The Relative Vitamin G Content of Dried Whey and Dried Skimmilk*

The Relative Vitamin G Content of Dried Whey and Dried Skimmilk*

The Relative Vitamin G Content of Dried Whey and Dried Skimmilk* V I C T O R HEiMANf Cornell University, Agricultural Experiment Station, Ithaca, New...

708KB Sizes 0 Downloads 11 Views

The Relative Vitamin G Content of Dried Whey and Dried Skimmilk* V I C T O R HEiMANf

Cornell University, Agricultural Experiment Station, Ithaca, New York (Received for Publication June 23, 1934)

T

• T h e research work reported in this paper was done in partial fulfillment of the requirements for the degree of Doctor of Philosophy at Cornell University and was made possible through the establishment of a special temporary fellowship by the Kraft-Phenix Cheese Corporation of Chicago, Illinois. t N o w at the State College of Washington, Pullman. The writer is indebted to L. C. Norris and G. F. Heuser for advice and suggestions in conducting this investigation.

relative vitamin G content. Accordingly, work on the fundamental phases of this problem was begun in 1931, followed by studies giving consideration to the practical application of the more fundamental results. Mussehl and Ackerson (1932) shortly afterwards reported that "there is no evidence of special nutritive value to be accredited to dried whey as compared to dried buttermilk when fed to growing chicks at the same level of protein." These workers compared dried whey with dried buttermilk on an equivalent protein basis and, since dried whey is a low-protein feed, it was necessary to include nearly three times the quantity of dried whey in the ration. Furthermore, in one experiment the diet was supplemented with yeast. These procedures, coupled with the fact that no control groups were included, made interpretation of their results impossible in terms of vitamin G. At the same time Bender and Supplee (1932) reported results with rats, showing that the vitamin G content of roller-process dried skimmilk and spray-process dried whey was not substantially different. These workers, however, called attention to abnormal fluctuations of the growth curves at the later stages of growth which were thought to be due to an unknown factor. Furthermore, they used only four rats per group and the results obtained were variable. At the beginning of this study a hypothesis was evolved based on the fact that vitamin G is water-soluble and therefore presumably not adsorbed on the milk sol-

[137]

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

HE extensive use of milk products and leafy green plants such as alfalfa in the poultry ration indicates that poultrymen have found them essential, but the underlying cause of the benefits derived from these products was not understood until recently. Now, with a more complete knowledge of the nutritive requirements of poultry, the vitamin G content of milk products and leafy greens is considered their chief contribution to poultry rations. Hauge and Carrick (1926) were the first to present evidence showing that a growthpromoting factor, now known as vitamin G, is required by poultry in addition to vitamin B. Later Norris and co-workers (1930, 1931, 1933) and Bethke et al. (1931) confirmed these results and called attention to the great requirement of poultry for this vitamin. On account of this the vitamin G content of milk by-products and other sources of this vitamin becomes an important practical consideration. Since dried skimmilk and dried whey (milk sugar feed) are two of the milk byproducts readily available for use in poultry rations, it. was felt desirable to study their

138

SCIENCE

heavy mortality, and severe pellagrous symptoms were obtained in all lots. The addition of even 20 percent of either of these milk products did not alter the severity of the disease. It was evident, therefore, that milk by-products are not rich in the pellagra-preventing factor of Ringrose et al. and that they are not of importance in practical feeding in preventing the development of this pellagra-like syndrome. As a result of this it was possible to limit this investigation to a study of the growthpromoting and anti-paralytic phases of vitamin G. A purified-casein diet was used for this purpose, since it was found to produce extremely poor growth in chicks, accompanied by nutritional paralysis of varying severity, mild or no pellagra-like lesions and somewhat higher mortality than is considered normal. The addition of sufficient dried whey or dried skimmilk to this basal ration resulted in normal growth and the disappearance of the accompanying abnormalities. EXPERIMENTAL In all growth studies involving this problem, the Cornell strain of White Leghorn chicks was used with the exception of Experiment IV, for which White Leghorn chicks were obtained from a nearby poultry farm. These were divided by random selection one day after hatching into the required number of groups of twenty-five chicks each. All chicks were individually weighed and identified with a numbered wing band. They were individually weighed weekly thereafter. Notations of abnormal symptoms were made at this time. All growth data was weighted for sex influence. The experimental period was eight weeks. The brooding equipment was the same for each group within any experiment. Brooders and runs were provided with hardware cloth floors to minimize possible ex-

