The Moa Seed Farm

The Moa Seed Farm

New Zealand’s first commercial seed farm

A Koanga Institute member, Mrs Ruth Davidson, recently sent us a copy of Seed Growing in Central Otago: the Moa Farm by Helen Fluit, which is essentially her family story.  I contacted the author, Helen Fluit, and she gave me permission to reprint parts of the book here. As the book is no longer available, and never was publicly available, I consider this precious information for those of us saving seed seriously for the long haul, and I wish to thank Mrs Fluit for not only collating and writing the book but also giving permission for parts of it to be printed here.

The Moa Seed farm was the first commercial seed operation in New Zealand, established as part of the Otago Expansion Leagues efforts to repatriate soldiers after WW1. It was established in 1918 and began operations in 1922. Mangolds, carrots, beetroot, onions, parsnips, potatoes, sweet peas, clovers, lucerne and various grasses were the main focus at the farm until its closure in 1966 .

Under the management of J.W. Hadfield, plant selection and seed production systems were established, and Helen Fluit has reprinted some of these details in her book, which we are reprinting parts of here for us as seed savers to be thinking about!


Mr. J.W.Hadfield was responsible for the introduction of Single Plant Selection methods at the Moa Seed Farm. Messrs Barron and Dance, having both worked under him, and realising the importance of these methods used them where applicable to all seed produced. They can be effectively applied to a large number of crops from the beginning of seed production.

The following text is taken from the New Zealand Journal of Agriculture, Vol. 92. 1956, and written by Don Wilson, Horticulture instructor, Department of Agriculture, Roxburgh. The data was supplied to him by Jack Dance.

First Season

A good strain of seed is secured and sown out in the normal way. In autumn when the roots are mature they are lifted and the crop thoroughly searched for a number of excellent species. These are then carefully inspected and by elimination the number reduced to say 24 even roots with all the desirable characteristics. These 24 roots are then carefully stored over winter. So that commercial seed production can be undertaken the following season a sufficient number of roots for requirements are then selected and stored over winter.

Second Season

In spring the 24 selected roots are planted out in the middle of a block of the same variety for commercial seed production to obviate outside influences. When the seed is mature a small portion of the seed heads from each of the 24 plants is collected and stored separately. The balance of the seed of the 24 plants is harvested and used for

sowing out in the following season’s commercial seed production area.

Third Season

In spring the seed selected from individual plants is sown out in separate rows under ideal conditions. A close watch is maintained throughout the growing season for behaviour, variation in type etc. Notes are taken and any row exhibiting undesirable characteristics is eliminated. In autumn these selection plots are lifted in one operation and the product of individual rows is carefully inspected. With the assistance of the notes taken during the growing season and by a process of elimination the best all-round row is left and from this row 24 roots are selected and stored over winter. The balance of the roots from these selection plots are stored and planted out in spring to supplement the commercial area.

Fourth Season

In spring the 24 roots selected the previous autumn from the best selection plot are planted out, again in the middle of a commercial seed production block. When the seed is mature a small quantity is collected from each plant and sown in selection plots as previously described. Here again the balance of the seed from the 24 selected plants is harvested and used for sowing out in the following season’s commercial seed production area.

After this system has been in progress a number of years the seed used for the sowing of commercial seed areas is only two generations removed from a single selected plant. Its continued use ensures that purity and quality of the seed produced are at least maintained if not improved, as without selection the proportion of undesirable off-types would be likely to increase.

Root crops are biennial; from seed to seed occupies two growing seasons, so that to harvest seed annually it is necessary to sow half of the seed originally secured the first year and the remainder the next…



J. W. HADFIELD, Manager, Moa Seed Farm, Dumbarton, Central Otago

Plant Selection

On every progressive seed-farm a considerable amount of time is necessarily devoted to the production of ‘stock’, ‘elite’, or ‘stud’ seed. This stock seed, which may become the most valuable property of the seed grower, is not sold, but is used exclusively by him for sowing the crops from which he produces commercial seed. The stock seed thus forms the very foundation upon which reputable seed production is built, while upon the care and initiative of the breeder depends the quality of the seed stock. The object of these notes is to outline what we are doing on the Moa Seed Farm in plant selection and the production of stock seed. In a subsequent article the writer proposes to deal briefly with our methods of producing commercial seed.

Soil and climate conditions play a large part in the appearance and viability of the seed; but the appearance of seed which is not itself used for consumption, such as carrot, parsnip, and mangold, except in so far as it indicates satisfactory and careful harvesting, is by no means of such importance as is generally supposed. Viability, though certainly of great importance, cannot become the one determining factor, because it sometimes happens that the highest-quality stock seed is somewhat lacking in this respect. As seed-growers we consider field trials, where it is possible to determine the most important quality or defect in the seed – namely, its varietal purity or trueness to type. This fact is becoming more generally recognised by the grower; mainly as a result of the activities of the crop inspection and certification services of such countries as Canada, the United States, and Denmark. There is also at the present day a definitive movement, backed by the seedsmen themselves, to eliminate the large number of synonymous varieties and bring our nomenclature to a simple straightforward basis. The National Institute of Agriculture Botany, at Cambridge, has done a considerable amount of work in this connection, and the United States Department of Agriculture is working on the same lines.

