Introduction
For many years, the use of dried skim milk powder (SMP) in a calf milk replacer (CMR) has been used as a sign of quality. The logic is easy to understand. Skim milk powder supplies casein as well as whey proteins, its protein pattern closely resembles whole milk, and casein forms a curd in the abomasum. By contrast, modern whey-based CMR contain little or no casein and don’t form the same firm curd.
The practical conclusion often goes one step farther: if some skim milk is good, then a CMR with a substantial percentage of skim milk should be better than one based on whey proteins. That conclusion has never been especially well supported by calf performance data. Older direct comparisons generally found little advantage to skim milk, and some studies actually favored whey protein concentrate (WPC) when compared to SMP.
There has been a reasonable objection to those older experiments, however. Many were conducted when calves were fed relatively restricted amounts of liquid, usually in two meals per day. Could a difference between casein and whey become important when calves are fed much more CMR? Two recent Journal of Dairy Science papers make that question especially timely.
What did the older comparisons show?
Before we get into the more recent data discussing SMP vs. WPC, some context is appropriate. One of the largest early comparisons came from Milk Specialties Company. Tomkins and Sowinski (1992) reported three trials conducted from 1989 to 1991 involving 605 Holstein bull calves. The results were published in the proceedings of the World Buiatrics Congress and American Association of Bovine Practitioners conference rather than in a peer-reviewed journal, an important limitation when considering the weight of the evidence.
Calves were randomly assigned to CMR containing 21% CP and 16.5% fat and were fed only CMR during the trials. A common WPC treatment was compared with SMP-containing formulations in which 23, 58, or 74% of the protein was supplied by SMP. Average daily gain for the SMP-containing versus WPC treatments was 0.79 vs. 0.76 kg/d in Trial 1 (245 calves), 0.57 vs. 0.56 kg/d in Trial 2 (240 calves), and 0.59 vs. 0.60 kg/d in Trial 3 (120 calves). None of the within-trial differences was significant; CMR intake and feed-to-gain were also comparable. Although the proceedings report lacks the detail of a full peer-reviewed paper, the large number of calves and the consistency among the three trials provide useful evidence that a greater contribution of SMP did not improve performance relative to WPC.
Later peer-reviewed studies at Penn State reached the same general conclusion. Terosky et al. (1997) compared four ratios of dried skim milk to WPC: 100:0, 67:33, 33:67, and 0:100. Performance of 16 Holstein bull calves was evaluated from shortly after birth to 8 weeks of age. Changing the SMP:WPC ratio did not affect growth, health, or overall apparent digestibility. Protein digestibility was, if anything, numerically slightly greater as WPC replaced skim milk.
Lammers et al. (1998) expanded the comparison to 125 calves in two trials. Calves were fed CMR at 10% of birth BW during the first 2 weeks and 12% of birth BW thereafter. In the trial in which calves received only CMR, calves fed the diets containing 67% or 100% WPC as the major protein source grew faster and used feed more efficiently than calves fed 100% SMP. When starter was offered, the source of milk protein did not affect growth or feed efficiency.
A later Finnish study added another useful comparison. Huuskonen (2017) assigned 30 Finnish Ayrshire bull calves, averaging 20 days of age at the start of the experiment, to CMR containing approximately 42%, 30%, or 0% SMP. Calves could consume up to 7.5 L of CMR daily from 20 to 62 days of age. Selected preweaning results are shown in Table 1. The 30% SMP treatment grew faster than the 0% SMP treatment, but the 42% SMP treatment was not different from either group. There were no treatment differences in feed conversion, diarrhea, cough, bloat, postweaning ADG, or overall ADG through 195 days of age.
Table 1. Selected preweaning results from Huuskonen (2017).
| SMP in CMR, % DM | Preweaning ADG, g/d | Total DMI, kg/d | DM/gain, kg/kg |
| 41.8 | 907ab | 1.69ab | 1.94 |
| 30.0 | 945a | 1.85a | 1.97 |
| 0 | 809b | 1.58b | 1.96 |
Within a column, values with different superscripts differ (P < 0.05).
Interpretation requires caution because the three CMR were complete formulations rather than identical diets differing only in SMP. As SMP declined, the amounts of whey products and hydrolyzed wheat protein changed, and the 30% SMP formula also contained more crude protein and metabolizable energy than the other formulas. Calves fed that treatment consumed more starter and total dry matter, resulting in greater energy and protein intake. Huuskonen therefore suggested that greater nutrient intake, rather than the presence of casein or clot formation itself, may have been the most important explanation for the preweaning growth difference.
