HistochemicalJournal, I (I969), 323-360

Fixatives and fixation: a review

D. HOPWOOD
Department of Human Morphology, University of Nottingham

Received 2I October I968, and in revised form I6 January I969

Contents
Introduction.
The chemistry of fixation.
Osmium tetroxide.
Ions of other metals.
Formaldehyde.
Glutaraldehyde.
Other fixation methods.
Loss of materials from tissue during fixation, dehydration and embedding
processes.
Lipids.
Mucosubstances.
Proteins.
Nucleic acids.
Low molecular weight substances.
The process of fixation.
Time between death and fixation.
Cellular and tissue changes during fixation.
Concentration of fixative.
Buffers and hydrogen ion concentration.
Osmolality of the fixative solution.
Added salts.
Duration of fixation.
Miscellaneous,
Comparisons of fixatives.
Secondary effects of fixatives.
Other phenomena associated with fixation.
Conclusions.

323
3~4 Hopwood
Introduction
Fixation in some form lies at the beginning of many morphological techniques
from embalming to electron microscopy. The processes involved have grown
up out of empirical systems or have been borrowed from other disciplines and
the understanding of the mechanisms involved is not great.
The present review deals with the literature between I96o and August
I968. A few papers which appeared before I96o are included. For the litera-
ture before I96o the reader is referred to the various excellent reviews on
fixation which appeared at about that time: Wolman, z955; Baker, I96o;
Gersh, I959; and Pearse, I96o. Since then have appeared reviews by Lojda
(I965), Pease (z964) and Trump & Ericsson (z965), the latter two dealing
mostly with electron microscopy. For the practical aspects of the subject the
reader is referred to textbooks of histochemistry and electron microscopy
(Pearse, z96o, I968; Pease, I964).
Fixation may be described phenomenologically, that is in terms of change
using living tissue as a standard. The major processes which tissue fixation
must ideally prevent have been listed by Baker (z96o): autolysis, attack by
bacteria, and change in volume and shape especially during subsequent
preparative treatment. To these one may add]oss of tissue constituents and
change in spatial relationship of organelles and macromolecules. Ideally then,
the fixed tissue should just 'mark time'.
Fixation may also be described in terms of molecular processes such as the
formation of a macromolecular network between the various cellular con-
stituents or in more specific terms in reactions between the fixative and some
chemical group. A complete definition of fixation is difficult or even impossible
as the changes brought about are necessarily compared with living tissues
which themselves are not by any means completely characterized. At the
present time, however, a number of fixatives are available each with its own
potential usefulness, but none of which is universally applicable.

The chemistry of fixation

The chemical events which occur during fixation are beginning to be under-
stood in greater detail. The fundamental reactions have been reviewed earlier
by Baker (I96o) and French & Edsall (I945). Some of the information is not
due to work directed to the understanding of fixation but is a by-product of
biochemical investigation into the structure and denaturation of proteins and
nucleic acids by various methods. Denaturation has been defined by Scheraga
(I96o) as an intramolecular change from an ordered to a disordered state with-
out rupture of covalent bonds, due to an altered environment. Under certain
conditions denaturation may be reversible. Other data are derived from
Fixatives and fixation: a review 325
tanning (Gustavson, I956) and virological studies. The information may be
usefully discussed under the following headings, osmium tetroxide, other
metal ion containing fixatives, formaldehyde, glutaraldehyde and other
methods.

Osmium tetroxide
The chief use of osmium textroxide is in fixation for electron microscopy
either alone or in combination with some other agent such as glutaraldehyde.
The best established reaction of osmium tetroxide in morphological studies
is the oxidation of unsaturated bonds. Chemically this is well known (Gun-
stone, I96o). In biological material one of the most important types of un-
saturated bond occurs in lipids. Bahr (I954) showed that osmium tetroxide
reacts readily with unsaturated compounds in model experiments. This
reaction has been investigated with I-decene, oleic acid and methyl oleate by
Riemersma (I963) and Korn (I966a,b). Korn showed that a his (methyl 9,Io-
dihydroxystearate) osmate was formed from methyl oleate both in vitro and
in vivo. The latter experiments were carried out on amoebae that were fixed
in osmium tetroxide. All unsaturated fatty acids were destroyed and at least
40% of the oleic acid was recovered as bis (methyl 9,Io-dihydroxystearate)
osmate. No 9,Io-dihydroxystearate was found. He concluded that osmium
tetroxide probably converted unsaturated fatty acids to stable glycol osmates.
More recently, Korn (I966c) has extended this work in model experiments
using methyl oleate, oleic acid and dioctadecenoyl choline with methyl 9,Ir-
dihydroxystearate and 9,xo-dihydroxystearic acid as controls in reactions with
osmium tetroxide. He found that in all cases the identified products were
diesters of osmic acid in which one molecule of osmic acid links two molecules
of fatty acid. These all have the common structure shown in formula I
(fig. I). This gives experimental evidence in favour of the hypotheses put
forward by Baker (r96o) and Stoeckenius & Mahr (I965). The reaction with
di-II-octadecenoyl phosphatidyl choline may either result in an internal
diester or the methanol-insoluble compound formed may represent a polymer
with the osmic acid as the cross-linking agent.
Riemersma & Booij (I962) investigated the reaction of osmium tetroxide
with lecithin. They showed that it reacted quantitatively with the double
bonds and that there was an associated change in the polar group manifested
as a part removal of the positive charge. Riemersma (I963) concluded that
the osmium may be deposited at the polar end of the molecule where it might
combine by coulombic forces with the quaternary ammonium group of
lecithin. The chemistry of the reaction between lipids and osmium tetroxide
and its relation to the electron microscopy of membranes have been reviewed
by Korn (I966c),
Bahr (I954) found no evidence for the reaction of osmium tetroxide with
326 Hopwood
native nucleic acids. Beer, Stern, Carmalt & Mohlenrich (I966) confirmed
this finding, but pointed out that osmium tetroxide will react with denaturatcd
DNA. The reaction carried out at room temperature was shown to be pre-
dominantly with the thymine bases. They also showed that in neutral solutions
osmium tetroxide reacts with deoxythymidine5'-monophosphate and cytidine
5'-monophosphate. The reactions with adenosine and guanosine 5'-mono-
phosphate were negligible. Burton (1967) confirmed these findings at o~
He also showed that the pyrimidines of yeast amino acid transfer RNA were
oxidized by osmium tetroxide, but more slowly than the corresponding
mononucleotides.
It was shown by Bahr (I954) that the reaction of osmium tetroxide with
proteins depended on their content of histidine, cysteine and tryptophan.
The reactions of proteins, peptides and amino acids with osmium tetroxide
have also been investigated by Hake (1965). He showed that oxidative de-
amination takes place with evolution of ammonia. Eddy & Johns (I965)
investigated the reaction of osmium tetroxide with the proteins of red cell
membranes. They found that amino acids vary in their ability to reduce
osmium tetroxide as opposed to merely binding it. This leads to differential
staining.
Lenard & Singer (1968) have compared the effect of osmium tetroxide,
potassium permanganate and glutaraldehyde on protein conformation by
circular dichroism. Glutaraldehydefollowed by post-osmication andpotassium
permanganate obliterated most of the helical structure of the proteins. The
changes were as great as those brought about by 8 M urea. They concluded
that the significance of electron micrographs produced by fixation by these
methods should be seriously questioned.

Ions of other metals

Mercuric chloride is well established as a fixative of proteins and nucleic
acids in Britain (Pearse, 1968), although it is not so widely used in continental
Europe (Lojda, 1965). Its properties as a fixative for electron microscopy
have been shown to be poor (Baker & Luke, 1963). It has been shown by
Yamane & Davidson (1961) that DNA forms a complex with mercuric ions
associated with a spectral change and a decrease in the intrinsic viscosity.
The viscosity change, they thought, did not imply a complete disordering of
the DNA. They also showed that magnesium, manganese (ous) and copper (ic)
ions also combined with DNA but in these cases the complex was with the
phosphate groups. All these changes were reversible. Lipsett (1964) showed
that one mercuric ion combined with every two guanidine residues. Katz &
Sanfilli (I96Z) found that tobacco mosaic virus nucleic acid could be in-
activated by mercuric chloride. This could be partly reversed as was shown
by some return of infectivity.
Fixatives and fixation: a review 327
Yamane & Davidson (1962) also demonstrated that silver ions will complex
with DNA without changes in intrinsic viscosity. Gordon (I965) has shown
that the alkali metals stabilize the macromolecular structure of RNA. The
effectiveness increased in the order: Li, Na, K, Rb, Cs. Bach & Miller (1967)
showed that calcium and magnesium ions increased the stability of the double
helix of DNA, and that copper (ic) had the reverse effect.
The use of potassium permanganate apparently as a fixative is well estab-
lished. Bradbury & Meek (196o), however, claim that the fixation of tissues
for electron microscopy by permanganate is brought about by ethanol during
dehydration. They found that RNA and histones were removed and phos-
pholipid-protein complexes unmasked. These workers believed that it was
the unmasked protein which gave rise to the high contrast of the membranous
structures. Hake (I965) has also investigated Ithe reaction between potassium
permanganate and amino acids, peptides and proteins. The reactions, he found,
were associated with evolution of ammonia again by oxidative deamination.
The fixative properties of permanganates of other salts have been investigated
including those of sodium, calcium, barium and zinc (Wetzel, 1961; Larson &
Lewis, 196z; Afzelius, I96z). The zinc and barium salts gave poor results.
Wetzel (196I) claimed that the sodium salt gave better preservation of mem-
branes than potassium. Afzelius (I962) found that the calcium salt showed
double membranes with a width of xoo A compared with 75 A produced by
the potassium salt.
The reaction of potassium dichromate on amino acids and proteins has been
studied by Hopwood (I969). Even though the redox potential of potassium
dichromate lies between that of osmium tetroxide and potassium perman-
ganate it caused only slight evolution of ammonia from sulphur containing
amino acids and none from the proteins tested. The oxidation of cate-
cholamines by potassium dichromate lies at the base of the chromaffin
reaction. The product is a chrome containing melanin, the reaction proceeding
through an aminochrome phase (Coupland, I955; Hopwood & Coupland,
1968). Wood & Barrnett (I964) used potassium dichromate to differentiate
between the noradrenaline- and adrenaline-storing cells of the adrenal medulla
at the electron microscopic level. Baker (1965) found there was little preserw-
tion of ultrastructure if potassium dichromate alone was used as a fixative.

Formaldehyde
Formaldehyde is known to react with proteins, lipids and nucleic acids.
Excellent reviews of the reaction of amino acids and proteins with formalde-
hyde have been published by French & Edsall (1945), Gustavson (1956) and
Walker (1964).
The reaction of formaldehyde with lipids is less well known. Heslinga
& Deierkauf (1962) working with human brain fixed from I to 24 years
H 3--Y
328 Hopwood
investigated the changes that took place. They found that the cerebrosides,
sulphatides, phosphoinositides and sphingomyelin were unaffected. Lecithin
and phosphatidylethanolamine were broken down to lyso compounds, fatty
acids and phosphatidic acid. The lyso compounds were further broken down
with the release of a second fatty acid. Thus there was a corresponding rise in
the fatty acid content of the tissue. The total lipid content decreased with the
formation of water-soluble phosphoryl compounds. Consequently the phos-
phorus content of the total lipid extracted was very much decreased.
Jones & Austin (I966) studied the reaction of formaldehyde with un-
saturated fatty acids in model experiments and during histological fixation.
They found that the esterified fatty acids reacted less easily than the free
fatty acids. The formaldehyde reacted with the double bonds in the presence
of an acid catalyst. This eventually gave z :3-glycols and z :3-dioxanes.
After fixation there was a decrease in the number of unsaturated bonds.
The pH sensitive reaction between formaldehyde and proteins is well
known. A variety of groups in the proteins take part in this reaction (see
Pearse, z96o, I968). Under the mild conditions and relatively short time of
fixation only a small part of the capacity of formaldehyde to form methylene
bridges and addition compounds will be used. The mechanism of the reaction
between formaldehyde and amines in aqueous solution has been re-examined
recently by Kallen & Jencks (I966).
It has been shown by various workers that the reaction between formalde-
hyde and proteins involves the formation of cross-links between the molecules
giving rise eventually to an insoluble product (French & Edsall, I945;
Gustavson, z956). The kinetics and mechanisms of the reactions between
formaldehyde, glyoxal and various gelatins have been studied by Davis &
Tabor (I963) by viscometry. They found that the rate of cross-linking was
proportional to the square of the gelatin concentration and some power of
the concentration of the cross-linking agent and inversely proportional to the
hydrogen ion concentration. Davis & Tabor (I963) suggested that the
reaction between aldehydes and gelatin proceeded initially through a gelatin-
methylol compound. With formaldehyde, two molecules may be involved in
forming the cross-link. It was pointed out that a single molecule may have
steric difficulties in forming such a link. Methylene bridges are 2.4-2. 5 )k
long. As far as glyoxal is concerned it was suggested that two side chain
amino groups were concerned with the formation of cross-links. It was
pointed out that when gelatin was used at low concentrations (below 2%)
intramolecular bonding took place predominantly. This may be of importance
in fixation in some situations.
The effect of various fixatives on electrophoretic mobility of various pro-
teins was investigated by Schneider & Schneider (z967a). They found that
after IO min reaction with proteins~ formaldehyde, osmium tetroxide and
Fixatives and fixation: a review 329
methanol produced very little denaturation which they equated with change
in electrophoretic mobility. A moderate amount of denaturation was produced
in this period by ethanol, Carnoy's fluid and Bouin's fluid whereas Susa's fluid
and glutaraldehyde caused much greater changes.
The reaction of formaldehyde with nucleic acids has been well investigated,
since it forms the basis for attentuating viruses. Contributions to the under-
standing of the reactions have come with the increased interest in nucleic
acids shown by biochemists.
The reaction between formaldehyde and adenosine has been followed by
Fel'dman (1964). Two compounds were formed. One of these is rapidly
formed and labile, the reaction being reversible by dilution. The final product
was a methylene-bis-adenosine, which was stable. The formation of this
product could be inhibited by the use of a great excess of formaldehyde.
Haselkorn & Doty (I961) noted that the reaction of formaldehyde with
adenosine 5'-monophosphate proceeded more slowly than with cytidine
5'-monophosphate. The reaction of formaldehyde with nucleosides has been
investigated by Eyring & Ofengand (1967). They found that hydroxy-
rnethylation occurred immediately but reversibly. The reaction with poIy-
nucleotides (Haselkorn & Doty, 1961) proceeded in two phases. The first
was a rapid denaturation and this secondarily permitted the slower reaction
of formaldehyde with the amine groups freed from hydrogen bonding in the
helical polynucleotides. There was no damage to the main phosphodiester
chain. The reaction was completely reversible. The Arrhenius constant for
denaturation was 35 kcal/mole whereas the reaction of formaldehyde with the
amino groups had a value of 17 kcal/mole. The denaturation of the poly-
nucleotides was measured by depression of the melting temperature. For a
discussion of the methods of measuring denaturation, see Wetlaufer (1962)
and Scheraga (I96O). Boedtker (1967) showed that the formaldehyde would
attack these helical regions which composed 60 to 75% of the RNA. These
double-stranded regions could be regenerated by removing the excess formal-
dehyde either by gel filtration or by dialysis.

