H.—34.

24

Brown-top.—(a) Broadcast trials : One hundred and four lots laid down in the spring of 1928 have been regularly mown all the year. The type from the poorer and drier soil-types of Canterbury still continues to make an excellent turf, and it would appear that there is little danger of this type proving undesirable from a lawn-seed export-trade point of view. For hill-country work, however, steps should be taken through certification to ensure that this form is not sold for secondary burns of hill country. Certainly, for this purpose, it is a less desirable type than the true Aijrostis tenuis type. One hundred and sixty-eight lines were sown out in the spring of 1929, and reported on as to type. One line contained a trace of red-top ; twenty-two lines were of the distinct pseudo-brown-top type from the poorer belts of Canterbury referred to above. By means of these plot trials it is hoped in a few years to map out definitely the various brown-top type regions. This will greatly assist certification. (b) Row trials : Sixty-three lines from various habitats and sources have been sown out in rows. These will provide material for single-plant type study and for systematic botanical work by the Botanist. Yorkshire Fog.—lt has long been felt that this grass has a place for the grassing of second-quality lands, and a preliminary type study is being undertaken this coming year. Ninety-three lines from various habitats and sources have been sown in rows, and these will be used as material for preliminary single-plant-study work. White Clover. —(a) Broadcast trials: One hundred and four lines sown in the spring of 1928 have been regularly mown throughout the year. Owing to volunteer white clover appearing everywhere throughout this area there was little difference to note for the first year of the trial. After January of this year, however, the volunteer type, together with many poor-recovery and lowpersistent lines, failed badly, and at the time of writing there are marked contrasting types showing up. A small percentage of lines —many from Hawke's Bay —are outstanding. These showed good summer production and very rapid recovery after autumn rains. It would appear we are on the verge of important disclosures in regard to white-clover types from varying sources of origin. (Ib) Row trials : In order to control volunteer white clover in further trials of this species, the double row, 2 ft. apart, was adopted rather than the broadcast plot. To ensure further freedom from volunteer white clover, some 400 lines were sown in sterilized soil. Ultimately the double row will join up as one broadcast plot, when both crown and tiller persistency can be studied. Eight hundred commercial lines from all over New Zealand are sown out in this trial with Kentish wild white clover and imported Dutch types as opposing control. (c) Row trials from various sources and habitats : One hundred and thirty-three lines additional to the above were sown out this autumn. Many of these are hand gathered and will be used for single-plant studies in comparison with commercial lots. (d) Single-plant studies: Approximately 4,000 plants are out for single-plant study as to type and crown persistency. These lines were sown in boxes in sterilized soil and were transplanted into their permanent positions 2 ft. apart each way. Kentish wild white and imported Dutch white clover are again used as opposing controls. Red Glover. —(a) Broadcast lots : One hundred and four lots sown in the spring of 1928 have been regularly cut. The weekly cut has eliminated virtually all the broad red and other non-persistent types, and the outstanding success of the Montgomery red-clover type in persisting under this treatment offers great hope in the possibility of ultimately working up a good persistent grazing type of red clover. (5) Single-plant study : Approximately 1,000 single plants of the three main red-clover types are under trial, and some 2,000 additional plants are pricked out in boxes ready for transplanting. (c) Row lots : One hundred and four lines from various habitats and sources were sown out this autumn. These will serve as additional material for single-plant study, the object of which will be to select up as rapidly as possible a uniform elite line of the Montgomery red-clover type. Miscellaneous Species.—Two hundred and thirty broadcast plots or rows are sown out with the following species: Crested dogstail, Timothy, meadow fescue, tall fescue, Poa trivialis, Chewings fescue, Poa pratensis, Italian rye-grass, Western Wolths rye-grass, Wimmera rye-grass, Poa alpina, Poa serotina, Poa hulbosa, African veldt-grass, Phleum alpinum, Phleum Micheli, Phalaris arundinacea, Phalaris tuberosa, golden oat-grass, tall oat-grass, Aira caespitosa, Bromus inermis, prairie-grass, Plantago serpentina, Danthonia pilosa, Danthonia semiannularis, woolly finger-grass, Lotus major, Lotus hispidus, Lotus corniculatus, English trefoil, clustered clover, Trifolium cernuum, Trifolium stellatum, Egyptian clover, and alsike. It is not hoped to do much with the foregoing species, but many of the lines are on trial on behalf of overseas research stations and farmers. (2) Field Research. The past year has been a busy one in the field and the enormous amount of work performed has been possible only through the close co-ordination and assistance of the Fields Extension Service and other specialists of the Plant Research Station. Three thousand five hundred and ninety plots of perennial rye-grass, 216 cocksfoot, 393 white clover, 401 red clover, 200 connected with rates of seeding, and 1,220 of miscellaneous grasses and clovers are now sown down in the field outside the central research area at Palmerston North. These are sown on the main leading soil-types throughout New Zealand and extend from North Auckland to Southland. A special area at Gore, and one at Lincoln, in Canterbury, have been chosen for particular trials of rye-grass strains to study the particular behaviour of the different rye-grass types when grown in Southland and Canterbury respectively, and compare these with the same lines grown at the central research area at Palmerston North. As far as Southland is concerned, the evidence to date would indicate that the behaviour of Hawke's Bay type, for example, is identical] in Southland with that in the North Island. The trials at Lincoln have not been sown sufficiently long to express an opinion.