SEED-TESTING.

New Zealand Journal of Agriculture, Volume VI, Issue 5, 15 May 1913, Page 481

SEED-TESTING.

THE DETERMINATION OF PURITY.

A. H. COCKAYNE.

Agricultural seeds probably vary more in quality, and therefore in value, than any other commodity that the .farmer has to buy. At the same time they are, weight for weight, the most expensive of any of his recurring purchases. It is consequently obvious that great care should be exercised in the buying of seeds, both with regard to their fitness for producing a satisfactory crop and : to determine which sample shows the best value for the price demanded ; or, in other words, which sample is the cheapest to sow. It can be taken as an axiom that- seeds offered below their ordinary market value are always expensive. The reverse is, unfortunately, not always true with normal and high-priced samples, and thus the determination of their relation of value to cost becomes an important consideration. At the present time New Zealand farmers almost always buy their seeds on a mere superficial examination of the general appearance, and this is in many cases comparatively valueless so far as determining their fitness for agricultural purposes is concerned. It is quite valueless in determining, the relation..of value- to - cost. This is especially true of such seeds as those of the grasses, clovers, turnips, and mangels, that vary more in quality than any others . that the farmer handles. In. most cases the price is used to determine the .quality rather than, as it should' be, the quality determining the price. Again,, with high-priced lines, the farmer nearly always relies on the general appearance rather than- on a determination of those, special qualities that should at least be the real reason for making such seed command a higher figure than lower-priced . samples. The only proper, method .of deciding the fitness of a sample for agricultural purposes and the proper relation of value to cost is by a thorough test of both its ..purity and germination.- The buying of seeds, on .their general appearance alone instead of with a knowledge ,of their purity and germination is very nearly equivalent .to the. purchase of a dairy cow on appearance instead of on her capacity for producing, butter-fat.

The . Department has now a small but fairly, well equipped laboratory for the purposes of seed-testing. , Farmers are- therefore

now enabled to have agricultural seeds tested free .of charge, and with considerably more expedition. than has formerly been the case. It is hoped that farmers will take full advantage of the facilities that are now being provided by the Department in this connection. Indeed, the large number of samples being received makes it apparent that the value of this work is being recognized. With regard to the methods adopted by the Department, some of these are at considerable variance with those in use in many of the main seed-testing stations in other countries, and a consideration of the reasons that have led to the rejection of certain of the foreign methods appears necessary.

The Functions of a Seed-analysis

Before discussing the. merits and demerits of the system employed by the Department in the testing of agricultural seeds it is well to briefly outline what the proper analysis of any sample should determine.

The functions that should be performed by a proper, and exhaustive analysis can be grouped under three headings, namely,—

(1.) The accurate determination as to whether the sample is- a suitable one for sowing.

(2.) The determination of the relative value of one sample to another;

(3.) The determination whether the price asked is a fair one, taking the average market price of the highest-grade seed as a guide.

In many cases, especially with regard to those seeds that represent distinct or improved strains, trueness to type is also a very important consideration ; but this cannot be determined by an analysis. • It necessitates the critical examination of the growing crop from which the seed is to be harvested.

The fitness of any - sample for - sowing depends primarily on the character and number of the foreign or extraneous seeds that it contains. If -these foreign seeds are those of plants that will in any way seriously affect the crop that is ■to be grown, the sample must be considered ' quite unfit for the farmer to use. They will not only cause a reduction in the crop, but may also act detrimentally on the productive. capacity of the land for many • years indeed, perhaps permanently reduce its value for agricultural purposes. Thus the accurate determination of the amount and kinds of foreign seeds present is essential in any seed-analysis.

The other factor that determines the fitness for sowing is the capacity of growth or germination. Samples that show a very low germination are comparatively valueless. They should always be rejected, as their price rarely compensates for the small percentage of living seed they contain. A low germination j however, need not

always preclude the seed being used, but the presence of dangerous weed - seeds should always condemn a sample, . -no matter whether the germination is satisfactory or not. A low germination can often be ' remedied by 1 increasing the amount of seed that is sown, but there is no method of profitably employing a sample . that contains dangerous and aggressive weeds. It is” thus seen that the ascertaining of germination, although very important, is not the essential feature that determines the fitness of a sample for sowing. The germination, however, does decide the amount of any sample fit for sowing that it is necessary to use in order to secure a satisfactory crop.