Biennal Root Crops: Methods of Selection

The method here described is, in principle, that adopted in the selection of mangolds, beets, carrots, parsnips, and onions. Any crops which have not been grown commercially at the Moa Seed Farm are not here discussed.

Mangolds, beets, and carrots, of which we grow several varieties, are very liable to cross-pollinate, and it is necessary to ensure against this by planting a sufficient distance apart. This is of particular importance in varieties of distinct type and colour, for while we might take a little risk of cross-pollination between, say, Ox-heart carrot and Early Horn, we must guard against any possibility of cross-pollination between either of these varieties with a long scarlet or white variety.

Commencing with a crop grown from the best commercial seed procurable, we make sufficient selections for our purposes when the roots are being dug. Taking one hundred as a convenient number, these are laid out on the ground and fifty are selected, these being those which conform most closely to our ideal. The remaining fifty are reserved as second-quality selections, to which reference will be made later. The fifty first selections are numbered, notes taken if desired, and are stored away over winter in a frost-proof root-store. In the spring, when the main crop is being planted, these selections also receive attention, and are planted, whenever possible, in an isolation plot sufficiently far removed to avoid any danger of cross-pollination with other plants of even the same variety. This is certainly the most desirable method, but, unfortunately, not always practicable on a farm where one’s resources are taxed to keep even the main seed crops sufficiently apart. Under these circumstances we plant the first-quality selected roots in the centre of the main crop, and surround them on all sides with the second-quality selections to which reference has already been made. This we find the best practical method.

When seed is formed the selections are examined and quite a number are generally discarded on account of poor seeding-properties. This culling is very necessary, since we find a decided tendency for high-quality roots to produce very littleseed, and, as we are primarily seed producers, we have to strike a balance between root quality and seed-production. A great deal of variation is to be observed in the seed and seed-bearing parts of carrot, parsnip, and mangold. Most of it is of no commercial significance, but some is. In parsnip, for example, the colour merges into a purple, while other plants produce almost seedless umbels; but the most objectionable property is the tendency of certain roots to produce puffed triangular clusters, the seeds of which contain no kernel.

When the remaining plants seed, a little from each of the first-quality selections is reserved, each in a marked packet, for sowing in the trial plots the following season. The remainder of the first-selection seed is bulked, and forms seed for sowing the main crop in the following season. The seed from the second-quality roots is also harvested, and forms a reserve in case of failure, or in case the seed from the first selections is not sufficient to meet requirements.

The seed-samples reserved for the trial plots are carefully examined, and sometimes germination tests are made. The following spring they are sown in well-prepared ground in rows about 1 chain long, each sample in a separate row. Thus each row is the direct progeny of one of the originally selected roots. The area reserved for the trial plots should be in the best possible condition. This is important, since unsatisfactory conditions will induce so much malformation as to completely swamp any slight differences which might otherwise be observed between one row and another. Thinning and cultivation must be done most carefully and uniformly, and an endeavour made to allow each row full and equal development.

When these rows are dug we endeavour to decide on the best and second best. This is by far the most difficult part of the work, and may entail hours of careful consideration before a final selection is made. We adopt different methods to arrive at a conclusion. Quite a number of rows can generally be discarded for possessing some undesirable character. For the remainder we find the most satisfactory method is for one man to sort each row into selections and culls, and, from the numbers of each, calculate a figure representing the proportion of selections as against culls. This generally reduces the number down to four or five, and from each of these the fifty best roots are taken and laid out in parallel rows, and a final selection made.

Having now the fifty first selections and the fifty second selections of roots we recommence by planting these in the main crop, or in an isolation plot as already described.

In the accompanying diagram, which shows the process graphically, it can be seen that in any series seed for the main crop is produced only in alternate years. The difficulty can be overcome by having a parallel series commencing one year later, or, what is very much simpler, by producing each year sufficient bulk seed to allow for planting two main crops.

Onions are far better transplanted in the autumn, after they are thoroughly dry and ripe, but with the hard Central Otago winter we prefer to plant our selections in the spring. Brown Spanish, which is our standard variety, is difficult to select, because the colour of the skin has, we believe, never been fixed.

Annual Root Crops

Radish is the only annual root crop we have grown at the Moa Seed Farm commercially. Its extreme liability to cross-pollination renders it unwise to grow more than a single variety on the farm. The method of selection we adopt is the same in principle as that used in the case of biennial roots. The roots, however, are lifted a few weeks after sowing, when the radish is in a proper state for eating. Trial rows are judged for flavour, quality, and appearance, and selected roots are topped and immediately transplanted. They will recommence to grow at once if the soil is moist or the roots are watered after transplanting.”

… something to ponder…  Kay