H.—34

and boron varied from 18 p.p.m. to 32 p.p.m. The amounts of lime and phosphoric acid showed relatively small variation. The work which has been done, although only of a preliminary nature, suggests the possibility of both magnesia and boron deficiency in certain samples, and emphasize the desirability of securing standard data for high-quality American and New Zealand leaf. Arrangements have now been made for the collection of typical samples of Riwaka leaf for chemical examination in respect to quality of tobacco. Seed-germination Studies. Previous studies of the germination of tobacco-seed used for the commercial crop in the Nelson District have shown great variation in the percentage germination of different lines of seed. The data have suggested 'that period of storage and the presence of seed-borne diseases may influence greatly the number of vigorous seedlings obtained from different lines of seed. With a view to obtaining more information on these points, studies have been undertaken relating to (a) the effect of age on the germination of tobacco-seed, and (b) the effect of disinfectants on seed-germination. (a) Effect of Age on Seed-germination. —For the purpose of the investigation, seed less than one year old, seed one year and two years' old, and seed eleven and fifteen years' old were available. Using a standard procedure in every germination test, the studies have shown that in the case of twenty sets of seed four months' old at the commencement of the experiments little difference in germination was found during an ensuing period of four months. In eight sets the final germination rate was approximately the same as the initial rate, in ten sets the final germination rate was appreciably greater, and in two sets the final germination rate was appreciably less. In the case of eight varieties of tobacco-seed imported fresh from the United States of America in 1937, germination tests a year later showed only minor differences in the germination rates for the two years. With one exception the germination percentage in the second year stood as high as 90 per cent. In the case of seed stored for eleven and fifteen years, respectively, marked reduction in the germination percentages was found. For nine varieties stored for a period of eleven years the germination percentage varied from 33 to 88, with an average of 65 per cent. This percentage would be still further reduced if the germination figures at fourteen days from sowing were taken. In the case of seed stored for fifteen years the percentage germination was 5-5 per cent, at fourteen days and 33 per cent, at 30 days from sowing. (b) Effect of Disinfectants on Seed-germination. —Alcohol, silver nitrate, and Uspulun have been compared in relation to the effect they exert on the germination of tobacco-seed after treatment with these chemicals. Six varieties of tobacco-seed of high, germinating-power were used for the experiments. The results of tests made on treated and untreated tobacco-seed showed that none of the chemicals exerted an appreciably adverse effect on the germination of the seed. On the contrary, alcohol gave an increase in germination rate in four out of six varieties, silver nitrate gave an increase in three out of six varieties, and Uspulun gave an increase in one out of six varieties. In no instance was the reduction after treatment with chemicals greater than 3-5 per cent. One important aspect of the studies has been the effect of storage after treatment with disinfectants on the germination of tobacco-seed. Two lines of tobacco-seed were used in this experiment, and germination tests were carried out at weekly intervals during a period of twelve weeks following disinfection of the seed. Storage of the seed prior to the germination tests under these conditions made relatively little difference to percentage germination of the seed. In five cases out of six the final figures showed a slight increase in percentage germination over the initial figures. Mosaic Investigations. (a) Factors affecting Mosaic Dissemination. —In co-operation with the Tobacco Research Officer, the results of last season's work relating to the incidence of mosaic disease were prepared for publication. The data pointed clearly to the tobacco-seedling bed as the most important factor in the spread of mosaic. (b) Effect of Manurial Treatment on Susceptibility of Tobacco-plants to Mosaic Disease— -It is well known that under certain conditions excessive use of nitrogenous manures increases susceptibility of plants to fungus attack, while the liberal use of potash, reduces the severity of certain diseases. It was considered possible that mosaic disease might show a somewhat similar reaction to variations in the manurial programme. A series of pots in the glasshouse and of field plots on adjoining land were arranged to give information on this point. The manurial treatments included in the test were (1) complete fertilizer (N +P + K); (2) complete fertilizer plus trace elements ; (3) complete fertilizer plus one additional dose of nitrogen ; (4) complete fertilizer plus one additional dose of potash ; (5) complete fertilizer with only one-half dose of nitrogen ; (6) complete fertilizer minus potash. Tobacco-plants of the Harrison Special variety were used throughout the tests. Inoculation with mosaic virus was carried out when the plants were 12 in. high. In the glasshouse all pots inoculated, irrespective of treatment, developed typical mosaic symptoms. In the field, out of 260 plants which were inoculated under different treatments, all but one developed mosaic. The experiments have shown fairly conclusively that variations in the manurial programme as outlined above do not inhibit the incidence of mosaic in tobacco-plants.

79