Article history:Accepted 12 December 2006Available online 15 December 2006
Towards themiddle of the 20th century, neuroanatomywas on the decline. It was revived bythe development of two newmethods. One was the Nauta-Gygaxmethod, which selectivelystained nerve fibers that had been caused to degenerate by experimental lesions. Thisallowed connections between various parts of the nervous system to be better determined.The second was electron microscopy, which allowed the structure of neurons and thesynapses between them to be examined in detail, and eventually this led to a revival of theGolgi impregnation methods. This occurred in the 1970s because of the desire of electronmicroscopists to determine the origins of the neuronal profiles they encountered in electronmicrographs of various parts of the central nervous system. Eventually this led to thedevelopment of Golgi/EM techniques, whereby individual impregnated neurons could firstbe characterized by light microscopy and then thin sectioned for detailed analyses.Examining the axon terminals of such impregnated neurons, especially those in the cerebralcortex, for the first time revealed details of intercellular connections and allowed neuronalcircuits to be postulated. However, Golgi/EM had only a brief, but fruitful existence. It wassoon superceded by intracellular filling techniques, which allowed the added dimensionthat the physiological properties of identified neurons could also be determined.
Towards the middle of the 20th century, anatomy, andespecially neuroanatomy, was in the doldrums. In neuroanat-omy a few basic stains were available for examining the centralnervous system and most parts of it had been studiedextensively using these stains. Some of the stains showed
er B.V. All rights reserved
neurons in their entirety,whileothersonly showedcertainpartsof neurons. To examine individual nerve cells in their entirety anumber of variations on the Golgi impregnation methods wereavailable, but theuncertaintyabout thequality andextent of thestaining produced after tissue had been impregnated led to the
257B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
Golgimethods being surroundedbymystique. In additionmanythought that there was not much point in using the Golgi stainsince all of the cell types had already been examined by Cajalanddescribed inhis 1911 volumes. Consequently only a handfulof investigators used theGolgi stain. Itwas also possible to showentire cells usingmethylene blue staining on live tissue, but thisnever became a popular stain.
Most investigators preferred to use sectioned material fortheir studies and a common technique to apply to them wassilver staining.Again therewere several silver stainingmethods,and each of them revealed all the nerve cells and processescontained within a section, so that the shapes of the nerve cellbodiescouldbe readilyvisualized, ascould thenumberandsizesof processes emerging from them. But unfortunately the silver-stained processes emerging from the neurons soon entered theneuropil, interlaced with other processes and went out of theplane of section, so that they could rarely be followed for anydistance. For revealing the sizes, shapesandpackingdensities ofnerve cells, Nissl stainswerepopular, andmyelin stains, suchasthe Heidenhain methods and the more routine luxol fast bluemethod, were available to follow the trajectories of nerve fibersthat were organized into tracts. However, it was rarely possibleto determine where the nerve fibers really ended, sincemyelinated nerve fibers axons usually lose their sheaths somedistance before they terminate. Even the use of lesions to bringabout chromatolysis could only be used when large neuronswere affected, and the consequencewas thatwhile theavailablestaining methods provided information about the kinds ofneurons and their cytology in various parts of the nervoussystem, the limitations of the stains frustrated attempts to gaininformation about the neuronal circuits. A new era began afterthe end of the Second World War, when two quite new, anddifferent, methods allowed more detailed information aboutconnections and cytology of neurons to be obtained. Thesemethods were the Nauta-Gygax stain and electron microscopy.
1. The Nauta-Gygax stain
The Nauta-Gygax method selectively stains nerve fibers thathave degenerated following experimental lesions (Nauta andGygax, 1954), leaving most of the unaffected nerve fibersunstained. Using this method, degenerating nerve fibers canbe traced quite close to their sites of termination, and it wasbelieved that even some degenerating axons terminals werestained (see Guillery, 1970). Some time later Fink and Heimer(1967) introduced a modification of the Nauta-Gygax stain thatdid allow degenerating nerve boutons to be selectively stained,so that the actual sites of termination of nerve fibers could bedetermined (Heimer and Peters, 1968). Of course these degen-eration techniques are only applicable to nerve fiber tracts inwhich the site of the lesion is some distance away from the siteof termination and interpretation of the results depends uponthe size and exact location of the lesions being used to bringabout the degeneration of fibers. Nevertheless, these degenera-tion techniquesopenedupaneraof tract tracing, because for thefirst time it became possible to determine the specifics aboutconnections between parts of the central nervous system. Ofcourse, themore recent tracts tracingmethodsusing radioactivelabels or dyes have now superceded even these degeneration
methods and have allowed finer details of connections to bedetermined, but these latter methods did not start to becomeavailable until the 1970s.
