The Journal of Agriculture.

New Zealand Journal of Agriculture, Volume XII, Issue 6, 20 June 1916, Page 421

The Journal of Agriculture.

WELLINGTON, NEW ZEALAND.

20TH JUNE, 1916.

CONVERSION OF FERN-LAND INTO GRASS.

A. H. COCKAYNE,

Biologist

INTRODUCTION.

THE sown grass-lands of New Zealand- not entering into any : special rotational system are of particular interest, owing to their being in general of ' quite recent origin. Although there are grass-lands in the Dominion that have been established over half a century, all the various types of pasture can be traced right from their primeval condition to their present - state. All the changes and modifications of vegetation that : occur in ’ their development can be studied step by step from the original plant-formation into grass-land, the derivation - of which it would be impossible to tell from its present condition alone. In all parts of New Zealand, but more ■ extensively in the North Island, the evolution of artificial pastures can be seen in all its ' varied phases, • and the origin of artificial grass-land can be investigated with a ■ detail that is impossible in older-settled ' countries, - where the origin of many of the types of permanent pasture is wrapped in obscurity.

Another point that is of surpassing interest from an ecological view is the fact that in primitive New Zealand there were no grass-lands that could be termed meadows and that could be cut for hay—that is, no grass-land showing seasonal growth, and where the herbage was more or less even-aged, was present. In fact, the only primitive grass-lands the tussock areas were composed mainly of grasses that, so far as their vegetative development is concerned, occupy a position nearer to shrubs than they do to the ordinary flat-leaved grasses which comprise the main vegetation of true meadows. The present artificial and artificially induced pastures of New Zealand are nearly all of the meadow type, and although not normally cut for hay to any appreciable extent (less than half of i per cent, are utilized for hay-production), they would be very largely so dealt with were our climate sufficiently rigorous as to necessitate general stall feeding during the winter months. How New Zealand has converted her primitive vegetation into meadow-land is a matter of intense scientific interest, bearing as it does on the evolution of grass-land in general. At the same time its study is of far-reaching practical importance, as the process is still in operation. A complete knowledge of the basic principles involved, and the influence and interaction of the various factors concerned, would enable the art of the conversion of primitive vegetation into grass-land to be applied with even greater certainty of success than has been possible in the past. Each year in New Zealand the conversion of over a quarter of a million acres of native vegetation into grass is undertaken, and at the present time some fifteen million acres have been so dealt with. A fair proportion of this area is now under some type of rotational system, but over ten million acres is still occupied by a vegetation derived from the original grass mixture with which it was sown, often profoundly modified according to the conditions to which it has been subjected. Owing to the topography of much of New Zealand inhibiting any general use of the plough, much of this land is destined to remain in grass “ for all time," and with this in mind the importance of success in the conversion of native vegetation into grass becomes clear. The effect of the failure of a crop under a rotation system is generally quite transient, but initial failure in the conversion of primeval vegetation into grass spells ruin to the land from a pasture point of view, or at least enormously increases the difficulty and cost of finally establishing a payable and permanent crop.

THE VARIOUS TYPES OF GRASS-LAND.

Excluding rotation grass, the principal grass-lands of New Zealand may be roughly classified under the following headings according to what type of vegetation they have replaced : Coastal- Grass-land. The area devoted to this class of pasture is comparatively small. Typical examples may be seen over a large portion of the Wanganui coastal places and . elsewhere along the coasts of both Islands. The majority of the grass-land of the Moumahaki Experimental Farm was of this type, but is

mainly now of a rotation character. Frequently coastal-forest grassland a few years after the original sowing turns gradually into Danthonia pasture, as can be well seen on the coastal hills around Wellington.

Rain-forest Grass-land. — This class of grass-land comprises many millions of acres,. mainly in. the North Island, although there are very large tracts in the South Island.. A feature of these grasslands is the general predominance of cocksfoot, although, of course, many other grasses occupy an important position. The majority

of “bush-burns” in general can be relegated to rain forest, and the grass-land derived from it comprises much of the very best of the permanent grazing-land of New Zealand.

Southern-beech Forest Grass-land— -This occupies a comparatively small area, and forest-land. of this type is amongst the most difficult of any in New Zealand to grass successfully. In many cases the earlier attempts resulted in failure, as can be well seen on the low hills • along the eastern side of the Hutt Valley, near Wellington, where the land has in many cases reverted to manuka.

