🔬 Discovery
Science describes approximately 18,000 new species each year — and yet estimates suggest that 86% of Earth's species remain unknown to science. The total number of species on Earth is estimated at 8.7 million (excluding bacteria and archaea), of which only approximately 1.2 million have been formally described. The gap between described and actual species diversity is greatest in the groups that are hardest to study: insects, nematodes, fungi, and marine invertebrates — taxonomically complex, often tiny, and abundant in habitats that are difficult to survey.
estimated species on Earth
new species described annually
of species unknown to science
to describe all species at current rate
The persistence of such vast unknown biodiversity reflects a combination of factors. Most of Earth's biodiversity is concentrated in the tropics — where habitat is vast, access is difficult, and taxonomic expertise is limited. The organisms comprising the majority of unknown species — insects, fungi, nematodes — are small, diverse, and taxonomically challenging, requiring specialist expertise that is in short supply globally. The number of taxonomists worldwide has remained approximately constant or declined in recent decades.
DNA barcoding — using short, standardised DNA sequences to identify species — is revolutionising the pace of species discovery. Where traditional taxonomy requires morphological examination of specimens by expert taxonomists, DNA barcoding allows rapid identification of species from tissue samples by comparison with a reference database. Environmental DNA (eDNA) methods take this further: collecting water or soil samples and sequencing the DNA they contain can detect hundreds of species simultaneously without collecting any individuals. These methods are dramatically accelerating the rate at which unknown species are identified — though the formal description of new species still requires morphological study and peer-reviewed publication.
The toolkit of the 21st-century taxonomist extends far beyond the morphological examination and museum collection that characterised species description for the first 200 years of Linnaean taxonomy. Environmental DNA (eDNA) — the genetic material shed by organisms into their environment as skin cells, mucus, faeces, and other biological material — can now be collected from water samples, soil cores, or even air samples and sequenced to detect the presence of species without ever observing the organism directly. eDNA metabarcoding — amplifying and sequencing a specific genetic marker (typically the mitochondrial COI barcode region for animals or ITS regions for fungi and plants) from environmental samples — can detect dozens to hundreds of species from a single water sample, revealing biodiversity that would take months of traditional survey effort to document. The technique has been particularly transformative for detecting rare or cryptic species: great crested newts have been detected in ponds from water samples in less than a week, replacing labour-intensive survey protocols that required multiple site visits over an entire season.
The rate of species discovery has accelerated dramatically in the genomic era — not because more biologists are exploring more habitats (though this is also true), but because molecular tools have revealed that many morphologically similar organisms are in fact reproductively isolated species with distinct genetic identities. Environmental DNA (eDNA) — the genetic material shed by organisms into water, soil, or air — allows the detection of species from their DNA traces without ever observing the organism directly: a water sample from a lake can reveal the presence of fish, amphibians, invertebrates, and even aquatic plants without a single organism being caught or seen. eDNA surveys have discovered populations of rare and cryptic species in habitats where traditional surveys missed them, and are transforming monitoring programmes for endangered species, invasive species, and poorly studied groups like fungi and meiofauna.
The formal description of a new species — the publication of a scientific paper providing a formal diagnosis, a type specimen, and a Latin binomial name — remains the gold standard for species discovery, but the backlog between collection and description is enormous. Taxonomists estimate that there are approximately 1-2 million undescribed species in museum collections worldwide — organisms collected, preserved, and accessioned into natural history collections but not yet formally named. For groups like beetles, nematodes, mites, and fungi, the ratio of undescribed to described species in collections may exceed 3:1. The development of rapid automated morphological imaging, machine learning for specimen identification, and genomic barcoding is beginning to address this backlog, but the global taxonomic community — approximately 7,000 active taxonomists worldwide — faces an essentially impossible task of describing all remaining species within any biologically meaningful timeframe without substantial changes in funding, training, and institutional support.
Get our latest science journalism delivered to your inbox.
✅ Thank you! You'll receive our next article in your inbox.
Dr. Al-Rashid has led over 40 scientific expeditions across six continents studying biodiversity discovery, species new to science, and field ecology in remote ecosystems.