Observations on Substances that react weakly tothe Periodic Acid/Schiff Test

By ARTHUR J. HALE(From the Histology Division, Institute of Physiology, The University, Glasgow)

With one plate (fig. i)

SUMMARY

It has been shown that certain mutinous substances which are weakly positive withthe periodic acid/Schiff technique stain more strongly after treating them with asolution of sodium hydroxide.

The deepening of the staining reaction is produced by making available, for aldehydeproduction by periodic acid, more .CHOH.CHOH., .CHOH.CHNH2., or . CHOH.CHNHR. groups. This is done without affecting those sulphate or metaphosphategroups responsible for metachromasia.

INTRODUCTION

IN histological sections there are present a number of substances looselydescribed as being of a mucinous nature. Examples of these are the ground

substance of cartilage and of other less dense connective tissues, sarcolemma,basement membranes, colloid of endocrine glands, and the mucus of epithelia.Histochemically they are related and the basis of the relationship is thepresence of a carbohydrate found either as a separate entity or in combinationwith protein or lipide. The biochemistry of these substances is as yet incom-pletely understood, but there are available histochemical methods which givesome indication of the composition of any one member of this family. Oneof these is the periodic acid/Schiff technique (McManus, 1946). Some of thesubstances mentioned above will give a strong colour reaction with thismethod, others will only give a weak reaction. Glycogen, the ground substanceof cartilage, and certain types of mucus are examples of the former; othertypes of mucus and sarcolemma are examples of the latter.

The periodic acid/Schiff (P.A.S.) technique (Lillie, 1947; Hotchkiss, 1948;McManus, 1948) is really a Malaprade reaction (Lillie, 1950). It identifies.CHOH.CHOH., .CH0H.CHNHa., and .CHOH.CHNHR. groups inpolysaccharides, mucopolysaccharides, mucoproteins, glycoproteins, andglycolipides by production of aldehydes, through periodic acid oxidation,which colourize the Schiff aldehyde-reagent.

Hale (1946), Gersh (1949), and Roberts and Jarrett (1950) have presentedmethods which may be of use in differentiating between substances whichgive a positive reaction with P.A.S. Hempelmann (1940), Friedenwald (1947),[Quarterly Journal of Microscopical Science, Vol. 94, part 3, pp. 303-313, Sept. 1953.]

304 Hale—Observations on the Periodic Acid\Schiff Test

and Grishman (1948) have introduced methods for differentiating tissueconstituents of this type which stain metachromatically. Dempsey and hisassociates (1946, 1947, and 1950) and Pearse (1949 and 1950) have usedmethodsfor identifying different mucopolysaccharides by their degree of basiphilia.

This report is the result of the finding that pre-treatment of histologicalsections with a weak solution of sodium hydroxide accentuates, in some cases,the depth of staining-reaction obtained with the P.A.S. method.

MATERIALS AND METHODS

As interpretation of the results depended on intensity of the colour re-action which developed, care was taken to subject all sections to identicalprocedures.

The periodic acid solution was prepared as described by Lillie (1947). TheSchiff solution was prepared as suggested by Lillie (1951a) and modified toconform with the findings of Longley (1952) and Atkinson (1952) in order tomake the results more easily reproducible.

The sulphite rinse between periodic acid oxidation and immersion inSchiff's solution has been shown by Lillie (195 ib) to reduce the intensity of thestaining reaction with periodic acid, and this step has been omitted through-out in order to maintain the strength of any possible positive reaction.

Acetylation blocks .CHOH.CHOH., CHOH.CHNH2., and .CHOH.CHNHR. groups (Gersh, 1949; McManus and Cason, 1950) and preventsthe production of aldehydes by subsequent periodic acid oxidation. De-acetylation returns the groups so blocked to their original potential reactivity.Acetylation of sections was carried out as described by Lillie (195 ib). It wasfound that deacetylation in 75 per cent, ethanol and ammonia was moreefficient than that in absolute ethanol and ammonia as used by Lillie.

Immersion of sections in hydroxylamine hydrochloride (Danielli, 1949),to block all aldehydes, was used to demonstrate the specificity of the Schiff'ssolution and to identify the nature of the reaction product.

The method for the Feulgen reaction, in so far as it could be carried out inparaffin sections, followed the instructions of Danielli (1949).