2. Electron microscopy
The other method that revived neuroanatomy was electronmicroscopy. This approach opened up a new era of under-standing the cytology of neurons and neuroglial cells. It alsomade it possible to determine the kinds of structures that arepostsynaptic to projection neurons, and for the first timemade it feasible to determine the local connections betweenneurons of the central nervous system.
The development of the first transmission electron micro-scope with a useful resolving power is generally attributed toErnst Ruska and Max Knoll, in Germany, in 1933, and Ruska'sefforts were recognized by him being awarded the Nobel Prizein Physics in 1986. In about 1935 Ruska joined the Seimens andHalskeCorporationandhelped todesign the first commerciallyproduced electron microscope, which came available in 1939,just as the Second World War began. But because of theimminence of war Germany decided to make these electronmicroscopes only available to its allies. However, Germanywasnot the only nation involved in the development of electronmicroscopes and the first useful prototype of a transmissionelectronmicroscope appears to be the one developed in 1935 inBritain by Metropolitan Vickers, while in North America, thefirst useful electronmicroscope seems to be the one developedby James Hillier and Albert Prebus in Toronto in 1939. Initially,theUnited States lagged behind in the development of electronmicroscopes, but by 1941 RCA had developed its own commer-cial transmission electron microscope, the RCA EM-B. Severalof these instruments were produced. Most were used in theUnited States although others were sent to abroad to allies.
Because of thewar few electronmicroscopeswere availablefor biological research. However, Keith Porter and AlbertClaude at the Rockefeller Institute in New York gained accessto an RCA EM-B microscope, which was operated by ErnestFullam at Interchemical Corporation, and they were able toexamine a cultured fibroblast from a chick embryo that hadbeen grown by Porter on a polyvinyl film. The fibroblasts werepeeled off the film and transferred to a wire specimen grid,after which they were treated with osmium tetroxide, washedand then dried to prevent evaporation in the vacuum of theelectron microscope. It is generally agreed that the resultingpictures were the first electronmicrographs of a cell (Porter et al.,1945) and they revealed the Golgi apparatus, and a lace-likereticulum that Porter later called the endoplasmic reticulum.
Although this showed that it was feasible to examine cellsby electron microscopy, it was obvious that to obtain moredetailed images it would be much better to use sections oftissue. Attempts to produce such sections led to a flurry ofactivity and the rapid development of new technology, ofwhich the following are the highlights. Producing thinsections for electron microscopy required a hard and stableembedding medium. The paraffin wax and celloidin that wereroutinely used to embed tissue for sectioning for lightmicroscopywere too soft to produce thin sections and paraffinwax evaporated in the electron beam. A major step towards
258 B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
the production of a useful embedding medium occurred in1949, when the plastics division at the US Bureau of Standardintroduced methacrylates. This was soon followed by Lattaand Hartmann (1950) showing that glass knives could be usedto cut thin sections and that they weremuch sharper than thesteel knives that had been used until that time. Then in 1952Palade showed that preservation of tissue by immersion couldbe much improved if small pieces of fresh tissue were fixedusing buffered osmic acid solutions, and in 1953 Porter andBlum introduced the first version of the Porter–Blum ultra-microtome that served for many years as the primarymicrotome in most electron microscope laboratories. Thenext important step took place in 1956, when Glauert et al.introduced epoxy resins for embedding tissue. This made thinsectioning easier and avoided the disintegration of blocks oftissue that frequently occurred when methacrylates polymer-ized. However, epoxy resins were more electron dense thanmethacrylates and so tissue in which osmium was the onlyelectron dense material had little contrast in the electronbeam. The increased contrast was provided by the use of leadstains such as lead hydroxide (Watson, 1958), lead tartrate(Millonig, 1961), and lead citrate (Reynolds, 1963), which areroutinely used today.
While small pieces of tissue fixed by immersion in bufferedosmic acid solutions and embedded in epoxy resins producedacceptable fixation of most tissues, the same was not true oftissue from the central nervous system, inwhich astrocytes andtheir processes swelled to produce unpleasant images. Initiallythis led to difficulties in the identification of the various types ofneuroglial cells and there were also problems with thepreservation of myelin. In an effort to solve this problem Palayet al. (1962) turned to the intravascular perfusion of rats withbuffered osmic acid solutions. This resulted in fixed brains thatwere black and hard and gave off fumes that fixed the olfactoryepitheliumand corneas of the investigators, but brains inwhichthe fixation of the tissuewasmuch superior to that produced byimmersion fixation of small pieces. Perhaps fortunately, in 1963Sabatini, Bensch and Barnett introduced glutaraldehyde as aless volatile fixative. And this led to brains being perfused withbuffered glutaraldehyde solutions that often contained paraf-ormaldehyde (Karnovsky, 1965), after which pieces wereremoved, osmicated and embedded in epoxy resins for thinsectioning. This is the method of preparation now usedroutinely in most laboratories doing transmission electronmicroscopic studies of the nervous system. The design ofultramicrotomes has also improved and it is common to usediamond knives to produce the thin sections. As a point ofinterest, it might be added that Fernández-Moran in Venezuelafirst advocated the use of cleaved diamonds to produce thinsections as early as 1950, but diamond knivesmadeby polishingcleaved surfaces did not become commercially available andwithin a reasonable price range until the 1970s.
