Researchers say that identifying these viral diseases, especially those that can spread to humans, could help to prevent future pandemics.
The team estimates that this could cost more than $6 billion, but says this is a fraction of the cost of dealing with a major pandemic.
The research is published in the journal mBio.
Prof Ian Lipkin, director of the Center for Infection and Immunity at the University of Columbia in the US, said: "What we're really talking about is defining the full range of diversity of viruses within mammals, and our intent is that as we get more information we will be able to understand the principles that underlie determinants of risks."
Flying fox
Nearly 70% of viruses that infect humans, such as HIV, Ebola and the new Middle East Respiratory Syndrome (Mers), originate in wildlife.
But until now, the scale of the problem has been difficult to assess.
To investigate, researchers in the US and Bangladesh looked at a species of bat called the flying fox.
This animal carries the Nipah virus, which if it spreads to humans can kill.
By studying 1,897 samples collected from the bats, scientists were able to assess how many other pathogens the animal carried.
They found nearly 60 different types of viruses, most of which had never been seen before.
The team then extrapolated this figure to all known mammals, and concluded there were at least 320,000 viruses that have not yet been detected.
The researchers said that identifying all of these would be crucial to keeping one step ahead of diseases that could become a threat to human health.
Prof Lipkin said: "Obviously we cannot survey every animal on the planet, but we can try and map as best as we can using a concept referred to as hotspots.
"We look at areas where we know, based on previous experience, there is a high likelihood that new infectious agents will emerge or will pose considerable threat to human health."
He said that this would take 10 years and would cost billions of dollars.
But he added: "Despite what looks like an extraordinary expense to pursue this kind of work, it really pales in comparison with what one might learn that could lead to very rapid recognition and intervention that could come to the fore with a pandemic risk.
"The idea is to develop an early warning system.
A related project called PREDICT has so far discovered 240 new viruses in areas of the world where people and animals live in close contact.
Commenting on the research, Prof Jonathan Ball from the University of Nottingham, said: "The authors focused on bats because they have been the original source of a number of virus outbreaks in people.
"But we should remember, bats adopt a lifestyle that's particularly helpful to viruses - they live in large communities, they are dispersed throughout the world and they fly very large distances."
"Whether or not other mammals carry a similar array of viruses is an important question to ask, and no doubt one that the researchers are looking into," he said.
"Will larger-scale studies like this help us predict or better control future virus outbreaks?"
He added: The number of potential virus reservoirs is huge - there are more than a thousand different species of bat alone - and adequately screening these and other animals for viral threats would be challenging to say the least."