In all cases, batches of sections containing one slide from each tissue werecarried through each procedure. Batches were placed in glass racks and timeswere accurately controlled. All sections, irrespective of the fixative used, weretaken through iodine and thiosulphate when being brought down to waterbefore staining. All solutions were renewed from standard stock bottles eachday. Washing was carried out in running tap water. Glasgow tap water is verypure and constant in quality (Garven and Gairns, 1952) and no harm appearsto be caused by washing sections in it. Before and after the periodic acid solu-tion the sections were washed in 70 per cent, ethanol. All sections weredehydrated in 90 per cent, ethanol and two changes of absolute ethanol,cleared in xylene, and mounted in D.P.X. Counterstains were not used.

The tissues used were known to give a weakly positive P.A.S. reaction in atleast part of their substance. They were:

Hale—Observations on the Periodic AcidjSchiff Test 305

Human tissues fixed in 10 per cent, neutral formalin: heart (lipofuscin),cervix, prostate, rectum, pancreas, trachea, sub-maxillary gland;

fixed in mercuric formalin: rectum, amyloid spleen, aorta,umbilical cord, amyloid liver, myxoma, heart valve;

fixed in Bouin's fluid: lung (pneumonic exudate), umbilicalcord, small intestine, stomach.

Rabbit tissues fixed in 10 per cent, neutral formalin: duodenum, kidney;fixed in Bouin's fluid: bone marrow.

Rat tissues fixed in 10 per cent, neutral formalin : eye, mesentery.

Representative batches of sections were carried through by each of thefollowing procedures after having been brought to water. Room temperature(approximately 200 C.) was used except where otherwise stated.

(1) Periodic acid 15 min.; wash; Schiff soln. 30 min.; wash.(2) Schiff soln. 30 min.; wash.(3) o-2N NaOH 15 min. at 22° C.; wash; periodic acid 15 min.; wash;

Schiff soln. 30 min.; wash.(4) o-2N NaOH 15 min. at 220 C ; wash; Schiff soln. 30 min.; wash.(5) Acetylation; wash; periodic acid 15 min.; wash; Schiff soln. 30 min.;

wash.(6) Acetylation; wash; deacetylation; wash; periodic acid 15 min.; wash;

Schiff soln. 30 min.; wash.(7) o-2N NaOH 15 min. at 22° C ; wash; acetylation; wash; periodic acid

15 min.; wash; Schiff soln. 30 min.; wash.(8) Acetylation; wash; o-2NNaOH 15 min. at22°C; wash; periodic acid 15

min.; wash; Schiff soln. 30 min.; wash.(9) Deacetylation 15 min.; wash; periodic acid 15 min.; wash; Schiff soln.

30 min.; wash.(10) Deacetylation 24 hrs.; wash; periodic acid 15 min.; wash; Schiff soln.

30 min.; wash.(11) Diastase 30 min.; wash; periodic acid 15 min.; wash; Schiff soln. 30

min.; wash.(12) 0-5 per cent, toluidin blue 30 min.; wash.(13) 0-2N NaOH 15 min. at 22° C ; wash; 0-5 percent, toluidin blue 30 min.(14) o-2N NaOH 15 min. at 22° C ; wash; periodic acid 15 min.; wash;

hydroxylamine 60 min.; wash; Schiff soln. 30 min.; wash.(15) Periodic acid 15 min.; wash; hydroxylamine 60 min.; wash; Schiff

soln. 30 min.; wash.(16) Periodic acid 15 min.; wash; hydroxylamine 24 hrs.; wash; Schiff

soln. 30 min.; wash.(17) Hydroxylamine 1 hr.; wash; o-iN HC1. 15 min.; wash; Schiff soln.

30 min.; wash.(18) Periodic acid 15 min.; wash; Schiff soln. 30 min.; wash in three changes

of sulphite soln., 5 min. in each change.

306 Hale—Observations on the Periodic AcidjSchiff Test

(19) o-2N NaOH 15 min. at 22° C.; wash; periodic acid 15 min.; wash;Schiff soln. 30 min.; wash in three changes of sulphite soln., 5 min. in each.

SIGNIFICANCE OF METHODS AND RESULTS

(1) This is the standard technique used to identify . CHOH. CHOH.,.CHOH.CHNH2., and .CHOH.CHNHR. groups by production of alde-hydes by periodic acid oxidation and identification of the aldehydes by re-colourization of Schiff's solution.