Taken together, these studies directly challenged the idea that a high proportion of SMP was required for good calf performance. A reasonable criticism of much of the older literature is that calves were fed more restrictively than many calves are today. Huuskonen is a partial exception because calves were offered up to 7.5 L/d, although that study began at about 20 days of age and changed several formulation components at the same time. The question of whether casein and clot formation become more important when very young calves consume large amounts of CMR therefore remained worth testing.
The volume argument
Casein (from SMP) and whey behave differently in the abomasum. Casein coagulates in the presence of acid and chymosin, trapping much of the casein and fat in a curd. Whey proteins, lactose, minerals, and water move more rapidly out of the abomasum with the liquid phase. Thus, a large meal of a whey-based CMR could theoretically deliver nutrients to the small intestine more rapidly than an equal meal containing substantial casein.
At a low feeding rate, that difference in nutrient flow may be biologically unimportant. The small intestine may have ample capacity to digest and absorb either pattern. With a much larger meal, however, one can reasonably ask whether slowing nutrient delivery by forming a curd might improve digestion, reduce osmotic load, or otherwise benefit the calf. This is a plausible hypothesis. Plausibility, however, is not the same as evidence of improved growth.
A 2025 study pushed intake much higher
Vorndran and Steinhoff-Wagner (2025) studied Brown Swiss calves during the first week of life. For the first six feedings, calves received their dam’s colostrum and transition milk. Beginning with the seventh feeding, calves either continued receiving dam’s milk or were changed to a commercial CMR. Feed was offered twice daily in essentially unrestricted amounts.
Calves consumed very large meals. By the later phase of the experiment, intake averaged about 10.3% of birth weight per feeding – roughly twice that amount per day. This is very different from the restricted feeding programs used in much of the older literature. Calves continuing on dam’s milk grew more rapidly and had lower fecal scores than calves switched to CMR.
At first glance, that result might appear to support the value of casein and clot formation when meal size is large. The problem is that whole milk and the CMR were not nutritionally identical, as milk supplied more protein, fat, and energy. Therefore, the experiment cannot separate the effect of clotting from the effect of nutrient density or other differences between whole milk and CMR.
There is another interesting detail. The CMR contained 50% skim milk powder and about 24% WPC. Related laboratory work from the same research group subsequently reported that this CMR failed to form a firm clot when tested with rennet under several conditions. Thus, a label statement showing 50% SMP did not guarantee that the finished CMR would behave like whole milk in the abomasum.
The new Górka study
Górka and Budziński (2026) approached the SMP vs. WPC question more directly. Ninety Holstein calves were assigned at 10 days of age to milk replacers containing 0, 25, or 50% dried SMP. The formulas were designed to contain similar concentrations of protein, fat, metabolizable energy, lysine, methionine, threonine, and tryptophan.
Importantly for the feeding-level argument, calves received 7.5 L of CMR per day containing 160 g of powder/L. That supplied 1,200 g of CMR powder daily – a substantial plane of nutrition. The daily allowance was divided into three 2.5-L meals. Starter and water were offered free choice.
Table 2. Growth and feed efficiency in the Górka and Budziński (2026) study.
| Skim milk in CMR | Final BW, kg | ADG, g/d | ADG, g / kg DMI |
| 0% | 81.9 | 741 | 429 |
| 25% | 83.0 | 763 | 440 |
| 50% | 79.8 | 698 | 403 |
The 25% skim treatment produced greater ADG and feed efficiency than the 50% skim treatment, and final BW was also greater. But the 0% skim treatment was not statistically different from either skim-containing treatment. There was no progressive improvement as skim milk increased. If anything, the treatment with the most skim milk had the lowest numerical growth rate.
That pattern does not support a requirement for SMP. It also does not prove that 50% SMP is detrimental. The treatments were complete formulations, not identical diets differing only in casein concentration. As skim milk increased, the proportions of WPC, delactosed whey, and sweet whey also changed. We will return to that problem in Calf Note #287.
Did the new study answer the feeding-level objection?
Largely, yes – at least with respect to daily intake. A calf consuming 1.2 kg of CMR powder each day is not on the old 0.45- to 0.55-kg/day restricted programs that shaped much of the historical literature. Despite the considerably greater nutrient supply, adding skim milk did not improve growth relative to the whey-based formulation.
