It was R. Cajal's student, Pio del Rio-Hortega, who coined the term microglia around 1920. He conducted the first systematic studies on this cell type. The cells had been previously described by F. Nissl and F. Robertson but Rio-Hortega is rightly considered the "father" of the microglia. Many of his observations are still valid (1). Resting microglia show characteristic elongated, almost bipolar cell bodies with spine-like processes that often branch perpendicularly.
There has been a decades-long debate about the "nature" and "identity" of microglial cells which lasted until immunocytochemical as well as lectin markers were discovered in the 1980s. It then became clear that microglia share phenotypic characteristics (as well as lineage properties) with bone marrow-derived monocytes/macrophages. In general, the immunophenotype of microglial cells appears to be "down-regulated" but the cells should not be considered "tamed" macrophages; microglia display a number of properties that are unique to these resident cells of the CNS which are about as numerous as neurons. One example is their involvement in synaptic plasticity under pathological conditions ("synaptic stripping", http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=5706753). During vascularization of the CNS, cells related to the mononuclear phagocyte system appear to invade the central nervous tissue and give rise to resting microglia. Unlike neuroectodermal glia, microglial cells are not electrotonically coupled, i.e. functionally connected via gap junctions. This may explain their very localized involvement in disease processes, which is of great diagnostic use including in neuroimaging (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12697620).
The presentation of antigen to lymphocytes is a likely function of microglia. Microglia are a resident CNS source of a number of "immune molecules" apart from MHC antigens, e.g. interleukins. There are many different names for microglia which exhibit considerable morphological plasticity. The term amoeboid microglia should be reserved to describe the cell's appearance in developing nervous tissue. Microglial rod cells representing an activated microglia phenotype are mainly found in the cerebral cortex typically in quaternary syphilis, subacute sclerosing panencephalitis and lead intoxication. Brain macrophages which are morphologically indistinguishable from macrophages of peripheral sources are found in early active multiple sclerosis lesions and in a cerebral infarcts. Some brain tumours (especially high grade gliomas) are very rich in what appear to be phagocytic microglia but it is at present impossible to be certain about their source.In addition to microglia, there are other macrophage phenotypes present within the bony confinements of the normal CNS. These include epiplexus and meningeal macrophages as well as perivascular macrophages (aka "perivascular cells" as they lack a macrophage phenotype in normal conditions) (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=2926837). The topic of brain macrophages and microglia is a difficult one due to historical nomenclatorial controversy and some still existing confusion. Terminological clarity is of utmost importance. As Wittgenstein said: if it is true it can be said clearly.
Reference not found in Medline: 1. Rio-Hortega (1932) Microglia. Cytology & Cellular Pathology of the Nervous System. Paul B. Hoeber, Inc., New York, pp 483-534