3. Analysis of electron micrographs of nervoustissue
Although some improvements in both techniques and preser-vation of tissue were yet to be made, by the beginning of the1950s investigators were examining thin sections of nervous
tissue. Attentionwas first paid to the peripheral nervous systemthat had fewer components than the central nervous system.The peripheral nervous system also had the added advantagethat when pieces of it were fixed by immersion reasonably goodpreservationwasobtained. Thiswasnot truewhenpieces of thecentral nervous systemwere fixed by immersion, because suchtreatmentusually led to a great deal of swellinganddistortionofthe components. Thus, as early as 1950 Fernández-Moran wasable to demonstrate that peripheral myelin has a lamellarstructure, and quite soon afterwards it was shown that thelamellae form a spiral and that the membrane forming thespiraled lamellae is in continuity with the plasmamembrane oftheSchwanncell (Geren, 1954;Robertson, 1955).However, itwasnot until 1960 that the myelin sheaths of nerve fibers in thecentral nervous system were demonstrated to have a similarspiraled structure (Peters, 1960; Maturana, 1960). Also, becauseof theproblems inobtainingadequate fixation the first articleonthe fine structure of a neuron in the central nervous systemwasnot published until 1955 (Palay and Palade, 1955), when it wasshown that the Nissl substance consists of arrays of granularendoplasmic reticulum and that neurons contained filaments,while the first description of a synapse in the vertebrate centralnervous systemwas not published until 1956 (Palay, 1956).
As the preservation of central nervous tissue improved andbetter images were obtained the main problem facing thosestudying the central nervous system by electron microscopywas how to identify the various kinds of profiles seen inelectron micrographs. It must be remembered that the stainsused in light microscopic studies were designed to show onlysingle nerve cells, or specific parts of nerve cells, but inelectron micrographs investigators were presented withimages in which profiles of all parts of neurons and neuroglialcells were represented. However, within a few years thecriteria for identifying profiles of dendrites and axons wereestablished and it was shown that there were at least twokinds of synapses present in the cerebral cortex (Gray, 1959),and it soon became apparent that there were even othermorphological types of synapses in other parts of the centralnervous system.
4. Resurrection of the Golgi techniques
The next problemwas to determine where the dendrites, axonsand axon terminals thatwere represented by profiles in the thinsections originated; and for the axon terminals what were thepostsynaptic structures with which they were synapsing.Interestingly, this led to a renewed interest in impregnatingtissue with the Golgi stain, because this was the principaltechnique for displaying the morphological features of indivi-dual neurons. One of the leaders in advocatingGolgi images as abasis for interpreting electron micrographs was Sanford Palay,who together with Victoria Chan-Palay effectively used thiscombined approach to interpret the organization, cytologicalfeatures and synaptic relationships of neurons in the cerebellarcortex (Palay and Chan-Palay, 1974). Because the cerebellarcortex containedonly a fewdifferent kindsof neurons andhas aregularly repeating morphology, interpretation of the finestructural images was relatively straightforward compared tothe problems associated with interpretation of fine structural
259B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
images obtained from the cerebral cortex and other brainregions.
In the cerebral cortex, the characteristics of pyramidal cellswere well known from Golgi preparations, but there was littleinformation about the other kinds of neurons present, beyondthose shown in the drawings made by Cajal. As pointed out byDeFelipe and Jones (1988), “With the exception of Lorente deNó's understandably superficial account of 1938, virtually noGolgi study of significance was published on the neocortex fornearly fifty years after 1922, the year in which Lorente de Nó,Cajal's illustrious pupil, published his first – and the masterhimself published his last – paper on the cortex of rodents.”Oneimportant exception to this statement that should not beoverlooked is the study carried out by O'Leary (1941) on thevisual cortex of the cat.