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

ids. If no adsorption occurs, then liquid whey, which is milk freed of a large portion of its protein and certain other solids, would be a slightly better source of vitamin G than an equal quantity of the whole milk from which it was made or an equal quantity of skimmilk. Furthermore, owing to the greater reduction of milk solids, 15 to 16 pounds of liquid whey are required to make a pound of dried whey, while only 10 to 11 pounds of liquid skimmilk are required for one pound of dried skimmilk. Therefore dried whey should represent about a 50 percent greater concentration of vitamin G than dried skimmilk. In conducting these studies of the relative vitamin G content of dried whey and dried skimmilk, the work was planned so as to prove or disprove this hypothesis. The problem was complicated at first by the probability that vitamin G is a complex consisting of two or more phases as pointed out by Ringrose and co-workers (1931) and later by Bethke et al. (1931). Although it has not been established that the first two phases are separate, the factors of the complex may be designated as follows: (1) A growth-promoting (hatchability-promoting) phase; (2) An anti-paralytic phase; (3) An anti-pellagric phase. Only the first two phases of this complex appeared of practical importance, since the pellagra-like lesions and concurrent slow growth in chicks as reported by Ringrose and co-workers, resulting from the feeding of egg-white and purified-casein diets, are phenomena still confined to the experimental laboratory. It was important, however, to establish whether or not the anti-pellagric phase was a factor involved in the experimental work reported here. Therefore a study of the value of dried whey and dried skimmilk in preventing pellagra-like lesions in chicks was made using the egg-white diet of Ringrose et al. In this experiment, poor growth,

POULTRY

MAY,

193 5.

VOL.

XIV, No. 3

gave a greater growth response, whereas at the higher levels no differences were obtained. On account of this, it was felt desirable to work out a curve of diminishing increment for this vitamin, using a standard source, as the relative vitamin G potency of different sources could then be accurately determined after plotting the gain over the control upon this curve, provided this gain was less than normal. Experiment I Experiment I was planned, therefore, to deter mine the effect of graded additions of vitamin G

1

i

/ ' /

l

// i

7

200

Ir

1

• /

f / /

1

:peri lent

cpari lent • * [peri lent

m E A E [peri Lent • R

/ 0

/ /

0.0

::i :II

rv

Pa: a boli of abolc of

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8,0

he si cond orde; he f r a t t r d e r

9.0

10.0 Per oent dried whey

FIG. 1. Effect of graded additions of vitamin G upon the rate of gain of chicks. upon the rate of gain. Dried whey was selected as the source of vitamin G to use in this experiment. Five samples were collected from different manufacturing plants in the Middle West. A uniform portion of each of these products was mixed into a composite sample. The levels of the composite sample of dried whey fed were 0.0, 0.63, 1.2S, 2.5, 5.0, and 10.0 percent. In addition, each of the indi-

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

perimental error induced by coprophagy. Water and feed were provided ad libidum. A fifteen-hour day was maintained by an automatic lighting system. The basal diet in all fundamental experiments reported here was of the same type. Feed ingredients sufficient for each experiment were obtained in advance, thoroughly mixed, and analyzed for protein and moisture. By varying the quantity of yellow corn meal and purified casein, the diets were adjusted to a protein content of 25 percent of the dry matter (approximately 22 percent air-dry basis). A typical control diet consisted of 60.25 percent yellow corn meal, 20.0 percent wheat flour middlings, 15.25 percent purified casein, 2.0 percent steamed bone meal, 1.0 percent pulverized limestone, 1.0 percent codliver oil, and 0.5 percent salt. Such a diet is deficient in vitamin G, but adequate in all other respects so far as the nutritional requirements of chicks are at present understood. A weekly record of feed consumption was made. The data obtained, however, were not used in compiling this report, as it was found that the effect of the cumulative increase in the feed consumption of the driedwhey groups was compensated by the decline in the relative effectiveness of dried whey. This appeared to be caused by a reduction in need for vitamin G during the latter part of the experimental period and the resultant coming into play of the law of diminishing increment (Spillman, 1924). A report of studies of the relationship of the rate of growth to the vitamin G requirement will be made later. In preliminary experiments it was found that in order to show differences in the growth rate induced by dried whey and dried skimmilk it was necessary to feed them at levels of five percent or less. At these levels it was found that the dried-whey group

