Butter is the fat of milk, containing small proportions of caseine, water, and salt (the latter mostly added), and possessing a somewhat granular structure. In its preparation the fat-globules of cream are made to coalesce by the process of churning, and are removed from the residual buttermilk. Its colour, due to lactochrome, varies from white to yellow, according to the breed and food of the cow. The fatty constituents of butter are butyric, caproic, caprylic, capric, myristic, palmitic, stearic, and oleic acids, which are combined with glycerine as ethers; the first four are soluble in hot water, the remainder insoluble. It is very probable that butter fat is composed of complex glycerides, i. e. tri-acid (presumably oleic, palmitic, and butyric) ethers, of the following character:—

The table on p. 64 exhibits a summary of the results obtained by various chemists by the analysis of numerous specimens of genuine butter.

Dr. Elwyn Waller found the following variations in the constituents of pure butter:—Fat, from 83 to 85; water, from 8 to 10; curd, from 1 to 3; salts, from 3 to 5 per cent.

Butter fat fuses at 28° to 37°, and at 37°·7 its specific gravity ranges from 0·91200 to 0·91400. The most common adulterations of butter consist in the addition of water, salt, colouring matters, and various foreign fats (notably oleomargarine). The first two admixtures are easily recognised by the proximate analysis; the detection of the last sophistication involves a somewhat elaborate examination of the fatty constituents of the butter.

Proximate Analysis.

About five grammes of the well-averaged sample are weighed out in a tared platinum capsule, and dried for three hours (or until constant weight is obtained) over a water-bath (or over a low flame, constantly stirring with a thermometer), and the decrease in weight (water) ascertained. As a rule, the proportion of water in genuine butter varies from 8 to 16 per cent. The residue in the capsule is then melted at a gentle heat, and the liquid fat cautiously poured off from the remaining caseine and salt, these latter being afterwards more completely exhausted by washing with ether. Upon now drying the residue, the loss in weight will give the amount of fat present. The caseine is next determined by the loss in weight obtained upon incinerating the matters left undissolved by the ether, the remaining inorganic matter being the salt contained.

The proportion of fat present in genuine butter ranges from 82 to 90 per cent.; it should never be below 80 per cent. The average amount of caseine is 2·5 per cent.; greater proportions, frequently occurring in unadulterated butter, render it more liable to become rancid. The ash should consist of sodium chloride, with some calcium phosphate; the amount of salt is quite variable, but it usually ranges from 2 to 7 per cent. The proportion of ingredients, not fat, in butter may be conveniently determined by melting 10 grammes of the sample in a graduated tube, provided with a scale at its lower end, which is narrowed, adding 30 c.c. of petroleum naphtha, and shaking the mixture. After standing a few hours, the non-fatty matters collect in the lower portion of the tube, and their volume is read off. Genuine butter is said to yield from 12 to 14 per cent. (assuming each c.c. to equal one gramme), while adulterated specimens may show 20 per cent. of matters not fat.

Examination of the Butter-fat.

The most common and important sophistication of butter consists in the addition of foreign fats, embracing both animal fats (oleomargarine and lard) and vegetable oils (cotton-seed, olive, rape-seed, cocoa-nut, almond, palm, etc.). Of these, oleomargarine is doubtless the most often employed. Oleomargarine is the more fusible portion of beef fat, and is prepared by straining the melted fat, allowing the oil thus obtained to stand for some time at a temperature of about 24°, when most of the stearine and palmitine will separate out, and cooling the remaining oil until it solidifies. This is next churned with milk, a little colouring matter (annato) being added, and the product is then chilled by mixing it with ice; salt is now added, and the mass is finally worked up into lumps.

It is stated that fifteen establishments in the United States are engaged in the manufacture of oleomargarine, the annual production in the State of New York alone being about 20,000,000 pounds. The rapid increase in the manufacture of oleomargarine is shown by the following statistics:—In 1880 this country exported 39,236,655 pounds of butter and 20,000,000 pounds of oleomargarine, while in 1885 the exportation of butter declined to 21,638,128 pounds, and the exportation of oleomargarine reached nearly 38,000,000 pounds. The present production is said to approximate 50,000,000 pounds per annum. The most characteristic difference in the composition of genuine butter and oleomargarine consists in the greater proportion of soluble fats contained in the former. This is illustrated by the following comparative analysis of the two products (MÉge MouriÈs):—

Lard is likewise occasionally used in the United States as an admixture to butter, the product, “lardine,” being sold either as oleomargarine-butter, or as the genuine article. Dr. Munsell mentions a factory in New York city where the weekly output of larded butter is 5000 pounds. As a result of the efforts of the New York State Dairy Commission, it has been estimated that the sale of imitation butter in this State in 1885 suffered a decrease of about 60 per cent., although the quantity manufactured in the United States showed an increase of 50 per cent.

