HOW MANY SPECIES ARE THERE ON EARTH? A FINAL, DEFINITIVE, AND PRECISE ANSWER
There are 42, exactly.
Key acronyms: JAL, EO, UW, CSU, MSU, ATBI, INBio, OUTBio, NTFC, and BINGO!
For what seems to be an interminable period of time biologists have been whining that all of the other sciences know more about their basic inventories than we do. Chemists have filled out the periodic table (except for rare reports of newly discovered, ridiculous elements, such as Whocaresium with a half-life of a picosecond); physicists have completed their quark menu (including all 31 flavors), astronomers have catalogued the stars (in fact, they have a few dozen catalogues suggesting that the project is pretty much old hat), and mathematicians have calculated pi to 1 million, mind-numbingly irrelevant digits. However, we biologists can’t seem to estimate the number of species on earth to the nearest order of magnitude.
The standard explanation for our ignorance is that we don’t get enough grant support. If only we had another 20 million taxonomists, all would be well in Nature (albeit hellish in departments of biology, entomology, botany, etc.). But the taxonomists are the problem, not the solution. Let’s face it, scientists who fail calculus become biologists and those who flunk arithmetic become taxonomists. If you doubt my assessment of the average quantitative skills of systematists, ask your local expert to explain the mathematical assumptions and processes that lie behind whatever software program s/he is currently using to crank out the next long-awaited, earth-shattering cladogram.
Clearly, the question of how many species live on earth needs the talents of an ecologist who has both quantitative skills and biological insights. Like me. So I used several widely accepted data collection methods and mathematical analyses to arrive at a final, definitive, and precise answer so that biologists could move onto something more important, like debating creationists, moving genes between unrelated organisms, sequencing long strings of nucleic acids, and building simulation models of ecosystems.
Step 1: Picking a Starting Point
According to E.O. Wilson, Terry Erwin claimed that two-thirds of all species live in tropical forest canopies (Wilson, 1992). Now if Terry said it (and he is an ecologist) and Ed repeated it, then the Gods have spoken. And I, for one, don’t want to mess around with Biological Scripture. However, only a fool would take this information and then decide to start inventorying life on earth by traveling through snake-infested jungles in order to “fog” the tropical canopy, so that he could then multiply this measure of species abundance by 1.5 to arrive at a global value. This strategy smacks of somebody who sneaked one too many snorts on the canopy fogger. A sane, unintoxicated, and clever ecologist would sample in areas that are not tropical canopies and then multiply the estimate of species richness by 3 (arriving, presumably, at the same number with a whole lot less work). So, I decided to determine the number of species in a non-tropical-forest-canopy habitat (NTFC).
Step 2: Determining the Number of Species in an NTFC Habitat
Given that the National Science Foundation sent all of its Biodiversity dollars down to the gang at INBio, they were unwilling to fund the OUTBio (Organization of Utterly Trivial BIOlogy, of which I’m the executive director) proposal for an ATBI of the French Riviera. As such, I decided to use local funds. The people of Wyoming have such implicit faith in their university faculty that they don’t even ask what we do with their money. On a particularly sunny day I sat on a bench outside my office and conducted a complete census of all life forms. After several hours of observation and one short period of inattention, I was able to find and identify seven species. In all, I saw one robin, one dog (black), one administrator (white), four students, 17 pine trees, 79 ants (red), and 1,000 blades of grass. Some biologists might argue that the administrator is conspecific with the students, but these scientists have obviously never spent any time with a university administrator (see also Darwyn 1987). Based on this data set, I decided to estimate the number of species using a really simple formula that even a taxonomist could follow (although the little bitty numbers above and below the letters might be off-putting):
where NT is the total number of species, N is the number of species observed, and I is the number of species observed only once. So, plugging in the numbers we find that:
Step 3: Determining the Completeness of the Census
One might ask how I know that there are really 12.25 species in my typical NTFC habitat. Well, I plotted the rate at which I found new species across observations, and by the time I hit the one thousandth blade of grass it was clear that discovery rate had reached an asymptote. So further observations would have failed to reveal more species. This clever mathematical move along with good ol’ common sense should be sufficient to dismiss any annoying reviewers with the chutzpah to suggest that my census was incomplete.
Step 4: Determining the Level of Endemism
While there are precisely 12.25 species on the campus of the University of Wyoming, you might wonder if this richness represents the total number of species in all non-tropical-forest-canopy (NTFC) habitats. To ascertain the universality of my finding, I called several colleagues at Montana State University, but the secretary reported that it was “huntin’ seezun for the boyz”, so they were all out blasting warm, fuzzy, gentle ungulates to strap across the hoods of their pickups and then hang in their garages.