(2) Immersion directly in Schiff's solution, which gives no reaction, showsthat none of the colour produced in (1) is caused by the presence of 'free'aldehydes in the sections or by spontaneous recolouration of the Schiffssolution.

(3) This is the method which increases the intensity of the staining reactionof certain substances which otherwise give only a pale colour with procedure (1).The maximum amount of change from the standard appearance occurred inthe first few minutes of incubation of sections in o-2N NaOH at 220 C. Timesor temperatures in excess of this produced no further change. When comparedwith the results from (1) the sections of human rectum and rabbit duodenumshowed the most striking changes. The Brunner's glands of the rabbit duo-denum changed from a faint pink colour with the standard technique to adeep magenta colour after treatment with the NaOH solution (fig. 1, A and B).The goblet cells of the human rectum changed from a pink to a very deepmagenta colour (fig. 1, c and D). The fibrin clot of the pneumonic exudate ofthe lung increased slightly its depth of colour after treatment. In all the tissuesthere was a slight increase in intensity of staining of reticulin, certain irregularareas of collagen and sarcolemma, and in the amount of diffuse pink connectivetissue background.

(4) Incubation of sections in NaOH, without subsequent periodic acidoxidation, followed by immersion on Schiff's solution was done to find out ifaldehydes were being produced. No colour was produced.

(5) The acetylation blocks the periodic acid reactive groups. Colour wasnot produced.

(6) This is a check on the specificity of (5) by removing the blocking agent,and it produced the same colour reaction as (1).

(7) The acetylation after incubation in NaOH was introduced to findwhether the alkali was releasing groups, other than those enumerated above,which were subsequently being oxidized to produce aldehydes. No colour wasproduced. This showed that any groups released by the NaOH were of theabove type.

FIG. I (plate), A, rabbit duodenum. P.A.S.B, rabbit duodenum. NaOH + P.A.S.c, human rectum. P.A.S.D, human rectum. NaOH + P.A.S.

The light-source, aperture, and exposure were the same in all four cases. Ilford nitersNos. 104 and 108 were superimposed.

Hale—Observations on the Periodic AcidjSchijf Test 307

(8) It was possible that the action of the NaOH might only be a deacetyla-tion, so this step was introduced. An excellent deacetylation, identical to thatproduced by the ethanol-ammonia mixture, was brought about, but in addition,increased staining of rectal goblet cell and Brunner's gland mucus was seen.

(9) As the NaOH can deacetylate this step was carried out to show that thecause of the increase in intensity was not of this nature. This procedure,giving the same reaction as (6) and (1), did not produce any increase in inten-sity of the staining reactions as enumerated in (3). Therefore the increasedcolour in (3) is not due to deacetylation.

(10) As in (9).(11) The diastase digestion was introduced to show that none of the in-

creased colour produced in (3) was associated with the presence of glycogen.The glycogen reactions of the squamous epithelium of the cervix and rectum,of the walls of the blood-vessels of the umbilical cord, of the macrophagesof the pneumonic exudate, and of the chondriocytes of trachea, were removed.

(12) This is the standard procedure for producing metachromasia.(13) It was thus shown that the NaOH did not remove those groups re-

sponsible for the production of metachromasia, as the appropriate metachro-matic reactions of all tissues used were obtained.

(14) (15), and (16) These procedures were carried out to show that therecolouring of the Schiffs solution was due only to the presence of aldehydes,whether procedure (1) or (3) was used. In all three procedures a pale reactionpersisted in what were normally strongly staining areas.

(17) Blocking of 'free' aldehydes and production of a Feulgen reaction werecarried out for comparison with the sites of colour production in the pro-cedures. A faint positive reaction was produced in the nuclei of the rectalepithelial cells, the retinal cells, and the pancreatic acinar cells. There wasalso a persistent positive reaction in the cartilage of the trachea and throughoutthe connective tissues and all nuclei of the rat mesentery.

(18) and (19) Rinsing sections in sulphite solution after leaving the Schiff'sreagent prevented the formation of non-specific colour formed by atmosphericoxidation of adsorbed Schiff's reagent. These sections showed no differenceas compared with those of (1) and (3) respectively.