The determining of the relative value and the price is in all cases based on the combined considerations of fitness for sowing and germination, or, in other words, on the comparative purity and germination of different samples. - The Composition of a Sample. Agricultural seeds that are offered for sale are, with the exception of those that are large and quite uniform in size, rarely if ever absolutely pure; and, in general, the composition of a' sample may thus comprise,— . - '

(1.) Fully developed seed of the- kind of which the sample is said to consist and capable of growing (living seed).

(2.) Fully developed seed of the kind of which the sample is said to consist and incapable of germination (dead seed). . -

(3.) Immature seed and empty husks of the kind being sold. Some of the immature seed may be capable of germination, but, presumably, they would produce weak inferior plants, (chaffy - seeds).

(4.) Seeds of plants other than that which- the sample is sold for and having some agricultural value (non-injurious foreign seeds).

(5.) Seeds of plants other than that of which the sample is said to consist and having •no agricultural value (Weed-seeds).

(6.) Fungus-spores, resting bodies, sclerotia, &c.

(7.) Insects, either living (such as mites and weevils) or dead (such as the dried bodies of numerous types).

(8.) Organic materials, such as pieces of • leaves, stems, glumes,

&c. (organic inert matter). . - (9.) Inorganic materials, such as rock and soil particles (inorganic inert matter). ‘ .

The Interpretation of the Term “Purity.” In the analysing of agricultural seeds it is customary to divide the sample into two groups—pure seed and impurities. , The purity is obtained by subtracting the percentage, of impurities from 100. Thus, a sample having 10 per cent, of impurities is said to be of

90 per cent, purity. In many cases there is no difficulty in deter 7 mining.- what portion of the , sample should be. included . among the impurities,. but with many seeds, and especially those of the grasses—where many seeds although true to name may be , more or . less devoid of kernels—there is a wide divergence, of opinion ( -.as.. to .what interpretation should be given to the term “ purity.” ■ In the . majority of . European seed-control stations all the imperfectly developed seeds, especially in the case-of the larger-seeded grasses—such as cocksfoot, rye-grass, - and the fescues — true to name," are included ■ under “impurities.” Again, at a few stations, notably the Irish one, these immature seeds are included under “ pure : seed.” As 'in . all cases the..- germination tests are only made from that seed left after all the impurities are removed,' it. is obvious that the results ; will be very different in accordance with what meaning has been given., to the term “purity.” Take an extreme case. Let us consider a sample of meadow foxtail that has been well cleaned and contains no foreign seeds or inert matter, but of which 30 per cent, of the foxtail-seeds are not properly filled. If we assume that all the filled seeds would germinate, then in the case where all the empty seeds are counted as impurities the germination of the selected seed would be 100 per cent.; but if the empty seeds were . counted as. pure seed the germination would be 70 per. cent. In the seed-testing done by this Department only foreign seeds, spores, &c., and inert matter are looked upon as impurities; and all imperfectly developed seed, provided it is true to name, is counted as pure seed, and is included in ( the germination tests. Here are the main reasons for adopting this interpretation of “purity” instead of . the ordinary European conception of the term : , . . , -

(1.) When only the- fully developed selected seed is used for the germination tests a misleading idea as to the value .of the seed is conveyed to the buyer, unless special stress is . laid on the amount of chaffy seed that is included in the percentage of - impurities. Where the germination, of the selected seed of a well-cleaned inferior line is high the seller would be induced to quote the germination test' only, and the farmer, seeing that the sample was. free from any real impurities such as foreign seeds, would be entirely misled as to the value, of the. sample.

(2.) In the case of high-grade lines containing little (if any) chaffy seed, there.- would be no difference -in -the germination of the seed if either of the above meanings of purity were applied. to the sample.

(3.) The complete separation of all the chaffy seeds is an exceedingly. difficult matter, and depends largely upon the opinion and skill of the seed-analyst, and in all' seed-testing the influence of the personality of the tester should be reduced to a minimum. . .

The Calculation of Foreign Seeds by Numbers.