The specific gravity and melting point of butter have been suggested as criteria for its purity; in most cases, however, these determinations possess a rather limited value; as already stated, butter fat, at the temperature of 37°·7, has a density ranging from 0·91200 to 0·91400.

The relation between the specific gravity of a fat and the proportion of insoluble acids contained was first noticed by Bell. This is shown in the following table which refers to pure butter fat.

The following results have been obtained by the analysis of samples of various animal fats, and oleomargarine butter.

It will be noticed that the fats mostly used to adulterate butter are of a lower density. Blyth regards a gravity below 0·911 (at 37°·5) as clearly pointing to the presence of foreign fatty admixture.

The specific gravity determination is made by means of the areometer, or by the gravity bottle; numerous indirect methods have also been proposed. P. Casamajor[26] suggests a process for distinguishing genuine butter from oleomargarine which is based upon the fact that the density of a liquid in which a body remains in equilibrium is the density of the body itself. As the result of his investigations it was found that pure butter at 15° would be held in equilibrium by alcohol of 53·7 per cent. (sp. gr. = 0·926), and that oleomargarine would remain in equilibrium, at the same temperature, in alcohol of 59·2 per cent. (sp. gr. = 0·905). If equal volumes of alcohol of 53·7 per cent. and 59·2 per cent. (i.e. an alcohol of 56·5 per cent.) are taken, and a drop of melted butter and of oleomargarine are delivered upon its surface, the former will sink to the bottom and the latter will remain at the top, so long as the two globules are warm and liquid. In case the temperature of the alcohol is about 30°, the butter will solidify and also rise to the top, whereas the oleomargarine may remain liquid. On now keeping the alcohol for a short time at a temperature of 15° the oleomargarine becomes opaque, but remains at the top, while the solidified butter will sink to the bottom. If alcohol of 59·2 per cent. is employed, oleomargarine will remain at the surface and genuine butter fall to the bottom at all temperatures above 15°, and at this temperature oleomargarine will be in equilibrium. Since not over 33 per cent. of butter is usually added to oleomargarine, it is proposed to use alcohol of 55 per cent., and consider as oleomargarine any sample which does not sink at 15°.

The foregoing method can be applied quantitatively by determining the strength of the alcohol which will keep in equilibrium a drop of the fat under examination. Since the difference between 59·2 and 53·7 is 5·5, the difference between the strength of the alcohol used and 53·7, divided by 5·5 (or multiplied by 0·18), will give the proportion of oleomargarine present. For example, if the globule is held in equilibrium at 15° in 57 per cent. alcohol, the sample contains about 60 per cent. of oleomargarine, for (57 - 53·7) × 0·18 = 3·3 × 0·18 = 0·594 or, say, 6/10.

The melting-point of butter is below that of most of its fatty adulterants; as previously stated, it varies from 28° to 37°. The determination is made either in the ordinary manner by means of a fine tube, or a little of the chilled sample is attached to a looped platinum wire, placed near the thermometer-bulb, in water which is gradually heated until fusion takes place. Blyth gives the following table of the melting-points of various fats:—

Numerous qualitative tests have been proposed by various authorities for the detection of foreign fats in butter, of which the following are perhaps sometimes of use. It should be added that the value of these tests, when applied to mixtures, is limited and very uncertain.

1. A little of the sample is heated in a test-tube: pure butter froths and acquires a brownish colour; with foreign fats there is but little foaming, and, although the caseine present darkens, the liquid itself remains comparatively clear.

2. If a sample containing oleomargarine is melted and the oil burned in an ordinary lamp-wick, a decided odour of burning tallow will be produced upon extinguishing the flame. Specimens of real butter, however, have been found to also emit a tallow-like odour.

3. The melted sample is filtered and treated with boiling ether; pure butter fat dissolves much more readily than do lard and tallow. Upon adding methylic alcohol to the solution the latter fats are precipitated, whereas pure butter will remain in solution.

