It's just what it sounds like: a patent held on a gene, a gene being a sequence of DNA which, when transcribed and translated, produces either a protein or an RNA "control sequence". In the modern day, patents are granted on genes which are not invented, but merely sequenced (or even just partially sequenced) by an individual or a team. However, the law governing it is convoluted: a gene in its "natural" state, as part of an organism, is not patentable, but the gene sequence when isolated, or the gene as a purified or possibly manufactured non-native stretch of DNA, is. In fact, your own, unique genetic material, derived from your own cells, can be patented if it is removed from your body and purified in a lab -- this has been legally upheld as per the precedent of Seattle businessman John Moore, whose cells have been patented by his doctors as the "Mo cell line" used in investigations of hairy cell leukemia, and as per a 1990 ruling, they are wholly the patenting doctors' property now, and Moore has no legal right to the cells themselves or the genetic material contained therein. Gene patents get ugly in a different legal way, however, when patents are issued for gene fragments rather than full genes -- some few of these have been done, many knowingly -- since multiple teams may be looking at separate stretches of the same gene, and also, many people argue against the legality of such patents on the ground that you cannot possibly have characterized the gene fully at that stage. One criterion of being granted a patent is that you are supposed to be able to describe what you are trying to patent *does*, and with an Expressed Sequence Tag (which is what these gene fragments are called) you simply don't know yet. The situation becomes even more ugly in that SNPs, Single Nucleotide Polymorphisms, can be patented -- this is a change in a single base unit of the sequence (which is common), and means that some genes are claimed by up to 20 patents, on the basis of these single-unit differences. If you think all this has disturbing implications, yes, you are absolutely correct. Because the human genome has been completely sequenced (as with many other genomes, now), and just about all the protein-producing genes have been identified, many of them are in fact patented. A survey published in the 14 Oct 2005 issue of Science noted that 20% of human genes are explicitly patented (4382 of the 23,688 of genes in the NCBI's gene database at the time of writing), many of them more than once (see SNPs, above), and 63% of these patents are held by commercial, profit-making companies as opposed to public-sponsored researchers. (See http://www.sciencemag.org/cgi/content/full/310/5746/239?ijkey=OL0pgwbL0K1eY&keytype=ref&siteid=sci for the survey and a discussion.) To make sure you understand, the sequences *in your body* cannot be patented; however, if the sequences in your body need to be tested, for a disease or to establish family relationship or for virtually any reason, then you may find that the sequence being examined may be patented by an organization, and you (or more generally, whoever is administering the test) must pay the patent-holder for making the comparison of your DNA to the baseline sequence or for using a test to test the sequence at all. As an example of how much of a problem this can be, there was a bitterly-fought court battle in Europe in 2004 over a gene strongly associated with breast cancer, BRCA1. A company called Myriad owned the patent on this gene. Because testing for a specific form of this gene is now part of a standard suite of tests for women considered at risk of breast cancer because of family history, Myriad could legally require that every doctor in every country which follows international patent law comply with their terms -- and their terms were that they were the only ones allowed to run the test, in their lab in Salt Lake City. Blood samples from women all over the world had to be submitted to Myriad, and they had to wait for Myriad to send back results, rather than practices and hospitals running the precise same test locally. (You can see why this made people unhappy.) Fortunately for the side of medical care, the patent was overturned on a technicality. (See a summary at http://www.nature.com/news/2004/040531/pf/429329a_pf.html if you are interested.) Patents are granted on the ground that they "stimulate research" by guaranteeing recognition of and return for those who make significant progress in genetics. However, aside from the problem outlined above, other effects of patents on genes may mean that people coming along after the patent is granted cannot even do research on the gene sequences specifically associated with diseases or aging, for investigation of effects or for experimentation for treatments, without being licensed by the patent holders. Many argue (me among them) that this does far more harm than good to medical and scientific progress. Outside the human genome, many gene patents are used to assert rights over medicines derived from plant sources, on the basis that the gene or genes in the plant which produce the specific bioactive compounds used are sequenced and described for the first time -- and in some, this is years after the drug itself has been developed, tested, and refined, generally by someone else. More commonly, gene patents over such sequences are used to remove control of "traditional" plant medicines from aboriginal peoples who may have provided the pharmaceutical companies with the first clues as to their existance and function. For a long, legal discussion of this problem, see http://www.iiirm.org/publications/Articles%20Reports%20Papers/Genetics%20and%20Biotechnology /Jim%20Zion%20Paper-Native%20Peoples%20Genetic%20Material%20Rights.pdf . Unfortunately, this (what I've written here) is only the tip of the iceberg for this incredibly contentious part of modern science. For a good "starting place" discussion of what gene patents really are and really cover, the legal issues and the rules on patentability, see http://www.ornl.gov/sci/techresources/Human_Genome/elsi/patents.shtml.