Glutaraldehyde
Glutaraldehyde was introduced as a fixative by Sabatini, Bensch & Barrnett
in 1963. Since then it has become increasingly used as a fixative for electron
microscopy and electron histochemistry.
The commercial preparation, which appears to be made only by Union
Carbide, contains various impurities which have been thought to include
glutaric acid, acrolein, glutaraldoxime and various polymers (Anderson, 1967).
Various methods have been suggested to remove these substances. Fahimi
and Drochmans (I965a) proposed a method of vacuum distillation and
Hopwood (1967d) noted that glutaraldehyde could be fractionated from the
330 Hopwood
impurities on Sephadex G-Io. Anderson (I967) showed that the impurities
could be absorbed with activated charcoal.
The structure of glutaraldehyde has been shown by Yokota, Suzuki & Ishii
(I965) to be temperature dependent between --60 and +4o~ The three
forms found are shown in formulae II-IV (below). At -- 60 to o~ 8o% of

- CH - CH -
I I
0 0
\ /
/ \ O
O s =
0 o
I I
- CH -- CH --
I

- - ~ O - -

II

-~ocH2(CH2!3CO~OCH2(CH2/,O]--~CO(CH,/3co]-
III

-CH [(CH2)3 CHO] O-

IV

glutaraldehyde is in form II. At temperatures above o~ form III pre-

dominates. Only small amounts of form IV are found.
The reaction of glutaraldehyde with protein and other biological materials
has been investigated by Bowes & Cater (I966). Glutaraldehyde appears to
react chiefly with the amino groups of lysine. The guanidino groups of
arginine only reacted at a pH greater than 9 to Io. Tyrosine was still found
in acid hydrolysates of tanned collagen and it was thought that there was
little reaction between this residue and glutaraldehyde. Similar findings were
reported by Quiocho & Richards (I966) who investigated the behaviour of
glutaraldehyde cross-linked carboxypeptidase A. However, evidence has been
brought forward by Hopwood (I968a) that glutaraldehyde reacts with the
amino acids tyrosine, tryptophan and phenylalanine. The reaction between
glutaraldehyde and proteins and noradrenaline is affected by pH (Bowes &
Cater, I966; Hopwood, I968a; Coupland & Hopwood, I966). At pH levels
greater than 8.o, glutaraldehyde undergoes rapid polymerization. At about
pH 6.5, there is a change in the gel filtration properties of glutaraldehyde sug-
gesting some intra- or intermolecular change (Hopwood, I967d).
Fixatives and fixation: a review 33 I
The reaction between proteins and glutaraldehyde may be followed by
changes in the ultra-violet spectrum between z5o and 3oo nm (Hopwood,
I968a). The reaction was not reversible to any extent and the kinetics were
pseudo-first order. The Arrhenius constants for the reactions between
glutaraldehyde and bovine serum albumin, casein and acid phosphatase were
about I x kcal/mole. This might imply that the proteins were not denatured
to any great extent during fixation as their values are about half those found
in denaturation (Casey, 1962; Haselkorn & Doty, 196I). However, against
this, Lenard & Singer (1968) have shown that glutaraldehyde causes con-
siderable obliteration of helical structure in bovine serum albumin, whale
apomyosin and proteins of red blood cell membranes.
However, there is other evidence from optical rotatory dispersion experi-
ments coupled with biochemical assays which shows that glutaraldehyde has
little effect on chloroplast proteins (Park, Kelly, Drury & Sauer, 1966). Also,
Kaldor & Weinbach (1965) found little change in the helical content of myosin
A fragments from enzyme studies which had been cross-linked with glu-
taraldehyde. There was also evidence of protein tertiary structure remaining
intact after glutaraldehyde fixation.
Changes in molecular structure during fixation have also been followed by
noting the decrease in enzyme activity after treatment with glutaraldehyde.
Sabatini et al. (1963), in semiquantitative experiments, thought this loss was
only slight.
Quantitative experiments have been carried out on the following enzymes in
tissues:/%galactosidase, N-acetyl-fi-glucosaminidase, acid phosphatase, fl-glu-
curonidase, catalase, various dehydrogenases and aminotransferases by Janigan
(1964, 1965), Hopwood (1967a) and Anderson (I967). In each case the loss
of enzyme activity with glutaraldehyde was about twice that after formalde-
hyde treatment. Quiocbo & Richards (1964) have used the cross-linking prop-
erty of glutaraldehyde to form stable crystals of carboxypeptidase A; X-ray
crystallography showed no evidence of a marked disordering of the crystal
and loss of activity varied with the crystal size. They thought that the loss of
activity was due to severe limitations in substrate diffusion. They also noted
that the solutions were yellow and attributed this to Schiff base formation.
Hopwood (1968a) investigated changes in albumin conformation after fixation
with glutaraldehyde, formaldehyde and e-hydroxyadipaldehyde by immuno-
logical techniques. Glutaraldehyde was found to produce much more change
in antigenic structure than formaldehyde or ~-hydroxydipaldehyde on reaction
with an anti-bovine serum albumin serum.
The ease of reaction between glutaraldehyde proteins and amino acids has
led to the production of an artefact which is of considerable importance in
autoradiography using labelled amino acids. Peters & Ashley (1967) showed
that glutaraldehyde caused the binding of 25 % of free leucine to serum albu-
332 Hopwood
min in solution compared with 0.5% binding caused by formaldehyde. Bind-
ing to liver slices in the presence of puromycin to prevent the incorporation
of leucine into proteins was thirty times higher with glutaraldehyde and six
times higher with osmium tetroxide compared with formaldehyde. Calcula-
tions showed that in autoradiographs prepared after fixation with glutaralde-
hyde, osmium tetroxide or formaldehyde, 62, 25 and 4% respectively of the
grains were due to the binding of free amino acids.
The reactions of glutaraldehyde with lipids appear to be slight (see p. 334,
where the reactions with mucosubstances are also discussed).
Other aldehydes have been introduced or discussed as fixatives. They
include acetaldehyde (Novikoff et al., 196o), methyl glyoxal, mono-, di- and
tri-glycodialdehyde, acrolein, crotonaldehyde (Halbhuber, 1966), ~-hydroxy-
adipaldehyde, malonaldehyde, malialdehyde and succinaldehyde (Gustavson,
1956; Sabatini et al., 1963; Sabatini et al., 1964). Many of these give satis-
factory fixation for electron microscopy. They can be arranged in a series
along with glutaraldehyde and formaldehyde with decreasing excellence as
ultrastructural fixatives and in cross-linking ability (Bowes & Cater, 1965).
The enzyme activity retained increases inversely.
Gustavson (1956) has pointed out that the tanning efficiency of the alde-
hydes appears to depend on the presence of reactive hydrogen atoms, this
was also suggested by Hopwr (i967d) in a study on the effect of pH on
the gel filtration of glutaraldehyde. Aldehydes show a marked tendency to
polymerize.

Other fixation methods

Since 196o, a variety of fixatives have been introduced. Various bifunctional
reagents have been introduced in the fields of protein chemistry and tanning;
Wold (1967) has reviewed some of these. Cross-linking in histochemical
fixation is a heteropolymerization, that is, the formation of intermolecular
links between proteins of different nature and other reactive molecules such
as amino acids. Various factors have been found to favour intramolecular
cross-linking. These include 'long' reagents, low protein concentration and a
high net charge on the proteins. Some of the reagents, such as the bifunctional
aryl halides, are unfortunately insoluble in water and will not be useful
histochemically. Needles (1967) investigated chemically the cross-linking of
gelatin by aqueous peroxydisulphates.
The chlorc-s-triazines have been used in the tanning industry and since
introduced as histological fixatives by Goland & Engel (1963) and Zerlotti
(1967); they give satisfactory results.
The difficulties of fixation of carbohydrates have been outlined by Curran
(1964). One of the most useful fixatives recently introduced for acid muco-
polysaccharides are the cetylpyridinium compounds (Scott, 1955). The use-
Fixatives and fixation: a review 333
fulness of these compounds were compared and contrasted against formalde-
hyde containing various additives by Conklin (1963) and Zugibe (1963).
Terzakis (1968) has found that uranyl acetate is a useful fixative for malarial
parasites.
The fixative properties of the widely used anti-histamine, promethazine
hydrochloride (Phenergan), were systematically investigated by MacDougall
(1961). This substance had been shown to inhibit gastric secretion by toxic
effect rather than by competitive inhibition. It was shown to be a useful
fixative at a concentration of 1%.
Malta (I962) introduced p-toluene sulphonic acid as a fixative. This
reagent is a potent protein precipitant which reacts with nucleic acids and is
lipid-soluble. The author recommended it for fixing the central nervous
system. Good cytological detail was obtained.
The use of diazonium salts for tissue fixation was described by Hess &
Pearse (I961/62) and Nagatsu et aL (I966). The latter authors found that
Fast Red B caused the retention of a buffer extractable portion of particle
bound leucine aminopeptidase. Quiocho & Richards (I966) used biphenyl
4,4'-bisdiazonium dichloride to cross-link carboxypeptidase A to the solid
state.
Gersh (1964) showed that the lipids in membranes could be preserved by
cross-linking after freezing by vapours of bis-phenylhydrazine.
Barnett & Cusick (1966) have advocated the use of controlled heat for
fixing blood smears and thin tissue sections. They used mercury heated to
55~ and found that various histochemical enzyme reactions remained posi-
tive. Flaming as used in bacteriology gave rise to uncontrolled temperatures.
Pinheiro & Plockner (1963) also advocate heat fixation for blood films, but
at I5o~
Pease (1966) found that tissues could be preserved in an unfixed state by
dehydration with various 'inert' substances such as glycerine and glucose.
These tissues then showed good cytological detail under the electron micro-
scope. This technique could be of value for electron histochemistry and
immuno-electron histochemistry.
Model experiments on the fixation of extracted and bacterial DNA were
carried out by Schreil (1964) using osmium tetroxide, uranyl acetate and
indium chloride. He found the last two agents gave an artefactual orientation
of the DNA.
Elbers et al. (1965) performed experiments on tricomplex fixation of
phospholipids. It was pointed out that there was little knowledge of the effects
of the normal preparative treatments for electron microscopy on membranes
by such agents as potassium permanganate or osmium tetroxide. Further
these agents can only fix unsaturated lipids. These workers developed a
system which allowed electron microscopic studies of phospholipid medels
334 Hopwood
to be made using ox brain phospholipids in which their structure was un-
altered. Links were formed between phosphatide amphoions by ion pairs of
calcium and heptamolybdate. However, Morgan & Huber (1967) found that
tricomplex fixation of guinea pig lung was associated with a 51% loss of
phospholipids.
Attempts have been made by some workers to perform histochemical
reactions in general without fixation. Wyllie (i965) recommended fixation of
sections after incubation. Altmann & Chayen (I965) reported that incubation
of sections in polyvinyl alcohol prevented diffusion of enzymes.
Another successful approach to the problem of fixation is freeze-substi-
tution. This method has been reviewed recentIy by Pearse (I968), who pointed
out that there were three major areas in the technique: conditions of freezing,
conditions of substitution and use of fixatives. Only the last is really pertinent
to the present review. It is agreed that the rate of fixation falls with tempera-
ture but the temperature below which fixation ceases is unknown. Various
anhydrous fixatives have been used including osmium tetroxide, mercuric
chloride and trichloracetic acid.

Loss of materials from tissue during fixation, dehydration

and embedding processes
There is an increasing literature on the loss of substances from tissue during
fixation and dehydration. The importance of this sort of study cannot be
over-emphasized. A large number of analytical techniques have been applied
to this problem.