139

140

P O U L T R Y

TABLE 1.—Summary of growth results, Experiment

Variable in diet

Lot

Neg. control 2.5% skimmilk 2.5% whey 5.0% skimmilk 5.0% whey 10.0% skimmilk 10.0% whey

1 2,3,4,5 6, 7, 8,9 10,11,12,13 14,15,16,17 18,19, 20,21 22, 23, 24,25

Mortality

S C I E N C E

II

Av. wt. at 8 wks. (grams)

0 1, 0, 0,1 0,0,0,1 1,1,1,1 0,0,0,0 0,1,0,0 0,0,1,1

252.6 + 9.76 380.9 + 6.75 451.3 + 7.27 536,0 + 7.01 580.3 + 6.33 642.9 + 5.71 640.2 + 5.66

The vitamin G ratio = l:1.55 + 0.26 at the 2.5 percent level.

Whe

W —C

G

M —C

1

W = the mean weight of a group of chicks at eight weeks of age on the basal diet previously described, supplemented with any suboptimum level of the dried whey to be tested. C = the mean weight of the control group at eight weeks. M — the mean weight of a group of chicks at eight weeks receiving dried skimmilk as the vitamin G supplement at the same level as fed to group W. G •— — the vitamin G ratio 1

In applying this formula to the data obtained in the preliminary experiment, a vitamin G ratio of 1:1.40 ± 0.23 at the 5.0 percent level was ob-

tained. This is in close agreement with the theoretical ratio of approximately 1:1.5 in which case dried whey contains 50 percent more vitamin G than dried skimmilk. Experiment

II

Having arrived at a suitable method for accurately determining relative vitamin G content, studies of the vitamin G potency of representative samples of dried whey and dried skimmilk were undertaken. Four samples of each of these products were obtained from different manufacturing plants. Each sample was fed in the basal diet previously described at levels of 2.5, 5.0, and 10.0 percent. This procedure resulted in four replicates of each level of both products. The variation in vitamin G content between the different samples of each product was not found to be great. Therefore, the growth data presented in Table 1 are the average weights of the four groups of chicks on each level of the two products. The vitamin G ratio found was in excellent accord with the theoretical ratio. Furthermore, at the 10 percent levels of dried whey and dried skimmilk, both products satisfied the vitamin G requirement, showing that only one factor was responsible for the increase in growth rate. Experiments III and IV In the work thus far it was felt that some criticism could be made concerning the method of obtaining samples of these products on account of variability in the raw milk and method of manufacture. To obviate these difficulties samples of spray-process dried skimmilk, spray-process dried whey, regular-process dried whey, and Swiss cheese were prepared from a single standardized lot of liquid whole milk. It was necessary, however, because of certain technicalities in the manufacture of regular-process dried whey, to use a large volume of liquid whey and in order to do this, part of

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

vidual samples was fed at the 2.5 percent level. The weights of these six groups, however, were averaged. The results of this experiment along with the pertinent results of Experiments II, III, and IV are presented graphically in Figure 1. Each point on the chart shows the grams gain over the negative control of a group of chicks eight weeks of age. It is apparent from these data that there is a curvilinear relationship between the increments of vitamin G fed and the grams gain over the control group. However, before the law of diminishing increment becomes strongly effective there is an almost arithmetical relationship at the critical levels of vitamin G intake, as shown by the line of best fit plotted from the 0.0 to the 5.0 percent levels. It is not a serious error then to assume that arithmetically progressive increments of vitamin G at suboptimum levels produce arithmetically progressive gains over the control group in growing chicks. With this simple relationship, the following formula may be readily applied to give the relative vitamin G content of dried whey and dried skimmilk using the vitamin G content of the latter product as unity:

141

the entire run of whey for the day was included in the sample of the regular^process dried whey used in Experiment III. In Experiment IV, however, these difficulties were overcome and it was possible to prepare all samples from the same lot of liquid whole milk. Since Experiment IV was simply a repetition of Experiment I I I the results of both series were combined for presentation in Table 2. The results with dried whey and dried skimmilk and the vitamin G ratios obtained confirmed the previous work. The samples of spray-process dried whey were included here to show that this process of manufacture did not effect the vitamin G content appreciably. Since the dried skimmilk and spray-process dried whey were dried by the same method and were obtained originally from the same sample, a further check on the vitamin G potency was obtained. The vitamin G content of Swiss cheese was found to be extremely low, since even at the 4.0 percent level the difference in mean weight above the control group was small and of questionable significance. The group receiving 2.0 percent Swiss cheese plus 2.0 percent dried whey was included to demonstrate that Swiss cheese contains nothing inhibitory to growth in chicks but was merely low in vitamin G. The mean weight of this group corresponded closely to that of the group receiving 2.0 percent of dried whey only. These results therefore indirectly confirmed the original hypothesis.

vitamin G is greenish-yellow water-soluble pigment, probably identical with lactochrome. If vitamin G were such a pigment, it seemed reasonable to expect that there should exist differences in the intensity of color of extracts from milk products which were shown by biological assay to contain varying quantities of vitamin G. In order to isolate the pigment from other interfering substances of milk which mask the color, the principle of diffusion into a gelatin membrane was used. A 1.5 percent gelatin solution was prepared and 2S cc. portions were put into petri dishes and allowed to solidify. One part of dried skimmilk or dried whey was then added to four parts of water, and the suspension was thoroughly stirred and allowed to stand overnight. Volumes of 25 cc. were then pipetted onto the solidified gelatin and allowed to stand in the refrigerator for forty-eight hours. At the end of this time the supernatant liquid was washed off with cold water. The gelatin was then carefully dissolved and stirred and finally transferred to test tubes. The intensity of the greenish yellow color could then be readily compared by observation. The intensity of color of extracts of the dried skimmilk and dried whey, fed in Experiment III, and the dried skimmilk, dried whey, and sprayprocess dried whey fed in Experiment IV were compared by this method. No differences in the intensity of color in any of the dried whey diffusates could be seen. However, the diffusates from the dried skimmilk samples showed considerably less pigmentation than the dried whey samples. These results were in agreement with those ob-

1935.

VOL.

Experiment V The recent work of Kuhn and co-workers (1933), Booher (1933), and others suggest that

TABLE 2.—Summary of growth results Experiment I I I Variable in diet Neg. control 2% dried skimmilk 2% dried whey 2 % spray dried whey 2% Swiss cheese 2% Swiss cheese 2% dried whey 4 % dried skimmilk 4 % dried whey 4 % spray dried whey 4 % Swiss cheese 8% dried whey

Mortality

Avg. wt. at 8 wks. (grams)

3 4 3

208.2± 6.82 312.4± 7.86 388.6 + 12.04

5 1 2 2 0 2 0

Experiment IV Mortality

Avg. wt. at 8 wks. (grams)

196.9± 8.56

4 0 1 0 1

230.9+ 8.50 318.3+ 7.49 395.1 + 13.82 419.7 + 11.44 230.9± 8.63

378.7± 9.67 426.5 + 10.90 515.8±13.76 470.2±15.64 238.0+ 6.09 537.8±15.48

0 1 0 1 0

436.0+11.95 508.6+14.67 508.5 + 17.53 277.7 + 14.59 545.3±10.77

Exp. III. The vitamin Gratio = 1:1.73 ±0.22 at the 2.Opercent level and 1:1.41 + 0.11 at the 4.0 percent level. Exp. IV. The vitamin Gratio = l : 1 . 8 8 ± 0 . 3 1 at the 2.0 percent level and 1:1.35 + 0.25 at the 4.Opercent level.