For the cerebral cortex, among the first to return to analysesusing Golgi staining were Szentágothai (1969) and Valverde(1971) and as an example of the state of knowledge of corticalneurons at that time, it is noteworthy that in a 1969 publicationSzentágothai divides neurons of the cerebral cortex into twobasic types, pyramidal cells and stellate cells. He shows bothGolgi and electron microscope images and while he was clearabout the distinguishing features of pyramidal cells, includingthe fact that their cell bodies and proximal dendrites receiveaxon terminals with flattened vesicles which were believedeven then to be inhibitory (Colonnier, 1968), the description ofthe stellate cell type is very vague and in retrospect appears toapply to a pyknotic neuron. In thepreviousyear, 1968, Colonnierhad carried out an in depth survey of the types of synapsespresent in the visual cortex of the cat using formaldehydeperfused material and reinforced Gray's (1959) conclusion thatthere are two basic types of synapses in cerebral cortex.However, Colonnier advocated a different nomenclature. Hesuggested that Gray's type 1 synapses, at which the axonterminals have round vesicles and synaptic junctions withprominent postsynaptic densities, should be referred to asasymmetric synapses, and that Gray's type 2 synapses, shouldbe referred to as symmetric synapses, because of the symme-trically disposed densities at the synaptic junction. It isimportant to note that the use of aldehyde fixation resulted inthe synaptic vesicles at symmetric synapsesbeingpleomorphic,so that while some synaptic vesicles had round profiles othershad oval, or elongated profiles. Moreover, Colonnier was able toshow that pyramidal cells have only symmetric synapses ontheir cell bodies,mostly asymmetric synapses on their dendriticspines and few synapses on their dendritic shafts. Otherneurons, which Colonnier believed to be stellate cells, hadboth symmetric and asymmetric synapses on their cell bodiesand both types of synapses also occurred on smooth dendriticshafts, which Colonnier believed to belong to the stellate cells.Because lesioning pathways such as the corpus callosum led tothe degeneration of axon terminals forming asymmetricsynapses, it was generally concluded that the axon terminalsforming asymmetric synapse originated from the plexuses ofpyramidal cells, and this was in concert with the belief thatasymmetric synapses are excitatory and symmetric ones areinhibitory.
Although it was known from Golgi studies that there areseveral different types of stellate, or nonpyramidal cells, in thecerebral cortex, when a cell body of one of these neurons was
encountered in thin sections prepared for routine electronmicroscopy, it was not possible to determine to what type ofnonpyramidal cell it belonged. Norwas it possible to be certainwhere the axon terminals forming symmetric synapsesoriginated, although it was generally believed on the basis ofGolgi preparations that they originated from intracorticalplexuses of nonpyramidal cells. This view was supported bythe studies of Szentágothai (1968) who showed that whenslabs of cerebral cortex are isolated and examined 2 monthslater, many of the Gray type 1 axodendritic synapses had beenlost, but the Gray type 2 synapses on the cell bodies ofpyramidal cells persisted, suggesting that they originate fromintrinsic neurons.
Another problem associated with studies of the cerebralcortex concerned the identity of the neurons postsynaptic tocortical inputs. For example although it was known fromdegeneration studies that thalamocortical axons terminate inlayer 4 of cortex and form asymmetric synapses, the identityof the postsynaptic elements could not be determined byroutine electron microscopy. To answer these and otherquestions the obvious solution seemed to be to examineGolgi impregnated neurons by electron microscopy.
5. Golgi electron microscopy
Among the early proponents of the Golgi electron microscopetechnique were Blackstad (1965) and Stell (1965) and anaccount of the early attempts to examine Golgi impregnatedmaterial by electron microscopy has been given by Blackstad(1970). The general approach was to fix tissue in aldehydesand then to carry out a Golgi impregnation en bloc, after whichthe block of impregnated tissue was sectioned, examined bylight microscopy, and suitably impregnated neuronsembedded for electron microscopy. The precipitate producedby the Golgi reaction is silver chromate, which fills animpregnated neuron and can make thin sectioning verydifficult. Even if the solid precipitate of silver chromate doesnot fall out of the thin section, it completely obscures thecytoplasm of the impregnated neurons and when the thinsection is hit by the beam of the electron microscope thedeposit usually heats up and melts the surrounding plastic,leading to the formation of holes. Nevertheless, study of suchpreparations produced some useful data, but the difficultiesinherent in using this procedure led to various attempts toreduce or alter the silver chromate precipitate. Accounts ofsome of these attempts can be found in articles by Blackstad(1965), Peters (1981) andWaterlood (1992). Themost successfulmodification was that developed by Fairén et al. (1977), inwhich goldwas substituted for the silver chromate. Essentiallythe impregnated neurons were gold toned by immersingsections in gold chloride, after which they were transferred tooxalic acid to reduce the gold, and finally deimpregnated insodium thiosulfate that removes the silver salts. The result ofthe gold toning is that small gold particles now label theimpregnated neurons and these particles do not impede thinsectioning or obscure the cytological details. An example ofthis approach is shown in Fig. 1.
