the fat that must be present in ice cream. Foreign fats have been ruled out in many cases where a tendency to use them became evident. Foreign fats will never be used in ice cream in the United States as long as we give the public a reasonable proportion of butterfat in ice cream. If the butterfat content is cut to a point where it is difficult to detect, butterfat substitutes might easily be used without public detection. We should not educate the public against the use of butterfat in ice cream; rather should we keep before it constantly the fact that it is used. Ice cream does not have the creamy, delicious flavor when made containing 8 to 10 per cent butterfat that it has when made containing 14 to 16 per cent of butterfat. No one has denied this fact in all the opinions that have been advanced on this subject. It has been stated that the sick, invalids, and the old will relish ice cream of a lower fat standard. This fact must be granted, but all do not come under this classification. The general public must be the judge, and ice cream made of a composition and at such a cost that it can be sold to meet with the approval of the greatest number of people in a given locality. This, in the writer's mind, is just what is slowly taking place with most of the manufacturers over the United States.

Burke's (7) tables, which indicate that the higher consumption of ice cream per capita takes place in those States having a low fat standard, do not tell the story of the public's preference. It is doubtful if the ice cream in any State in the Union, where an 8 per cent standard is in force, will average so low. Many ice-cream manufacturers tell us that it is the laboring class of people that eat the bulk of ice cream made in the United States. This factor certainly would have to be considered in looking at the consumption in States such as Pennsylvania, Ohio, and Indiana, where the laboring class predominates. A study of economic conditions indicates that with strikes, lack of demand for labor, etc., ice-cream consumption drops. If it were possible to make a survey of the quality of ice cream made in the United States, the writer ventures to state that this factor would be found to be one of the most important factors affecting the consumption of ice cream per capita in the States. Pennsylvania is known to make a high-quality ice cream, particularily in its larger cities. The State's standard for plain ice cream is 8 per cent, yet the average in the State is nearer to 11.5 per cent fat and 36 per cent total solids. It is not unusual in Pennsylvania to find ice cream testing 38 per cent total solids. Perhaps it is this factor that makes Pennsylvania the leading State in the Union both as to the production of ice cream and the consumption per capita.

The American public delights in seeing the game of competition. The American business man seems to thrive upon it. Low State standards have given us keen competition, and after the mist has cleared we will learn that after all the public was the judge, and on the pedestal of success will stand its favorite. High quality will rule.

Though much discussion has been given to the composition of the ice-cream mix, too little attention has been given in the past to the quality of dairy products used in ice cream as well as the methods of preparing and handling the mix. Uniform methods will never be found in the United States until more uniform raw products are

produced. A few States require Pasteurization of the ice-cream mix, while in one or two instances where Pasteurization was a law the law was repealed. Only a few States give any attention to viscolization or homogenization, permitting these procedures. The bacterial content of ice cream has been largely enforced by our larger cities; no States have taken up the enforcement of this factor. More attention should be given to methods and sanitary conditions in the future. It is doubtful whether we should invite action through our State governments on these matters. However, Pasteurization should be enforced in every State in the Union, as it adds materially to the safety of the product as a food. The demand for high quality will bring about other changes in the industry as the public is willing to pay for them.

CONCLUSIONS.

1. The composition of the ice-cream mix should be that which meets with the public demand.

2. The public should not be allowed to lose sight of the fact that ice cream is a dairy product.

3. As a rule, the public prefers a rich, mellow ice cream, such as can made from a mix containing 14 per cent butterfat, 9 per cent milk solids not fat, 15 per cent sucrose, and 5 per cent gelatin. A 12 per cent fat standard seems very much in favor in the United States.

4. The entire mix should be Pasteurized at a temperature of at least 145° F. for 30 minutes.

5. Experimental work is badly needed on methods of care and handling the mix in the ice-cream factory.

6. The quality of ice cream must be at least maintained if consumption of ice cream is increased in the future as it has in the past.

REFERENCES.

1. MOJONNIER, TIMOTHY, and TROY, HUGH C. The technical control of dairy products. 1922.

2. FISK, WALTER W. 3. WASHBURN, R. M.

v. 13, no. 3, p. 29. 4. WASHBURN, R. M. v. 5, no. 13, p. 162. 5. WASHBURN, R. M.

v. 6, no. 9, p. 78.