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

XIV, No. 3

MAY,

-.42

POULTRY

SCIENCE

TABLE 3.—Summary of results on egg production and maintenance of mature birds, Experiment VI (29 pullets per lot)

Factor Production Eggs per bird Avg. egg weight (grams) Feed consumed per bird (grams) Avg. initial body weight (grams) Avg. final body weight (grams) Avg. body weight for 10 periods (grams) Total mortality

Lotl. Control 57.0% 159.7 51.8 28,174 1,767 1,697

Lot 2. 2.5% dried skimmilk

Lot 3. 2.5% dried whey

Lot 4. 5.0% dried skimmilk

Lot 5. 5.0% dried whey

59.8% 167.3 51.7 29,069 1,747 1,671

50.9% 142.5 51.7 28,350 1,804 1,785

51.6% 144.3 51.2 26,192 1,721 1,657

58.1% 162.7 52.5 28,460 1,765 1,810

1,733 16

Experiment VI Studies of the relative vitamin G content of dried whey and dried skimmilk were also made with laying hens. A uniformly raised lot of 145 twenty-eight-week-old White Leghorn pullets of the Cornell strain were divided into five groups of twenty-nine birds each by random selection. Identical housing facilities were provided for all groups. The pens were steam heated to provide a more nearly uniform temperature throughout the winter months. The birds were trap-nested and daily individual egg records were kept. The total weight of the eggs from each pen was recorded daily. The weekly feed consumption by pens was determined. The condition of the birds throughout the experiment was ascertained by general observations and individual weighings at four week intervals. The duration of the experiment was 280 days (ten periods) .

/

The dried skimmilk was obtained upon the open market and only a single brand of this product was used. The dried whey came from the same manufacturing plant. Enough feed was mixed from each lot of ingredients to last eight weeks. Fresh samples of dried skimmilk and dried whey were used for each mixture. Each ingredient of this ration was analyzed for protein and the protein content of the ration adjusted to 16 percent, air-dry basis. A typical ration consisted of 47.9 percent yellow corn meal, 20.0 wheat flour middlings, 20.0 wheat bran,

1,788 13

-

1,852 11

1,721 16

5.6 purified Argentine casein, 3.5 oyster shell flour, 1.5 steamed bone meal, 1.0 cod liver oil and 0.5 percent salt. Table 3 gives the summary of the egg production, egg size, feed consumption, and mortality obtained in this experiment. The differences in results do not show any relationship with the vitamin G content of the diets. This was confirmed by applying analysis of variance to the egg production data. The results of this analysis are presented in Tables 4 and 5. The standard error of the mean egg production was 4.1 percent. By applying this value to the mean production of the various groups, the standard error was found to be ± 0.65, ± 0.69, ± 0.58, ± 0.59, and ± 0.67 eggs for the mean periodic production of groups 1, 2, 3, 4, and 5 respectively. TABLE 4.—Results of applying analysis of variance to egg production data, Experiment VI Sums of squares

Degrees of freedom

Periods Lots Error

462.44 64.20 68.13

9 4 36

Total

594.77

49

Variance 51.38 16.05 1.89

It will be noted in this analysis that the D / E values between lot 1 (the negative control) and the other lots, receiving various increments of vitamin G, show doubtful or no statistical significance. The other groups, however, show significant variations among themselves, but they are of an inconsistent nature. It is, therefore, justifiable to conclude that, although there are statistically significant differences in the egg production of certain groups, these differences cannot be attributed to the vitamin G content of the diet.

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

tained in biological assays. They indicate the possibility that lactochrome, the greenish-yellow pigment of milk serum, is the growth-promoting phase of the vitamin G complex, discussed previously in this paper. Preliminary determinations, by R. A. Sullivan of this laboratory, of the lactochrome content of these milk by-products by means of a photoelectric cell also showed correlation with biological vitamin G value.

1,749 12

143

TABLE 5.—-D/E values between the mean egg production of various groups, Experiment VI

responding group receiving dried skimmilk. This indicates, therefore, that at the 5.0 percent levels of dried whey or dried skimmilk, the vitamin G requirement for hatchability was satisfied. Since each hatch from the five groups of pullets was conducted at the same time under identical conditions, Student's method of statistical analysis was applied to the experimental results. The average difference in hatchability between lots 2 and 3 was found to be 9.9 percent with a z value of 1.81 which in terms of odds is greater than 87 to 1. It is, therefore, a significant difference. If it is assumed that the same relation of vitamin G to growth holds for the hatchability of eggs, the formula for determining the vitamin G ratio of dried skimmilk and dried whey may be applied to the hatchability results. This was found to be 1:1.S8 at the 2.5 percent level. This ratio is in

Between lots 1 & 2 lotsl&3 lots 1 & 4 lots 1 & 5

1935.

VOL.