The strength of the Golgi/electron microscope (Golgi/EM)approach is that neurons can be first examined in the light
Fig. 3 – A gold-toned, Golgi impregnated plexus of achandelier cell in layer 3 of rat visual cortex. The plexusconsists of vertical strings of axonal bouton (arrows) thatextend along the initial axon segments of pyramidal cells.Scale bar=50 μm.
Fig. 1 – A transversely sectioned, gold-toned, Golgiimpregnated apical dendrite (Ap) of a layer 5 pyramidal cell.The small gold particles mark both the main shaft of theapical dendrite as well as the spines (sp) emanating from it.Scale bar=1 μm.
260 B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
microscope to determine their overall features, after whichthey can be thin sectioned and specific parts of the impreg-nated neurons examined in the electron microscope. Usingthis technique it was shown that in the cerebral cortex theaxons of both pyramidal and spiny stellate cells have termi-nals that form asymmetric synapses, while the axons of non-pyramidal cells form symmetric synapses (e.g., LeVay, 1973;Parnavelas et al., 1977; Peters and Fairén, 1978). An example ofa gold-tonedaxon terminal froma smooth stellate cell forminga symmetric synapse with the cell body of a pyramidal cell isshown in Fig. 2. Also, for the first time it was possible toexamine the synaptic relationships between two impregnatedand specifically identified neurons, namely a stellate cell thatformed synapses with a pyramidal cell (Peters and Proskauer,1980). It also was shown, as only previously suspected, that allof the axonal boutons formed by an individual neuron are ofthe same morphological type. The weakness of the Golgi/EMapproach is that the kinds of neurons that can be examineddepend upon the vagaries of the Golgi impregnation. There isalso the added problem that myelinated axons do not im-
Fig. 2 – A gold-toned axon terminal (At) from a smoothstellate cell in rat visual cortex forming a symmetric synapse(arrow) with the perikaryon of a pyramidal cell (pyr). Scalebar=1 μm.
pregnate, which is the reason that some investigators usingthe Golgi methods prefer to use young animals in which theaxons have not yet fully myelinated.
For the cerebral cortex, most of the focus of Golgi/EMwas onthe nonpyramidal neurons, of which very little was known,especially about where their axons terminate. In general terms,these studies showed that some nonpyramidal neurons, suchas the multipolar and bitufted cells, form symmetric synapseswith the cell bodies and dendritic shafts of both pyramidal andnonpyramidal neurons (e.g., LeVay, 1973; Peters and Fairén,1978; Peters and Proskauer, 1980; Fairén and Smith-Fernández,1992), while the basket cells prefer to synapse on the cell bodiesand proximal dendrites of pyramidal and nonpyramidal cells(e.g., DeFelipe and Fairén, 1982; Somogyi et al., 1983). On theother hand a type of double bouquet cell examined by Somogyiand Cowey (1981, 1984) preferred to synapse with the dendriticshafts that appeared to belong to other nonpyramidal cells. Butthe nonpyramidal cell typewith themost specific terminationswas shown to be the chandelier cell (see Figs. 3 and 4), whichsynapses almost exclusively with the initial axon segments of
Fig. 4 – Light microscopic image of the axon terminal (At)from the axon plexus of a chandelier cell forming asymmetric synapse (arrow) with the initial axon segment(ais) of a layer 3 pyramidal cell. Scale bar=1 μm.
Fig. 5 – The spine (sp) of a Golgi impregnated and gold-tonedapical dendrite (Ap) in rat visual cortex forming anasymmetric synapse with a degenerating geniculo-corticalaxon terminal (d). Scale bar=1 μm.
261B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
groups of pyramidal cells (e.g., Somogyi, 1977; Fairén andValverde, 1980; Peters et al., 1982; Somogyi et al., 1982). Theoutcome of these studies was that for the first time some of theintrinsic connectivityof the cerebral cortexwas revealed, so thatneuronal circuits could be proposed.
In another set of studies, Golgi/EM was combined withaxon terminal degeneration (Peters et al., 1977). Thus afterlesions had been made in the lateral geniculate nucleus ofthe rat, and the visual cortex Golgi impregnated, it wasshown that the degenerating thalamic afferents formsynapses with all neuronal elements contained in layer IVof visual cortex that are capable of forming asymmetricsynapses. This includes the spines of basal dendrites of layerIII pyramidal cells, the spines of the apical dendrites of layerV pyramidal cells (Fig. 5), the spines of spiny stellate cells,and the perikarya and dendrites of smooth multipolar andbipolar cells (Peters et al., 1979; Peters and Kimerer, 1981). Asimilar result was obtained when this approach was used toexamine the thalamic afferents to mouse somatosensorycortex (White, 1978; Hersch and White, 1981) and to catvisual cortex (Hornung and Garey, 1981), thus showing thatthe thalamic afferents to cerebral cortex do not terminate onspecific cortical cell types.
