of ice and a continued concentration of lactose, into the metastable area. We might even under suitable conditions pass through a supercryohydric point into the labile area without lactose separation. In this case it is quite probable that the whole mass would solidify, without crystal formation, into an amorphous mass.

Freezing experiments upon lactose-water solutions covering a wide range of concentration show that it is actually possible, if nuclei are absent, to carry these solutions well into the labile state with the separation of ice alone. Lactose solutions of 10 parts of lactose to 100 of water may be frozen solid without the visible appearance of the lactose crystals. It has been found possible to freeze a solution containing 30 parts of lactose without the separation of lactose until a few minutes after freezing starts. In one instance, as judged by the temperature at which crystallization took place, we had a solution containing 70 parts of lactose to 100 of water. In a number of instances, even when the lactose crystals formed, they were so fine that they could be made to pass through a filter paper.

This brief exposition of the significance of the freezing point and solubility curves serves to show us roughly the conditions which exist in ice cream, particularly if the solubility of lactose in the water of an ice-cream mix is not markedly affected by the presence of the other constituents, and as far as we know it is not. A few solubility results are given in the following table:

TABLE 2.--Solubility of lactose in an ice-cream mix containing 12 per cent fat, 10 per cent milk solids not fat, 14 per cent sugar, and 4 per cent gelatin.

It is quite evident, then, that, provided there are no lactose crystals present in an ice-cream mix to furnish neuclei for crystal growth, ice cream may be carried well into the labile state. It will also be true that the tendency of the lactose to crystallize after freezing will be small as long as sufficiently low temperatures are maintained. This is in keeping with Hallimond's statement that the tendency to crystallize is lessened when a certain degree of supercooling is exceeded. If crystallization has taken place, the crystals may be so fine as to escape detection, and they will of course not tend to increase in size if low temperatures are maintained. With rising temperatures, as the melting point is approached the tendency to lactose crystallization in either case will increase. We thus see why concentration and temperature-the same factors which largely control the separation of the milk sugar from ordinary solution- are factors in governing the development of sand in ice cream.

So far we have called attention to the effect of temperature, concentration, and seeding on lactose crystallization from aqueous solution, endeavoring to show the bearing of this on the separation of milk sugar from concentrated milk and ice cream.

When we are dealing with milk solutions, however, the conditions are not so simple as with pure aqueous solutions. The viscosity becomes a factor in controlling the rate of diffusion of the lactose to the growing crystal. The preliminary treatment of the milk may have been such as not only to affect viscosity, but to produce nuclei favoring crystallization. For example, we have already shown that the forewarming of a milk can produce nuclei in the form of insoluble salts which may induce lactose crystallization in both metastable and labile areas, and it is well known that forewarming markedly affects the viscosity of both evaporated and condensed milks.

Lactose does not usually give trouble in ice cream unless the concentration is high (that corresponding to a milk solids not fat concentration of about 12 per cent). It will probably never be possible to use high lactose concentrations in ice cream without producing sand. The work on ice cream now in progress at the Dairy Division laboratories has for its aim the determination of the conditions which will permit the presence of this maximum concentration of lactose without crystallization. In other words, we have determined the major factors governing the separation of lactose from aqueous solution; we are now engaged in determining the minor factors controlling the separation of milk sugar from the ice cream.

REFERENCES.

1. HUDSON, C. S. Jour. Am. Chem. Soc., v. 30, pp. 960, 1767. 1908.

2. GILLIS, J. Rec. trav. chim., t. 39, no. 1, p. 88. 1920.

3. HALLIMOND, A. F. Jour. Iron and Steel Inst., v. 105, no. 1, p. 359.

1922.

4. MIERS, H. A., and ISAAC, F. Jour. Chem. Soc., London, v. 89, p. 413. 1906. 5. OSTWALD, W. Ztschr. f. physikal. Chem., Bd. 22, S. 289. 1897. Jour. Chem. Soc., London, v. 93, p. 1739.

6. JONES, B. M.

1908.

7. MARC, R. Ztschr. f. physikal. Chem., Bd. 79, S. 71. 1912.

Chairman MARTIN. I am sure many of you have had actual experience with sandy ice cream and I know you have been interested in Mr. Leighton's discussion as to how it may be avoided.