L/p~
Loss of lipid is a problem to workers investigating nervous tissue. Roozemond
(I967) investigating the hypothalamus found that about 50% of the lipid was
extracted by water. Formaldehyde with calcium chloride decreased this loss.
The lipid which remained in the tissues was mainly in the form of phos-
pholipid. Levy et al. (I965) found that glutaraldehyde had little effect on
extraction of brain lipids compared with the control. The effect of formalde-
hyde on rat and human brain lipids was studied by Deierkauf & Heslinga
(~962). They showed that lipids can still be extracted with a chloroform :
methanol mixture after fixation of the brain with formaldehyde plus calcium
chloride or uranyl nitrate. Sphingomyelin cerebroside and sulphatides were
not affected, but lecithin and phosphatidyl-ethanolamine showed marked
losses. As a result, cholesterol, cholesterol esters, glycerides and fatty acids
increased in amount proportionally.
Lipids were also shown to be lost from a variety of other sites during
fixation and the subsequent preparation for microscopy. This loss has bcea
Fixatives and fixation: a review 335
demonstrated largely by the use of radioactive labelled lipid. Dermer (x968)
demonstrated that the loss of fed ~H-oleic acid and its metabolites fromhamster
intestine during osmium tetroxide fixation was about I6% - mainly as neutral
tipids and phospholipids. The effect of various lipid solvents used in tissue
preparation on liver intracellular osmiophilic droplets was studied by Ash-
worth et al. (I966). They found that formaldehyde- or glutaraldehyde-fixed
liver that was post-osmicated revealed the same osmiophilic dense particles
as liver fixed primarily in osmium tetroxide. When the aldehyde-fixed material
was extracted with ethanol, acetone or a chloroform : methanol mixture and
then post-osmicated, the membranes disappeared whilst many other cyto-
plasmic structures remained undisturbed. Chemical studies showed that after
aldehyde-fixation 43 to 93% of the lipid and I to 6% of the protein was
extracted by lipid solvents. After osmium tetroxide fixation, the lipid loss was
reduced to approximately 7%. Morgan & Huben (I967) investigated the loss
of 3H-phosphatidylcholine from the lungs of new-born guinea pigs using
four methods of fixation. Osmium tetroxide fixation was associated with a loss
of 67% radioactive lipid, formaldehyde followed by post-osmication 47%,
glutaraldehyde followed by post-osmication 39 %, and tricomplex by 51%.
Korn & Weisman (I966) studied the effect of different fixatives and the loss
of lipids from amoebae during preparation for electron microscopy. The
fixatives studied were glutaraldehyde, potassium permanganate and osmium
tetroxide alone and in combination. After glutaraldehyde fixation alone, the
lipids were almost unchanged but were almost completely extracted by
ethanol. This is in agreement with the findings of Levy et aI. (I965). After
permanganate fixation, all neutral lipid and 25% of the phospholipid was
extracted by dehydration. Osmium tetroxide allowed the extraction of some
phospholipid and most neutral lipid whilst it destroyed all unsaturated fatty
acid.
Cope & Williams (I968), in a comprehensive study, investigated the loss of
lipids from peritoneal macrophages during ten different fixing, dehydrating
and embedding schedules. These included different fixatives, dehydrating
agents, methacrylate and epoxy resins used at varying temperatures. No pro-
cedure allowed the retention of all neutral lipid. They stressed that electron
micrographs of tissue prepared by several different techniques should be
compared. There is general agreement between this work and that of Korn &
Weisman (I966) but disagreement with that ofAshworth et al. (I966). Cope &
Williams (x968) concluded that better preservation oflipids might be achieved
if temperatures of about --2o~ were used, or alternatively cold curing of the
epoxy resins.
A successful attempt to overcome some of these problems has been made
by Witske & Ross (I965) in an investigation into the autoradiographic
!oc~lization of lipid- and water-soluble compounds, namely 3H-asprin and
336 Hopwood
ZH-oest~adiol. Rat tissues were quick-frozen, frozen-dried and fixed in
osmium tetroxide vapour and vacuum-embedded in Epon.
Studies of membrane structure also demand that the lipid is not removed if
their morphological integrity is to remain. This has been achieved by freeze-
etching. The results of such studies on cell membranes have recently been
reviewed by Glauert (I967), who discussed the presence of substructures in
close-packed forms with pores in between; this arrangement is compatible
with many of the findings of transport through membranes.

Mucosubstances
The loss of mucins from rat submaxillary glands has been investigated by
Tock et al. (I966), who found that freeze-drying followed by formaldehyde
vapour gave good fixation. These findings were confirmed chemically. The
general difficulty in fixing mucosubstances has been pointed out by Curran
(i964) and is reflected in the large number of fixatives recommended.
Kugler & Wilkinson (I964) investigated the effect of various fixatives on
glycogen. The best fixative was ice-cold 8o0/0 ethanol. All solutions, except
Susa's and Zenker's fluids, preserved more than 75 ~o of the tissue glycogen.
Hopwood (I967a), using rat liver, also found that formaldehyde preserved
75% of the glycogen and glutaraldehyde 65%. It appears that the most suc-
cessful way of preventing the loss of glycogen from tissues is the use of
Rossman's fluid.

Proteins
The loss of proteins from biological material is well known. Ostrowski et al.
(I96I) investigated quantitively the loss of proteins from liver and kidney
treated by different chemical and physical means. The loss of proteins after
fixation with various agents were: formaldehyde, o; Carnoy, 1.4%; ethanol,
8.3 o/o; acetone, 13 ~o. Freezing techniques resulted in the loss of much protein
(32 to 39~/o). Using chemical and electron microscopic techniques, Amster-
dam & Schramm (I966) have demonstrated that protein is lost from pancreatic
and parotid zymogen granules when fixed in osmium tetroxide but not when
fixed in glutaraldehyde. Dallam (I957) also found a loss of protein from
isolated organellles of liver, kidney and heart during fixation in osmium
tetroxide. The value of glutaraldehyde in binding tissue proteins to the
tissues has been noted by Hopwood (r968a), who found it very much more
efficient than formaldehyde or 0~-hydroxyadipaldehyde. Stiles & Crane (I966)
have shown that the mitochondrial elementary particles, which are visible
after negative staining, are destroyed after glutaraldehdye fixation in the sub-
sequent stages to sectioning.
The solubility of proteins and DNA in non-aqueous solvents is well estab-
lished biochemically (Singer, r96z).
Fixatives and fixation: a review 337
Nucleic acids
Merriam (I958) compared the loss of proteins and nucleic acids from tissues
during fixation with formaldehyde and acetic acid-ethanol. After formalde-
hyde fixation, 6~o protein was lost from liver, 4% from muscle and I2~o from
the brain. Acetic acid-ethanol resulted in no loss of nucleic acids but the same
amount of protein was lost. Holtzman (I965) investigated the solubilities of
histone fractions in perchloric acid, ethanol-hydrochloric acid, hydrochloric
acid, and formaldehyde. He concluded that they provided a way of extracting
the different fractions which probably depended upon their arginine/lysine
content. The loss of nucleic acids, proteins and lipoproteins from human brain
has been investigated by Schneider & Schneider (I964, I967b).

Low molecular weight substances

Loss of low molecular weight substances into fixing fluid is a serious problem.
Schneider & Schneider (1967b) have noted the loss of purines, phosphate esters,
carbohydrates and indole derivatives after fixing brain in formaldehyde or
Carnoy's fluid. The loss of catecholamines from the adrenal medulla has been
investigated by Couptand & Hopwoed (I966) in rat. Using a glutaraldehyde
fixative they found that adrenaline, a secondary amine, was not fixed at all
whereas more than 9O~o of noradrenaline, a primary amine, remained in situ.
Hopwood (i~67b) studied the loss of adenosine triphosphate (ATP) and
catecholamines from adrenals fixed in formaldehyde. Fixation alone caused
the loss of 30 to 4o% of these compounds. Dehydration brought the loss of
ATP and catecholamine to 9O~o. Similar losses of 1~C- and 3H-labelled cate-
cholamines were recorded by Hempel (I965) in mice after formalin-calcium
fixation. Fixation in dichromate, however, retained 60 to 8o~ of the label. In
a series of experiments using isolated ox chromaffin granules, Hopwood
(I968c) showed that the loss of ATP, catecholamines and soluble proteins
was dependent on pH.
The shortcomings of the various fixatives in immobilizing substances
within the tissue is obvious from the discussion above. This loss also raises
problems of false localization within tissues and cells due to movement such
as diffusion. The proper study of tissues by morphological techniques should
include an analysis of the fixing, dehydrating and embedding media as has
been suggested by Arnold & Schneider (I966) and Cope & Williams (I968)
and implied by many other workers.

The process o f fixation

Although some fixatives may be used in the vapour phase, e.g. formaldehyde
or osmium tetrox~de, they are mostly used as aqueous solutions; these often
contain various other components especially in fixation for electron micro-
scopy. The additional components usually comprise a buffer, some osmotic-
338 Hopwood
ally active substance and various salts. However, some workers such as
Malhotra (I96z) have used a simple aqueous solution of osmium tetroxide
to examine ultrastructure. The general appearance was claimed to be the same
as with buffered osmium tetroxide. Afzelius (I962) has studied the use of
other solvents including acetone, carbon tetrachloride, glycerine and pyridine.
Some of the results obtained using carbon tetrachloride were very similar to
those with aqueous osmium tetroxide.

Time between death and fixation

The interval between the death of the tissues and the onset of fixation is
important. To reduce this time to the minimum, perfusion fixation was intro-
duced for electron microscopy by Palay et aL (I962) although as these workers
pointed out, this technique had been used previously by various neuro-
anatomists. The effect of this interval on ultrastructure has been investigated
by various workers. Karlsson & Schultz (I966) investigated the rat brain
using glutaraldehyde perfusion with delays of up to r hr after exsanguination
of the rat before perfusion.
The post-mortem changes have also been investigated in the liver by Ito
(I962) and Trump et al. (I962). These studies indicated that if the temperature
is reduced, the onset of autolysis is not as rapid as has been commonly sup-
posed. The organelles, however, may become increasingly fragile.

Cellular and tissue changes during fixation

The response of cells and tissues during fixation is of interest in the light it
may throw on the final product. Elbers (I966) has studied the effect of fixa-
tives on the ion permeability of the egg of Limnea stagnalis L. using a micro-
conductimetric technique. Within a few seconds of fixation with osmium
tetroxide, the cell membrane became completely permeable to ions, thus
questioning the necessity or value of tonicity additives for osmium tetroxide
buffer systems. Fixation with glutaraldehyde, however, permitted ions to
leave the eggs only very slowly. Elbers pointed out that osmium tetroxide
treatment would mean that cytoplasmic colloids were fixed in the absence of
ions which were normally present and had a determining influence on the
spatial relations of the colloids. Further, ions of the buffers would enter the
cells and would be either totally foreign or in unphysiological concentrations.
After fixation the cytoplasmic contents had a net negative charge. Tooze
(I964) made similar observations working with various proteins.
Jard et al, (I966) made an interesting comparative study on the effect of
various fixatives on frog bladder. They studied the net transfer of sodium ions
and water and the effect of stimulation with hormones before and after
fixation. They also observed the different morphological changes. They found
Fixatives and fixation: a review 339
that fixatives may be classified into three groups according to their physio-
logical effect on the bladder.
(i) complete permeability - ethanol and potassium permanganate.
(ii) part preservation of impermeability to sodium and no response to
oxytocin on water permeability - osmium tetroxide and Bouin's fluid.
(iii) preservation in part of all the properties of the epithelium - formalde-
hyde and glutaraldehyde.
A comparison between the effects on permeability and the morphological
preservation showed that the two aspects were not apparently correlated. The
continuity of the epithelial cell membrane was preserved in each case.

Concentration of fixative
The effect of varying the concentrations of fixatives has been investigated by
various workers. Maunsbach (I966) found very little difference in the effect-
iveness of different concentrations of glutaraldehyde provided the buffer
system had the appropriate osmolality (see below). He noted that glutaralde-
hyde was an effective fixative down to o.z5%. This has been confirmed by
Hopwood (I967d). Schultz & Karlsson (i965) obtained similar results whilst
investigating the brain.
Baker & McCrae (I966) noted the effect of various concentrations of
formaldehyde on the fixation of exocrine cells of mouse pancreas. They found
that low concentrations (0.25 %) or short times of fixation gave irregularities
in the endoplasmic reticulum but mitochondria appeared to be insensitive
to changes in concentration of formaldehyde from I to I6~o. In general,
providing the time exceeded 45 min, there was little difference in results.
These workers also investigated the effect of temperature on fixation. They
found little difference in tissue morphology when fixation was performed at
a temperature between o and 45~ although fixation for electron microscopy
is traditionally carried out in an ice-bath at o to 4~ Palade (I956) suggested
that the reduction in extraction of materials and decreased rate of cytolysis
were justifications for the adoption of this temperature. However, higher
temperatures may be useful for some areas of investigation (Thomas, I962).
Monis et al. (1965) investigated the effect of different concentrations of
glutaraldehyde (I to I2%) on fixation of tissues for aminopeptidase activity.
They found that low concentrations of glutaraldehyde required longer fixa-
tion times and that this resulted in shrinkage of the tissue and diffusion of
the enzyme. It was concluded that 4 to 6% of the fixative gave optimal results.
This is at variance with Fahimi & Drochmans (I965b), who obtained the best
results by fixing with 1.7 to 3.3% glutaraldehyde.
340 Hopwood
Buffers and hydrogen ion concentration
As Palade (I956) has pointed out, when osmium tetroxide is used as a fixative
its penetration through the tissues is preceded by a wave of acidity which
alters the structure of the cells. In order to prevent this type of phenomenon,
buffers are commonly used during fixation. Their individual uses are to be
found in books of histochemistry and electron microscopy (Pearse, I96o,
I968; Pease, I964). Care must be taken that the buffer does not react with
the fixative as this will remove the buffering power, e.g. barbiturate buffers
with aldehyde fixatives (Holt & Hicks, x96Ib). The buffer may react with part
of a histochemical medium, e.g. phosphate with lead salts. Alternatively the
buffer may inhibit the enzyme system being investigated, e.g.o.I M phosphate
completely inhibits glucose-6-phosphate dehydrogenase (L6hr & Waller,
I963).
Schultz & Karlsson (I965) have investigated the effect of pH on glutaralde-
hyde fixation in the central nervous system between pH 2 and n . The
extremes in this range were detrimental to the ultrastructure, but between
pH 6 and 8, there was little change in morphology. Malhotra (I963) studied
the effect of low pH (I. 5 to 3.5) on osmium tetroxide fixation and metha-
crylate embedding. The membranes and ribosomes retained some structure
even after prolonged periods at pH 1.5 although other cell structures were not
well preserved. The various merits of commonly used buffer systems for
osmium tetroxide were assessed by Wood & Luft (I965) and Trump &
Ericsson (I965). They assessed the stainability, penetration, cytology and ease
of sectioning for the following buffers: phosphate, s-collidine, arsenate,
bicarbonate, veronal-acetate and chromate. The different buffer ions appears
to produce different fixative effects with osmium tetroxide and this was
related to the loss of protein. They also found that the vehicles penetrated in
front of the fixative.
Although most fixation is mainly performed near neutrality some workers
have obtained satisfactory results at other pH levels. Helander (I962) observed
optimal fixation of gastric mucosa at pH 5.5. Coupland (I965) found the most
intense chromaffin reaction occurred when tissues were fixed at pH 5.8.

