Ajith Harish and Charles Kurland at the Uppsala and Lund Universities in Sweden have in the article “Mitochondria are not captive bacteria” shown that the probability that mitochondria originated from proteobacteria is 44 orders of magnitude lower than the probability they were created locally. The import of a bacterium to become a mitochondrion in a primitive eukaryote was held as the most central event in the endosymbiosis theory. It is still possible that plant chloroplasts are endosymbiotic cyanobacteria, but it is very unlikely. There must be very overwhelming evidence for chloroplasts as endosymbionts to make the theory survive. Without such evidence we must conclude that mitochondria are created in the host. There are then three possibilities for the origin of bacteria: either they are the original, they are reduced eukaryotes or they are escaped organelles, as described in the Organelle Escape Theory. Of these, it is only the latter that can explain the similarities between the translation apparatus in organelles and that found in eubacteria.