DISCUSSION

The P.A.S. reaction gives a positive result, with a varying degree of in-tensity, with polysaccharides, mucopolysaccharides, mucoproteins, glycopro-teins, and glycolipides, as stated before.

The main representatives in each of these groups, in mammals, are (Meyer,1945):

polysaccharides: glycogenmucopolysaccharides: neutral: blood-group substances

acid: [simple): hyaluronic acid(complex): mucin, heparin, chondroitin sul-phate

308 Hale—Observations on the Periodic AcidjSchiff Test

mucoproteins: submaxillary mucoid, serum mucoid, seroglycoidglycoproteins: serum albumin and globulinglycolipides

Substances which stain metachromatically, at least as far as they are en-countered in histological material, are polysaccharides containing uronicacids in addition to sulphate groups (Lison, 1935), or polymeric metaphos-phate groups (Bank and Bungenberg de Jong, 1939; Michaelis and Granick,1945; Michaelis, 1947). The causation of metachromasia is incompletelyunderstood, but it is generally considered to depend on polymerization of thedye molecules in the presence of these chemical groups (Kelly and Miller,1935; Bank and Bungenberg de Jong, 1939; Hempelmann, 1940; Michaelisand Granick, 1945; Michaelis, 1947; Landsmeer, 1951; Massart and others,1951). The substances of interest in this investigation are the mucopoly-saccharides, such as rectal mucus or heparin, and the glycoprotein amyloid.

Glycogen can be eliminated by diastase treatment. Glycolipides are notusually evident in the paraffin sections in the areas studied. This can be con-firmed by a chloroform-methanol extraction method (Pearse, 1951). We arethus left with the necessity of differentiating between mucopolysaccharides,mucoproteins, and glycoproteins.

Davies (1952) states that hyaluronic acid, a simple acid mucopolysaccharide,considered to be the main constituent of the 'ground substance' in connectivetissues, will only stain very faintly with the P.A.S. technique according to theformula of Meyer and Fellig (1950). He himself found that it neither stainsmetachromatically nor with periodic acid if it is properly purified. My ownfindings with completely purified and incompletely purified hyaluronic acidpreparations confirm this. With regard to chondroitin sulphuric acid, acomplex (i.e. sulphate-containing) acid mucopolysaccharide, Davies pointsout that according to the formula given by Haworth (1946), the reaction withP.A.S. should be strong; but according to that of Meyer, Odier, and Siegrist(1948), the reaction should be very faint.

Lillie (1949) reports that the mucin (? hyaluronic acid) of the connectivetissue of the umbilical cord stains metachromatically but gives no colour reac-tion with periodic acid. Sylven(i94i, 1945, and 1950), Davies (1943), Wislockiand others (1947), Campani and Reggiani (1950), Mancini (1950), and Wislockiand Sognnaes (1950) report the presence of metachromatic ground substancein granulating tissue and other actively growing connective tissues. Meyer(1946) states that'young growing fibroblasts secrete hyaluronic acid, which is followed by the secre-tion of chondroitin sulphate and of a precursor of collagen, the latter a non-fibrousand soluble protein. By local acidification in the immediate neighbourhood of thefibroblasts, the precursor is denatured by the polysaccharides, the latter acting asanionic detergents rolling up the peptide chains along the acidic groups of the fibrouspolysaccharide molecules. Most of the hyaluronate is removed enzymatically,leaving the more firmly bound chondroitin sulphates as a network on the surface ofthe fibres.'

Hale—Observations on the Periodic AcidjSchiff Test 309

Thus it appears that the metachromatic ground substance stained bynumerous investigators is likely to be free chondroitin sulphate which ispresent in actively multiplying connective tissue. Thus as hyaluronic acid isP.A.S. negative and not metachromatic it does not come into the initial classi-fication of P.A.S. positive substances.

If the strongly P.A.S. reacting substances are left out in Meyer's classifica-tion it is now necessary to differentiate between those mucopolysaccharideswhich react weakly and the persistently more weakly reacting mucoproteinsand glycoproteins.

Chondroitin sulphates, which are bound in collagen to form a mucoprotein,or to globulin as in amyloid to form a glycoprotein (Krakow, 1897; Lillie,1950), show varying degrees of metachromasia and of reactivity with theP.A.S. method. Free chondroitin sulphate and that bound in cartilage show astrong metachromasia and a strong reaction with P.A.S.