4. If the filtered fat is distilled with a mixture of alcohol and sulphuric acid, the distillate will possess the odour of butyric ether in a very marked degree, in case it consists of butter.[27]

5. The strained fat is treated with a solution of carbolic acid (1 part acid and 10 parts water): genuine butter dissolves to a clear solution; beef, mutton, and swine fat form two layers, the upper one becoming turbid upon cooling.

6. If the sample consists of butter or oleomargarine, and is mixed with about ten parts of glycerine and the emulsion digested with a mixture of equal parts of ether and alcohol, two layers of solution will be produced, without any deposit of solid matter between them; if, however, lard, suet, or starch is present it will become deposited between the layers.

It has already been mentioned that butter differs from some of its fatty adulterants in containing a considerable proportion of fatty acids which are soluble in hot water, the acids present in most foreign fats being, on the other hand, almost entirely insoluble. The estimation of the relative amounts of soluble and insoluble acids contained in a fat possesses therefore much importance; indeed, more significance attaches to this determination than to any other. The processes most frequently employed in the quantitative examination of butter fat are those of Koettstorfer, Hehner, and Reichert.

Koettstorfer’s method[28] is based upon the fact that, as butter fat contains the fatty acids, having a smaller molecular weight than those present in other fats, it must contain more molecules of acid, and will therefore require a greater amount of an alkali to effect saponication. The process is executed as follows:—One or two grammes of the filtered fat are weighed out in a narrow beaker and heated over a water-bath with about 25 c.c. of one-half normal alcoholic solution of potassium hydroxide. The saponification of the fat is assisted by repeated stirring; when it is completed the beaker is removed from the bath, a few drops of alcoholic phenol-phthaleine added for an indicator, and the excess of potash used titrated back with one-half normal hydrochloric acid. It has been found that pure butter fat requires from 221·4 to 232·4 milligrammes of potassium hydroxide for saponification. The following are the number of milligrammes of alkali necessary for the saponification of one gramme of various other fats:—

Taking 227 milligrammes as the average amount of potassium hydroxide required to saponify one gramme of pure butter fat, the following formula has been suggested for the estimation of the proportion of admixture in a suspected sample:—

(227 - n) × 3·17 = x,

in which n represents the number of milligrammes of potassium hydroxide used, and x the percentage of foreign fat added. In the Paris Municipal Laboratory, 221 milligrammes of K(OH) are regarded as a standard for the saponification of one gramme of genuine butter.

Cocoa-nut oil, unfortunately, requires a figure (250 mgr.) considerably above that of butter, and it is quite possible to prepare a mixture of this oil and oleomargarine, that by the foregoing test would show a result almost identical with that afforded by pure butter. Hehner’s process,[29] which is often employed for the determination of the insoluble fatty acids, is as follows:—About 4 grammes of the melted and strained sample are dissolved in 50 c.c. of alcohol, containing two grammes of potassium hydroxide in solution, and the mixture is heated until complete saponification takes place. The alcohol is removed by evaporation, the residue dissolved in 200 c.c. of water, and the fatty acids precipitated by adding dilute sulphuric acid to distinct acid reaction. The fatty acids are next melted by heating the liquid and are then allowed to cool, after which the insoluble acids are poured upon a tared filter and repeatedly washed with hot water until the washings cease to show acidity. The filter and contents are finally cautiously dried and weighed. In genuine butter the proportion of insoluble fatty acids ranges between 86·5 and 87·5 per cent.; it should not be above 88 per cent.[30] Oleomargarine, lard, mutton, beef, and poppy, palm, olive, and almond oils contain about 95·5 per cent. of insoluble acids.[31]

The preceding process is also imperfect in not effecting the detection of cocoa-nut oil, which affords only about 86 per cent. of insoluble fatty acids, and although the presence of any considerable proportion of this oil in butter would probably be indicated by the decreased melting point of the admixture, an estimation of the soluble fatty acids is by far the most reliable means for its detection. For this determination Reichert’s method[32] is eminently adapted. In this process advantage is taken of the facts that the amount of soluble acids in a mixture of fat bears a direct relation to the proportion of genuine butter present, and that, if the aqueous solution of a saponified fat is decomposed by an acid and heated to boiling, the greater portion of the soluble acids escape with the watery vapours and can be collected and determined in the distillate. The details of this method are essentially as follows:—2½ grammes of the filtered sample are introduced into an Erlenmayer flask together with 1 gramme of potassium hydroxide and 20 c.c. of dilute (80 per cent.) alcohol, and the mixture is heated over the water-bath until complete saponification is effected, and the alcohol entirely removed. The soap thus formed is dissolved in 50 c.c. of water, and decomposed by adding 20 c.c. of dilute sulphuric acid (1:10). The flask is next connected with a Liebig’s condenser and the contents carefully distilled until 50 c.c. have passed over. The distillate is now freed from any insoluble acids possibly present by filtration; it is then titrated with decinormal soda solution, a few drops of litmus solution being employed as an indicator. As the result of numerous tests, it has been found that genuine butter, when examined by the above method, requires from 13 to 15 c.c. of the decinormal solution. The following are the number of c.c. required by various other fats:—