Between

D/E

lots 2 & 3 lots2&4 lots 5 & 3 lots5&4

0.80 1.98 1.78 0.32

D/E 2.73 2.55 2.27 2.07

At the twentieth week of the experiment, two male birds were placed in each pen. These were rotated among the pens daily to avoid any influence that might accrue from individual differences among the males. After a ten-day preliminary period, eggs were saved for weekly intervals and held in a refrigerator at 45 to 50°F. Five hatches were incubated in the same incubator under conditions as nearly comparable as possible.

TABLE 6.—Summary of hatchability results, Experiment VI Lot 2 2.5% dried skimmilk

Lot 1 Control Total eggs set Fertility Dead germs on 7th day* Dead germs on 18th day* Hatchability f

225 85.3% 11.5% 50.0% 21.9%

383 86.7% 5.7% 38.9% 38.9%

Lot 3 2.5% dried whey

Lot 4 5.0% dried skimmilk

241 88.4% 4.7% 21.6% 48.8%

Lot 5 5.0% dried whey 404 81.4% 4.0% 9.1% 70.5%

270 80.7% 2.8% 5.5% 69.7%

* Expressed as percent dead germs of fertile eggs, t Percentage good chicks of fertile eggs set. The average results of these hatches are presented in Table 6. The average fertility varied somewhat with each group, but there seemed to be no relationship between the fertility and the vitamin G content of the diets. The hatchability of the fertile eggs, however, appeared to be directly related with the vitamin G content of the diet. In groups 4 and S receiving the 5.0 percent levels of dried skimmilk and dried whey the hatchability was the same. In the groups receiving the 2.5 percent level of these products, the pullets receiving the dried whey produced eggs which had a consistently greater hatchability than the eggs from the cor-

excellent agreement with those obtained in growth studies and again confirms the original hypothesis. A further study was run in order to demonstrate that the vitamin G reserve of the chicks hatched was related to the vitamin G content of the diet of the dams. The chicks obtained from the fifth hatch of the above hatchability experiment were used for this study. Each chick was weighed and wing-banded and all five groups were then combined in a single large brooder pen. These chicks received the purified casein negative control diet previously described. The chicks from all groups of pullets were thus treated in exactly the

TABLE 7.—Summary of growth results, Experiment VI Lot

Diet of dams

Number chicks at start

1 2 3 4 5

Control 2.5% skimmilk 2 5% whey 5.0% skimmilk

11 22 20 23 24

5.0% whey

Total mortality 8 13 10 6 8

Avg. wt. at 6 wks. (grams) 76.0 + 79 0 + 121.0 + 122.0 + 133.3 +

2.81 5.03 6.86 4.80 6.82

-7-,

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

XIV, No. 3

MAY,

144

Experiment VII The work reported thus far has dealt with the relative vitamin G content of dried whey and dried skimmilk from a more or less fundamental standpoint. It was felt desirable, therefore, to check the results thus far obtained by making practical applications. In undertaking studies of this character compensation had to be made in the practical rations used for the lower protein content of dried whey by resorting to proteins of presumably lower quality than casein. In the course of several experiments it was found that sardine meal could be satisfactorily used as the sole source of supplementary protein. The protein level fed was 25 percent of the dry matter. The vitamin G ratio obtained at the S.O percent level of dried whey and dried skimmilk was 1 ;1.63