The outcome of these studies was that for the first time thesites of termination of afferents to the cerebral cortex wererevealed, and some of the intrinsic connectivity of the cerebralcortex was understood. Consequently it became possible tospeculate meaningfully about cortical neuronal circuits. How-ever, although the Golgi/EM studies made important contribu-tions to our understanding on neuronal circuits, the techniquehad only a short life. Itwas soon superseded by the technique ofintracellular filling, which was first introduced in 1976 (e.g.,Culheim and Kellerth, 1976; Kitai et al., 1976; Light and Durvic,1976).
The technique of recording from neurons that aresubsequently filled to reveal their morphology has rapidlyexpanded and this approach has advanced our knowledge ofcortical circuitry immensely (e.g., Somogyi et al., 1998). Andin retrospect it has become evident that Golgi impregnationsdo not impregnate neurons randomly, as had been believed
since the time of Cajal. For example, prior to intracellularfilling and the introduction of antibody labeling, whichshowed double bouquet cells to be quite common (e.g.,Peters and Sethares, 1997), only a few double bouquet cellshad been visualized (see Somogyi and Cowey, 1984).Furthermore, it has also become evident that the Golgimethod does not show the complete axonal plexuses ofmost neurons, because the axonal plexuses revealed byintracellular filling are generally much richer than thoseimpregnated by the Golgi method (e.g., Martin, 1988).
What is the future of the Golgi technique? Because ofincreased ease of use and the greater amount of informationthat can be obtained from intracellular filling, one can suspectthat few investigators will be using the Golgi method in thefuture. And interestingly electron microscopy now finds itselfin the doldrums and it has become little more than ahandmaiden to other approaches. Furthermore, while theelectron microscopists of the 1970s and 1980s strove toproduce high quality images, this is no longer a worthwhilegoal, since most editors of Journals believe that words speaklouder than pictures and reduce images to such an extent thatno details can be seen in the electron microscopic and otherimages presented in many scientific Journals. It is also truethat the fine structure of most parts of the normal centralnervous system of normal animals has been examined, sothere is little new to be seen. The future of electronmicroscopylies in detailing the binding sites of antibodies, and inexamining the brains and other tissues of animals that havebeen affected by such events as disease, increasing age, andalterations in their genetics, because electron microscopy isstill the best means to determine subtle alterations inmorphology.
Acknowledgments
I wish to thank Claire Folger for her help with preparing theillustrations and Mike Bowley for his comments on an earlyversion of this article.
R E F E R E N C E S
Blackstad, T.W., 1965. Mapping of experimental axondegeneration by electron microscopy of Golgi preparations.Z. Zellforsch. 67, 819–834.
Blackstad, T.W., 1970. Electron microscopy of Golgi preparationsfor the study of neuronal relations. In: Nauta, W.J.H., Ebbesson,S.O.E. (Eds.), Contemporary Research Methods inNeuroanatomy. Springer-Verlag, New York, pp. 162–172.
Colonnier, M., 1968. Synaptic patterns on different cell types in thedifferent laminae of the cat visual cortex: an electronmicroscope study. Brain Res. 9, 268–287.
Culheim, S., Kellerth, J.O., 1976. Combined light and electronmicroscopic tracing of neurons including axons and synapticterminals after intracellular injection of horseradishperoxidase. Neurosci. Lett. 2, 307–313.
DeFelipe, J., Fairén, A., 1982. A type of basket cell in superficiallayers of cat visual cortex. A Golgi-electron microscope study.Brain Res. 244, 9–16.
DeFelipe, J., Jones, E.G., 1988. Cajal on the Cerebral Cortex. An
262 B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
Annotated Translation of the Complete Writings. OxfordUniversity Press, New York.
Fairén, A., Smith-Fernández, A., 1992. Electron microscopy ofGolgi-impregnated interneurons: notes on the intrinsicconnectivity of the cerebral cortex. Microsc. Res. Tech. 23,289–305.
Fairén, A., Valverde, F., 1980. A specialized type of neuron in thevisual cortex of cat. A Golgi and electron microscope study ofchandelier cells. J. Comp. Neurol. 194, 761–779.
Fairén, A., Peters, A., Saldanha, J., 1977. A new procedure forexamining Golgi impregnated neurons by light and electronmicroscopy. J. Neurocytol. 6, 311–338.
Fink, R.P., Heimer, L., 1967. Two methods for selective silverimpregnation of degenerating axons and their synapticendings in the central nervous system. Brain Res. 4, 369–374.
Geren, B.B., 1954. The formation from the Schwann cell surface ofmyelin in the peripheral nerves of chick embryos. Exp. Cell Res.7, 558–562.
Gray, E.G., 1959. Axo-somatic and axo-dendritic synapses of thecerebral cortex. An electron microscope study. J. Anat. 93,420–433.