So, I called a buddy down at Colorado State University (home of the fabled Natural Resources Ecology Laboratory and dozens of ecological modelers who are highly unlikely to be out hunting). Being paragons of productivity, the CSU faculty are always willing to lend a hand. I managed to catch my esteemed ecological colleague after his 2-hour lunch, just as he was leaving for a racquetball game. I asked him to conduct a census of all biological species on his campus. Being a theoretical ecologist, he expressed serious reservations about the possibility of having to “go outside”. I suggested that he could collect the data by looking out the window, which he did. He reported seeing one lawn, one dog (brown), one mosquito, two birds (“medium-sized with red tummies”), four trees (probably pines, but he wasn’t sure), 22 students, and 114 ants (actually these were consuming his sandwich on the window ledge but we included them in the census).
Incredibly enough, the species he reported appeared to be identical to those recorded in Wyoming. The only controversial classification might be the conspecificity of the administrator (WY) and the mosquito (CO). But given that they are both blood-sucking, small-brained parasites with overwhelming urges to make more of their own kind, I decided that they could be safely subsumed under a single species. Such is the advantage of thinking in terms of ecological function rather than morphological similarity. To mathematically determine the overlap between the species in Colorado and Wyoming, I used Jaccard’s index:
Only an ecologist really grasps what the horseshoes mean, but this is clearly a complex—and hence true—bit of mathematical legerdemain. Using some really cool software that I “borrowed” from another ecologist (like all of your software was purchased?) and my brand-new laptop computer I was able to determine the extent of overlap was 99.9%. You might think that two apparently identical sets would have 100% similarity, but this merely suggests that you don’t understand statistical variation, formal logic, tensor calculus, and computer software that is written by people who are smarter than you. As such, accept that each non-tropical-forest-canopy (NTFC) university campus would add an additional 0.1% to the total species inventory completed at the University of Wyoming. The precision of this finding allows me to conclude, by the way, that my final value was not destined to be one of those namby-pamby “estimates” but the real, authentic, incontrovertible determination of the number of species on Earth.
Step 5: Determining the Total Number of Non-Tropical-Forest-Canopy Species on Earth
Because university campuses are built all over the earth, it is really easy to figure out the total number of species living in NTFC habitats across the entire planet. There are exactly 1,750 universities located in non-tropical forest canopies (Smith 1994), with each one having 0.1% endemism. Thus, these habitats collectively add an additional 1.75 species (i.e., 0.001 x 1,750) to the recorded total of 12.25. This calculation leads to the irrefutable conclusion that there are precisely 14 species on all non-tropical-forest-canopy habitats.
Step 6: Determining the Total Number of Species on Earth
Since Erwinian ecology dictates that one-third of all species live in non-tropical-forest-canopy habitats, we (you’ve read this far, so the first-person plural is appropriate) can simply multiply our species richness of 14 by 3. Thus, we find that the final, definitive, and precise number of species on Earth is 42.
The greatest mind in the universe concurs with my finding:
“All right,” said Deep Thought. “The Answer to the Great Question …”
“Yes …!”
“Of Life, the Universe and Everything …” said Deep Thought.
“Yes …!”
“Is …” said Deep Thought, and paused.
“Yes …!”
“Is …”
“Yes …!!!… ?”
“Forty-two,” said Deep Thought, with infinite majesty and calm.
It can be no mere coincidence that the most powerful computer in the Universe also arrived at the value of 42 (Adams 1979). Indeed, my finding substantiates the fact that all other scientific disciplines are irrelevant and petty. At long last Ecology has discovered what the Ultimate Question of Life, the Universe and Everything, for which the answer is most assuredly 42 (Adams 1979). I could go on, but when you’ve revealed the Greatest Question in the Universe and validated its answer, what else is there to say?
I would like to thank the editor of The Science Creative Quarterly for not cashing my check for “annual dues” until this manuscript was accepted. And as for funding, I have nothing but contempt for various grant agencies that couldn’t recognize a solid scientific proposal if it bit ‘em on the butt.
None. You can’t really call what we publish in scientific journals “literature”.
Adams, D. 1979. The Hitchhiker’s Guide to the Galaxy. Harmony Books, New York.
Wilson, E.O. 1992. The Diversity of Life, Norton, New York (Note: yeah, sure there’s newer stuff, but the biodiversity-thing is really just a reiteration of the same old stuff for the last 15 years as we come no closer to knowing what a species is, how many there are, or what can be done to save them).
Darwyn, C. 2004. On the contraspecificity of Homo administratus and Homo sapiens: Implications for the end of civilization. Journal of Social Parasitism 37: 867-898.
Smith, T. R. 2007. Guide to the World’s Universities. Vol 2. Campuses located in Non-Tropical-Forest Canopies. Nonacademic Press, 893 pp.
JAL, from the University of Wyoming, is an ecologist of international repute. If you’ve not heard of him, then don’t share your ignorance with others as this will only lead them to conclude that you’re out of the loop regarding the movers and shakers in science. Among his other honors, he is a lifetime member of Modelers Anonymous (“National Science Foundation please grant me the funds to model that which I can simulate, the serenity to accept that which is stochastic, and the software to tell the difference”). At present he works in the department of philosophy (or at least as much as one can call doing philosophy actual work), which is a really long story involving academic intrigue, administrative ineptitude, and professorial metamorphosis.