Human rectal mucus and the mucus of Brunner's glands in the rabbit duo-denum are consistently strongly metachromatic and weakly positive withperiodic acid. Rabbit Brunner's glands are characteristic in this respect (Chap-man, 1952). Lillie (195ib) has pointed out, in discussing intestinal mucus andits staining reaction with the Bauer (1933) and Casella (1942) techniques, thatthere appears to be an inverse relationship between strongly P.A.S. positivesubstances and metachromatically positive ones; for example, gastric mucusstains very strongly with periodic acid but is only very weakly, if at all, meta-chromatic. Rectal mucus gives the reverse reaction. He suggests that a mucusweakly P.A.S. positive but strongly metachromatic has fewer groups availablefor aldehyde production with the P.A.S. technique because of a longerchain structure.

Mucoproteins such as sub-maxillary mucoid are consistently weaklyP.A.S. positive.

Procedure (3) causes an increase in the P.A.S. staining reaction of humanrectal mucus and of the mucus of the rabbit Brunner's glands. The first is acomplex acid mucopolysaccharide. The exact nature of the second is notknown. An increased depth in staining also occurs in reticulin and on thesurface of collagen, both of which are considered to be mucoprotein (Meyer,1952). The change in reaction occurs partially in the clot of pneumonicexudate, which is probably a mixture of serum mucoprotein and glycoprotein.It does not take place in sub-maxillary mucoid, which is a mucoprotein. Itdoes not take place in amyloid, which is a glycoprotein (Krakow, 1897; Lillie,1950). Gersh (1949) intimates that o-ooooiN sodium hydroxide, acting forone hour at an unspecified temperature, weakens the P.A.S. staining reactionof the 'glycoprotein' of the Golgi apparatus. He does not report the effect on thesurrounding tissues.

Meyer and Rapport (1951), in discussing the nature of the bonds betweenthe mucopolysaccharide and the protein in the mucoproteins of connectivetissue, intimate that they extract the mucopolysaccharides with a O-33Nsolution of NaOH at o° C. Meyer (1952) suggests that mucoproteins of

310 Hale—Observations on the Periodic AcidjSchiff Test

connective tissue are mucopolysaccharides bound to protein by strong electro-valent forces. Therefore the staining reaction change which I have obtainedmay be due to the breaking of weaker bonds in certain mucoproteins.

As the change in reaction appears more clearly in mucus, it may be that itis occurring more selectively in mucoitin sulphates as distinct from chon-droitin sulphates. It is thus a possibility that the variation in reaction maybe due to the different hexosamine present; acetylglucosamine in mucoitinsulphates and acetylgalactosamine in chondroitin sulphates. Morgan and Elson(1934) have shown that alkaline hydrolysis of acetylglucosamine and acetyl-galactosamine (acetylchondrosamine) can be used to give quantitativecolorimetric estimations of their products with paradimethylaminobenz-aldehyde. Aminoff, Morgan, and Watkins (1952) show that by this methodacetylchondrosamine gives only 23 per cent, of the degree of colour reaction ofacetylglucosamine, and blood group mucoids give only 7-10 per cent. Theexact mode of reaction is not known but it is suggested that the alkali forms anoxazoline or a pyrazine with the hexosamine. It is not known whether theseproducts are P.A.S. positive or not. It has been stated (Mann, 1902; Dempseyand others, 1947) that fixation of tissues in formaldehyde will prevent thiscolour reaction. Frey-Wyssling (1948) states, in discussing the chemical natureof cytoplasm, that formaldehyde is thought to fix tissues by forming bridgesbetween neighbouring polypeptide chains. So it may be that the difference inthe type of hexosamine present might influence the P.A.S. reaction afteralkaline hydrolysis, though the effect of the formalin fixatives must be bornein mind.

Mucopolysaccharides, according to Meyer's definition (1945), are poly-saccharides which contain hexosamine as one component, whether theyoccur free or bound to substances of higher molecular weight. Mucoproteinsare those substances which contain a mucopolysaccharide in firm chemicalunion with a peptide, where the hexosamine content is greater than 4 per cent.Proteins which contain less than 4 per cent, hexosamine are classified asglycoproteins. Thus the selectivity of the change in reaction may be due to therelative amount of hexosamine present.

