One of the joys of the last 50 years of biology is learning how evolutionary relationships we long took for granted need revision.
First it was confirmation about Margulis’ endosymbiotic origin of mitochondria and chloroplasts. Then the discovery and classification of archaea. Now it’s upending our understanding of fungal phylogenetics! For people interested in biology, it’s a wonderful time! (Well except for massive species die-off, loss of biodiversity, destruction of ecosystems, etc…)
Regarding the "confirmation about Margulis’ endosymbiotic origin of mitochondria and chloroplasts", I assume that OP has meant that after this event (her paper was published in 1967) it became clear that the original view of the evolution of the living beings as along a tree has been overly simplistic.
Even if the most common form of evolution is the appearance of several new species by differentiation from a common ancestor, there are also many cases when a new species appears as a hybrid of 2 previously distinct species.
This hybridization happens most frequently between closely related species, by the normal process of syngamy between reproductive cells (i.e. fusion of gametes), and this does not change much the form of the evolution tree, but in a few cases the hybridization has occurred between extremely distantly related living beings, as a consequence of the evolution of a symbiosis relationship towards the integration of 2 distinct cells, until they have merged into a hybrid cell, which reproduces as a unit and which can no longer be separated into the former independent constituents without causing death.
Because of these very long range hybridization cases, the evolution tree has merged branches, so it is no longer a tree in the graph theory sense, even if most smaller parts of it are still trees.
Besides these cases of complete hybridization, the form of the evolution tree is complicated by partial hybridizations, i.e. by the so-called lateral transfer of genes, which is frequently mediated by viruses, between unrelated species, e.g. between snakes and frogs, as reported right now in another thread on HN.
So, like in a code repository, in the evolution of the DNA the most frequent event is creating forks, which are then edited separately (i.e. separate DNA mutations in distinct species), the next most frequent event is to copy and paste some code fragment from a branch to another branch (i.e. lateral transfer of genes between distinct species), and the least frequent event is the merge of 2 distinct branches, which is less and less frequent the more the branches have diverged since forking, because that increases the chances of fatal bugs in the merged code (i.e. species hybridization events).
It should be noted however, that while Margulis’ supposition about the endosymbiotic origin of mitochondria and chloroplasts has been confirmed beyond any reasonable doubt, she got carried away by the enthusiasm of understanding these important events in the history of the living beings and she had also made a third proposal, that also the flagella of the eukaryotes are the result of an endosymbiotic event. This third supposition was wrong.
Mutations also happen in one individual. This gene then enters the pool of the population, and it can spread. The genes from a hybrid individual diffuse into the population when it reproduces. Speciation happens when a population accumulates enough changes to be distinct from other populations of the same original species.
First it was confirmation about Margulis’ endosymbiotic origin of mitochondria and chloroplasts. Then the discovery and classification of archaea. Now it’s upending our understanding of fungal phylogenetics! For people interested in biology, it’s a wonderful time! (Well except for massive species die-off, loss of biodiversity, destruction of ecosystems, etc…)