All seed-control stations calculate the percentage of impurities by weight, no matter what interpretation is given to the term purity.” This is not wholly satisfactory, as the germination percentage .is always calculated by numbers. The ideal way, of course, would: be to reckon both percentages in the same terms; but, so far. as germination .is concerned, no method has been, or is likely to be, devised to ascertain it by weight. Again, with regard to • the foreign seeds present in a sample, their ■ weight often gives no proper indication of their relative abundance. It frequently occurs that these are slightly smaller and weigh less ' individually than the averagesized seeds of the kind being sold, and therefore when the percentage by numbers is taken it is found to be greater than the percentage by ' weight. • It has been shown that the essential feature in the determination of purity is the number and character of the foreign seeds, and that this determination clearly indicates whether the- sample .is a suitable one for sowing. The Department therefore calculates the foreign ' seeds •by ■ numbers, and not by weight. Thus, a sample containing three foreign seeds per hundred is said to be of ; 97 per cent, purity. It will be ! at once' said that the weakness of this system is that the inert matter, which is also an impurity, is not taken into consideration. This • weakness is, however, l in the majority of cases of comparatively minor importance. .. The Calculation of Inert Matter ■ not essential. In the first place, in all machine-dressed samples the percentage by weight of the inert material' is inappreciable, and if it does occur in excessive quantities it is a certain indication that the cleaning has been faulty. Again, the inert matter does' not in any way depreciate the value of the pure seed ' for sowing, ■ nor has it any prejudicial effect' on the resultant crop. The presence of inert- matter certainlyreduces the price that' should be paid, but only exactly by that percentage in which it occurs in a sample. ' Thus, in two samples of equal germination and . freedom from objectionable foreign seeds, in which one contains 2 per cent, more inert matter than'the other, the difference in price should be 2 per cent;; and as the percentage of inert matterat any rate, in New Zealand machine-dressed seed rarely exceeds that'figure, it need in practice hardly be taken into consideration. Whenever a sample contains a large amount of such material a special note is made of the fact, ' and mention is made in the report : submitted to the sender. With' regard to fungusspores,' living insect's,'. &c., their presence 'is ' also always ’ indicated, and if they are likely to prove of detriment they' are especially emphasized; 1 . ■ ' 11 '-

“ Real Value.”

In all European seed-testing great, stress is often laid on the necessity for ascertaining what is commonly termed the “ real value ” of a • sample.' By “real value is' meant the percentage of pure and properly developed seed capable of - germination. This is determined by multiplying the percentage of purity by the percentage of g'ermination and dividing by 100; thus, PerCemta - Be Py£itya.P^entog e germination = real value. In order that- the determination of real value should be in any way correct it is essential that no empty seeds should be included in the germination tests, as improperly developed seeds would cause a considerable error. This is due to the fact that seeds devoid, or more or less devoid, of kernels weigh individually less than properly matured ones; and, as purity is calculated by weight arid germination by number, the presence .of any chaffy seeds ■ in. the germination tests will seriously affect the correct “ real value.” In fact, in order that the real value should be absolutely correct* it would be necessary for the seeds that did not germinate to be of equal weight individually to those that sprouted. When it is considered that no sharp line of demarcation exists between fully and imperfectly developed grass-seeds, • but that the one kind merge gradually into the other, the great difficulty of determining the real value with any degree of accuracy can be appreciated. That this difficulty is one that is almost impossible to overcome is suggested by the fact that all seed-control stations allow a considerable margin of errorgenerally set down at 5 per cent.—in making this calculation.

In all cases where ‘ the impurities do riot include -all the light and chaffy seeds, and are calculated by weight, the real value cannot be in any way correctly determined, owing to the great variation in weight of the seeds used in the germination tests and to the .living ones being calculated by numbers. It is entirely due -to this that the European definition of impurity is made to include all those seeds which although correct ■in name may be more or less devoid of kernels.

The ascertaining of real value by weight is of great assistance to seed-merchants „in making their bulk purchases, as it enables any differences that may occur in sale and delivery samples being accurately adjusted between buyer and seller, and that without the arbitration by individual opinion as to the difference in value that may exist.. But so far as the farmer is concerned, and especially the New Zealand farmer, real value is not the- essential feature of seed-testing that the European school of seed-analysts would lead one' to suppose. In the first place, it is quite likely to be incorrect unless the greatest care is taken in ' separating out the chaffy seeds,