Osmolality of the fixative solution

The effect of osmolality on fixation has been investigated by various workers.
Schultz & Karlsson (I965) found that although hypertonic solutions perfuse
the central nervous system well they give rise to shrinkage and the appearance
of large extracellular spaces. Isotonic fixatives had a similar effect to hypotonic
fixatives on brain and both led to swelling of cells and resulted in poor
preservation. Hypotonic solutions also gave poor results with perfusion due
to the swelling of the vascular endothelium. The range of osmolalities investi-
gated were I5 o to I2OO mOsM. Fahimi & Drochmans (I965b) came to similar
Fixatives and fixation: a review 341
conclusions. They favoured a slightly hypertonic solution of 4oo to 450 mOsM
('isotonic' solutions are 34o mOsM). They also found that after perfusion with
glutaraldehyde, the tissues still remained labile; this has also been noted by
Hommes et al. (I966). The findings of Schultz & Karlsson (i965) correspond
to those of Elfvin (I 962) on cat splenic nerve fibres.
Doggenweiler & Heuser (i967) investigated the ultrastructure of prawn
nerve sheaths and the effect of fixative and osmotic pressure in the formation
of vesicles from thin cytoplasmic laminae. They concluded that in fixation,
artefactual vesicle formation was governed by the interplay of three factors,
viz. direct chemical effect of the fixative on the polar layers of adjacent unit
membranes, osmotic forces applied to membranes and the pre-existing natural
relationship between the membranes. They found that vesicles could be pro-
duced with any fixative providing the other conditions were varied. They
stressed the importance of comparing the results of various fixatives so that
systematic artefacts are not described as being characteristic of the natural
state.
The relationships between pH, osmolalities and concentration of fixative
solutions have been expressed in graphical form by Maser et al. (I967). Five
commonly used buffer systems were given, viz. cacodylate, s-collidine, Mill's
and S6rensen's phosphate and veronal-acetate. Measurements of osmolality
have been made either with a depression of freezing point apparatus (Tah-
misian, I964; Fahimi & Drochmans, i965b ) or with membrane osmometers
(Schultz & Karlsson, I965).

Added salts
Some fixation mixtures have added salts. Recent biochemical work on the
effect of various ions on proteins and nucleic acids is of relevance here.
Robinson & Grant (I966) found that various anions had a denaturing effect on
DNA. These included the sodium salts of trichloracetate, thiocyanate, per-
chlorate and iodide. Cations had only a small denaturing effect. This sort of
effect has also been described for various proteins. Von Hippel & Wong
(I964) studied the effect of neutral saks on thermally induced denaturation
of ribonuclease, DNA, collagen and myosin. The effects were similar in each
case. Calcium chloride and potassium thiocyanide were shown to be potent
structural destabilizers and denaturants. Salts such as ammonium sulphate
and potassium dihydrogen phosphate strongly stabilized the native conforma-
tion of the proteins. The resuks were not explained other than that salts do
react with proteins and this may alter their ordered stability. Bello & Bello
(I966) have also investigated this problem using viscometry, calorimetry and
crystallography. They concluded that lithium and calcium ions probably
acted directly on peptide or other polar groups rather than working by alter-
ing the solvent structure.
342 Hopwood
Some experiments performed by Robbins (1961) on the fixation of meta-
phase chromosomes may be relevant to this discussion. He fixed cell cultures
at pH 7.o with osmium tetroxide but in the presence of cations of different
valency and in solutions of different dielectric constants. These caused pro-
found changes in metaphase chromosomes through a flocculated form to com-
plete disappearance. Conventional fixation for electron microscopy caused
them to disappear almost completely when they were viewed by phase contrast
microscopy.

Duration of fixation
Most workers seem to use arbitrarily standardized times for fixation with
osmium tetroxide or glutaraldehyde of x to 4 hr, although longer times have
been used by some (e.g. 6 days by Filshie & Rogers, 1962). The optimum time
probably varies from tissue to tissue, with the size of the blocks and the
various buffers used. Prolonged fixation causes the loss of lipids (Heslinga
& Deierkauf, 1962) and proteins (Palade, 1962) from tissues. With increasing
length of fixation there is a proportional loss of enzyme activity (Hopwood,
i967a; Janigan, 1964; Pearse, 1968).

Miscellaneous
Some workers have advocated the addition of chloral hydrate to fixative
solutions, notably Fishman and his colleagues (Baker et al., 1958). These
workers found chloral hydrate and formaldehyde very useful in preserving
fl-glucuronidase and extended these findings to acid phosphatase, esterase
and lipase but not alkaline phosphatase. They concluded that chloral hydrate
probably reacts with protein in a similar manner to formaldehyde but
irreversibly. Chauncey, Kronman & Levitt (1964) also investigated the effect
of a formaldehyde-chloral hydrate mixture on enzyme fixation. After fixation
there was a considerable reduction of the activities of lyo enzymes, but those
of the desmo enzymes remained at a high level.
Hayashi & Freiman (I966) investigated the effects of fixation on myosin
adenosine triphosphatase (ATPase), succinate dehydrogenase and ~-glycero-
phosphate dehydrogenase. They found that if the fixative solution contained
the substrate and other additives, then the enzymes sensitive to fixation sur-
vived. For myosin ATPase they found that formaldehyde plus ATP and
sugar was successful. This may in part have been due to a decrease in the
concentration of formaldehyde by reaction with ATP.
Person (I963), investigating cholinesterases, found that a fixative containing
formalin, sucrose and ammonia gave improved localization and preservation
of the enzyme compared with the standard formalin saline fixative. The addi-
tion of the ammonia increases the pH to nearer 7 which, along with the
sucrose, will lessen cellular distortion. The greater cross-linking ability of
Fixatives and fixation: a review 343
formaldehyde at higher pH levels will permit better localization of the
enzyme (Bowes & Cater~ I965). Birge & Tibbits (1961) found that fixatives
containing sodium chloride considerably lessened cellular distortion and
shrinkage as compared with those containing sucrose.

Comparison of fixatives
During the period under review, most comparisons of the effects of fixatives
on tissues have been at the electron microscopic level. However, when
Sabatini et aL (1963) introduced various aldehydes as fixatives for electron
microscopy they also commented on their histochemical and electron histo-
chemical properties in a semiquantitative study. Formaldehyde and ~-hydroxy-
adipaldehyde preserved enzymes best but gave only fair fixation for
ultrastructure. Holt & Hicks (I96Ia), however, recommended formaldehyde
fixation for electron histochemistry of various hydrolytic enzymes. At the
present time, formaldehyde generated from paraformaldehyde has become
much used for electron histochemistry, e.g. Bradbury & Stoward (1967).
Glutaraldehyde and glyoxal give moderate enzyme preservation and good
ultrastructural morphology. Acrolein preserved only traces of some enzymes
but fixed tissues well for electron microscopy (Sabatini et al., 1963). The
excellence of morphological preservation was proportional to the number of
cross-links formed by the fixative with protein (Bowes & Cater, 1965). The
preservation of enzyme activity was, however, inversely proportional to the
number of cross-links. Quantitative studies have also confirmed that formalde-
hyde preserves various enzymes better than glutaraldehyde (see p. 331).
Comparisons of the effect of various fixatives on glycogen have been under-
taken by Trott (1961). He estimated the remaining glycogen histochemically
either by the PAS reaction o1" by Best's Carmine after fixation in various
formaldehyde solutions, ethanol, or an ethanol-acetic acid-formaldehyde
mixture, or in Bouin's, Carnoy's, or Rossman's fluids. He concluded that an
ethanol-acetic acid-formaldehyde mixture gave the best fixation and that
aqueous fixatives cannot be relied upon. Kugler & Wilkinson (1964) made
some quantitative estimates on glycogen fixation. They found that some
solutions (e.g. absolute alcohol, Gendre's fluid) made glycogen unstainable
whereas others (e.g. Zenker's & Susa's fluids) destroyed much of the glycogen.
The best fixative in their experience was ice-cold 80% ethanol. Swigart,
Wagner & Atkinson (196o) also investigated the preservation of glycogen
using a quantitative assay technique. They concluded that the best method
was to fix the tissue in Rossman's fluid and to float the sections out on the
slide in this fluid. The fixation of glycogens is due to removal of water mole-
cules rather than by action on the associated protein (Pearse, 1968).
A few papers have appeared which report the effects of various fixatives
H J--Z
344 Hopwood
on a single tissue at the light microscopic level. Belanger & Bois (1964) com-
pared the effects of fixatives on proteins, arginine, carbohydrates and iodinated
materials in the thyroid. Similarly Heinke & Hager (1966) studied the loss of
adrenal medullary melanins, produced by the chromaffin or iodate reaction.
Few workers appear to have compared the effects of various fixatives on the
electron microscopic appearance of tissues. The importance of such studies
in establishing a baseline of normal appearance is great, especially when
attempts are made to describe the changes brought about by disease pro-
cesses. One publication emphasizing this point is by Trump & Ericsson
(I965), who investigated, systematically and comprehensively, the effects of
various fixative solutions, including osmium tetroxide, aldehydes and potas-
sium permanganate under various conditions, on the ultrastructure of the
proximal convoluted tubule of rat kidney. They found that at a constant pH,
the vehicle used with osmium tetroxide exerted a profound influence on the
structures observed. This variation, they suggested, may be due to differ-
ential extraction of materials during the preparative processes. Variations in
form induced by different buffers was also noted by Wood & Luft (x965).
Different buffers with aldehyde fixatives did not seem to produce this effect.
Trump & Ericsson (1965) also point out that when any structure is being
investigated for the first time, the results of aldehyde fixation and osmium
tetroxide fixation with diffelent buffers should be compared. For a detailed
discussion on the effect of fixatives and buffers on the individual cellular
components, the reader is urged to consult their paper.
Potassium permanganate was introduced as a fixative by Luft in x956,
Since then its usefulness for demonstrating membrane structures with great
clarity has become well known. The effect is probably due to the extraction
of other materials (Bradbury & Meek, 196o). Rosenbluth (i963) , using toad
spinal ganglion, has shown that osmium tetroxide produces chains of vesicles
near the cell membrane. In permanganate-fixed material, these vesicles were
absent and in their place were continuous invaginations of the cell membrane.
The preservation of membrane systems by permanganate was also noted
by Sedar (1962) in frog gastric glands and by Tormey (I964) in ciliary
epithelium. Permanganate has been extensively used in studies on the ultra-
structure of plant cells and the parameters of fixation have been varied experi-
mentally (Mollenhauer, 1959). The differences between permanganate and
osmium tetroxide-fixed plant cells were found to be similar to those des-
cribed above for animal material. More recently, aldehyde-osmium tetroxide
has been found to give good fixation in botanical specimens (Gunning, x965).
Various papers have appeared in which the fixative properties of osmium
tetroxide and aldehydes have been compared. They begin with the work of
Sabatini et al. (x963). The ability of glutaraldehyde to fix protein was shown
both biochemically and electron microscopically by Amsterdam & Sehramm
Fixatives and fixation: a review 345
(1966). Osmium tetroxide caused the release of most soluble proteins during
fixation from isolated zymogen granules. When the granules were examined
by electron microscopy after shadow casting, they were quite flat. Granules
fixed in glutaraldehyde retained both their protein content and their spherical
shape. The sensitivity of membranes has also been compared by R6hlich
(1966). He found that in degenerating and regenerating planarian retinal
clubs, vesicles were formed at the site of a microvillous border. After glu-
taraldehyde fixation, however, the vesicles were absent and regular microvilli
were observed. The formation of membraneous whorls in embryonic hepato-
cytes has been investigated by Curgy (1968) after aldehyde-post-osmication,
aldehyde, osmium tetroxide and potassium permanganate fixation. The mem-
branous whorls were present only after double fixation. Acid phosphatase was
not found in them specifically, thus suggesting that they were not lysosomes.
Curgy suggested that the whorls formed as a result of the mobilization of lipid
during aldehyde fixation. In an attempt to overcome some of these artefacts,
Trump & Bulger (1966) suggested the use of a mixture of glutaraldehyde plus
osmium tetroxide.
Torack (1965, 1966) has compared the fixative properties of osmium
tetroxide, glutaraldehyde and ~-hydroxyadipaldehyde on the central nervous
system. He found that the intercellular space produced by glutaraldehyde or
osmium tetroxide fixation was about 15o A. Fixation with ~-hydroxyadipalde-
hyde produced a larger extracellular space, 250 A. He concluded that this was
probably a chemical effect of fixatives on proteins.
One of the earliest comparisons of the fixative abilities of osmium tetroxide
and glutaraldehyde is by Franzini-Armstrong & Porter (1964). After glu-
taraldehyde fixation they found the T system in fish muscle fibres to be a
tubular sarcolemmal derivative. After osmium tetroxide fixation, the T
system appeared as a string of vesicles. Further, the myofibrils were thicker
after glutaraldehyde fixation, as observed by negative staining, than after
osmium tetroxide. This observation was interpreted as being due to a loss
of material occurring during osmium tetroxide fixation, compared with a
smaller amount lost during glutaraldehyde fixation.
Ericsson et al. (I965) compared cellular ultrastructure after fixation with
osmium tetroxide, dichromate, glutaraldehyde and formaldehyde in the
presence of various buffers and salts. They found that lysosomes showed the
greatest changes with fixative. Clumping of chromatin along the nuclear
membrane and around the nucleous occurred with the aldehyde and dichro-
mate plus osmium tetroxide whereas osmium tetroxide gave a uniform distri-
bution. They found that dichromate reacted specifically with the granules in
the juxtaglomerular cells.
346 Hopwood
Secondary effects of fixatives
Besides their usefulness as fixatives, various of the reagents discussed above
also have secondary effects which are of use in histology and cytology. Some
of these will now be discussed.
Adams et al. (I967) claimed that osmium tetroxide could be used as a
specific histochemical reagent for demonstrating the ethylene double bonds
of liquid or solid cis-unsaturated lipids or the A5 double bonds in cholesterol.
These workers accepted that certain reducing groups of amino acid such as
sulphydryl can reduce osmium tetroxide but pointed out that steric con-
siderations largely preclude the reaction of both oxygen atoms of osmium
tetroxide with reducing groups in the rigid system of proteins found in tissue
sections. Thus the reaction with reducing groups on proteins could be ruled
out. However, Hopwood (I968b) has shown that the adrenaline storing cells
in the ox reacted strongly with osmium tetroxide. He suggested this was due
to the reduction of osmium tetroxide by the adrenaline which, in the chrom-
atfin granules, must fulfil the steric requirements mentioned by Adams et al.
(1967). Another exception was put forward by Wigglesworth (i964). Although
agreeing that the greater part of osmium tetroxide is taken up by unsaturated
lipids he showed that protein is also capable of reacting. Furthermore, when
tissues were exposed to a weak acid, chromosomal and cytoplasmic nucleic
acids took up large amounts of osmium tetroxide. Hake (i965) has also noted
the reaction of proteins with osmium tetroxide.
The ability of osmium tetroxide to form osmium black with certain com-
pounds has been exploited for electron histochemistry. The product is a
co-ordinated polymer of osmium which is amorphous and insoluble in water
and organic solvents. It has been used to enhance the contrast in lipid-
containing membranes following their treatment with osmiophilic thio-
carbohydrazide (Seligman et al., I966). It has also been used to demonstrate
the localization of cytochrome oxidase by a modified Nadi reaction (Seligman
et al., I967). Osmiophilia was made use of by Bradbury & Stoward (I967) in
a periodic acid-p-fluorophenylhydrazine reaction to demonstrate muco-
substances.
Formaldehyde may be used to demonstrate biogenic amines. Er~ink6
(1955) found that noradrenaline-storing cells in the adrenal medulla could be
made to fluoresce when frozen sections were placed in a formaldehyde solu-
tion. An extension to this basic technique, using frozen-dried material treated
with formaldehyde vapour, was introduced by Caxlsson et al. (1962). This
has been extensively exploited for investigating nervous tissue by Swedish
workers, e.g. DahlstriSm & Fuxe (1965). For a review of this method, see
Corrodi & Jonsson (1967). Modifications to make the method technically
easier to perform have been introduced; one is the method of E1-Badawi &
Schenk (1967).
F i x a t i v e s a n d f i x a t i o n : a review 347
Aldehyde fixation has been found to induce a basophilia in the adrenaline-
storing cells of the ox adrenal medulla (Hopwood, I967c ). This is probably
due to the loss of adrenaline from these cells and the unmasking of the acidic
protein chromogranin associated with catecholamine storage.
Norton et al. (1962) have found that fixation in formaldehyde or acrolein
destroyed 50 to 60% of tissue plasmalogen in 6 hrs as estimated histo-
chemically and biochemically.
Glutaraldehyde has also been used to differentiate between the noradren-
aline- and adrenaline-storing cells of the adrenal medulla at the electron
microscopic level. Two similar techniques were introduced independently
by Coupland et aI. (I964), Coupland & Hopwood (I966) and Tramezzini et
al. (1964) in which the noradrenaline-storing granules are stained intensely
while the adrenaline-storing granules do not react. Both groups showed that
the effect was due to the fact that most of the adrenaline had been leached
from the tissue during fixation and dehydration.
Joniau et al. (1968) used glutaraldehyde to cross-link protein to form an
immuno absorbent. This is useful in the isolation of pure antibodies. Blough
(1966) found that glutaraldehydes selectively destroyed the haemagglutinin
of certain strains of influenza viruses whilst preserving more than 5o% of the
neuraminidase activity in myxovirus-infected cells. The opposite effect was
noted with paramyxoviruses.
Van Duijn (1961) found that acrolein fixation followed by Schiff's reagent
will demonstrate amino groups in proteins. The mechanism involved is
believed to be the reaction between amino groups of proteins and unsaturated
bonds of acrolein; there are then induced in the tissue, carbonyl groups which
may be demonstrated with Schiff's reagents. Mayor & Jordan (1963) found
that acrolein preserved both the ultrastructure and antigenicity of viruses.

