Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 56, 1926
This text is also available in PDF
(487 KB) Opens in new window
– 738 –

The Quality of New Zealand Wheats and Flours.

[Abstract.]

On the whole, New Zealand wheats are good; they yield well from healthy plants, and are well suited to local climatic conditions. These are important factors from the farmer's point of view. The scantiest regard has, however, been paid to the quality, or capacity to produce the best loaf of bread, of the different varieties favoured by wheat-growers in this country.

Milling Tests.

The first investigation of the milling properties of local wheats was due to the courtesy of Mr. J. C. Bruennich, Chief Chemist, Department of Agriculture, Queensland, who milled a number of samples in 1909 (N.Z. Jour. Agric., 1910, vol. 1, p. 20), and again in 1912 (Aston, N.Z. Jour. Agric., 1912, vol. 4, p. 410). Commenting on these at the same time, Aston (ibid.) said that while it was not altogether fair, without knowing more about the origin of the samples, to compare these wheats with Queensland wheats, it might be stated that the thirty New Zealand samples tested in 1912 gave an average of 79.4 points (per 100), compared with 83.3 points

– 739 –

gained by forty-eight Queensland wheats tested in 1909, and 83.8 points by fifty samples in 1910.

Work was begun locally in 1922 by the Chemistry Section, Department of Agriculture, when milling tests were carried out with local wheat varieties, and chemical analyses made of the flours obtained. In addition, baking tests have been made in a number of cases.

The milling tests show a fairly wide variation in the percentage of flour obtained. Bruennich found the amount of flour in the New Zealand samples he milled to range from 68.1 to 76.1 per cent. Among the wheats so far tested on the Chemical Laboratory's experimental mill a range from 67.1 to 77.1 per cent has been observed.

Varieties differ considerably in the amount of flour they produce, but, owing to the influence of factors such as locality, season, soil, &c., samples within a variety differ much amongst themselves. The following table, however, calculated over three years' figures, gives the averages obtained from the better-known local varieties.

Table 1.—Averages of Flour obtained in Milling Tests on New Zealand Wheats (1923–24–25).
Variety. Average Flour
per Cent.
Number of
Samples milled.
Victor 74.5  8
Tuscan 72.7 26
Hunters 72.7  8
Velvet 72.5 18

When localities are compared some fairly consistent results have been obtained from year to year, though as yet they are necessarily very incomplete; a considerable number of samples from Canterbury have now been examined, however, and a certain significance may be attached to them. Such results as the limited data allow are given in the following table:—

Table 2.—Averages of Flour obtained in Milling Tests of Wheats from certain Wheat-growing Localities.
Locality. Average Flour per Cent. Number of Samples.
1923. 1924. 1925.
Canterbury 73.2 72.8 73.0 69
Southland 70.7 70.2  7
Lake County 72.1 72.5 16

It should be possible, as information on milling tests is gradually collected, to obtain useful figures not only for land districts but even for each county, and eventually for each wheat-growing district.

Quality of Flour.

Much work has been done on improving the yield per acre of local wheat varieties, but until the present work was initiated no very definite information was available as to their quality. Amos (Processes of Flour manufacture, 1920, p. 32) includes New Zealand wheats amongst the weak varieties, but gives no details as to the basis of his conclusions.

Quality, or strength, in wheat and flour is meant to indicate possession of good breadmaking properties—that is, properties which enable the baker to produce loaves of good volume, shape, texture, and colour. Naturally,

– 740 –

the baking test must be the final criterion of quality, and in experimental baking tests the volume of the loaf is generally taken as the best single factor on which judgment is based. Any results obtained from analyses of flours must, if they are to be of any use, necessarily be capable of correlation with baking tests.

Protein in Flour.

It has been found in other countries that, on an average, the protein content of a wheat-flour is a fair indication of its baking-qualities; exceptions are often found, but amongst a large number of samples the above statement remains true. It applies equally to New Zealand flours (N.Z. Jour. Agric., 1923, vol. 27, p. 167; 1926, vol. 32, p. 26). Moreover, different varieties are, as a whole, characterized by different average percentages of protein, though here again there are wide variations within a variety. The following table gives the average protein content of the commoner varieties tested by the Department, together with several varieties not extensively grown but nevertheless of considerable interest:—

Table 3.—Table of Averages of Protein Content of some New Zealand Wheats tested in 1923, 1924, and 1925.
Variety. Average Protein Content per Cent.
Essex Conqueror 13.75
Marquis 12.40
Velvet* 12.19
Dreadnought 10.73
Yeoman 10.50
Hunters* 10.50
Tuscan* 10.40
Victor*  8.58

The more usually grown varieties are marked with an asterisk.

