H.—34.

Clover Nodule Bacteria.—Continuation of this work has suggested two major conclusions : — (a) Efficient strains of bacteria should be selected from the district in which inoculation is intended ; and (b) Media on which the bacteria are grown should contain sufficient available calcium, as where calcium deficiency pertains the organisms are liable to lose their power to promote high nitrogen content and growth of host-plants. Other experiments have shown that nodule bacteria of various legumes are capable of forming nodules on host-plants of different genera. Linen-flax Diseases. Disease surveys of the commercial areas under linen flax (approximately 14,000 acres) have been made during the growing season and the following diseases found to be present. The first three are carried with the seed. Browning (I'olyspora lini). —General in crops in Southland, but troublesome only in those which were sown in October. It has been collected also on Linum monogynum. Rust (Melamspora lini). —Present in most crops in Marlborough and Canterbury, but not seen in any Otago or Southland area. Infection was slight so little, if any, damage to the fibre was reported. The rust is also present on the indigenous Linum monogynum. Wilt (Fusarium lini). —The fungus has been isolated from diseased material taken from one crop at Oxford in Canterbury and three in the Wintou district of Southland. No trace of the disease has been noted in Marlborough or South Canterbury. Foot-rot (Cause unknown). —First noted in crops when 3 in. to 6 in. tall. This condition appears to have been present to some extent in all lines examined, save those grown from certified Canadian seed. Despite many attempts, a pathogen has not been isolated, so that in all probability foot-rot has been induced by unfavourable soil conditions, possibly aggravated in many cases by contact of plants with superphosphate in the soil. Pasmo (Sphaerdla linicola). —Isolated from plants grown at Owairaka from seed procured from commercial lines grown at Blenheim. Symptoms so closely resemble those of browning that it is possible that pasmo may have been mistaken for the former in the field during the course of the survey. Fruit-tree Diseases. (See Fruit Research Report, p. 42.) Small-fruit Diseases. Tomato Diseases.—(a) Tomato-necrosis: Following inoculations with severe-streak-infected tomatoes a combination of tobacco-mosaic and an unidentified necrotic virus appeared. The latter has been studied as to host-range, longevity in vitro, dilution end-point, and thermal death-point. (b) Tomato Leaf-mould (Cladosporium fulvum) : Tests were made in the new commercial type glasshouse of three varieties of tomatoes said to be immune to mould. The standard commercial variety Kondine Red was used as a check, duplicate plots of which were sprayed with Shirlan Ag. to ascertain the effect of the spray on mould control. Results are most significant, Kondine unsprayed producing 2 lb. 5 oz. per plant, Kondine sprayed 31b. 4 oz., one immune variety 51b. per plant, a second 5 lb. 11 oz. One immune variety proved to be shy-setting and was discarded. A second appears to be unfixed as about 5 per cent, of plants developed mould. Selections have been made, and they, together with the most productive variety, will be grown in numbers sufficient to produce commercial quantities of seed. (c) Verticillium-wilt (Verticillium albo-atrum) : Control of this soil organism was undertaken in a glasshouse the soil of which had been artificially inoculated. In the check plots 100-per-cent. infection was secured, whereas in those treated with formalin solution 1-50 and 1-80 complete control was attained. It was found necessary to apply 50 gallons of solution to 15 square yards of soil as the ground must be saturated to give adequate results. 11. Plant Protection. Therapeutant Testing. Mercury Compounds.-—Two proprietary organic mercurials were tested against the standard acidulated mercuric chloride steep for control of Corticium vagum. Both proved ineffectual, as they did not kill the sclerotia even when employed at concentrations in excess of those recommended by the manufacturers. They did not injure tubers, however, as did the standard treatment. Copper Compounds. —Three were tested against standard bordeaux mixture. Copper naphthenate was found to be about equal in toxicity per unit of copper, but too costly to employ commercially. A proprietary copper oxide and copper oxychloride have given promising results, the former proving such a useful bordeaux substitute that it has been certified as a spray for tomatoes. Sulphur Sprays. —A number of proprietary compounds have been tested biologically. Work has been continued on the study of particle size as affecting toxicity, a matter which has a practical bearing on disease control in the field. The work has confirmed field trials in demonstrating the comparative worthlessness of the coarse sulphurs which are still being sold to orchardists. Denis Products. —Work is proceeding on standardization of biological testing of these products, since it has not been possible to ascertain the rotenone content by chemical analysis. Considerable numbers of insects have been tested, and several of the. most suitable have been chosen for toxicity trials. Correlated field trials have shown that commercial derris products are too costly to be employed on a field scale if applied in the form of dusts. Certification of Therapeutants. (See Fruit Research Report, p. 42.) Improvement in Spray Programmes. (See Fruit Research Report, p. 42-)

4—H. 34.

25