Tussock Grass-lands. — These are mainly located in the South Island, and are of various types. Much of the land-originally covered with the lowland tussock type is now under, rotational grass, nearly all the cereal-growing districts having , been lowland tussock in the past.' In Southland large areas of tall tussock (Danthonia Raoulii) moor- have been converted by drainage and liming into both rotation and permanent grass-land. . '

Reclaimed-swamp Grass-lands. — This type of pasture is scattered throughout both Islands, and many of the areas comprise the best fattening-land in the Dominion. Their conversion has been attended with much difficulty, and the great expenditure necessary to drain them sufficiently to grow grass has been a bar to their rapid conversion into grass-land.

Heath Grass-lands. — These have been derived from two main typeswet heath, being very closely related to moor, and dry heath, mainly comprising fern-land and manuka scrub.

The above classification is an extremely rough one, but it serves to show from what varied types of original vegetation our grasslands have been derived.

THE CONVERSION OF FERN-LAND INTO GRASS.

In dealing with the conversion of fern-heath into grass it is well to understand that there are two main types of fernland. The first is what I term natural fern-land, and which has been continuously in fern, even prior to the time that the first Europeans landed in the country. The other type, which I term artificially induced fern-land, did not carry fern before being brought under occupation. Fern-land in New Zealand occupied very large areas prior to . the advent of settlement; but, arguing from what has occurred over a very large extent of bushburn country, I am . inclined .to think that much of the land now occupied by what appears to be a .primitive vegetation became clothed with fern subsequent to extensive forest fires occurring in

the pre-European days of this country. It is a notable fact that many types of New Zealand forest-land when burnt and • then left unstocked do not as a normal rule reproduce forest, but turn rapidly into fern. This feature is of such an outstanding nature that it would appear safe to generalize and say that all fern-lands in New Zealand, both primitive and induced, with a few notable exceptions, occupy the site of former forest, and that a plant covering of fern is one of the stages through which, under natural conditions, the land must pass before forest is again developed.

When this view is taken it is seen that there is not much difference in origin between natural and artificially induced fernlandthe destruction of forest, either naturally or artificially, being the all-essential requisite for the production of both types.. If the history of the vegetation were carried back far enough 'to the plant - colonization. of actually virgin ground, it seems feasible to expect that fern in all cases preceded the development of forest. A study of the fern-lands of the Central Volcanic Plateau seems to fit in well with this hypothesis.

There is, however, a very signal difference between natural and artificially induced fern-lands. The vegetation of the former is

of very considerable age, whereas that of the latter is of very recent origin, and its actual commencement can be seen in many parts of New Zealand. In addition to this, the vegetation of natural fern-land is much easier to eradicate than that of artificially induced fern-land. In certain cases quite light stocking of primitive fern-land has resulted in the eradication of the fern, as, for instance, in the Te Kuiti district, where fern-land has frequently been successfully broken in and the fern got rid of by stocking with milkingcows alone. Artificially induced fern, when properly established, could not possibly be dealt with in this way. The greater difficulty of dealing with induced fern is probably due to the fact that much natural fern-land, especially where tutu is becoming abundant, is in one of its later stages of development before the land again becomes clothed with forest, and the fern is naturally not so aggressive as when in the stage immediately following its successful invasion of previously forested country. ' - ..

The grass-lands that have been converted from the type of heath I term natural fern-land are of very considerable extent, comprising some two million acres, the larger portion being located in the east coast region of the North Island, and also in the southern portion of the Auckland Province. Much . that has been converted for many years —and this is particularly true of large areas in Hawke’s Bayno longer shows any signs of ever having been covered with a dense growth of fern, but the ready development of the plant on unutilized land adjoining such grass-lands shows clearly the previous ! condition of 'the land. During the past few years there has been a great extension' in ■ the work of converting fern-land in the King-country, and large areas still remain to be transformed -into- pastures. —ln- -this district all - stages in the conversion of fern-land into grass can be seen —from those where the dominance of the fern is quite unaltered to those where fern is apparently quite non-existent.