6. RUEHE, H. A. 1923.

7. BURKE, A. D.

1922.

The book of ice cream, p. 3. 1919.

Fat standards and food values. Ice Cream Trade Jour.. 1917.

What is a fair standard for ice cream? Ice Cream Rev..
1922.

The ice cream standard question again.
1923.

Ice Cream Rev

Ice cream standards. Ice Cream Rev., v. 6, no. 9, p. 78.

Who eats ice cream? Ice Cream Rev., v. 6, no. 4, p. 92.

8. WILLIAMS, O. E., and CAMPBELL, G. R.

The effect of composition on the

palatability of ice cream. U. S. Dept. Agr. Bull. no. 1161. 1923. 9. AMBROSE, A. S. A study of the relation of the composition of the mix to the quality of the finished ice cream. Jour. of Dairy Science, v. 6, no. 5, pp. 446-454. 1923.

SANDY ICE CREAM.

CHESTER D. DAHLE, instructor in dairy husbandry, University of Minnesota, University Farm, St. Paul, Minn.

Sandiness in ice cream has caused more worry to the ice-cream manufacturer than any other technical problem. This defect of ice cream is characterized by the appearance of gritty, sand-like sub

stances occurring in the finished product. Ice cream so affected is unpalatable and unsalable.

Sandiness in ice cream has been a source of worry ever since the manufacturer recognized the possibility of increasing the total solid content of the ice-cream mix by the addition of milk-serum solids obtained from condensed milk or powdered milk. It was evident that this practice would increase the profits because of the improved condition of the finished product, the increase in overrun, and the replacement of some of the fat in the mix with cheaper milk solids. Additional serum solids made it possible for the manufacturer to make a desirable product from a mix having a lower fat content.

Coincidently, the origin of sandy ice cream dates back to the inception of condensed milk as an ingredient in the ice-cream mix. The first cases of sandiness that occurred were baffling. Sandiness appeared in almost every locality where condensed milk was used. It was not consistent in its appearance. One retailer might report a case of sandy ice cream, while another dealer supplied from the same source might not. Ice cream which became sandy in the retailer's pack but which did not become sandy in the hardening room offered another problem. Usually sandiness was not noticed until the ice cream was a few days old. Sometimes it occurred in the hardening room before it went out to the trade, but more often it occurred after having been placed in a retailing cabinet.

Although the cause of sandiness was baffling, one thing was cer tain-sandiness was unknown until condensed milk was used. When condensed milk was eliminated from the mix the trouble ceased. Not wishing to return to earlier practices for obvious reasons, the ice-cream manufacturer continued to add the extra serum solids cautiously, at the same time making note of all conditions relative to sandy occurrence.

In some

Inasmuch as the cause was attached to the condensed milk, examinations were made of some samples of condensed milk. samples of highly condensed milk crystals were found which had the characteristics of sand. Condensed milk containing these crystals has been termed "sandy" by Hunziker (1). These crystals were more often found in the bottom of containers which had stood at a low temperature. Some believed sandy ice cream was caused by adding these crystals to the mix before freezing; others found that Pasteurization of the mix eliminated these crystals. Reducing the concentration of the condensed milk tended to decrease sandy occurrences for a time. In the course of a short time, however, sporadic cases occurred which increased the worries of the manufacturer.

The only remedy for the difficulty seemed to lie in decreasing the amount of milk-serum solids added to the mix, and in more rapid consumption of the ice cream. This was apparently the only remedy if condensed milk was to be used. The remedy worked very successfully until a spell of cool or rainy weather occurred, causing the ice cream to be held longer in the pack, a condition which favored sandiness. Epidemics of sandy ice cream often followed prolonged cold spells.

CAUSE OF SAND CRYSTALS.

Up to this time no sure or certain relief from this trouble was in sight. Almost every theory had been disproved. The first investi

gator to offer real enlightenment on the subject was F. H. Bothell (2). This investigator experienced the trouble from the beginning. His first important contribution to the cause was to offer experimental proof that the sandiness was the result of crystal growth at low temperatures subsequent to the freezing of ice cream. He mentioned that any crystals or sandiness existing in condensed milk before addition to the mix would dissolve on Pasteurization of the mix, though they may reappear after freezing. He added that these crystals were lactose crystals, though no proof of this statement was offered. Another point brought out by this investigator was to the effect that the concentration of condensed milk added to the mix was not so important as the amount of serum solids added in the condensed milk. He pointed out that decreasing the concentration of condensed milk while still using the same gallonage may eliminat? the trouble, but this would be due to lowering the amount of serum solids added.