It may be said quite definitely that up to the present Velvet has proved the best of the better-known varieties. The little-known variety Essex Conqueror has for three years proved the best in protein content of the miscellaneous varieties grown at the Ashburton Experimental Farm. Marquis is the famous Canadian wheat. Victor, which mills excellently, is proved to produce a flour of inferior quality.

Effect of Environment (Central Otago Wheats).

The effect of environment on strength of wheat is considerable. It has long been observed that comparatively high temperatures, long days, and absence of excessive moisture during ripening hasten maturation of the grain and increase its protein content; it was to be expected, therefore, that wheats grown in the drier districts of New Zealand would show differences from the average. Such has been the case with samples of wheats tested from Central Otago. In 1923 nine samples were received from the Tuapeka and Upper Taieri (Central Otago) districts, and all were found to possess very good strength. Some of these varieties were more or less unknown to wheat-growers, but amongst them were three samples of Velvet, all of which were prominent for strength in this collection of good wheats. Again in 1925 seven samples from Lake and Vincent Counties (Central Otago) produced loaves of very good volume. These particular flours, being low in protein content, were exceptions to the general rule that good strength is associated with good protein content; nevertheless, one of them (a sample of Tuscan) produced a loaf which compared favourably

– 741 –

with the best of the flours yet tested. The matter of low protein content and good loaf volume is referred to later (see “Degree of Buffering”).

In connection with the question of protein content it may be added that local samples of good strength have been compared with samples of wheat of accepted good strength grown in Kansas (the great wheat-growing centre in the United States) and have lost nothing in comparison (N.Z. Jour. Agric., 1923, vol. 27, p. 173).

Chemical Analyses of Flours.

The baking test is, of course, the final test of quality; but even baking tests are not infallible, and it is a matter of difficulty to obtain consistent results with every sample; since all tests must, therefore, be done at least in duplicate, the process tends to become tedious. It would be a matter of importance if some chemical method could be found which would be of more general application than the present simple one of protein-determination. Various methods have been suggested at different times, and some of them have been applied to local wheats and flours. A considerable amount of work has been done locally in this connection during the last three years, and there follows a brief summary of the results obtained.

Amended” Gliadin.—Efforts have long been made to correlate the ratio of gliadin to glutenin (the two principal proteins in wheat-flour) with strength, but without much success. Recently, however, Martin (Jour. Soc. Chem. Ind., 1920, vol. 39, p. 246) has shown that previous methods of determining gliadin were faulty; he has found that flours with high bakers' marks were those in which the percentages of what he calls “amended” gliadin were high. The writer has determined the “amended” gliadin in ten New Zealand wheat-flours, and has found that, with one exception, Martin's “amended” gliadin was a fair indication of strength (results unpublished).

Ratio of Wet to Dry Gluten.—In the International Review of the Science and Practice of Agriculture (1922, p. 1331) it is said that the breadmaking value of a flour is determined by the following ratio:

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Moist gluten/Dry gluten x 100

—that is, the more water a wet gluten (the isolated crude protein) retains under certain defined conditions the better is the breakmaking value of the flour from which it is obtained. This ratio has been determined in every sample tested, but has not been found of any practical use in determining the baking-value of New Zealand flours.

Absorption of Water.—Guthrie (N.S.W. Science Bull. No. 7, 1912) says that the proportion of water taken up by a flour may be regarded as a good indication of strength. This is true to a certain extent with local flours, but there are so many exceptions that it has no practical value.

Composition of Ash.—The U.S.A. Experimental Station Record (1923, vol. 48, p. 29) gives an abstract of data from a study of North Dakota wheats which indicates that the calcium and magnesium content of the ash varies with the loaf-volume. In a series of analyses of New Zealand flour-ash no such relationship could be found (results unpublished). When, however, the percentage of these constituents was calculated on the amount of flour, and not on the amount of ash, a certain parallel was found between the amount of calcium in the flour and the amount of protein; and also between the magnesium in the flour and the amount of protein, and the

– 742 –

ratio of wet to dry gluten. A distinct relationship was found between the percentage of phosphoric anhydride (P2O5) in the flour and the amount of ash in the flour.

Colloidal State of Flours.—A great deal of attention has been directed recently to the physical chemistry of wheat-flours in an endeavour to correlate their behaviour as colloids with their baking-properties. Gortner and Doherty have done a considerable amount of work on this problem, and they say (Jour. Agric. Research, 1918, vol. 13, p. 389) that the hydration capacity (that is, its behaviour as a hydrophyllic colloid) of a gluten under certain well-defined conditions is closely related to its strength.