If all the preceding arguments are reconsidered it appears that the selectivityof the reaction may be due to one or more of the following causes:

(a) The number of reacting molecules available, perhaps depending on thechain length of the substance.

(b) The degree of chemical union with protein.(c) The type of hexosamine present.(d) The relative amount of hexosamine present.

All that can be said regarding the nature of the extra groups being madeavailable, by the caustic solution, for aldehyde production by periodic acid,is that:

(a) No aldehydes are being produced by the alkali as shown by procedure(4), where no colour is produced.

Hale—Observations on the Periodic Acid I Schiff Test 311

(b) The groups which are being made available are .CHOH.CHOH.,.CHOH.CHNH2., or .CHOH.CHNHR. groups, as shown by pro-cedure (7), in which all colour reaction is blocked by acetylation.

(c) The action of the alkali is not just a deacetylation as suggested by pro-cedure (8), in which the full colour reaction is produced, because procedures(9) and (10) restore the normal colour reactions to the tissues but do not inten-sify the mucus-reactions.

\d) The sulphate or metaphosphate groups thought to be responsible formetachromasia are not affected; this is shown by procedure (13), where themetachromasia remains after alkaline treatment.

[e) None of the increased colour intensity is due to absorbed leucofuchsinwhich has been subsequently oxidized in the atmosphere, as shown by pro-cedures (18) and (19).

(/) An attempt to show by procedures (14), (15), and (16) that the increasedcolour reaction was due to the eventual production of aldehydes was preventedby the retention of a slight degree of colouration in normally strongly reactingareas in all three procedures. It is known that the condensation of hydroxyl-amine with aldehydes is dependent to a certain extent on the pH of the reaction;as the pH rises, within certain limits, so does the rate and degree of condensa-tion (Olander, 1927). It may be that these areas, which are known to be ofacidic nature (Dempsey and others, 1947), prevent complete condensationand thus leave a certain amount of aldehyde available to recolourize theSchiff solution.

CONCLUSIONS

It has been shown that certain substances that are weakly positive to theperiodic acid/Schiff reaction can be made to stain more strongly by firstexposing them to a solution of sodium hydroxide.

It is suggested that this may provide a method of differentiating betweenweakly staining long chain mucopolysaccharides, mucoproteins, and glyco-proteins. The first deepen markedly in colour, the second show an inconstantsmall increase in depth of colour, and the last remain unchanged.

The relationship of the strength of the colou r reaction to the varying chemicalstructure is not definite; the colour may depend on the chain length, the de-gree of conjugation with protein, or possibly on the type or amount ofhexosamine present.

From the notes on methods and results it can be shown that the differencein staining reaction is produced by making available more . CHOH. CHOH.,.CHOH.CHNH2., or .CHOH.CHNHR. groups for aldehyde productionby periodic acid without affecting those sulphate or metaphosphate groupswhich give certain tissues their metachromatic affinities. Whether the requisitegroups for aldehyde production are made available by protein extraction or bysome other reaction is not yet known. The process does not appear to be adeacetylation.

The investigation is being pursued with a wider range of tissues and other

312 Hale—Observations on the Periodic AcidjSchiff Test

methods which are being developed. It is hoped to eliminate fixation errors bythe use of frozen-dried material.

I wish to thank Professor R. C. Garry and Dr. H. S. D. Garven for theircontinued interest and encouragement. I wish to thank Dr. H. Munro andDr. G. T. Mills for their advice, and Dr. G. T. Mills and Dr. E. Smith formaking available some extracts of hyaluronic acid. I also wish to thank Mr.R. Carswell for his technical assistance.

REFERENCES

AMINOFF, D., MORGAN, W. T. J., and Watkins, W. M., 1952. Biochem. J., 51, 379.ATKINSON, W. B., 1952. Stain Tech., 27, 153.BANK, O., and BUNGENBERG DE JONG, H. G., 1939. Protoplasma, 32, 489.BAUER, H., 1933. Z. mikr.-anat. Forsch., 33, 143.CAMPANI, M., and REGGIANI, O., 1950. J. Path. Bact., 62, 563.CASELLA, C, 1942. Anat Anz., 93, 289.CHAPMAN, I. M. M., 1952. Personal communication.DANIELLI, J. F., 1949. Quart. J. micr. Sci., 90, 67.DAVIES, D. V., 1943. J. Anat. Lon., 77, 160.