It appeared now that a solution was found for the difficulty which had become so widespread, but it was a year before a complete proof was presented fixing firmly the cause of the sandy crystals and associating it with the lactose of the condensed milk. Zoller and Williams (3) isolated the crystals and examined them microscopically. They worked with a mix containing 10 per cent fat, 14 per cent sugar, and 12 per cent serum solids. The raw products were 40 per cent cream (ripened), evaporated skim milk, and pure cane sugar. The mix was frozen with a 10° F. brine, and the ice cream packed in 2-quart tins and placed in the hardening box. The tins were unpacked every other day, and on the fourth day fine crystals appeared. The cans of ice cream were "heat shocked" by setting them in contact with air for 20 minutes, causing the crystals to grow larger. The crystals grew in size throughout 20 days. Upon examination these crystals were found to have the characteristic shape of lactose hydrate crystals. These crystals present a wedge-shaped or tomahawk-shaped appearance. These authors experimenting on crystallization of lactose noted with interest "the rapidity with which these crystals formed in the presence of complex protective substances which ordinarily tend to check crystallization of many other compounds." Such complex protective substances as colloids, acid, and cane sugar were referred to. The Purdue Station (4) reported in 1920 some evidence that sandy occurrences may be inhibited by certain gums, gelatin, and coagulants. The work done indicates that these substances acted as protectors against crystal growth. The writer will refer to this phase under the heading "Experimental."

Thus far the cause of sandiness in ice cream has been established beyond doubt. The next important item is to consider what factors will influence the occurrence of the crystals. Zoller and Williams have already stated that heat shocking helped materially to cause greater and more rapid crystal growth. Evans (5) explains this phenomenon as an "increase of diffusibility, resulting consequently in a more rapid formation of lactose crystals."

The concentration of milk sugar in the mix offers another angle of observation. Bothell chose 10 per cent lactose, based on the water

of the mix, as the maximum amount to use. He used this figure because he found that this concentration of pure lactose in water gave no crystals when frozen. Williams (6) placed 5.85 per cent lactose, on the basis of total mix, as the maximum to use. In another paper (7) the writer recommended 10 to 12 per cent serum solids as the maximum amount for a mix containing 12 per cent fat, 14 per cent sugar, and 0.5 gelatin, based on the total mix. The mix containing 10 per cent serum solids would contain 5.41 per cent lactose on the basis of the whole mix, and 8.52 per cent on the basis of the water. The mix containing 12 per cent serum solids would contain €.49 per cent lactose on the basis of the total mix and 10.55 per cent on the basis of the water in the mix. The latter mix would be in danger of becoming sandy in a shorter time.

To set a definite standard which will satisfy all conditions and mixes can not be recommended at this time. Mixes vary in amount of different ingredients, which makes this difficult. If a given amount of lactose on the basis of the mix is satisfactory in one case it may not be in another, since the per cent on the basis of water may vary. Table 1 bears this out.

TABLE 1.-Two mixes containing the same lactose content but varying in per cent of lactose on the basis of the water in the mix.

It will be noted that both mixes represent about the extreme amounts of fat and sucrose for practical mixes. Both, however, contain 5.84 per cent lactose on the basis of the total mix, but do not contain the same amount of lactose on the basis of the other ingredients, such as water and total solids. The mix No. 2 would be in danger of becoming sandy sooner than mix No. 1. In the first place, there is more lactose in the water of the mix, and, in the second place, the total solids are higher, causing more substances, such as the sucrose, to compete for the same water as the lactose.

SOME PROPERTIES OF LACTOSE.

Zoller and Williams's explanation of sandiness is based on knowledge of the properties of lactose. To quote:

The solubility of lactose in water at 10° C. is about 17 grams per 100 grams of water. The solubility of sucrose at 12° is given as 198.6 grams in 100 grams of water. When these solutions are cooled they become supersaturated, provided they contain the above-mentioned quantities of the respective sugars. In case of lactose, because of its low solubility in water compared with sucrose, it is more difficult to form supersaturated solutions of any great magnitude. When its solutions are concentrated it crystallizes without much excitation.