Guillery, R.W., 1970. Light- and electron-microscopical studies ofnormal and degenerating axons. In: Nauta, W.J.H., Ebbsson,S.O.E. (Eds.), Contemporary Research Methods inNeuroanatomy. Springer-Verlag, New York, pp. 77–105.
Heimer, L., Peters, A., 1968. An electron microscope study of asilver stain for degenerating boutons. Brain Res. 8, 337–346.
Hersch, S.M., White, E.L., 1981. Thalamocortical synapsesinvolving identified neurons in mouse primary somatosensorycortex: a terminal degeneration and Golgi/EM study. J. Comp.Neurol. 195, 253–263.
Hornung, J.P., Garey, L.J., 1981. The thalamic projection to catvisual cortex. Ultrastructure of neurons identified by Golgiimpregnation or retrograde horseradish peroxide transport.Neuroscience 6, 1053–1068.
Karnovsky, M.J., 1965. A formaldehyde-glutaraldehyde fixative ofhigh osmolarity for use in electron microscopy. J. Cell Biol. 27,137A–138A.
Kitai, S.T., Kosis, J.D., Preston, R.J., Sugimori, M., 1976.Monosynaptic input to caudate neurons identified byintracellular injection of horseradish peroxidase. Brain Res.109, 601–606.
Latta, H., Hartmann, J.F., 1950. Use of a glass edge in thin sectioningfor electron microscopy. Proc. Soc. Biol. Med. 74, 436–439.
LeVay, S., 1973. Synaptic patterns in the visual cortex of the catand monkey: electron microscopy of Golgi preparations.J. Comp. Neurol. 150, 53–86.
Light, A.R., Durvic, R.G., 1976. Horseradish peroxidase. Animprovement in intracellular staining of singleelectrophysiologically characterized neurons. Exp. Neurol. 53,847–853.
Martin, K., 1988. From single cells to simple circuits in the cerebralcortex. Q. J. Exp. Physiol. 73, 637–702.
Maturana, H.R., 1960. The fine structure of the optic nerve ofAnurans—An electron microscope study. J. Biophys. Biochem.Cytol. 7, 107–120.
Millonig, G., 1961. A modified procedure for lead staining thinsections. J. Biophys. Biochem. Cytol. 11, 736–739.
Nauta, W.J.H., Gygax, P.A., 1954. Silver impregnation ofdegenerating axons in the central nervous system: a modifiedtechnique. Stain Technol. 29, 91–93.
O'Leary, J.L., 1941. Structure of the area striata of the cat. J. Comp.Neurol. 75, 131–164.
Palay, S.L., 1956. Synapses in the central nervous system.J. Biophys. Biochem. Cytol. 2, 193–202 (Suppl).
Palay, S.L., Palade, G., 1955. The fine structure of neurons.J. Biophys. Biochem. Cytol. 1, 69–88.
Palay, S.L., McGee-Russell, S.M., Gordon, S., Grillo, M.A., 1962.Fixation of neural tissues for electron microscopy byperfusion with solutions of osmium tertroxide. J. Cell Biol. 12,385–410.
Parnavelas, J.G., Sullivan, K., Lieberman, A.R., Webster, K.E., 1977.Neurons and their synaptic organization in the visual cortex ofthe rat. Electron microscopy of Golgi preparations. Cell TissueRes. 183, 499–517.
Peters, A., 1960. The structure of myelin sheaths in the centralnervous system of Xenopus laevis (Daudin). J. Biophys. Biochem.Cytol. 7, 121–126.
Peters, A., 1981. The Golgi-electron microscope technique. In:Johnson, J.E. (Ed.), Current Trends in MorphologicalTechniques, vol. 1. CRC Press Inc., Boca Raton, Florida,pp. 187–212.
Peters, A., Fairén, A., 1978. Smooth and sparsely-spined stellatecells in the visual cortex of the rat: a study using a combinedGolgi-electron microscope technique. J. Comp. Neurol. 181,129–172.
Peters, A., Kimerer, L.M., 1981. Bipolar neurons in rat visual cortex.A combined Golgi-electronmicroscopic study. J. Neurocytol. 10,921–946.
Peters, A., Proskauer, C.C., 1980. Synaptic relationship between amultipolar cell and a pyramidal cell in rat visual cortex. Acombined Golgi-electron microscopic study. J. Neurocytol. 9,163–183.
Peters, A., Sethares, C., 1997. The organization of double bouquetcells in monkey striate cortex. J. Neurocytol. 26, 779–797.
Peters, A., White, E.L., Fairén, A., 1977. Synapses betweenidentified neuronal elements. An electron microscopicdemonstration of degenerating axon terminals synapsing withGolgi-impregnated neurons. Neurosci. Lett. 6, 171–175.