With regard to the eradication of fern on artificially induced ferndand much work has been accomplished, and many hundreds of thousands of acres where fern had attained the mastery over grasses sown on bush-burns have been more or less cleared of the weed. In many places, however, such as the Marlborough Sounds, artificially induced fern-land covers large areas, and the eradication of the fern is yearly becoming more difficult. Theoretically, induced fern should not exist, provided proper methods to avoid its establishment were carried out, and it is probable that with a better knowledge of the necessary initial treatment of bush-burns the development of artificially induced fern could be enormously diminished.

Before detailing the methods of fern-land conversion it will be well to give some details concerning the life-history of fern and the general principles underlying its eradication.

THE LIFE-HISTORY OF FERN. The dominant plant of fern-land, both natural and induced, is the bracken fern (Pt&ridium esculentum (Forster, F.), L. Cockayne). This plant is closely related to the European bracken, but is considered to be a distinct species. Like all ferns, it reproduces itself by spores, and not by seeds, as is the case in the higher plants. The spore on germination does not give rise to an individual

resembling the plant from which, it was produced, . but develops into a plant barely visible to the naked eye and > known as the prothallus. . On the latter sexual organs . are produced, and from the fertilized ovum contained in. the female ■ organ . the true fernplant is later developed. This method of propagation, known as the sexual method, is often overlooked in the case.of bracken fern, on account of the tremendous power of vegetative reproduction with which the ordinary spore-bearing fern-plant is endowed. .

The two most conspicuous features of the bracken fern —or simply fern,” as it. is generally termed in New Zealandare its fronds that rise above ground, and its thick underground stem that penetrates deeply into the soil. This stem is popularly spoken

of as the root,” but the true roots are comparatively slender organs arising from the stem, especially at some distance behind the white-tipped growing point. The underground stem contains very large quantities of starch, which is . manufactured in the aerial portions of the plant and transferred to the stem. This starch is utilized in the production of fresh underground-stem development, and especially for the maintenance of the young fronds until such time as they uncurl. The underground stem gradually dies back from its older end, but considerably less slowly than new growth, is produced at its growing points. It will be noted that the development of the underground stem of fern is analogous to that of the ordinary - Californian thistle; and the plant when once established can spread. over a very large area of ground in a comparatively short space of time. However, there is an extremely important difference between the underground stem of fern and the underground stem of Californian thistle. From any portion of the underground stem of the thistle, especially if cut, buds can be developed, and each cut portion can thus give rise to a separate plant.' When the fern-stem is cut the case is. very different, and it is only those portions that have a frond developed on them or possess a growing point that can develop into separate plants. Any portion of the stem devoid either of a frond or of a growing point will die when cut in pieces. This fact— -important one—explains how ploughing or disking fern will thin it out considerably, while the same treatment on Californian thistle will only increase the vitality of the plant. Another extremely important character of fern is the fact that if the frond, is cut no further growth of that particular frond can take place. A fresh frond has to be developed from below the base of the cut one, and in this operation, again, a great deal of .the reserve food in the underground stem is employed. It is thus seen that the whole of a frond of fern, no matter how large or small it may be, is merely a single leaf, and this fact has an immense bearing on successful eradication. A further character of New Zealand bracken is that it is what it is termed a light-demander, and the growing of a crop that completely shades it' will cause its . early extinction. This point is not of any value so far as the average farmer is concerned, but from a forestry standpoint .it is of especial significance. The fronds, however, have a great power of elongation, and in moderately shaded forests and on their outskirts the fronds, in their efforts to reach the . light, may attain a height of 20 ft. or more before they uncurl. Such growth, however, is an immense tax on the reserve food-supply, and death, rapidly occurs unless the efforts

of the fronds to. reach the light are successful. This explains the complete absence of bracken in dense forest. Wherever forest becomes open, and on the outskirts, fern rapidly puts in an appearance to the point where, the vertical depth of the forest shade is too great for the fronds to penetrate. Another feature of fern and this an important one from the eradication aspect —is the fact that the development of fronds more or less ceases from the late autumn to the late spring or early summer. In certain districts the fronds virtually die in the winter, but this is by no means universal, and in many localities,

especially in the North, the plant is truly evergreen. This cessation of growth is of importance, taken in conjunction with its other peculiarities that have been mentioned, but is not of itself sufficient to make the' plant comparatively easy to eradicate under certain conditions. Californian thistle also undergoes a somewhat, similar dormant period, but that does not make it any the easier to destroy.. Bracken does not seem at all particular as to what type of ground it occupies, and it is found abundantly on all classes of New Zealand soils. It is frequently asserted that fern is an indication of acid soils; and that it exerts a toxic effect on other types of vegetation, but more particularly on grass, is a common belief.