Dr. Elwyn Waller[33] modifies the foregoing method of procedure by adding 50 c.c. of water to the contents of the flask remaining after the first distillation, and again distilling off 50 c.c., the process being repeated until the final distillate neutralises only 0·1 c.c. of the decinormal alkali. With butter fat, it was found that the first distillate contained about 79 per cent. of the total volatile acids present. By means of this modification, a distinction between the rate of distillation of the volatile fatty acids of different fats is possible. The non-volatile acids left in the flask are washed several times with water, in order to remove the glycerine and potassium sulphate present, and are then dried and weighed.

For estimating the percentage of pure butter fat in a sample of mixed fat, Reichert employed the formula: B = 7·3 (m - 0·3), in which m is the number of c.c. of soda solution used in the titration.

Baron HÜbl[34] has recently suggested a method for butter testing, which is founded upon the fact that the three series of fats (acetic, acrylic, and tetrolic), unite in different proportions with the halogens (iodine, bromine, and chlorine), to form addition products. Iodine has been found especially well adapted to the examination of fats. The standard solution employed is prepared by dissolving 25 grammes of iodine in 500 c.c. of 95 per cent. alcohol, and adding to the solution a solution of 30 grammes of mercuric chloride in 500 c.c. of alcohol. The reagent is then standardised by means of a solution of 24 grammes of sodium hyposulphite in 1 litre of water. The test is applied as follows:—1 gramme of the sample under examination is introduced into a flask and dissolved in 10 c.c. of pure chloroform. The iodine reagent is then gradually added from a burette, the mixture being well shaken, until the coloration produced indicates that an excess is present, even after standing for about two hours; 15 c.c. of a 10 per cent. potassium iodide solution and 150 c.c. of water are then added and the excess of iodine present determined by means of the sodium hyposulphite solution, and deducted from the total quantity used. The amount of iodine (in grammes) absorbed is calculated to 100 grammes of the fat; this is termed the iodine number. The examination of numerous samples of genuine butter and oleomargarine, and other fats, made at the laboratory of the New York State Dairy Commissioner, furnished the following results:[35]—

It has been proposed to differentiate between butter and oleomargarine by a determination of the proportion of glycerine contained. LiebschÜtz[36] employs the following process for this estimation: 10 grammes of the sample are saponified by heating with 20 grammes of barium hydroxide, until the water of crystallisation has been almost entirely expelled. Alcohol is then added with constant stirring; saponification quickly takes place, and is completed by evaporating the mass nearly to dryness. The glycerine is extracted with boiling water, the solution filtered, and the barium contained removed by means of sulphuric acid. The filtrate from the barium sulphate is then concentrated by evaporation, and the excess of sulphuric acid present neutralised by adding a little barium carbonate. The filtered liquid is now again evaporated to a small volume, and most of the salts present precipitated by addition of absolute alcohol. After filtration the alcoholic solution is evaporated over the water-bath, then dried at 100° until constant weight is obtained. It is finally ignited and the proportion of glycerine contained estimated by the loss in weight sustained. This process is certainly far from being exact, owing principally to the volatilisation of glycerine that occurs in the evaporation of its aqueous and even alcoholic solutions. The following results were obtained upon treating genuine butter and oleomargarine according to the above method:—

Gelatine is said to have lately been used as an adulterant of butter, more especially of artificial butter. Its detection is a matter of some difficulty. The following method has been suggested. A considerable quantity of the suspected butter is boiled with water, the solution strained, a drop of acetic acid and a little potassium ferrocyanide added, and the liquid boiled until the precipitate formed becomes bluish in colour. The solution is then filtered hot and the filtrate examined for gelatine by adding tannic acid to, or conducting chlorine gas through it.