1952. Stain Tech., 27, 65.DEMPSEY, E. W., and Singer, M., 1946. Endocrinology, 38, 270.DEMPSEY, E. W., and WISLOCKI, G. W., 1946. Physiol. Rev., 36, 1.DEMPSEY, E. W., BUNTING, H., SINGER, M., and WISLOCKI, G. B., 1947. Anat. Rec, 98, 417.DEMPSEY, E. W., SINGER, M., and WISLOCKI, G. B., 1950. Stain Tech., 25, 73.FREY-WYSSLING, A., 1948. Submicroscopic morphology of protoplasm audits derivatives. London

(Elsevier Publ. Co.).FRIEDENWALD, J. S., 1947. Cold Spr. Harb. Symp. quant. Biol., 12, 141.GARVEN, H. S. D., and GAIRNS, F. W., 1952. Quart. J. exp. Physiol., 37, 131.GERSH, I., 1949. Arch. Path., 47, 99.GRISHMAN, E., 1948. Bull. int. Ass. med. Mus., 28, 104.HAI-B, C. W., 1946. Nature, Lon. 157, 8oz.HAWORTH, W. N., 1946. Proc. Roy. Soc. A, 186, x.HEMPELMANN, L. H., 1940. Anat. Rec, 78, 197.HOTCHKISS, R. D., 1948. Arch. Biochem., 16, 131.KELLY, E. G., and MILLER, E. G., 1935. J. biol. Chem., n o , 119.KRAKOW, A., 1897. Arch, exper. Path. Pharmak., 40, 195.LANDSMIEH, J. M. F., 1951. Acta physiol. pharm. neerl., 2, nz .LILLIE, R. D., 1947. J. Lab. clin. Med., 33, 910.

1949. Bull. int. Ass. med. Mus., 29, 1.1950. Anat. Rec, 108, 239.1951a. Stain Tech., 26, 163.19516. Ibid., 26, 123.

LISON, L., 1935. Arch. Biol. Paris., 46, 595.LONGLEY, J. B., 1952. Stain Tech., 27, 161.MCMANUS, J. F. A., 1946. Nature, 158, 202.

1948. Stain Tech., 23, 99.MCMANUS, J. F. A., and CASON, J. E., 1950. J. exp. Med., 91, 651.MANCINI, R. E., 1950. Rev. Soc. argent. Biol., 26, 139.MANN, G., 1902. Methods and theory of physiological histology. Oxford (Clarendon Press).MASSART, L., COUSSENS, R., and SILVER, M., 1951. Bull. Soc. Chim. Biol. Paris, 33, 514.MEYER, K., 1945. Advances in protein chemistry. New York (Academic Press Inc.).

1946. Amer. J. Med., i, 675.1952. Proc. Faraday Soc. (Proof), August 1952.

MEYER, K., and RAPPORT, M. M., 1951. Science, 113, 596.

Hale—Observations on the Periodic AcidlSchiff Test 313

MEYER, K. H., and FELLIG, J., 1950. Experientia, 6, 186.MEYER, K. H., ODIER, M. E., and SIEGRIST, A. E., 1948. Helv. chim. acta., 31, 1400.MICHAELIS, L., 1947. Cold Spr. Harb. Symp. quant. Biol., 12, 131.MICHAELIS, L., and GRANICK, S., 1945. J. Amer. chem. Soc., 67, 1212.MORGAN, W. T., and ELSON, L. A., 1934. Biochem. J., 28, 988.OLANDER, A., 1927. Z. physiol. Chem., 129, 1.PEARSE, A. G. E., 1949. J. clin. Path., 2, 81.

1950. J- Path. Bact., 62, 351.1951- J. clin. Path., 4, 1.

ROBERTS, G. B. S., and JARRETT, W. F. H., 1950. J. Anat. Lond., 84, 407.SYLVEN, B., 1941. Acta chir. scand., 86, Suppl. 66.

1945- Acta radiol. Stockh., Suppl. 59.195°- Exper. Cell. Res., 1, 582.

WISLOCKI, G. B., BUNTING, H., and DEMPSEY, E. W., 1947. Amer. J. Anat., 81, 1.WISLOCKI, G. B., and SOGNNAES, R. F., 1950. Amer. J. Anat., 87, 239.

Fie. i

A. J. HALE