Whether fern is generally indicative of acid soils or not I cannot say, but it certainly does not object to what in • New Zealand is. termed limestone country. In the Te Kuiti district it grows abundantly actually on the limestone outcrops; and in a large part of that neighbourhood red clover flourishes amazingly on the fern-lands during the process of their conversion into permanent grass. If fern is taken as an indication of acid soil, then the soil on the sides of the railway from New Plymouth to Wellington —covered as . it. is with an almost unbroken line of induced fern, and adjoining some of, the most highly prized agricultural land of the Dominion-should be of an acid nature. It would be interesting to ascertain the acidity of the soil along the railway where this induced fern is growing—say, in the vicinity of Hawera -—and compare it with the grass-land devoid of ; fern immediately adjacent.

A factor, however, that does act detrimentally on the growth of fern is altitude, as can be clearly seen .by an examination of the Mount Egmont district, where the altitudinal limit of bracken is well, under 3,000 ft. In places I know it ascends higher, but in general it can be looked upon as a . serious pest only between sealevel and 2,500 ft.

The great, and rapid development of the aerial portions of the bracken fern and the general unpalatability of the expanded fronds, together with the deep - and powerful underground stem filled with reserve food, would appear to make this plant almost impossible of eradication. Yet with all its obvious advantages for holding possession of the ground it possesses four fatal defects that make its control -possible.. These are the incapacity of the underground stems to develop fronds except in specially defined places, the inability of the frond to grow when once broken off, the extreme ease with which the frond may be broken before it expands or when it is in what is known as the curl stage, and the long dormant season extending from autumn to late spring

. In the conversion of the fern-land into grass all these defects should be taken into' consideration, but in many cases the measures necessary to direct against' these vulnerable points cannot be practically carried out, and therein lies the difficulty of controlling fern over ‘ all areas where it is found.

PREPARATION OF THE SEED-BED FOR GRASS. < For various reasons it is well, to discuss first the methods of natural fern-land conversion, and afterwards give some attention to those of induced-fern country, as certain special factors operate in

the greater part of the country of the latter .type. In the conversion of natural fern the preparation of the seed-bed is the first essential. It is only in certain instances, and where by quite ordinary, stocking methods the fern will finally disappear and grass-land assert itself almost naturally, that no attention need .' be paid to this matter. In general, the first procedure to adopt is to burn the fern. Care should be taken to isolate the area to be burnt, so as to

avoid burning a larger area than was intended. There is a r general consensus of opinion that fern should be burnt only in the autumn, but there is no really valid reason for this. so far as the fern itself is concerned. In fact, midsummer burns, provided the grass sown can come away rapidly and be stocked before the fresh ferngrowth develops, would be preferable to autumn burns. The reason is that in the summer the fern-stems are less provided with reserve food than in the autumn, and, provided each growth of fresh fronds can be broken off or eaten by stock while yet in the curl stage,

the eradication of the. fern must be sooner accomplished than if burning is delayed. This system was in past times extensively followed by the late Archdeacon Williams on the east coast of the North Island; but, of course, the practice is unusual. The reason is obvious. In general, midsummer sowing of surface-sown grass is not likely to be uniformly successful, and heavily stocking a bad take of grass punishes the stock too much for such an operation to be entertained at the present day. Thus autumn burning not earlier than the end of February can be looked upon as the general practice for surface-sown fern-lands.

On fern-land that is ploughed the burning often takes place 'earlier, and the ground is stocked for a few days after the young fern appears after . ploughing, . the ground being thereby consolidated before the seed is sown in the autumn. On fern capable of being ploughed this operation should always be carried out and the land worked up in the ordinary manner. Many consider disking ’is preferable. It certainly has the merit of cheapness, and the cutting of the fern “root-” is also highly desirable. Ploughing and disking, however, would be better. Where land cannot be ploughed the seed should -be sown after the land has been somewhat consolidated by rain, or a temporary stocking to harden the ground should take place. '

THE SEED MIXTURE.