Other phenomena associated with fixation

The penetration of the fixative either into a gel or tissue is clearly an important
phenomenon. Medawar (1941) showed that fixatives obeyed the diffusion
laws, that is, the depth penetrated (d) was proportional to the square root of
time (t). The constant k, the coefficient of diffusibifity of the fixative, in the
following equation was specific for each fixative.
d = kV~
This relationship was investigated further by Dempster (196o) who showed
that the full equation was
t = k.d e
where e is an exponent relating to the rate of diffusion of the fixative. For
many fixatives, e is not far from two, but this does not apply to formaldehyde
348 Hopwood
and ethanol. Values for the constant k for the more recently introduced
fixatives are given in table I. The low values obtained for glutaraldehyde are
noteworthy. MacDougall (I96I) investigated the effect of fixative concentra-
tion on the coefficient of diffusability of promethazine hydrochloride between
0.25 and Io.o% . He found that it rose with increasing concentration. Ericsson
& Biberfeld (I967) and Chambers et al. (I968) obtained similar findings with
glutaraldehyde and Burkl & Schiechl (r968) with osmium tetroxide. Com-
paring the values of k obtained for glutaraldehyde by Ericsson & Biberfeld
(I967) and Hopwood (I967a), it can be seen that the constant also increases
with temperature. The results of Chambers et al. (I968) confirm this.
Table I. Coefficients of diffusability k for some recently introducect fixatives
Fixative Concentration of k Temp. (~ Source
fixative (%)

Glutaraldehyde 6 0.25 4 Ericsson et al. I967

3 o.I9 4
4 0.34 zo Hopwood, I967a
1.2- 4 O.5-I.O 20 Chambers et al., I968
1.2-4 0.33-0.5 4
Promethazine r o.88 20 MacDougall, I96I
hydrochloride
p-Toluene sulphonic r 7 4.5 20 Malm, I962
acid*

* This value ofk was obtained using an albumin gel. All others determined from tissues.
Flitney (I966) has shown that fixation depends on two factors, viz. rate of
diffusion of fixative and the speed of reaction between the fixative and the
protein. The rate of diffusion he claimed to be very rapid. Glutaraldehyde,
crotonaldehyde and acrolein reacted very quickly whereas ~-hydroxyadipalde-
hyde, glyoxal and acetaldehyde reacted slowly and so were not good in pre-
venting protein loss.
The formation of various zones of fixation in tissues has been studied in
the case of glutaraldehyde and paraformaldehyde by Ericsson & Biberfeld
(x967). With glutaraldehyde, they found an intermediate zone between the
well fixed and unfixed tissues. Tissue changes here did not resemble autolysis.
Fixation produced by paraformaldehyde was not found to follow its rate of
diffusion. Similar zones have also been found in tissues with osmium tetroxide.
Burkl & Schiechl (I968) have investigated quantitatively the gradient of
osmium tetroxide within gelatin and tissue blocks. They also studied the
effects of rising osmium tetroxide concentrations on living cells, as with this
slowly diffusing fixative, there will be a time when they are exposed to con-
centrations which neither kill nor fix. It is a less effective respiratory inhibitor
than formaldehyde.
Fixatives and fixation: a review 349
Volume changes associated with fixation are a well-known phenomenon.
The mechanism causing swelling of the cells is not completely understood and
is probably due to several factors. Amongst these are changes in membrane
permeability (see p. 338), inhibition of respiration, and changes in sodium
transport activity (Robinson, 196o). Christensen & Cassel (I967) found that
when collagen is denatured it occupies a greater volume. Swelling associated
with fixation in osmium tetroxide has been thoroughly investigated by Bahr,
Bloom & Friberg (1957). Tooze (1964) also found that osmium tetroxide at
pH 7.4 caused a swelling of cells (erythrocytes) by 37%. However, some
shrinkage occurred during dehydration and embedding. Bahr et al. (I957)
also investigated the change in volume brought about by formaldehyde fixa-
tion and found that it was inversely proportional to formaldehyde concentra-
tion. If specimens were left in the fixatives for prolonged periods they showed
secondary shrinkage. In both cases, lower fixation temperature eventually
resulted in great shrinkage. Lodin et al. (1967) studied the changes in volume
of nuclei in the central nervous system after various treatments, and as a
standard, compared these with the size of nuclei in smears. Cryostat sections
showed a decrease of IO~o. Formaldehyde followed by ethanol dehydration
gave a 4~ to 50% shrinkage and this increased if dehydration was carried out
in the tertiary butanol. Glutaraldehyde fixation followed by post-osmication
and paraffin embedding produced no change in size, whereas epoxy embedding
gave rise to an increase in size of about 7o~o. Glutaraldehyde alone (after
Sabatini et al., 1963) was found to cause shrinkage of tissue by Edwards et al.
(1964); this finding was confirmed by Hopwood (I967a) using rat liver and
albumin gels.
Bernard & Wynn (1964) investigated shrinkage and weight gain in slices
of various tissues and albumin-gelatin gels during fixation in formaldehyde.
The weight gain was found to be proportional to the solute concentration but
formaldehyde was apparently not osmotically active. Gels responded in the
same way as tissues but to about twice the extent. The weight gain varied with
the pH but was similar to buffer solution alone. To prevent swelling of the
tissues, the solutions had to be hypertonic. Bernard & Wynn (1965) also in-
vestigated the effect of chromate solutions on tissue slices.
Another possible effect of fixation on tissues is an interference with enzymes
used as histochemical reagents. Jonsson & Lagerstedt (1959) investigated the
inhibitory effect of formaldehyde on ribonuclease which is used in control
experiments in various histochemical methods for RNA (Pearse, 196o).
Ribunuclease was 8o% inactivated when fixation was carried out at pH 7.0,
but there was little inactivation when fixation was carried out at lower hydro-
gen ion concentrations. As has been observed for many of the reactions
between formaldehyde and biological materials, Jonsson & Lagerstedt found
that there was a rapid initial phase followed by a slow second phase. Sasse
35~ Hopwood
(I965) in some work on glycogen demonstration found that osmium tetroxide
inhibited diastase. This was not due to its oxidative effects but to its heavy
metal component.
An important feature of aldehyde fixatives is that they not only react with
amine containing biological materials but also with amine containing buffers,
e.g. barbiturates and tris. Holt & Hicks (x96xb) pointed out that the former
reacts with formaldehyde to form methylene-b# compounds which has no
buffering power in the physiological pH region.
Mayhew & Nordling (I966) found that the electrophoretic mobility of fixed
tissue culture cells differed from that of living cells. These workers noted that
formaldehyde and osmium tetroxide increased the mobility by Io to 50%.
Glaeser & Mel (1964) investigated the effectof osmium tetroxide and potassium
permanganate on the electrophoretic behaviour of rat red cells. They found
that osmium tetroxide-fixed cells behaved as unfixed cells, both having an
isoelectric point at pH 1.6. Potassium permanganate considerably altered the
pH-mobility curve. Tooze (x964) noted that osmium tetroxide altered the
electrophoretic mobility of various proteins including haemoglobin. Amino
groups were destroyed and the isoelectric point decreased. This mechanism
probably explains the changes in tissue culture cell mobility with fixation but
the findings of Glaeser & Me1 (I964) remain obscure in their interpretation.

