Remember when I told you that bacteria have circular DNA? Well, we have it too, only not in the nucleus where the rest of our DNA sits. It's a rather interesting story, one that biologist Lynn Margulis proved in 1967 : our cells contain organelles called mitochondria, which originally were separate organisms (prokaryotes), and at some point entered a symbiotic relationship with eukaryotic cells through endosymbiosis. As a result, they contain their own, circular DNA called mitochondrial DNA or, in short, mtDNA.
Circularity is not the only fascinating thing about mtDNA. It contains 37 genes, and, because it's not found in the nucleus, non-nucleated cells like precortical cells (found in hair shafts) can't be used for DNA analysis, but can indeed be used to extract mtDNA. However, whereas nuclear DNA is unique to each individual, mtDNA is not. That's because it's inherited exclusively through the maternal lineage. As you know, paternal and maternal chromosomes undergo recombination and then fuse together to make the unique DNA of a new individual. However, mtDNA does not undergo recombination and the only variation happening is due to random mutations when the cell splits. These are quite rare and in fact, it's not unusual to share identical mtDNA with our siblings, and/or to inherit it unchanged from our mothers.
Maternal mtDNA inheritance occurs in most eukaryotic species, indicating that, from an evolutionary point of view, it's an old and conserved mechanism. One might argue that paternal gametes (sperm) are much smaller than maternal gametes (eggs) and therefore contribute a limited amount of mitochondria, which then get lost. In fact, the general belief was that, at least in some species, paternal mitochondria was excluded due to the fact that only the head of the spermatozoon enters the oocyte's cytoplasm. Well, that doesn't quite explain the whole story, and the mechanism that allows the clearance of paternal mitochondria during early embryonegesis was not understood until recently.
Two studies published in the November 25 issue of Science [2, 3] used a Caenorhabditis elegans model to show that the degradation of paternal mitochondria is achieved through involvement of autophagosomes, double membrane vesicles that recruit the organelles, engulf them, and then destroy them. Rawi et al. also proved that autophagy is triggered in the mouse too, within minutes after fertilization, whereas, in the absence of autophagosomes (which they induced artificially in some animals), the paternal mitochondria persist in the embryos.
This is a great step forward, but many questions remain unanswered, as Levine and Elazar note in the accompanying perspective:
"The findings of Sato and Sato and Al Rawi et al. help to explain how paternal mitochondria and mtDNA are destroyed, but why they are destroyed remains a mystery. Is heteroplasmy, the occurrence of more than one mtDNA genotype, dangerous for the developing embryo? Or is the degradation of paternal mitochondria merely a primitive defense in which the fertilized oocyte views the paternal mitochondria as a potentially dangerous intruder that must be destroyed?"
 Sagan, L. (Margulis, L.) (1967). On the origin of mitosing cells Journal of Theoretical Biology, 14 (3) DOI: 10.1016/0022-5193(67)90079-3
 Sato, M., & Sato, K. (2011). Degradation of Paternal Mitochondria by Fertilization-Triggered Autophagy in C. elegans Embryos Science, 334 (6059), 1141-1144 DOI: 10.1126/science.1210333
 Al Rawi, S., Louvet-Vallee, S., Djeddi, A., Sachse, M., Culetto, E., Hajjar, C., Boyd, L., Legouis, R., & Galy, V. (2011). Postfertilization Autophagy of Sperm Organelles Prevents Paternal Mitochondrial DNA Transmission Science, 334 (6059), 1144-1147 DOI: 10.1126/science.1211878