In . certain cases on a first-ploughed fern-burn turnips are. sown, and : at times even two successive crops are taken off . before the ground is laid down to grass. Such methods, however,. are applicable only on land where the fern is. not particularly vigorous. lam inclined to think that in nearly all instances on ploughed fern-land a seeding with a temporary grass mixture is preferable to immediate conversion into permanent pasture. This is now a regular practice, and the mixture used should consist of about 15 lb. Italian rye-grass, 8 lb. cow-grass, 2 lb. white clover or alsike, and perhaps a little cocksfoot. In a couple of seasons under cattle-stocking such a pasture will be almost entirely cow-grass, and a preliminary crop of this description leaves the ground in excellent condition for the later permanent mixture sowing, and also even for the production of annual crops, both of a forage or cereal nature. The permanent mixture may be sown after the ploughing of the clover sod, but excellent results also follow harrowing the clover in the autumn and sowing almost immediately after. . The temporary clover may remain from . two to six years before being resown, its duration depending entirely on the condition of the pasture when the permanent mixture is sown. As much as •30 lb. of seed

per acre should be used, of which at least io lb. should be cocksfoot. The actual species used in the permanent mixture varies in different localities. Crested dogstail, Poa nemoralis, and perhaps timothy, should always be included in addition to clovers and ryegrass. Meadow-fescue is also well worth a place, and not less than 5 lb. per acre should be sown.

Ploughable fern-land is generally of quite good quality except in areas containing large amounts of “ raw ” pumice, such as is found on the Central Volcanic Plateau. The grassing of fern-land of

the latter type is not yet in a satisfactory state, and many local problems present themselves. On the better-class fern-land, however, the ultimate aim should be pastures largely composed of cocksfoot and white clover, such as are the dominant pastures of ordinary rainforest grass-land.

On surface-sown ploughed fern - burns the permanent mixture has to be sown as soon after burning as suitable, and should be similar to-that sown on the ploughed land. If the land is poor, grasses such as Danthonia -pilosa and red- or brown-top should be

included, and also about | lb. Lotus angustissimus. I should' also advise including at least 2 lb. of Poa pratensis on all unploughable fern-burns. ' .

The following table of mixtures, compiled from seed-merchants’ order-booksall with the exception of the first three having been actually used during the past two seasons indicates the wide range of variations that occur in fern-country sowings. This is to be expected when the very different classes of country occupied by fern is taken' into consideration. It will be noticed that in the majority of the permanent mixtures . cocksfoot occupies pride of place, and, in fact, is only excluded in one purely temporary mixture. In one mixture it will -be noted that sixteen different species are included, many 'in small quantities evidently of a purely experimental nature. In many cases yarrow, from f oz. to J oz. per acre, is also included in the mixtures. I am highly dubious of the value of this plant on fern, country.. Again; several of the mixtures represent very light sowings. This is a great mistake, and is done in nearly every case because the farmer is not sure whether he will be able to carry out the necessary measures to control the. fern; accordingly he is apt to make his mixture as cheap as possible, so that ,if failure follows the monetary loss' is not great.

THE CRUSHING OF FERN. After autumn sowing of grass - seed takes place on fern - land there is virtually no growth at all of the fern till the late spring. The exact time the young fronds begin to shoot through the ground varies with the district. In the North the first growth is considerably earlier than at the same altitude in the South. In general, however, it can be said that the first growth will not be

pushing through the ground till the middle of October. In the winter and spring following the sowing hoggets may be grazed on the young grass, but as soon as the fern begins to show up, and before the fronds begin to uncurl, heavy stocking with cattle from eighteen months to two years of age, or sheep (for preference heavy wethers in good condition), has to be carried out. The object of heavy stocking is either to feed off or break off the young fern. While yet in the curl stage the main stem or petiole of the leaf is very easily broken off, apart from being in the young state quite relished by stock. Where there is a fair growth of grass no evil effects appear to result. At times, however, cattle when fed solely on young fern are