Conclusions
During the period under review, various new fixatives have been introduced.
The most important of these is glutaraldehyde which has become very widely
used. The processes involved in fixation are becoming better understood in
chemical and physical terms, largely due to developments in biochemistry and
tanning. This understanding will become increasingly important as the resolu-
tion of macromolecutes by the electron microscope increases. The images
seen will then be able to be interpreted in the light of biochemical knowledge
of the macromolecules and thus the gap between morphology and biochem-
istry will be bridged.
Some of the fixatives have secondary uses in that they may demonstrate
chemical substances. Others apparently destroy certain biological activities
selectively.
Increasing attention is being paid to the loss of substances from biological
material during preparation. The importance of such studies cannot be over-
emphasized.
No one fixative is ideal for all situations. By comparing the results obtained
by a number of fixatives, and assessing the loss of substances during pro-
cessing, it is far less likely that artefacts will be described as constant features.
Fixatives and fixation." a review 35I

References
ADAMS, C. W. M., ABDULLA, Y. H. & BAYLISS, O. (1967). Osmium tetroxide as a
histochemical and histological reagent. Histochemie 9, 68-77.
AFZELIUS, B. A. (1962). Chemical fixatives for electron microscopy. In The interpreta-
tion of ultrastructure, Vol. I (ed. R. J. C. Harris). New York: Academic Press.
ALTMAN, F. P. & CHAYEN~ J. (1965). Retention of nitrogenous materials in unfixed
sections during incubation for histochemical demonstration of enzymes. Nature,
Lond. 207, 12o5-6.
AMSTERDAM~ A. & SCHRAMM, M. (1966). Rapid release of zymogen granule protein
by osmium tetroxide and its retention by glutaraldehyde, ft. Cell Biol. 29, 199-2o 7.
ANDERSON, P. 3. (1967) 9 Purification and quantitation of glutaraldehyde and its
effects on several enzyme activities in skeletal muscle, ft. Histochem. Cytochem. I5,
652-61.
ARNOLD, M. & SCHNEIDER, R. (1966). Dtinnschicht chromatographische Kontrolle
der Extraktions Wirkung yon Fixierungmitteln. Histoehemie 7, 26o.
ASHWORTH~ C. T.~ LEONARD, ~. S.~ EIGENBRODT~ E. H. ~ WRIGHTSMAN~ 17. J.
(1966). Hepatic intracellular osmiophilic droplets. Effects of lipid solvents during
tissue preparation.ft. Cell Biol. 31, 3o1-18.
BACH, D. & MILLER, I. R. (1967). Polarographic investigation of binding of Cu++
and Cd ++ by DI'qA. Biopolymers 5, 161-72.
BAHR~ G. F. (1954). Osmium tetroxide and ruthenium tetroxide and their reactions
with biologically important substances. Electron stains III. Expl Cell Res. 7, 457-79.
BAHR, G. F., BLOOM, G. & FRIBERG, U. (1957). Volume changes of tissues in physio-
logical fluids during fixation in osmium tetroxide or formaldehyde and during
subsequent treatment. Expl Cell Res. 12, 342-55.
BAKER, 5. K. (1960). Cytological technique: The principles underlying routine methods.
4th Ed. London: Methuen.
BAKER, 5. R. (1965). The structure produced in cells by primary fixatives. 2. Potas-
sium dichromate. Q.ftl microse. Sci. lO6, 15-21.
BAKER, J. R., HEW, H. & r IS HMAN, W. H. (1958). The use of chloral hydrate formalde-
hyde fixative solution in enzyme histochemistry.ff. Histochem. Cytochem. 6, 244-5o.
BAKER, I. R. & LUKE~ B. M. (1963). The fine structure produced in ceils by primary
fixatives. I. Mercuric chloride. Q.ftlmicrosc. Sci. lO4, l o l - 6 .
BAKER~ J. R. & MCCRAE~ 5. M. (1966). The fine structure resulting from fixation by
formaldehyde: the effects of concentration duration and temperature, ftl R. microsc.
Soc. 85, 391-9 9
BARNETT, C. H. & CUSICK, E. T. (1966). Controlled heat fixation of smears and
thin tissue slices in liquid mercury. Stain Technol. 41, 239-42.
BEER~ M.~ STERNs S.~ CARMALT~ D. & MOHLENRICH, K. H. (1966). Determination
of base sequence in nucleic acids with the etectron microscope. The thymine-
specific reactions of osmium tetroxide with deoxyribonucleic acid and its com-
ponents. Biochemistry, N . Y . 5, 2283-8.
BELANGER, L. F. & BOIS, P. (1964). A histochemical and histophysical survey of
the effects of different fixatives on the thyroid gland of rat. Anat. Ree. 148, 573-8o.
BELLO, ~. HASS & BELLO, H. R. (I966). Interactions of protein-denaturing salts with
model amides. Biochemistry, N.Y. 5, 2539-48.
BERNARD, G. R. & WYN/,Z, G. G. (1964). Weight responses of tissue slices and albumin-
gelatin gels during formaldehyde fixation with observations on the effect of pH.
Anat. Rec. I5O, 463-72.
352 Hopwood
BERNARD, G. R. & WYNN, G. G. (r965). T h e behaviour of tissue slices in chromate
solutions: a study of the comparative importance of the tonicity and osmotic
concentration of fixative systems. Q. Jl microsc. Sci. xo6, 23-3o.
BrRGE, W. J & TIBBITS, E. O. (I961). T h e use of sodium chloride-containing fixatives
in minimizing cellular distortion in histological and cytochemical preparations.
J. Histochem. Cytochem. 9, 4o9-14.
13LOUGH, H. A. (1966). Selective inactivation of biological activity of myxoviruses by
glutaraldehyde. J. Bacteriol. 92, 266-8.
o ED Tr~ER, H. (r 967). The reaction of ribonucleic acid with formaldehyde. I. Optical
absorbance studies. Biochemistry, N.Y. 6, 2718-28.
BOWES, J. A. & CAT~.R, C. W. (I965). Crosslinking of collagen. J. Appl. Chem. (Lond.)
x5, 296-3o4.
BOWES, J. A. • CATER, (3. W. (I966). T h e reaction of glutaraldehyde with proteins
and other biological materials. Jl R. microsc. Soc. 85, 193-2oo.
BRADBURY~ S. & MEEK~ G. A. (1960). A study of potassium permanganate 'fixation'
for electron microscopy. Q. ffl microsc. Sci. IOI, 241.
BRADBURY~ S. & STOWARD~ P. J. (1967). T h e specific cytochemical demonstration
in the electron microscope of periodate-reactive mucosubstances and polysac-
charides containing vic-glycol groups. Histochemie n , 7r-8o.
BtrRXL, W. & SCHIECHL, rL (r968). A study of osmium tetroxide fixation. J. Histo-
chem. Cytoehem. I6, I57-6I.
BuRToN, K. (1967). Oxidation of pyrimidine nucleosides and nucleotides by osmium
tetroxide. Biochem. J. IO4, 686-94.
CARLSSON, A., FALK, B. ~: HILLARP, N.-A. (I96~). Cellular localization of brain
monoamines. Aeta physiol, scan& 56, Suppl. I96.
CASEY, v.. j. (1962). Biophysics: concepts and mechanisms. New York: Reingold.
CHAMBERS, R. W., Bowr.ING, M. C. & GRIMr.EY, P. M. (I968). Glutaraldehyde
fixation in routine histopathology. Archs Path. 85, r8-3o.
CHAUNCEY~ H. H., KRONMAN~ J. H. ~2 LEVIT, M. A. (1964). Preservation of total
lyo- and desmo-enzyme activity in mammalian salivary glands following chloral
hydrate formalin fixation. J. Histochem. Cytochem. 12, 647-53.
CHRISTENSEN;R. G. & CANNEL, J. M. (1967). Volume changes accompanying collagen
denaturation. Biopolymers 5, 685-90.
CONKLIN, J. L. (I963). Staining reactions of mucopolysaccharides after formalin
containing fixatives. Stain Teehnol. 38, 56- 9.
coPE, G. H. & WILLIAMS, U. A. (1968). Quantitative studies on neutral lipid pre-
servation in electron microscopy, f i R. microsc. Soc. 85, 259-77.
CORRODI, I-I. & JONSSON, H. (1967). T h e formaldehyde fluorescence method for the
histochemical demonstration of biogenic amines. J . Histochem. Cytochem. 15, 65-78.
COUI'LAND, R. E. (1965). The Natural History of the Chroma)fin Tissue. London:
Longmans.
COIIPLAND, R. E. & HOPWOOD, D. (1966). The mechanism of the differential staining
reaction for adrenaline- and noradrenaline-storing granules in tissues fixed in
glutaraldehyde. ~. Anat. IOO, 227-43.
COUPLAND, R. E., PYPER, A. S. & HOPWOOD, D. (1964). A method for differ-
entiating between noradrenaline- and adrenaline-storing cells in the light and elec-
tron microscope. Nature, Lond. 2Ol, 124o-2.
CURGY, J. J. (1968). Influence du mode de fixation sur la possibilit6 d'observer des
structures my61iniques dans les h6patocytes d'embryons de poulet. J . Microscopie
7, 63-80-
Fixatives and fixation: a review 353
CURRAN, g. C. (I964). The histochemistry of mucopolysaccharides. Int. Rev, Cytol.
17, 149-212.
DAHLSTROM, A. & FUXE, K. (1965). Evidence for the existence of monoamine-
containing neurons in the central nervous system. Acta physiol, scand. 62, Suppl.
232.
DALLAM, R. D. (1957). Determination of protein and lipid lost during osmic acid
fixation of tissues and cellular particulates. J . Histochem. Cytochem. 5, 178-81.
DAVIS, P. & TABOR, B. E. (1963). Kinetic study of the crosslinking of gelatin by
formaldehyde and glyoxal. J . Polym. Sci. A I , 799-815.
DEIERKAUF, F. A. & HESLINGA, F. J. M. (1962). The action of formaldehyde on rat
brain lipids. J. Histochem. Gytoehem. IO, 79-82.
DEMPSTER, W. T. (1960). Rates of penetration of fixing fluids. J. Amer. Anat. lO7,
59-72.
DERMER, G. B. (1968). An autoradiographic and biochemical study of oleic acid
absorption by intestinal slices including determinations of lipid loss during pre-
paration for electron microscopy. J. Ultrastruet. Res. 22, 312-25.
DOGGENWEILER, C. F. & HEUSER~ J. E. (1967). Ultrastructure of the prawn nerve
sheaths. Role of fixative and osmotic pressure in vesiculation of thin cytoplasmic
laminae. J. Cell Biol. 34, 4o7-20.
EDDY, A. A. & IOHNS, P. (1965). Preliminary observations on the proteins of the
erythrocyte membrane with special reference to their cytological behaviour. Soc.
Chem. Ind. Monograph 19, 24-35-
EDWARDS, S. R., PAUL, W. ~. & POMERAT, C. M. (1964). The effect of various
fixatives and dehydrating fluids on tissue culture cells. Anat. Rec. 148 , 278.
EL-BADAWI, A. & SCHENI(, E. A. (1967). Histochemical methods for separate con-
secntive and simultaneous demonstration of acetylcholinesterase and norePine-
phrine in cryostat sections. J . Histochem. Cytoehem. 15, 58o-8.
ELDERS, P. F. (1966). Ion permeability of the egg of Limnea stagnalis L. on fixation
for electron microscopy. Biochim. biophys. Acta 112, 318-29 .
ELDERS, P. F., VERVERGAERT, P. H. J. T. & DEMEL, R. (1965). Tricomplex fixation
of phospholipids. J. Cell Biol. 24, 23-30.
ELFVi N, L. G. (I 962). Electron microscopic studies on the effect of anisotonic solutions
on the structure of unmyelinated splenic nerve fibres of the Cat. J. Ultrastruct. Res.
7, 1-38.
ERANK6, O. (1955). Distribution of adrenaline and noradrenaline in the adrenal
medulla. Nature, Lond. 175, 88- 9.
ERICSSON, ~. L. E. & BIBEREELD, P. (1967). Studies on aldehyde fixation. Fixation
rates and their relation to fine structure and some histochemicat reactions in liver.
Lab. Invest. 17, 281-98.
ERICSSON~ J. L. E., SALADINO~ A. I. & TRUMP, B. F. (1965). Electron microscopic
observations of the influence of different fixatives on the appearance of cellular
ultrastructure. Z. Zellforseh. mikrosk. Anat. 66, 161-81.
EYRING, E. J. & OFENGAND, J. (1967). Reaction of formaldehyde with heterocyclic
imino nitrogen of purine and pyrimidine nucleosides. Biochemistry, N.Y. 6, 2500-6.
FAHiMI, H. D. & DROCHMANS~ P. (I965a). Essais de standardization de la fixation au
glutarald6hyde I. Purification et d&ermination de la concentration du glutarald6-
hyde. J . Microscopie 4, 725-36.
PAHIMI, H. D. & DROCHMANS, P. (I965b). Essais de standardization de la fixation
au glutarald6hyde II. Influence des concentrations en aldehyde et de l'osmolatit~,
J. Microscople 4, 737-48.
354 Hopwood