There is one remaining nuance. Daily intake and meal size are not the same thing. Górka et al. divided 7.5 L into three meals of 2.5 L. A modern calf fed 8 to 10 L in only two feedings may consume 4 to 5 L at one time. If clot formation has an advantage, it could conceivably appear only when the abomasum is challenged by very large individual meals. The Górka experiment was a high-daily-intake study, but it was not an especially large-meal study.
That hypothesis is worth testing directly. A useful experiment would use closely matched whey- and casein-containing formulas at the same high daily solids intake, but feed each diet either in two large meals or four smaller meals. The protein-source by meal-size interaction would tell us much more than another simple comparison of commercial formulas.
What about clotting in the Górka study?
The Górka paper did not measure curd formation. The SMP used in the study was classified as medium heat by WPNI, and the authors discussed processing as one possible explanation for the performance of the 50% treatment. However, they did not determine whether the 25% or 50% skim formulas actually formed an abomasal curd. We should therefore resist saying that the Górka skim-containing diets were nonclotting. We simply do not know.
This distinction matters. The Vorndran example shows that even a CMR containing 50% skim milk powder may fail an in vitro clotting test. Conversely, some carefully processed skim-containing formulas can form a firm curd. Percentage skim milk and clotting ability are related concepts, but they are not interchangeable.
Where does this leave us?
The older literature, the Huuskonen study, and the new Górka experiment point in the same general direction. A well-formulated whey-based milk replacer can support excellent calf growth, and there is no consistent evidence that adding a specified minimum percentage of skim milk improves performance. Huuskonen is useful because calves were offered up to 7.5 L/d, although the formulas also differed in whey products, wheat protein, protein concentration, and energy. The Górka study is especially useful because the calves received substantially more CMR solids than calves in many older comparisons.
This conclusion should not be turned into the opposite dogma. Skim milk powder is an excellent feed ingredient when it is properly manufactured and used in a balanced formula. The evidence simply does not justify using the percentage of skim milk on a feed tag as a stand-alone measure of CMR quality.
The next question is more fundamental: perhaps the advantage attributed to skim milk was never about the ingredient itself, but about the curd it can form in the abomasum. Calf Note #287 will examine experiments that tried to separate the effect of clotting from the many other differences between skim- and whey-based formulas.
Summary
Direct comparisons conducted for decades have not demonstrated that a high proportion of SMP is required for good calf growth. This includes a large 605-calf set of Milk Specialties trials reported in conference proceedings as well as later peer-reviewed studies. Huuskonen (2017) offered calves up to 7.5 L of CMR daily and found no overall growth or feed-efficiency advantage from increasing SMP, although interpretation was complicated by differences among the complete formulas. Vorndran and Steinhoff-Wagner (2025) showed that very large meals can make differences between whole milk and CMR biologically important, but their diets differed in nutrient density and therefore did not isolate clotting. Górka and Budziński (2026) fed 1.2 kg of CMR powder daily and still found no performance advantage of 25 or 50% skim milk over a whey-based formula. The remaining question is whether very large individual meals – rather than high daily intake – interact with protein source or clot formation.
References
Górka, P., and W. Budziński. 2026. Effect of inclusion of skim milk in milk replacer on growth performance of newborn dairy calves. J. Dairy Sci. doi:10.3168/jds.2026-29094. Link
Huuskonen, A. 2017. Effects of skim milk and whey-based milk replacers on feed intake and growth of dairy calves. J. Appl. Anim. Res. 45:480-484. doi:10.1080/09712119.2016.1217868. Link
Lammers, B. P., A. J. Heinrichs, and A. Aydin. 1998. The effect of whey protein concentrate or dried skim milk in milk replacer on calf performance and blood metabolites. J. Dairy Sci. 81:1940-1945. doi:10.3168/jds.S0022-0302(98)75767-4. Link
Terosky, T. L., A. J. Heinrichs, and L. L. Wilson. 1997. A comparison of milk protein sources in diets of calves up to eight weeks of age. J. Dairy Sci. 80:2977-2983. doi:10.3168/jds.S0022-0302(97)76264-7. Link
Tomkins, T., and J. S. Sowinski. 1992. Impact of modern milk replacer formulations on calf health and performance. Proc. World Buiatrics Congress and American Association of Bovine Practitioners Conference. Vol. 3:124-126.
Vorndran, A. M., and J. Steinhoff-Wagner. 2025. Effects of high-volume voluntary feed consumption on growth and health of Brown Swiss calves: Continued dam’s milk feeding versus early introduction of milk replacer during the first week of life. J. Dairy Sci. 108:5954-5967. doi:10.3168/jds.2025-26405. Link