poisoned, or at least rapidly lose condition. Fern poisoning is, however, rare when the number of stock employed in keeping down fern is considered. The prevention of the fern getting beyond the curl stage is the essential feature of fern control. When once the frond expands it becomes far more difficult to break off than when in the curl stage, and if sheep are being used to crush the fern the land must be stocked heavily before any sign of uncurling of the leaf takes place. If cattle are used the fern can be successfully crushed at a little later stage than it .can with sheep, and this is the main reason why cattle are often looked upon as more efficient fern-

crushers than are sheep on land suitable for cattle. I should always advise the employment of cattle, as then a delay of a few days in the first heavy stocking in the late spring does not matter much, although it is always better to get the beasts on the ground before any of the fern has reached beyond the curl stage. On many types of ground cattle-stocking is not practicable, and sheep must then be used. In crushing fern subdivision into comparatively small paddocks of not more than 50 to 100 acres greatly facilitates the process, and even smaller paddocks are desirable. The area annually broken in should never be so large as to necessitate insufficient fencing, on account of expense, as a small acreage thoroughly dealt with in the season following sowing is much more satisfactory than a larger area on which the fern is only partly kept in check. Every fernfrond that is allowed to expand only increases the difficulty of control, for by its activity in manufacturing food materials it restores vitality to the underground stems and renders them capable of producing fresh fronds. When stock are turned into an area ready for crushing, the quicker the land is cleaned up the better. As soon as they have eaten down or broken off by trampling all the young fronds the stock should be removed to another area and the process carried but there. It has been said that with sheepcrushing four days’ stocking and eight days’ restingj-.of the paddocks should be carried but during the first season, but such an intensive treatment is not really necessary. . -zUA. When once the - first great growth of fern has been checked in the spring a considerable time often elapses before the next growthdeveloped from buds arising from The fronds previously eaten downmakes its appearance above ground. As soon as this occurs restocking has at once to be taken in hand and the operation repeated each time the fern reappears. If stocking is properly carried out the first season after sowing there should be very little difficulty experienced in the following years. Under no circumstances, however, should the fern be allowed to get away after the first season, and if this occurs the area may rapidly become fern-infested again. In paddocks where in the second or third season fern gets away in patches it should, if possible, be mown so as to force up fresh growth, and then be crushed out while the fern is young. . The crushing-out of fern, provided plenty of stock are available, presents .no difficulties; but, unfortunately, in certain cases the conversion of fern-land is attempted without proper provision being made for an adequate supply of stock for crushing purposes during the initial, crucial stages.

THE CONTROL OF INDUCED FERN. The control of induced fern presents difficulties that are not so noticeable when breaking in natural fern country. Induced fern generally follows either an unsuccessful burn or else develops on bush-burn country where the grass mixture sown fails to permanently keep its hold on the land—that is, where a mixture not suitable for the district is used. On comparatively flat land induced fern should never become particularly troublesome, for the same methods as applied against natural fern can be carried out. But on steep, rugged country induced fern is often extremely bad,

and when once well established is exceedingly difficult to eradicate. On hilly bush-burn country likely to turn into fern within a few years of sowing very particular attention has to be paid to the mixture used. Cocksfoot and rye-grass are generally the predominating seeds that are sown, and for some years the land, may keep up a two-sheep standard. In a few years, however, the rye-grass disappears, and soil-creep occurring on the bare ground leaves the isolated cocksfoot plants raised above the general level, in which condition they are very liable to be actually pulled out by sheep. Fern then begins to appear, and, the carrying-capacity having dropped, sufficient stock cannot be maintained to keep it down. In a few years the fern attains the mastery, and frequent firing of the fern, followed by insufficient stocking, only makes its dominance

more pronounced. The farmer meanwhile has probably kept up. his total number of stock by fresh bush-burns, which in turn finally become fern-clad.