The writer investigated thirty-one samples of wheat-flour in 1925, and found that the hydration capacity (which is intimately connected with viscosity) of the moist gluten of these New Zealand flours was not any useful indication of the quality of the flour as revealed by baking tests (N.Z. Jour. Agric., 1926, vol. 32, p. 93). Similar results have recently been obtained by some American investigators (Cereal Chemistry, 1925, vol. 2, p. 191).

Hydrogen-ion Concentration and Degree of Buffering * of Flours.—In investigating those flours which, in 1925, were found to prove exceptions to protein content being a fair indication of strength, it was found that in thirty-one wheats examined the degree of buffering of each flour was closely associated with its baking-properties. To eliminate the effect on loaf-volume of protein content, the flours were arranged first of all in series in order of approximate protein content. In each series (with very few exceptions, and these probably within the experimental error) there was a decided increase of loaf-volume coinciding with a decrease in degree of buffering. The following table, taken from one series out of a total of six, will illustrate this point:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table 4.—Relationship of Loaf-volume to Relative Degree of Buffering.
No. Variety. County. Protein (approx.). Loaf-volume. Relative Degree of Buffering.
Per Cent. Per Cent.
708 Tuscan Lake 8 700 1.4
711 Tuscan Lake 8 670 1.3
714 Velvet Vincent 8 650 1.2
710 Tuscan Lake 8 620 1.1
101 Hybrid W Eyre 8 580 0.9

The very good baking-properties of the Tuscan wheat (No. 708) from Lake County, Central Otago (mentioned in a previous paragraph), might thus be explained by its very light degree of buffering. It was found to

[Footnote] * The degree of buffering may be defined as the resistance to change in hydrogen-ion concentration. The buffer value of a dough is determined by several of its ingredients—proteins, milk, salt, &c.

[Footnote] † Change in pH of 25 c.c. of flour suspension + 2.5 c.c. N/50 hydrochloric acid: i.e., the greater the degree of buffering the smaller the change in pH and the smaller the figure in this column.

– 743 –

be generally true that flours more lightly buffered than the average produced good loaves even though they were low in protein content. The reverse was also true: highly buffered flours, even though possessing high protein contents, produced poorer loaves than such protein content suggested (N.Z. Jour. Agric., 1926, vol. 32, p. 98). It seems that this method may give good results with local wheat-flours.

The initial pH * of flour-doughs was also investigated, but it was found that in the thirty-one samples investigated loaf-volume and initial pH of the dough were not correlated; only a slight correlation was found between initial pH and amount of ash, and between initial pH and degree of buffering. No correlation could be found between initial pH and percentage absorption of water.

Summary:

Summarizing the results obtained during 1923, 1924, and 1925 with different varieties of wheats grown in New Zealand and the flours obtained from them, it may be said that—

(1.)

New Zealand wheats when milled experimentally produce good amounts of flour. Some varieties produce better amounts than others; the average is 72–73 per cent.

(2.)

Wheats of definitely good quality (or breadmaking value) can be grown in this country. Some of the varieties now grown produce, on an average, strong and medium-strong flours. Some of the more climatically suitable wheat-growing areas (e.g., Central Otago) grow wheat which is usually of very good strength.

(3.)

The protein content of local wheats is a fair average indication of their quality.

(4.)

Methods of determining strength depending upon the percentage absorption of water, or on the ratios of wet to dry gluten, have little application to New Zealand flours.

(5.)

Analysis of ash of flours failed to show any relationship between the amounts of calcium or magnesium in the ash and the baking-value. A certain parallel was found between the amounts of calcium and magnesium in the flour and the protein content, and between the magnesium in the flour and the ratio of wet to dry gluten. A distinct relationship was found between the amount of phosphoric anhydride (P2O6) in the flour and the amount of ash.

(6.)

A determination of the hydration capacity (which is connected with viscosity) of gluten from New Zealand flours has not proved of any practical value in determining their baking-values.

(7.)

The relative degree of buffering amongst flours of approximately the same protein content has been a good indication of baking-value, and better than any other method. Its determination should prove, with local flours at least, a useful guide to baking-values.

(8.)

The initial pH of the flour-dough has so far not been an indication of baking-value.

[Footnote] * pH is used to express the hydrogen-ion concentration, or intensity factor of acidity, and is the log. to the base 10 of the hydrogen-ion concentration, the negative sign being omitted: i.e., pH = log. 1/(H+. A solution is said to be normal m respect to hydroden ions when it contains 1.0 grm. of hydrogen ions per litre. An N/10 solution of HCl is expressed as pH 1.0; exact neutrality, pH 7.0.