FEL'DMAN, M. YA. (I964)- A non-enzymic model of the inhibiting action of excess
reagent. Biokhimiya 29, 616-21.
FILSHIE, B. K. • ROGERS~ G. E. (1962). An electron microscope study of the fine
structure of feather keratin. J. Cell Biol. I3, 1-12.
FLITNEY, F. W. (1966). The time course of the fixation of albumin by formaldehyde,
glutaraldehyde acrolein and other higher aldehydes. Jl R. microsc. Soc. 85, 353-64.
FRANZINI-ARMSTRONG, C. & PORTER, K. (1964). Sarcolemmal invaginations con-
stituting the T system in fish muscle fibres. J. Cell Biol. 22, 675-96.
FRENCH, D. & EDSALL, J. T. (1945). The reaction of formaldehyde with amino acids
and proteins. Adv. Protein Chem. 2, 277-335 .
GERSH, i. (1959). Fixation and staining. In The cell I, pp. 21-66 (eds. J. Brachet &
A. E. Mirsky). London: Academic Press.
GERSH, I. (1964). A cytochemical method for fixation and staining of lipids in frozen
dried mouse pancreas for study with light and electron microscopes. Histochemie
4, 322-9.
GLAESER~ R. M. & MEL~ H. C. (1964). The electrophoretic behaviour of osmium
tetroxide-fixed and potassium permanganate-fixed rat erythrocytes. Biochim.
biophys. Acta 79, 606.
GLAUERT, A. M. (1968). Electron microscopy of lipids and membranes. Jl R. microsc.
Soc. 88, 49-70-
~OLAND, P. & ~NGEL, M. (I963). Fixation of cells and tissues by chloro-s-triazines.
J. Histochem. Cytochem. I1, 751-62.
GORDON, J. A. (1965). Stabilizing effectiveness of alkali metals toward ribonucleic
acid from bacterial virus RI7. Biopolymers 3, 5-14.
GUNNING, B. E. S, (1965). The fine structure of chloroplast stroma following alde-
hyde osmium-tetroxide fixation. J . Cell Biol. 24, 79-93.
GUNSTONE, F. D. (I960). Hydroxylation methods. Adv. org. Chem. I, lO3-43.
GUSTAVSON, K. H. (1956). The chemistry of Tanning Processes. New York: Academic
Press.
HARE, T. (1965). Studies on the reactions of OsO 4 and KMnO4 with amino acids,
peptides and proteins. Lab. Invest. 14, 12o8-12.
HALBHU•ER, K. I. (1966). Zur Verwendung von Crotonaldehyde als Reagens zum
histochemischen Eiweissnachweis. Acta histochem. 24, 385-6.
HASELKORN, R. & DOTY, I'. (1961). The reaction of formaldehyde with polynucleo-
tides. J. biol. Chem. 236~ 2738-45.
HAYASHI, M. ~z FREIMAN, D. G. (1966). An improved method of fixation for formalin
sensitive enzymes with special reference to myosin adenosine triphosphatase.
J. Histochem. Cytochem. I4, 577-8I.
HEINK~, W. & HAGER, G. (1966). Der Einfluss der Fixierung auf den histochemischen
Nachweis der Catecholamine in Paraffinschnitt. Aeta histochem. 24, 337-44-
HELANDERj H. F. (1962). On the preservation of gastric mucosa. In Proceedings of th~
5th International Congressfor Electron 2kIicroscopy,Vol. 2, pp. P-7. (ed. S. S. Breese).
New York: Academic Press.
HEMPEL, K. (1965). Zur Fixierung yon radiaktivmarkierten Catecholaminen bei
autoradiographischen Untersuchungen. Histochemie 4~ 5o7-13.
HESLINGA, F. J. M. & DEIER~:AUF, F. A. (1962). The action of formaldehyde solutions
on human brain lipids. J. Histoehem. Cytochem. Io, 704-9.
HESS, R. & PEARSE, A. G. E. (1961/62). Enzyme immobilization. Substantive dia-
zonium salts in enzyme histochemistry. Enzymol biol. din. x, 87-113.
Fixatives and fixation: a review 355
HOLT, S. I. & HICKS, R. M. (I96Ia). Studies on formalin fixation for electron micro-
scopy and cytochemical staining purposes. J. biophys, biochem. Gytol. lid 31-45.
HOLT, S. ]. & HICKS, R. M. (I961b). Use of veronal buffers in formalin fixatives.
Nature, Lurid. 191, 832.
HOLTZMAN, E. (1965). A cytochemical study of the solubilities of the histones of fixed
Necturus liver. J. Histochem. Gytochem. 13, 318-27.
HOMMES, O. R., WARSHAWSKY~ H. & LEBLOND, C. P. (1966). Fixation of brain and
incisor teeth for light microscopy; glutaratdehyde perfusion followed by immer-
sion in Bouin's fluid. Stain Technol. 4I, 261-6.
HOt'WOOD, D. (I967a). Some aspects of fixation with glutaraldehyde. A biochemical
and kistochemical comparison of the effects of formaldehyde and glutaraldehyde
fixation on various enzymes and glycogen with a note on penetration of glutaralde-
hyde into iiver.J. Anat. I o i , 83-92.
HoeWOOD, D. (1967b). T h e effect of formaldehyde fixation and dehydration on ox
adrenal medulla with respect to the chromaffin reaction and post-chroming.
Histoehemie To, 98-1o6.
HOeWOOD, D. (I967c). Adrenal medullary basophilia in Ox, Pig and Sheep. A histo-
chemical, immunohistochemical and cell fractionation study. Histochemie 11,
268-79.
HOeWOOD. D. (I967d). T h e behaviour of various glutaraldehydes on Sephadex
G - l o and some implications for fixation. Histochemie Ix, 289-95.
HOPWOOD D. (1968a). Some aspects of fixation by glutaraldehyde and formaldehyde.
J. Anat. lO3, 581.
HOPWOOD D. (I968b). A histochemical analysis of adrenal lipids in ox and sheep.
Histochemie 14, 27o-81.
HOI'WOOD D. (I968c). T h e effect o f p H and various fixatives on isolated ox chromaffin
granules with respect to the chromaffin reaction. J. Anat. IO2, 415-z4.
HOPWOOD D. (1969). Amino acids in the adrenal medulla and the effect of potassium
dichromate on them. ft. Anat. IO4, 71-8o.
HOrWOOD D. & COUPLAND, R. E. (1968). T h e effect of p H on the chromaffm
reaction. J. Anat. lO2, 585-6.
ITO, S. (1962). Light and electron microscopic study of membranous cytoplasmic
organelles. I n The Interpretation of Ul~rastructure, Vol. I (ed. R. J. C. Harris).
London: Academic Press.
JANIGAN, D. T. (1964). Tissue enzyme fixation studies. I. T h e effects of aldehyde
fixation on /3-glucuronidase, /3-galactosidase, N-Acetyl-/%glucosaminidase and
/3-glucosidase in tissue blocks. Lab. Invest. 13, lO38-5o.
]'ANIGAN, D. T. (1965). T h e effect of aldehyde fixation on acid phosphatase activity
in tissue blocks.J. Histochem. Cytochem. 13, 473-83.
]'ARD, S., BOURGUET, V., CARASSO, N. • FAVARD, P. (1966). Action de divers
fixateurs sur la permfiabifit6 et l'ultrastructure de la v6ssie de grenouille. J . Micro-
seopie 5, 31-5o.
IONES, D. & AUSTIN, a. (1966). Reaction of formaldehyde with unsaturated fatty
acids during histological fixation. Nature, Lond. 2IO~ 1386-8.
JONIAU, M., ONKELINX, E. ~: LONTIE, R. (1968). Glutaraldehyde polymerized
proteins as immunosorbents. Archs int. Physiol. Biochim. 76~ 182-3.
JONSSON, ~,T & LAGERST~T, S. (I959). T h e effect of formaldehyde containing
fixatives on ribonuclease activity. Histoehemie 1, 251-6.
KALDOR~ G. • WEINBACH~ S. (1965). Use of toluidine diisocyanide and glutaric
356 Hopwood
dialdehyde in the conjugation of heavy meromyosin with light meromyosin. Archs
Biochem. Biophys. xxz, 448-52.
KALLEN, R. G. & JENKS, W. l'. (I966). Equilibria for the reaction of amines with
formaldehyde and proteins in aqueous solution, ft. biol. Chem. 241, 5864-78.
KARLSSON~ U. • SCHULTZ~ R. (I966). Fixation of the central nervous system for
electron microscopy by aldehyde perfusion. III. ft. Ultrastruct. Res. 14, 47-63.
KATZ, S. & SANTILLI, V. (1962). The reversible reactions of Tobacco Mosaic Virus
nucleic acid and mercuric chloride. Biochim. biophys. Acta 55, 621-6.
KORN, E. D. (1966a). Synthesis of his (methyl 9,1o-dihydroxystearate) osmate from
methyl oleate and osmium tetroxide under conditions used for fixation of biological
material. Biochim. biophys. Acta 116, 317-24 .
KORN, E. D. (I966b). Modification of oleic acid during fixation of Amoebae by osmium
tetroxide. Biochim. biophys. Acta. 116, 325-35.
KORN, E. D. (I966c). Structure of biological membranes. Science 153, 1491-8.
KORN, E. D. (1967). A chromatographic and spectrophotometric study of the products
of the reaction of osmium tetroxide with unsaturated lipids ft. Cell Biol. 34, 627-38.
KORN, E. D. & WEISMAN, R. A. (1966). Loss of lipids during preparation of Amoebae
for electron microscopy. Biochim. biophys. Acta 116, 309-16.
KtrGLER, J. H. & WILKINSON, W. ~. C. (1964). Quantitative studies with the solutions
used for the fixation of glycogen. Acta Anat. 56, 184-95.
LARSON, D. A. & LEWIS, C. (1962). Cytoplasm in mature, non-germinated pollen.
In Proceedings of the 5th International Congress for Electron Microscopy, Vol. 2,
pp. W - I I (ed. S. S. Breese). New York: Academic Press.
LEONARD, J. & SINGER, S. J'. (1968). Alterations of the conformation of proteins in
red blood cell membranes and insolution by fixatives used in electron microscopy.
ft. Cell Biol. 37, 117-21.
LEVY~ W. A., HERZOG~ I., SUZUKI, K., KATZMAN, R. & SCHEINBERG~ L. (1965).
Method for combined ultrastructural and biochemical analysis of normal tissue.
97. Cell Biol. 27, II9-32.
LIPSETT, M. N. (I964). Aggregation of guanine oligoribonucleotides and the effect
of mercuric salts. 97. biol. Chem. 239, 125o-5.
LODIN, Z., MARE.t, V., KARASEK, I. & SKf~IVANOVA, P. (1967). Studies on the
effect of fixation on nervous tissue. II. Changes of sizes of nuclei of nervous ceils
after fixation and after further histological treatment of the nervous tissue. Acta
histochem. 28, 297-312.
LSHR, G. W. & WALLER, H. D. (I963). In Methods of Enzymatic Analyses (ed. H. U.
Bergmeyer). London: Academic Press.
LOJDA, Z. (1965). Fixation in histochemistry. 17olia Morph. 13, 65-83.
LUFr, I. H. (1956). Permanganate - a new fixative for electron microscopy. 97. biophys.
biochem. Cytol. ~, 799-8Ol.
MACDOUGALL, J. D. B. (1961). Promethazine hydrochloride as a histological fixative.
Scot. Med. 97. 6, 457-64.
MALHOTRA, S. K. (1962). Experiments on fixation for electron microscopy. 1. U n -
buffered osmium tetroxide. Q. ffl microsc. ScL 1o3, 5-16.
MALHOTRA, S. K. (I963). Experiments on fixation for electron microscopy. 3. A note
on fixation in acidified osmium tetroxide. Q.97l microsc. Sci. Io4, 123-7.
MALM, M. (I962). p-Toluene sulphonic acid as a fixative. Q. 97l microsc. Sci. 1o3,
163-7I.
MASER, M. D., POWELL, T. E. & PHILPOTT, C. W. (1967). Relationships among
pH, osmolality and concentration of fixative solutions. Stain Technol. 4~, 175-82.
Fixatives and fixation: a review 357
MAUNSBACH, A. B. (1966). The influence of different fixatives and fixation methods
on the ultrastructure of rat kidney proximal tubule cells. II. Effects of varying
osmolality ionic strength, buffer system and fixative concentration of glutaraldehydc
solutions. J. Ultrastruet. Res. 15, 283-309.
MAYHEW, E. & NORDLING, S. (1966). The electrophoretic mobility of fixed tissue
culture cells with formaldehyde and osmium tetroxide. Expl Cell Res. 43, 72-6.
MAYOR, H. D. & JORDAN, L. (1963). A c r o l e i n - a fine structural fixative for viral
cytochemistry. J. Cell Biol. 18, 207.
MEDAWAR, P. B. (1941). The rate of penetration of fixatives. Jl R. mierosc. Soc. 61,
46-57.
MERRIAM, R. W. (I958). Standard chemical fixations as a basis for quantitative investi-
gations of substances other than deoxyribose-nucleic acid..7. Histoehem. Cytochem.
6, 43-51 .
MOLLENHAUER, H. H. (1959). Permanganate fixation of plant cells. J. biochem.
biophys. Cytol~ 6, 431-6.
MONIS, B., WASSERKRUG, H. & SELIGMAN, A. M. (1965). Comparison of fixatives
and substrates for aminopeptidase. J. Histochem. Cytochem. 13, 503-9 9
MORGAN, T. X. & HUBER, G. L. (1967). Loss of lipid during fixation for electron
microscopy. J. Cell Biol. 32, 757-6o.
NAGATSU, I., MASON, S. & GLENNER, G. G. (1966). Tissue fixation by diazonium
salts of a particle-bound rat kidney aminopeptidase. 3*. Histochem. Cytochem. 14,
663-8.