This concludes the list of papers whose authors are present. Unless specially requested the remaining papers will be read by title.

Does anyone desire any discussion from the speakers who have read their papers this morning?

Mr. CHARLES H. SNOW (representing the Illinois Association of Ice Cream Manufacturers). I would like to ask Mr. White if he has had any experience regarding the effect of homogenizing at different. temperatures in the pan on the casein and albumin content, and I would like to hear what it is.

Mr. WHITE. Just in what way do you
Mr. SNOW. As it affects the albumin.

mean?

Mr. WHITE. I am afraid I can't give you any information on that point at all. I have had no experience along that line.

Mr. SNOW. We are preparing cream in that manner, and our experiments show there is considerable variation in final results obtained.

Professor MORTENSEN. We have carried on some experiments along that line. We have found that the greatest factor there is the agitation of the milk. You will never have the same viscosity. And that is what I mentioned to have been the greatest factor-the

clumping together of the different globules. You can take a combination of casein and you can homogenize it and you will not have any increase in the viscosity.

There is another factor of breaking up the fat clumps. We have physical chemists, and I always let them talk on that subject, because I am not prepared. But it is evident that there is more fixed serum on account of the fat globules because there is more surface exposed. But the big factor is the clumping together of fat globules. By stirring that mix (and you homogenize the mix first) you will have a great deal of viscosity. If you are stirring that mix you are breaking up the fat clumps and you are reducing the viscosity in proportion to the way you are breaking up those fat clumps.

Mr. SNOW. In our particular method of handling we do not agitate after homogenizing the mix any more than the agitation caused by the flowing over the cooler. Whatever agitation there is is all done before homogenizing. And yet our experiments with condensed milk, even skimmed condensed milk, show that under some conditions of temperature and pressure, different conditions arise as to the viscosity which are not controlled by the acidity or other conditions. Another thing. I guess it is the experience of everyone who has handled condensed milk, that homogenizing will destroy the viscosity of superheated condensed milk. You can obtain it in the pan by superheating and passing it through the homogenizer and thus reduce that viscosity. That seems to mean that those larger casein particles that have been thrown out by the superheating process are either redissolved or so finely divided as to reduce their viscosity and thickening quality in the condensed milk.

It seemed to me that in the homogenizing of both the fat and the milk in the same mixture at the same time we were carrying on two contradictory processes. In other words, the homogenization of the fat increases the viscosity, but I have also been led to believe that the homogenizing of the milk solids not fat at the same time under conditions which would give maximum viscosity from fat alone might reduce the viscosity obtainable from the milk solids not fat.

There seem to be two different types of ingredients in the mixture: and in our process when both are carried on at the same time, we find they are contradictory in their results. I wonder if any one has done any experimental work along that line.

Chairman MARTIN. Mr. Mortensen is a practical ice-cream man. I know it. I have seen him make ice cream.

Professor MORTENSEN. I am not able to answer those questions, because I haven't had the experience with the condensed milk as you have had. We have a man on those problems now, but it takes a long time to get results. The people in Washington are pretty well along in that line of work.

Mr. SNOW. I would like to say that I am an infant in this preparation of ice-cream mix in the vacuum pan and putting the whole mix through the vacuum pan. We have been doing it only this past season. We installed the vacuum pan in May, and that is all the experience I have had in it. But in the time we have had it we have had several things arise that we have had to solve, and this is one of them: We attempted to make a mix which would give satisfactory results without the use of any quality of that kind, and we found that homogenizing seemed to undo the benefits which are ordinarily obtained by the use of superheated condensed milk. When we dis

covered that we then made some experiments on superheated condensed milk to see what the effect of the homogenizing process was on it. I was surprised to find that under certain conditions of temperature and pressure you can very nearly destroy entirely the benefit of superheating by the use of a homogenizer on that product.