Peters, A., Proskauer, C.C., Feldman, M.L., Kimerer, L., 1979. Theprojection of the lateral geniculate nucleus to area 17 of the ratcerebral cortex: V. Degenerating axon terminals synapsingwith Golgi-impregnated neurons. J. Neurocytol. 8, 331–357.
Peters, A., Proskauer, C.C., Ribak, C.E., 1982. Chandelier cells in ratvisual cortex. J. Comp. Neurol. 206, 397–416.
Porter, K.R., Claude, A., Fullam, E., 1945. A study of tissue culturecells by electron microscopy. Methods and preliminaryobservations. J. Exp. Med. 81, 233–241.
Reynolds, E.S., 1963. The use of lead citrate at high pH as anelectron-opaque stain in electron microscopy. J. Cell Biol. 17,208–212.
Robertson, J.D., 1955. The ultrastructure of adult vertebrateperipheral myelinated nerve fibers in relation tomyelinogenesis. J. Biophys. Biochem. Cytol. 1, 271–278.
Somogyi, P., 1977. A specific “axo-axonal” interneuron in thevisual cortex of the rat. Brain Res. 136, 345–350.
Somogyi, P., Cowey, A., 1981. Combined Golgi and electronmicroscopic study of the synapses formed by double bouquetcells in the visual cortex of the cat and monkey. J. Comp.Neurol. 195, 457–566.
Somogyi, P., Cowey, A., 1984. Double bouquet cells. In: Peters, A.,Jones, E.G. (Eds.), Cerebral Cortex, vol. 1. Plenum Press, NewYork, pp. 337–360.
Somogyi, P., Freund, T.F., Cowey, A., 1982. The axo-axonicinterneuron in the cerebral cortex of the rat, cat and monkey.Neuroscience 7, 2577–2608.
Somogyi, P., Kisvarday, Z.F., Martin, K.A.C., Whitteridge, D., 1983.Synaptic connections of morphologically identified andphysiologically characterized large basket cells in the striatecortex of cat. Neuroscience 10, 261–294.
Somogyi, P., Tamas, G., Lujan, R., Buhl, E.H., 1998. Salient featuresof synaptic organisation in the cerebral cortex. Brain Res. Rev.26, 113–135.
Stell, W.K., 1965. Correlation of retinal architecture andultrastructure in Golgi preparations of goldfish retina. Anat.Rec. 153, 389–397.
263B R A I N R E S E A R C H R E V I E W S 5 5 ( 2 0 0 7 ) 2 5 6 – 2 6 3
Szentágothai, J., 1968. The possible histological basis of inhibition.In: Asratyan, E.A. (Ed.), Progress in Brain Research, 22.Amsterdam, Elsevier, pp. 148–160.
Szentágothai, J., 1969. Architecture of the cerebral cortexIn: Jasper, H., Ward, A.A., Pope, A. (Eds.), Basic Mechanisms ofthe Epilepsies. Little Brown, Boston, pp. 13–28.
Valverde, F., 1971. Short axon neuronal subsystems in the visualcortex of the monkey. Int. J. Neurosci. 1, 181–197.
Waterlood, F.G., 1992. Techniques for converting Golgi precipitatein CNS neurons into stable electron microscopic markers.Microsc. Res. Tech. 23, 275–288.
Watson, M.L., 1958. Staining of tissue sections for electronmicroscopy with heavy metals: II. Application of solutionscontaining lead and barium. J. Biophys. Biochem. Cytol. 4,727–729.
White, E.L., 1978. Identified neurons in mouse Sml cortex whichare postsynaptic to thalamocortical axon terminals: Acombined Golgi-electron microscopic and degeneration study.J. Comp. Neurol. 181, 626–642.
F U R T H E R R E A D I N G
Cajal, S. Ramon y, 1911Histologie du SystèmeNerveux del'Hommeet des Vertebres. Trans. L. Azoulay Paris, Maloine.
Fernández-Moran, M., 1950. Electron microscope observations onthe fine structure of the myelinated nerve fiber sheath. Exp.Cell Res. 1, 143–149.
Glauert, A.M., Glauert, R.H., 1958. Araldite as embedding mediumfor electron microscopy. J. Biophys. Biochem. Cytol. 4,191–194.
Glauert, A.M., Rogers, G.E., Glauert, R.H., 1956. A new embeddingmedium for electron microscopy. Nature 178, 803.
Palade, G.E., 1952. A study of fixation for electron microscopy.J. Exp. Med. 95, 285–297.
Sabatini, D.D., Bensch, K., Barnett, R.J., 1963. Cytochemistry andelectron microscopy. The preservation of cellularultrastructure and enzyme activity by aldehyde fixation. J. CellBiol. 17, 19–57.