NEEDLES, H. L. (1967). Cross linking of gelatin by aqueous peroxydisulfate. 3*. Polym.
Sci. AS, 1-13.
NORTON, W. T., GELFAND, M. & BROTZ, M. (1962). Studies in the histochemistry
of plasmalogens : i. The effect of formalin and acrolein fixation on the plasmalogens
of adrenal and brain. 3*. Histoehem. Cytochem. IO, 375-86.
NOVIKOFF, A. B., SHIN, W. Y. & DRUCKER, J. (196o). Cold acetone fixation for
enzyme localization in frozen sections. 3*. Histochem. Cytochem. 8, 37-4 o.
OSTROWSKI, K., KOMENDER, J. • KWARECKI, K. (1961). Quantitative investigations
on the solubility of proteins extracted from tissues fixed by different chemical and
physical methods. Annls Histochim. 6, 5Ol-6.
PALADE, G. X. (1952), A study of fixation for electron microscopy. 3*. exp. Med. 952
285-99.
PALADE, G. X. (I956). The fixation of tissues for electron microscopy. In Proceedings
of 3rd International Conference on Electron Microscopy. pp. 129-42. London: Roy.
Micro Soc.
PALAY, S. L.0 MCGEE-RUSSELL, S. M., GORDON2 S. ~ GRILLO, M. A. (1962). Fixa-
tion of neurological tissues for electron microscopy by perfusion with solutions of
osmium tetroxide. 3*. Cell Biol. 122 385-41o.
PARK, R. B., KELLYO J., DRURY, S. & SAUER, K. (1966). The Hill reaction of chloro-
plasts isolated from glutaraldehyde-fixed leaves. Proc. hath. Aead. Sci. U.S.A. 55,
1056-62.
PEARSE, A. ~. E. (1960). Histoehemistry, theoretical and applied. 2nd Ed. London:
Churchill.
PEARSE, A. G. E. (1968). Histochemistry, theoretiealandapplied, Voi. I. 3rd Ed. London:
Churchill.
PEARSON, C. K. (1963). A formalin sucrose ammonia fixative for cholinesterases.
3*. Histochem. Cytochem. I12 665-6.
358 Hopwood
PEASE, D. C. (1964). Histological techniques for electron microscopy. London: Academic
Press.
v EASE, D. C. ( 1966). The preservation of unfixed cytological detail by dehydration with
'inert' agents. J. Ultrastruct. Res. i6, 356-78.
PETERS, T. & ASHLEY, C. A. (I967). An artefact in radioautography due to binding of
free amino acids to tissues by fixatives. J. Cell Biol. 33, 53-60.
PINHEIRO~ P. & PLOCKNER~ P. (1963). Standardized heat fixation for blood smears.
Stain Technol. 38, 227-9.
QUIOCHO, F. A. & RICHARDS, F. M. (1964). Intermolecular crosstinking of a protein
in the crystalline state: carboxypeptidase-A. Proc. natn. Acad. Sci. U.S.A. 52,
833-9.
QUIOCHO, F. A. & RICHARDS~ F. M. (1966). T h e enzymic behaviour of carboxy-
peptidase-A in the solid state. Biochemistry, N . Y . 5, 4062-76.
RIEMERSMA, J. C. (1963). Osmium tetroxide fixation of lipids: nature of the reaction
products. J . Histochem. Cytochem. Ix, 436-42.
RIEMERSMA, J. c. & BOOII, H. L. (1962). T h e reaction of osmium tetroxide with
lecithin; application of staining procedures, ft. Histochem. Cytochem. IO, 89-95.
ROBBINS, E. -(1961 ). Some theoretical aspects of osmium tetroxide fixation with
special reference to the metaphase chromosomes of cell cultures. J. biophys, biochem.
Cytol. I I , 449-55.
ROBINS ON~ D. R. & GRANT, M. E. (1966). The effect of aqueous salt solutions on the
activity coefficients of purine and pyrimidine bases and their relation to the de-
naturation of deoxyribonucleic acid by salts. J . biol. Chem. 24I, 4030-42.
ROBINSON, J. (196o). Metabolism of intracellular water. Physiol. Rev. 4o~ 112-49.
R6HtLICH, P. (1966). Sensitivity of regenerating and degenerating planarian photc-
receptors to osmium fixation. Z. Zellforsch. mikrosk. Anat. 73~ 165o-73.
ROOZEMOND, R. C. (1967). T h i n layer chromatographic study of lipid extraction from
cryostat sections from rat hypothalamus by some fixatives, ft. Histochem. Cytochem.
I5~ 526-9.
ROSENBLUTH~ J. (I963). Contrast between osmium-fixed and permanganate-fixed
toad spinal ganglion. J. Cell Biol. 167 143-57.
SABATINI, D. D., BENSCH, K. • BARNNETT, R. J. (1963). Cytochemistry and electron
microscopy. T h e preservation of cellular ultrastructure and enzymatic activity by
aldehyde fixation. J . Cell Biol. 17, 19-58.
SABATINI, D. D., MILLER, F. & BARNNETT, R. J. (1964). Aldehyde fixation for mor-
phological and enzyme histochemical studies with the electron microscope. J. Histc-
chem. Cytochem. 12, 57-7I.
SASSE, D. (1965). Untersuchungen zum cytochemischen Glykogennachweis. VII.
Mitteihing versuch zur OsO4 - bedingten Diastaseresistenz. Histoehemie 57 378-83.
SCI-IERAGA~ H. A. (1960). Protein structure. London: Academic Press.
SCHNEIDER, G. & SCHNEIDER, G. (1964). l~ber Stoffverluste bei FormoI-Fixierung
yon menschlichen Gehirnwebe. Histochemie 4, 348-50.
SCHNEIDER, G. & SCHNEIDER, G. (I967a). Elektrophoretiscke Untersuchungen fiber
die Wirkung yon Gewebe-Fixierungsmitteln auf Eiweissl6sungen. Histochemie 8,
132-54.
SCHNEIDER, G. & SCHNEIDER, G. (I967b). Qualitative und quantitative Unter-
suehungen fiber Stoffverluste bei Formol- und Carnoy-Fixierung yon menschlichem
Hirngewebe. Acta Histochem. 28~ 227-42.
SCHREIL, W. (1964). Studies on the fixation of artificial and bacterial D N A plasms
for the electron microscopy of thin sections. J. Cell Biol. 22, 1-2o.
Fixatives and fixation: a review 359
SCHULTZ~ R. L. • KARLSSON~ U. (X965). Fixation of the central nervous system by
aldehyde perfusion II. Effect of osmolarity, p H of perfusate and fixative concentra-
t i o n . J . Uhrastruet. Res. Ie, 187-2o6.
SCOTT, J. E. (I955). The solubility of cetylpyridinium complexes of biological poly-
anions in solutions of salts. Biochim. biophys. Acta I8, 428-9 .
SEDAR, A. W. (I962). Electron microscopy of the oxyntic cell in the gastric glands of
the bullfrog Rana eatesbiana III. Permanganate fixation of the endoplasmic reticu-
lum. J. Cell Biol. I4, I52-6.
SELIGMAN~ A. M.~ PLAPINGER~ R. E.~ WASSERKRUG~ H. L., DEB~ G. ~ HANKER~
I- S. (1967). Ultrastructural demonstration of cytochrome oxidase activity by the
Nadi reaction with osmiophilic reagents. J. Cell Biol. 34, 787-8oo.
SELIGMAN~ A. M., WASSERKRUG, H. L. & HANKER, J. s. (1966). A new staining
method (OTO) for enhancing contrast of lipid-containing membranes and droplets
in osmium tetroxide-fixed tissue with osmiophilic thiocarbohydrazide (TCH).
J. Cell Biol. Bo, 424-32.
SINGER, S. J. (1962). The properties of proteins in non-aqueous solvents. Adv. Protein
Chem. I7, 1-68.
STILES, J. W.~ & CRANE, F. L. (~966). Demonstration of the elementary particles of
mitochondrial membranes fixed with glutaraldehyde. Biochim. biophys. Acta i~6,
179-81.
STOECKENIUS~ W. & MAHR~ S. C. (I965). Studies on the reaction of osmium tetroxide
with lipids and related compounds. Las. Invest. *4, I I96-2o 7.
SWIGART, R. H.~ WAGNER~ C. E. • ATKINSON~ W. B. (1960). The preservation of
glycogen in fixed tissues and tissue sections. J. Histochem. Cyzochem. 8, 74-5.
TAHMISIAN, T. N. (I964). Use of the freezing point method to adjust the tonicity
of fixative solutions. J. Ultrastruct. Res. Io, 182-8.
TERZAKIS~ J. A. (x968). Uranytacetate, a stain and a fixative. J. UItrastruct. Res. z2,
I68-84 .
THOMAS, R. S. (1962). Demonstration of structure-bound mineral constituents in
their sectioned bacterial spores by nhramicroincineration. In Proceedings of the
5th International Congress for Electron Microscopy, Vol. 2, pp. RR-11 (ed. S. S.
Breese). New York: Academic Press.
TOCK~ E. P. C.~ WOODWARD~ E. M. 8~ PEARSE~ A. G. E. (I966). Preservation of tissue
mucins by freeze-drying and vapour fixation. II. Quantitative chemical studies.
Histochemie 7, 224-9.
TOOZE, J. (I964). Measurements of some cellular changes during the fixation of
amphibian erythrocytes with osmium tetroxide solutions. J. Cell Biol. zz, 551-63.
TORACK, R. M. (1965). The extracellular space of rat brain following perfusion fixation
with glutaraldehyde and hydroxyadipaldehyde. Z. Zellforsch. mikrosk. Anat. 66~
352-64 .
TORACK, R. M. (I966). The penetration of thorotrast into brain spaces following
osmium, glutaraldehyde and hydroxyadipaldehyde fixation. J. Ultrastruct. Res. 14,
590-60 I.
TORMEY, 3. McD. (1964). Differences in membrane configuration between osmium
tetroxide-fixed and glutaraldehyde-fixed ciliary epithelium. J. Cell Biol. 23, 658-64.
TRAMEZZAN1, J. H., CHIOCCH10, S. & WASSERMAN, G. F. (1964). A technique
for light and electron microscopic identification o f adrenalin- and noradrenalin-
storing cells. J. Histochem. Cytoehem. Is, 89o- 9.
TROTT, I. R. (I96I). An evaluation of methods commonly used for the fixation and
staining of glycogen. J. Histochem. Cytoehem. 9, 7o3 - I o .
H ~--AA
360 Hopwood
TRUMP, B. F. & BULGER, R. E. (I966). New ultrastructural characteristics of cells
fixed in a glutaraldehyde-osmium tetroxide mixture. Lab. Invest. 15, 368-79.
TRUMP, ~. F. & ERICSSON, J. L. E. (1965). T h e effect of the fixative solution on the
ultrastructure of cells and tissues. A comparative analysis with particular attention
to the proximal convoluted tubule of the rat kidney. Lab. Invest. 14, 19,45-1325.
TRUMP, B. •., GOLDBLATT, P. J. & STOWELL, R. E. (1962). An electron microscopic
study of early cytoplasmic alterations in hepatic parenchymal cells of mouse liver
during necrosis in vitro (autolysis). Lab. Invest. I I , 986-1o15.
VAN DUIJN, I'. (1961). Acrolein-Schiff, a new staining method for proteins. J. Histo-
chem. Cytochem. 9, 234-41.
VON HIPPEL, P. H. & WONG, K. Y. (1964). Neutral salts: The generality of their
effects on the stability of macromolecular conformations. Science 145 , 577-80.
WALKER, J. F. (1964). Formaldehyde. 3rd Ed. London: Chapman & Hall.
WETLAUFER, D. B. (1962). Ultraviolet spectra of proteins and amino acids. Adv. Protein
Chem. 17, 303-9 ~ 9
WETZEL, B. K. (1961). Sodium permanganate fixation for electron microscopy.
J. biophys, biochem. Cytol. 9, 711-16-
WIGGLESWORTH, V. B. (1964). T h e union of protein and nucleic acid in the living
cell and its demonstration by osmium tetroxide. Q. Jl microsc. Sci. lO5, 113-22.
WlLSKE, K. R. & ROSS, R. (1965). Autoradiographic localization of lipid- and water-
soluble compounds: a new approach. J. Histoehem. Cytochem. 13, 38-43.
WOLD, P. (1967). Bifunctional Reagents. I n Methods in Enzymology, xi, pp. 617-4 ~
(ed. C. H. W. HiTs). London: Academic Press.
WOLMAN, M. (1955). Problems of fixation in cytology, histology and histochemistry.
Int. Rev. Cytol. 4, 79-1o2.
WOOD, ]'. G. & BARRNETT, R. I. (1964). Histochemical demonstration of norepine-
phrine at a fine structural level. J . Histoehem. Cytochem. 12, 197-2o 9.
WOOD, R. L. & LUFT, ]'. N. (1965). T h e influence of buffer systems on fixation with
osmium tetroxide. J . Ultrastruet Res. I2, 22-45.
WYLLIE, R. G. (1965). Fixation in enzyme histochemistry. Nature, Lond. 2o7, 93-4.
YAMANE, T. & DAVIDSON, N. (1961). On the completing of deoxyribonucleic acid
(DNA) by mercuric ion. J. Am. chem. Sue. 83, 2599-26o7.
YAMANE, T. & D AVIDSON, N. (1962). On the compleTing of deoxyribonucleic acid by
silver (I). Bioehem. biophys. Acta 55, 6o9-21.
YOKOTA~ K., SUZUKI, Y. & ISHII, Y. (1965). Temperature dependence of the
polymerization modes of glutaraldehyde. Chem. Abstr. 65, 13835 (Abstract of
Iapanese article in KogyoKagaku Zasshi 68, 2459-63).
ZERLOTTI, E. (1967). Crosslin!dng of rat tail tendons with chloro-s-triazines. Nature,
Lond. 214, 13o4-6.
ZUGIBE, F. T. (1963). T h e demonstration of the individual acid mucopolysaccharides
in h u m a n aortas, coronary arteries and cerebral arteries I. T h e methods. J. Histo-
chem. Cytochem. IO, 441- 7.