SCIENCE

± 0.28 in favor of the dried whey group. The groups receiving optimum levels of both milk products were normal in all respects and their mean weights were approximately the same. In a further experiment using 10.0 percent meat scrap with purified casein to bring the protein content of the diet to 25 percent of the dry matter, a vitamin G ratio of 1:1.53 ± 0.39 at the 1.25 percent level was obtained. In further work it was found that soybean oil meal could also be used satisfactorily in conjunction with animal proteins to compensate for the low protein content of dried whey. With these results as a basis, Experiment VII was conducted as a final demonstration that dried whey could be used to replace dried skimmilk in the diet of growing chicks. The basal diet used in this experiment contained a combination of feeds which can be regarded as forming a complete ration for chicks with the exception of the vitamin G content. The ingredients of the negative control diet consisted of 38.2 percent yellow corn meal, 20.0 wheat flour middlings, 10.0 wheat bran, 10.0 ground oats, 5.0 meat scrap, 5.0 fish meal, 8.3 soybean oil meal, 2.0 pulverized limestone, 1.0 cod liver oil, and 0.5 percent salt. The protein content of this diet was 19.1 percent (air-dry basis). The other diets were adjusted to this level of protein by decreasing the quantity of soybean oil meal and corn meal. It was assumed that the relative vitamin G content of dried skimmilk and dried whey was established at 1:1.5 using dried skimmilk as unity. The samples of these products used to supplement the above diet were the same as were used in Experiment IV. It will be recalled that these samples were prepared from the same original lot of liquid whole milk and were, therefore, strictly comparable. The dried whey was included in the diet at levels of 2.0, 3.0, 4.0, 5.0, and 6.0 percent while the dried skimmilk was fed at levels of 3.0, 4.5, 6.0, 7.5, and 9.0 percent. It is apparent from this experimental outline that no significant growth differences should appear between any two comparable lots provided the basic assumption was correct. This experiment was also planned to show how much vitamin G in the form of these milk-products is required for capacity growth under practical conditions. The results of this experiment are summarized in Table 8. It is evident from these data that in no case did the differences in mean weight at eight weeks of age on comparable levels of dried whey and dried skimmilk approach statistical significance. These results showed further that no advantage in weight increase was gained by increasing the amounts of dried whey in the diet above 4.0 per-

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

same manner, and were reared under identical conditions. Individual weekly weighings were made and observations were recorded at this time. The results of this experiment are summarized in Table 7. It was not felt that these results were quantitative because of the limited number of individuals used in this experiment. The influence of the vitamin G content of the diet of the hen on the growth and livability of the resulting chicks, however, is clearly demonstrated. These results indicate that vitamin G is stored in the egg. This is in agreement with results reported by Bethke et al. (1933) which showed that vitamin G is stored in the egg in accordance with the amount present in the laying ration. A direct vitamin G assay of the egg whites and egg yolks from eggs produced by these pullets was also made. The results, however, were not as convincing as those already presented. The differences in the gains of groups of chicks on eggs from the pullets, receiving large and small amounts of vitamin G in their diets, was relatively small. It was observed, however, that the dried egg whites contained considerably more vitamin G than the dried egg yolks. It was also observed that the hatchability of these eggs and the growth of the chicks hatched was related to the degree of pigmentation of the egg whites. This pigmentation was in turn correlated with the vitamin G content of the rations fed the hens. The difference in the amount of pigment in the white of eggs produced by hens, receiving a purified casein diet and those receiving the same diet plus S.O percent of dried whey, was discernible. This is interpreted as supporting the opinion of Kuhn et al. (1933) that lactochrome and the pigment of egg white are identical.

POULTRY

1935.

145

cent and the amount of dried milk above 6.0 percent. The groups on these levels of milk products were normal in all respects and the mortality of all groups was negligible.

comparable owing to the rigid method of sampling employed, that the actual vitamin G ratio of dried skimmilk to dried whey was very close to 1:1.5. The original hypothesis was given further support by indirect evidence when it was shown that cheese, made from the same milk as the comparable milk by-products, contained only a very small quantity of vitamin G, which could be accounted for in part by the normal whey content of the cheese. The number of eggs laid by pullets fed a diet deficient in vitamin G was approximately normal, indicating that the requirement for this vitamin for egg production is very small. The hatchability of eggs produced by pullets fed suboptimum levels of dried whey and dried skimmilk was found to be significantly higher in the group receiving dried whey and resulted in a vitamin G ratio of 1:1.58. The rate of growth on a vitamin G deficient diet of chicks hatched from pullets receiving dried whey and dried skimmilk was roughly proportional to the hatchability results and to quantity of vitamin G in the diets fed. Evidence was obtained that the vitamin G content of the eggs produced and the greenish-yellow pigment of the white of the eggs was directly related to vitamin G content of the diets and that the dried egg whites were relatively richer in vitamin G than the dried yolks. Methods developed for estimiting the relative amounts of lactochrome, the greenish yellow water-soluble pigment of milk, showed that there was a larger amount of this pigment in dried whey than in dried skimmilk and that the quantity present was correlated with growth response and hatchability. The results obtained in feeding dried whey and dried skimmilk to chicks receiv-

VOL.