A sample lately imported under the name of “butter preservative” was found by the author to consist of a dilute solution of phosphoric acid. The use of this agent does not, however, appear to be prevalent to any great extent.[37]

Artificial Colouring.—The list of colouring matters said to be added to butter includes the vegetable dyes, annato, carotin, fustic, turmeric, marigold, and saffron; the coal-tar colour, Victoria yellow (potassium dinitrocresylate), and Martius yellow (potassium dinitronaphthalate), and the mineral pigment chrome yellow (plumbic chromate). Of the foregoing, annato and carrot colour appear to be most commonly employed. Mr. Edward W. Martin[38] has proposed a method for the isolation of the former which consists in dissolving the butter in carbon disulphide, and shaking the solution with a dilute solution of potassium hydroxide, in which the colouring matter dissolves; it is subsequently identified by further tests. According to Mr. R. W. Moore,[39] the presence of carotin in butter may be detected by first agitating the carbon disulphide solution of the fat with alcohol, which fails to extract this colour. Upon now adding to the mixture a drop of dilute ferric chloride solution, again shaking the liquid and then putting it aside for a short time, the alcoholic solution dissolves the carrot colour, and if no other colouring matter is contained in the butter, leaves the carbon disulphide colourless.

The artificially coloured butter may be dissolved in alcohol and tested with the following reagents:—

(a) Nitric acid: greenish coloration, saffron.

(b) Sugar solution and hydrochloric acid: red coloration, saffron.

(c) Ammonia: brownish coloration, turmeric.

(d) Silver nitrate: blackish coloration, marigold.

(e) Evaporate the alcoholic solution to dryness and add concentrated sulphuric acid: greenish-blue coloration, annato; blue coloration, saffron.

(f) Hydrochloric acid: decolorisation, with formation of yellow crystalline precipitate, Victoria or Martius yellow.

(g) Separation of a heavy and insoluble yellow powder, chrome yellow (see p. 130).

Microscopic Examination.—The microscopic examination of butter has lately received considerable attention as a means for the detection of the presence of foreign fats. Genuine butter generally exhibits under the microscope a crowded mass of globules of fat, fatty crystals being commonly absent. In oleomargarine a more crystalline structure is observed, with pear-shaped masses of fat and but few globules. While the presence of crystals in a sample may justly be regarded as suspicious, it is by no means a positive evidence of adulteration, since, under certain circumstances, pure butter may present the same indications. In applying the microscopic test, a small portion of the fat is made into a thin layer on the slide, and then protected with a glass cover, applied with rather gentle pressure.

Plate VI.[40] represents the microscopic appearance of genuine butter and oleomargarine. It will be observed that in butter (Fig. 1) numerous globules but no crystals of fat are presented, the crystals present being those of salt. In oleomargarine (Fig. 2) the distinctive pear-shaped masses of fat, accompanied by only a small number of fatty globules, are to be seen. Dr. Thomas Taylor (of the U.S. Department of Agriculture), has made an elaborate investigation of the microscopic appearance of various fats when viewed by polarised light. He regards the presence of peculiar globular crystals and the black cross commonly termed St. Andrew’s cross as characteristic of genuine butter.[41] Lard, beef, and other fats are said to exhibit different and, to some extent, distinctive crystalline forms. Prof. Weber,[42] however, affirms that mixtures of lard and tallow fat, under certain conditions, cannot be distinguished from butter by means of this method of examination. More recently, Dr. Taylor states that the distinguishing difference between butter and other fats under the microscope is that the former, when observed by polarised light through a selenite, exhibits a uniform tint, whereas the latter shows prismatic colours. Although the results of these investigations cannot as yet be considered as perfectly satisfactory or conclusive, they certainly are entitled to rank as a highly valuable and important step in advance of the optical processes hitherto employed.

Plate VII. exhibits the appearance of butter, oleomargarine, beef, and some other fats, when viewed by the microscope and polarised light. It will be noticed that, while a discrimination between lard and butter is readily made, oleomargarine presents the St. Andrew’s cross, stated to be characteristic of genuine butter. These photomicrographs represent the results of investigations made in the Chemical Division of the U.S. Department of Agriculture.

The question of the sanitary effects of oleomargarine and other substitutes for butter, has been studied by many scientists, and with very discordant results. Doubtless the great divergence of opinion which at present exists, is largely due to the fact that the artificial products examined have been made according to different processes, and with varying regard to the quality of the fats used in their manufacture, and to the degree of care and cleanliness observed. The attention of the American public has very lately been directed to the oleomargarine question, by the recent enactment of a national law imposing a tax upon the manufacture of the article.