a matter which has been shown to depend very largely on the personal ability of the seed-analyst. Secondly, it gives no indication of the relative value of samples so far as the presence of foreign seeds and their amount by numbers 'is concerned. -Theoretically, a sample with a real value of 90 per cent.— one containing 90 lb. of. pure living seed per 100 lb.;is much more valuable than one of 80 per cent, real value— one containing 80 lb. of pure living seed per 100 lb. In practice, however, the sample of lower value may be far preferable to sow, as the higher-valued one might contain a far greater number of objectionable foreign seeds, and in such a case the lower-valued one would be immeasurably the better sample for the farmer to employ. ' Again, as the foreign seeds are calculated by weight, the percentage by' number might easily

be much greater if the weed-seeds are individually lighter than the seed that is being sold. With our method of calculating the foreign seeds' by • percentage in numbers a comparatively true “ real value,” based on numbers, can always be obtained with those samples' that do not contain an

appreciable percentage of ' inert matter. Thus, for example, a ryegrass sample might have ten foreign seeds per hundred, and a germination 'of 90 per cent. - If the purity is multiplied by the germination and. divided by 100 the real value by numbers is 'Bl per cent., which means that in every hundred seeds of the sample there are eighty-one rye-grass seeds, capable of germination. ' As the weight of a thousand mixed seeds is also always determined, it is easy to calculate the number of seeds per pound, and by using the real value by' numbers ' the farmer can tell • accurately the number of living seeds per pound of the .variety which he wishes to sow. It is, however, probable that real value calculated on numbers may in certain cases be misleading. This will occur in those instances where the germination and purity are more or less identical, but the number of seeds per pound varies considerably. Thus, take the case of two rye-grass samples of equal real value per hundred seeds, but where one sample has, say, 200,000 living seeds per pound and the other 250,000 living seeds per pound real value being the same—the difference in the number of living seeds per pound is due to the individual seeds of one sample being heavier than those of the other; and, as heavy seeds invariably produce more vigorous plants than light ones, the heavier-seeded line is the more preferable to sow. In the ' majority of cases investigated,, however, the better samples have generally more seeds per pound capable of r growing than, have the inferior ones.

Real value calculated on numbers lias the great - advantage of showing accurately the amount of seed that is necessary to sow to an

acre in order to secure a satisfactory crop. The calculation of the amount to sow' based on weight, and not on numbers, is no more scientifically sound than would be the planting of an orchard by weight instead of by certain definite numbers. • The only error that occurs in our method of calculating real value by numbers lies in the fact that it ’ does not take into consideration the inorganic impurities, but these are rarely of sufficient moment in properly -machine-dressed seed to.) in any way seriously prejudice the result. It can* also ■be claimed that no undressed seed should really bemused by a farmer, and that, so far as he is concerned, the Department looks with disfavour on the testing of undressed samples. Of . course, with regard to the merchant dealing . with seed in the rough, and wishing to know what loss will occur in bringing the seed up ’to a certain purity, germination, and weight, the European system for the determination of purity and germination has to be adopted; but these are cases which do not affect in any way the system that is being adopted for the testing of seed that is offered to the farmer for sale. Where a sample contains a good deal of inert matter in addition to foreign seeds, as is the case with the .bent grasses-'and danthonia, it ’is often ' necessary to weigh this material separately,, and when this occurs the real value by numbers cannot be given. In all cases, however, foreign seeds are calculated by numbers, and not by weight, as it is considered that, so far as purity is concerned, the essential feature to determine accurately is the kinds and actual number of the foreign seeds that are contained in a sample. The presence of inert matter, although it should lessen the price that is asked for any seed, does not prejudicially affect the seed after it is sown; and its determination, especially when not present in large quantities, ’is not of particular moment so far as the fitness of the sample for agricultural purposes is concerned.

■ “In Auckland there are 230 schools giving agricultural instruction (out of 594 schools and 108 -part-time schools), and as this work was proving so satisfactory it had been decided to appoint another instructor.” So said the Chairman of the Education Board (Mr. G. J. Garland) at the opening of the new Te Papapa School last week. — Zealand Graphic.

No industry is so vital -to the well-being of a nation as agriculture, and nothing is so vital to agriculture as the soil. How to use and not abuse the soil is the most important problem which faces the farmer of to-day— one worthy of the best efforts of our most profound and learned scientists, for upon its solution depends the future prosperity of the nation. — Scientific American. 1 ' ■ ,