicide glyphosate (Roundup) kills P. falicparum.

The herbicide kills plants by disrupting the chloroplast-based shikimate pathway, which synthesises the amino acids tryptophan, phenylalanine and tyrosine.

The 175-odd species of Plasmodium have an oddly eclectic menagerie of vertebrate hosts that includes humans, non-human primates, bats, porcupines, squirrels, birds, and reptiles.

All are obligate blood parasites that never see the light of day, yet McFadden's discovery of an unpigmented chloroplast showed Plasmodium had evolved from a light-harvesting ancestor. What could it have been?

The answer has now been found at the bottom of Australia's most famous harbour.

As a PhD student with Associate Professor Dee Carter's research group of the School of Molecular and Microbial Biosciences at the University of Sydney, Robert Moore was attempting to isolate a dinoflagellate alga, Symbiodinium, from a scleractinian coral, Plesiastrea versipora, that grows in Sydney Harbour at a depth of around three metres.

All coral polyps host symbiotic zooxanthellae, specialised photosynthetic dinoflagellates that live in their tissues, harvesting light and synthesising carbon compounds for their hosts, in return for a steady supply of carbon dioxide. Symbiodinium is the dominant genus among the zooxanthellae.

"At the time Bob was very interested in trying to develop a system to culture Symbiodinium," Carter says.

"We were analysing the population genetics of Symbiodinium to look at sexual recombination in the genus, to determine if we were dealing with a single species with diverse genetics, or a population of different species with diverse genetics.

"Bob was picking out single cells, to see if he could find anything different, and found an unusual dinoflagellate. He started sequencing its DNA, and it looked very interesting, so he continued the work during his first postdoctoral appointment in the US."

The divergent DNA sequence indicated they had discovered an unrecognized genus of zooxanthellae. In a recent paper in Nature, they formally named it Chromera velia.