Without entering to any great extent into the subject of the wholesomeness of artificial butter as it is generally met with in commerce, it will be of interest to refer to the conclusions reached by two or three sanitarians who have devoted particular attention to this aspect of the question. Prof. W. O. Atwater[43] summarises the results of his investigation of oleomargarine as follows:—

“1. The common kinds of imitation butter, oleomargarine, butterine, etc., when properly made, agree very closely in chemical composition, digestibility, and nutritive value with butter from cow’s milk.

“2. In fulfilling one of the most important functions of food, namely, that of supplying the body with heat and muscular energy, they, with butter, excel in efficiency all, or nearly all, our other common food materials.

“3. Considering the low cost at which they can be produced, as well as their palatability and nutritive value, they form a food product of very great economical importance, and one which is calculated to greatly benefit a large class of our population whose limited incomes make good dairy butter a luxury.

“4. Imitation butter, like many other manufactured food materials, is liable (but in actual commerce has been found not to be so) to be rendered unwholesome by improper materials and methods of manufacture. It is also open to the especial objection that it is largely sold as genuine butter. The interests of the public, therefore, demand that it should be subjected to competent official inspection, and that it should be sold for what it is, and not as genuine butter.”

Dr. S. B. Sharples[44] states: “When well made, it (oleomargarine) is a very fair imitation of genuine butter; being inferior to the best butter, but much superior to the low grades of butter too commonly found in the market. So far as its influence on health is concerned, I can see no objection to its use. Its sale as genuine butter is a commercial fraud, and as such, very properly condemned by the law. As to its prohibition by law, the same law which prohibited it should also prohibit the sale of lard and tallow, and more especially all low-grade butters, which are far more injurious to health than a good sweet article of oleomargarine. A good deal has been said in regard to the poor grade of fats from which the oleomargarine is made. Any one making such assertions in regard to the fats is simply ignorant of the whole subject. When a fat has become in the least tainted, it can no longer be used for this purpose, as it is impossible to remove the odour from the fat after it has once acquired it.”

Per contra, Dr. R. B. Clark, in an exhaustive report on butter,[45] affirms with great decision, that artificial butter is not a wholesome article of food, for the following reasons:—

“1. On account of its indigestibility.

“2. On account of its insolubility when made from animal fats.

“3. On account of its liability to carry germs of disease into the human system.

“4. On account of the probability of its containing, when made under certain patents, unhealthy ingredients.”

The two last grounds for condemning oleomargarine are evidently affected by, and, in fact, dependent upon the character of the fat and the exercise of care employed in its manufacture. In regard to the relative digestibility of butter and its imitations, actual experiments have been made by several chemists. A. Mayer,[46] from the results of feeding human beings for three days on butter and on oleomargarine, found that 1·6 per cent. less of the latter was absorbed by the system than of the former, and inclines to the opinion, that with healthy persons this proportion is so inconsiderable, that it is of little or no importance. Dr. Clark considers these experiments of too limited duration to be regarded as conclusive, although, so far as they went, the results reached coincided with those obtained by him by a more exhaustive investigation. Dr. Clark has made an examination of the artificial digestion of butter as compared with oleomargarine and other fats, including beef and mutton suet, and lard, cotton-seed, sesamÉ, and cod-liver oils. The method of examination pursued was as follows:—About 2 grammes of the melted fat was added to a digestive fluid consisting of 0·33 gramme of “extractum pancreatis,” and 0·33 gramme of sodium bicarbonate, dissolved in 10 c.c. of distilled water. This mixture was introduced into a test-tube, well shaken, and then exposed to a temperature of 40°. The contents of the test-tube were microscopically examined at the lapse of intervals of one, four, and twelve hours. It was found from these tests that cod-liver oil exhibited the most perfect state of emulsion, after which came genuine butter, next lard oil, and then commercial “oleo.” Plate VIII. represents the results obtained from the experiments made with butter and commercial oleomargarine, as presented at the end of one, four, and twelve hours. The globules of butter-fat, it will be observed, are smaller in size and more uniform in appearance. Dr. Clark likewise instituted experiments which tended to demonstrate the relative insolubility of the fats used in the preparation of artificial butter.

PLATE VIII.

ARTIFICIAL DIGESTION OF FATS.