It has been the practice of many ice-cream manufacturers, and it was previously our practice, to place in a large vat the condensed milk which had previously been superheated, to add the other ingredients to it, and to homogenize the entire product. It seems we were wrong in adding that condensed milk in the vat and having it homogenized. Of course, the acidity of the mix controls to a great extent the effect of the homogenized product too, and it is probable we have, through the use of superheated condensed milk, cream, and whole milk in our Pasteurizers, homogenizing the entire quantity, been obtaining the benefits of a high acid content on the final viscosity; while, when we shift it to this other process, using milk as it comes from the producers, the acid content runs lower than in the same products after passing through another condensery, and more goes into the products, with the result that, from an acidity standpoint, our mix is coming through much lower in acidity than it was when we were using the former method. And I ascribe to that factor the decreased viscosity which we got when we changed. to this pan method, rather than to the fact that we are not superheating the mix we put in the pan. Our own experiments show that the homogenizing of condensed milk puts it in the condition in which it would be if it had not been superheated. We stumbled onto that discovery entirely because we thought if a homogenizer might be. good in one place it might be good in another.

Professor MORTENSEN. At what temperature do you homogenize? Mr. SNow. In this case we dropped it from 105° to 100°-at that temperature.

Professor MORTENSEN. Would that have any effect on the lactose? Mr. SNOW. The pressure approximately is 2,500 pounds per square inch. The viscosity is less at a Pasteurizing temperature. The final result is a lower viscosity. The maximum viscosity seems to occur at either a low temperature or a high temperature. You can use temperatures from 90° to 105° if you want a higher viscosity.

The acidity content of the mixes has a great deal to do with the final viscosity in the homogenizing process. We also have homogenized a great deal of sweet cream, some of it from butter and whole milk; and we find that the acidity content of the sweet butter varies a great deal, and that variation will cause a variation in the final viscosity under identical conditions. So that seems to have quite an effect.

Mr. ZOLLER. I would like to ask the gentleman whether he uses gelatin in his mix or not.

Mr. SNOW. Yes, indeed.

Mr. ZOLLER. Gelatin plays an important part in the physical conditions under which the mix is handled prior to the time it is allowed to set in the vat. I have a little different conception of what we term viscosity, and I don't intend to burden you with that now.

But I would like to say we have been working on that problem for some time, and it seems to me that in the case of the super

77612-24- -32

heated milk in the superheated process you have a relation, and by passage of that product through the homogenizer you will generally disperse that coagulated mass into gelatin; that gelatin tends to prevent the recombination of those particles, and the tendency is an apparent lower viscosity.

But I take it that we are interested in the viscosity from two standpoints. One is its effect on the acid content and the other is the effect on the overrun capacity of the mix, and our experience has been that the homogenizer will perform upon the mix the same function that superheating will perform upon the mix, providing the two mixes are constant and held under the same conditions. I figure from that basis that the heat effect in the coagulation of the other materials is simply due to the high pressure effect on the coagulation of the proteids in the other materials. And while you will not get exactly the same degree of viscosity, you will get the same kind of viscosity in both cases.

Mr. SNOW. I might say that in addition to the gelatin we make mixes both ways. Some of our mixes contain no gelatins. But there is an apparent difference in the viscosity of the mix prepared without being superheated and passed through the homogenizer, and another mix, prepared by the mixture of the cream and superheated, prepared without gelatin, and to which sugar might be added.

Mr. ZOLLER. Your experience is, then, that the two mixes are different as regards their power to yield?

Mr. SNOW. Yes; in making the viscosity. As to the power to yield, I might say that apparently the mix prepared without superheating will yield more than the one prepared by heating, but it does not have the power to retain its incorporated air so well. It will tend to make it more foamy than ice cream which will liberate the air from it. But it does not have a good texture and firmness in the product. The ability, though, to secure overrun is apparently not one of our troubles at all. But the viscosity of the final product that overruns rapidly, leaves the can and causes very rapid melting, that is a difficult condition.

Chairman MARTIN. Is there any further discussion on any of the papers or the part that has been discussed? It is regrettable we can't bring more discussion out of this meeting.

(Adjournment.)

(Papers read by title):

STANDARDIZING THE ICE-CREAM MIX.

WILLES BARNES COMBS, associate professor of dairy manufactures, Pennsyl vania State College, State College, Pa.

The term "standardize" as used in connection with dairy products may be construed to have several meanings. Webster's New Interna tional Dictionary of the English Language defines the term "standardize" as follows: "To reduce to or compare with a standard; to render standard; to determine the strength, scale, value, etc.. of, by comparison with some standard; as, to standardize a solution; to standardize a voltmeter." Dairy specialists for a number of years have used the term as referring to an alteration or modification of the chemical composition of milk or cream. Mojonnier and Troy (1),