TABLE 8.—Summary of growth results, Experiment VII Lot

Variable in diet

Mortality

Avg. wt. at 8 wks. (grams)

1 2 3 4 5 6 7 8 9 10 11

Control 3.0%skimmilk 2 . 0 % whey 4.5%skimmilk 3.0% whey 6.0%skimmilk 4 . 0 % whey 7 . 5 % skimmilk 5.0% whey 9.0% skimmilk 6.0% whey

0 0 0 1 0 0 0 0 0 2 0

570.7 + 13.60 628.2 + 12.39 636.4+ 6.90 631.6+17.11 641.5+ 8.91 661.4+ 8.91 670.6+10.48 681.7+ 8.76 693.8+ 9.97 657.1 + 10.77 679.7+ 6.72

It is necessary, however, to regard the value of so-called practical experiments in the study of the vitamin G requirement for growth as exceedingly limited. This is especially true because of variability in the vitamin G content of the ingredients of the diet. It is, therefore, impossible to state exactly what minimum quantity of any vitamin G supplement will be necessary in a practical chick ration to furnish the optimum growth response unless the vitamin G content of the other ingredients is known. SUMMARY

This investigation was based upon the hypothesis that dried whey contains about 1.5 times as much vitamin G as dried skimmilk. It was established in preliminary work that the phases of the vitamin G complex of practical importance in milk were the growth-promoting (hatchability-promoting) and anti-paralytic phases. The relationship of the quantity of vitamin G in the diet to the resulting growth response was established and a formula was deduced with which to calculate the relative vitamin G content of dried skimmilk and dried whey, using the vitamin G content of the former as unity and expressing the results as a vitamin G ratio. It was found by repeated assays of several samples, some of which were strictly

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

XIV, No. 3

MAY,

146

POULTRY

ing a simplified diet containing purified casein as the supplementary source of protein were confirmed with diets of a more practical nature. A practical diet containing dried whey as the chief source of vitamin G was evolved which gave satisfactory growth results with chicks. Dried whey or dried skimmilk can be used interchangeably in this diet with the same results, provided the proper protein adjustments are made.

Bender, R. C. and G. C. Supplee, 1932. The vitamin " B " and "G" content of dry skimmilk and dry whey. Jour, of Dairy Sci., 15 :445-450. Bethke, R. M., P. R. Record and D. C. Kennard, 1931. A type of nutritional leg paralysis affecting chicks. Poul. Sci., 10:355-368. Bethke, R. M., P. R. Record and D. C. Kennard, 1933. Relation of the vitamin G complex to hatchability and nutritive value of eggs. Poul. Sci., 12 :332. Booher, Lela E., 1933. The concentration and probable chemical nature of vitamin G. Jour. Bio. Chem., 102:39-46.

Hauge, S. M. and C. W. Carrick, 1926. A differentiation between the water-soluble growth-promoting and antineuritic substances. Jour. Bio. Chem. 69:403-413. Kuhn, R., P. Gyorgy and T. Wagner-Jauregg, 1933. tiber Lactofiavin, den Farbstoff der Molke. Ber. deutsch chem. Ges., 66:1034-1038. Kuhn, R. and T. Wagner-Jauregg. uber die aus Eiklar und Milch isolierten Flavine. Ber. deutsch. chem. Ges., 66:1577-1582. Mussehl, F. E. and C. W. Ackerson, 1932. The growth promoting values of dried buttermilk, dried skimmilk and dried whey for chicks. Poul. Sci., 11:69-73. Norris, L. C , G. F. Heuser, and H. S. Wilgus, Jr., 1930. Is the chief value of milk for feeding poultry due to the presence of a new vitamin? Poul. Sci., 9:133-140. Norris, L. C , G. F. Heuser, A. T. Ringrose, H. S. Wilgus, Jr., and Victor Heiman, 1933. The vitamin-G requirement of poultry. Report of the Fifth World's Poultry Congress, Rome, Italy, Section 2a, No. 40. Ringrose, A. T., L. C. Norris, and G. F . Heuser, 1931. The occurrence of a pellagra-like syndrome in chicks. Poul. Sci., 10:166-176. Spillman, W. J., 1924. The law of diminishing increment. The Law of Diminishing Returns, World Book Co., Chicago, Part 1:1-77.

Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 20, 2015

REFERENCES

SCIENCE