Such a condition of affairs is, happily, not usual, but many areas can be seen that have been treated in this way. On this class of country I hold it to be better to -grass with a mixture that is likely to produce permanent one-sheep grazing rather than virtually sacrifice the land for a few years’ higher carrying-capacity by means of grasses that by themselves are not likely to remain permanent. . I should advise the cutting-out of perennial rye-grass altogether . and making a fairly liberal use of such grasses as Poa pratensis, Chewings fescue, crested dogstail, and Danthonia, with probably brown-top in the northern parts of the Dominion. Amongst these grasses Poa pratensis is one of the most important ; it binds the association together well, gives excellent feed, and stops soil-creep —a very important consideration on steep hills. A generalized .mixture, somewhat as follows, should give good results on country where fern is likely to attain the mastery : Cocksfoot, 8 lb. ; Danthonia pilosa, 5 lb. ; Poa pratensis, 3 lb. ; Chewings fescue, 5 lb. ; crested dogstail, 2 lb. ; Poa nemoralis, 2 lb. ; white clover, 1 lb. ; Lotus angustissimus, | lb. : total, 26J lb. In addition, a few pounds of Italian rye-grass, to give an immediate bite, might with advantage be included. The objection to : a mixture of the above type is its expense, but this consideration should in no case determine the farmer to use a lower-priced but unsatisfactory sowing. In all cases the area of the bush-burn must be regulated so that an efficienteven if high-pricedmixture can be sown. On induced-fern country that can be subdivided easily, and where the ground is not too hard for grass-seed to germinate, methods similar to those employed on natural fern country can be carried out. It is important, however, to let the fern develop a good growth before being fired. The burning each season of induced fern not followed by sowing is to be deplored, as in this way the chance of ever getting a good seed-bed is seriously reduced. Where it is found absolutely necessary to burn induced fern without previous subdivision an attempt should be made to turn the land into Danthonia country. ’ After burning, a mixture somewhat after the following nature should be used : 8 lb. Danthonia pilosa, 8 lb. Yorkshire fog, and 3 lb. suckling-clover. Such a mixture, when even partly established, will survive periodical burning of the fern, and should later develop into exceedingly useful pasture. < On certain types of induced fern the utilization of pigs in the rooting-up and destroying of the underground stems has been applied

in many cases. Certainly the effect of pigs is often extremely satisfactory, but the main objection to this method is that only small areas can be dealt with, and the expense of fencing is Very considerable.

CONCLUSION. A consideration of the conversion 7 of fern-land into grass shows clearly how whole-heartedly the farmer will attempt to control a serious weed provided practical and payable methods for its eradication are known. The control of fern has in most instances been shown to be highly remunerative, and no effort is spared to attain that end. How different it is with weeds whose control has not yet been shown to be a profitable proposition ! Fern, although one

of the worst of our weeds, -and in the control of which very large sums of money are annually expended, has not been deemed worthy of inclusion in the Noxious Weeds Act, nor would its inclusion be of any aid in its repression. The pioneer farmers of New Zealand showed that the control of fern under most conditions was both feasible and payable, and the average farmer stands in no need of compulsion in the conversion of fern into grass-land. In certain instances where fern is spreading seriously and its. control is but half-heartedly carried out the main reason for the apparent apathy with regard to its eradication is because in such cases really payable methods of control have not yet been demonstrated. ’ -

Rye-grass, perennial .. 2 6 2 6 5 4 63 5 4 4 51 9 4 6 . 4 Rye-grass, Italian I i5 i5 2 2 2 2 3 3 3 3 3 3 4 4 IO IO 5 . 3 . 2 2 2 2 3 3 2 2 Cocksfoot I • 2 5 3 8 8 8 6 7 6 4 IO 8 6 16 IO Crested dogstail I I ■ 4 I ■ 2 I I i I I 2 I ; 2 Timothy I 1 2 4 I 2 2 Poa pratensis I I I . 1 1 I T Meadow-fescue 2 I 2 2 2 Sheep’s fescue 1Sheep’s fescue I * I I i I Chewings fescue I I. I V 1 I. 1 I i i Poa nemoralis I 1 4" Danthonia pilosa 4 4 i ll 1 3 3 1 4 1 Brown-top : i 3 4 3 • • Cow-grass i 31 6 3 2 3 3 4 4 1 I. 2 2 T White clover 2 I i I 2 2 2 i I I I 4 1 I 4 I I I 1 I 1 Alsike I I 3 3 .. Trefoil .. I I I I T . 1 1 Lotus major .. I i I 1 2 Lotus angustissimus X X I 1 J- — — — — — — 231 261 251 261 —— Total pounds per acre 26J 91 261 .— Total pounds per acre II 91 III II 15 ill 15 15 23 15 231 23 231 231 25 25' 25 25 251 26 20 261 28J 29

GRASS-SEED MIXTURES ACTUALLY USED ON FERN-LAND. (Pounds per Acre.)