September 3, and drawn at 3.30 p. m. The ashes and embers were removed and the trenches lined with a layer of green grass. They were then filled with Havana leaf-the ripened top leaves on the quarter acre devoted to the Snow process being used, together with some ripe seed leaf. The leaves were placed smoothly and closely. The filling was completed at 5 p. m. Over the top was placed a layer of ripe meadow grass, building paper, and 6 to 8 in. of earth. The temperature in the trench, as shown by a garden thermometer, rose to over 120° F., and between 5 p. m. September 3, and 4 p. m. September 5, ranged from 93 to 104° F. When the leaves were removed September 5, 4 p. m., they emitted an odor like decayed tomato vines, and gave off considerable ammonia. Both faces of the leaves were very wet. The sap appeared to be removed from the leaf tissue, but not from the rib. The leaf tissue was dark, translucent, and tough. The leaves were dried in an ordinary shed, when they assumed a purplish black tint. The loss in weight in drying by this process was 90.4 per cent.

Comparison of the several curing processes (pp. 237-257).-The first examination was made by experienced tobacco packers at the curing barn in October, at which time "only the product of the 2 Snow proeess cures and some air-cured leaf from an entirely different experiment were in condition for careful examination. . . . The tobacco was found to be well cured and of excellent color," but not fine. "The leaves were thick and greasy,' not thin, elastic, and silky as is required in high-grade wrappers." The tobacco cured by the usual process was better than the other; but no comparison was made of any single sample from the same plat and cured by both processes.

Samples were also submitted to tobacco experts in New York: one considered the air-cured sample the best, two the Snow cured, and one thought each best for particular uses.

As to curing leaf alone vs. curing leaf on stalk, the authors found that on the average 72.70 per cent of the whole weight of the plant was leaf; and that the cured leaf constituted 15.8 per cent of the weight of the original leaf with the stalk-cured tobacco, and 14.4 per cent with the leaf cured; likewise the leaf tissue was 11.1 per cent in the former and 10.3 per cent in the latter case.

The authors conclude that "there is little difference in the losses of substance during curing by the 2 methods."

As to air curing rs. artificial heat, the authors say that the final thickness of the leaf seems to be dependent upon prior conditions of growth rather than upon the methods of curing.

Tobacco cured by the "Cuban" process gave off no tobacco aroma when smoked; the taste was sharp and tarry, and the effect on all who tested it was nauseating.

Experiments on the burning quality of tobacco, A. SCERHATI (Jour. Landw., 43 (1895), No. 4, pp. 379-458).-The author reviews and

discusses the work of previous experimenters. A number of series of both pot and plat trials were made and tests of the burning quality of tobacco grown in these trials and elsewhere were made, aggregating in number over 3,300.

The method adopted for determining the burning quality was to cut the tobacco leaf in two, crosswise, in the middle, and note the time of burning after lighting the whole cut edge of one-half in a gas jet. By means of a chronograph the duration of the glow was exactly determined. The influence of the variety of tobacco, the climate, weather, and soil, and of the different sorts, qualities, and combinations of manures on the burning quality of tobacco are discussed in connection with the tabulated data given.

The author concludes in part as follows: Nessler's assertion that tobacco which contains more than 0.4 per cent of chlorin and at the same time less than 2.5 per cent of potash does not burn well is not proven, nor that the burning quality depends primarily on the content of chlorin or potash. The burning quality varies with the position of the leaf on the tobacco stalk, the leaves growing in the center of the stem usually burning best. The variety also has considerable influence. A moist climate and warm moist weather promote the burning quality while drought hinders it. The physical properties of the soil have primarily an important influence, a warm, moist, fertile soil producing the best burning tobacco under otherwise equal conditions. Barnyard manure applied in the spring is not as injurious to the burning quality of tobacco as many assert, and generally has little influence. With a one-sided application of nitrogen, sulphate of ammonia injures the burning quality, while nitrate of soda does not. Phosphoric acid is generally not injurious to the burning quality of tobacco, and potash manures have by no means the favorable effect that is generally believed; they are without effect in clay soils, and of greatest advantage in sandy soils. In soils poor in lime liming is especially advantageous for the burning quality of tobacco.

Experiments with wheat, E. DAVENPORT and W. J. FRASER (Illinois Sta. Bul. 41, pp. 147-155).-Previous work in this line was published in Bulletin No. 34 of the station (E. S. R., 6, p. 412).

A meteorological record is given for the years 1888 to 1895 and notes and tabulated data during 4 years on the comparative weights of grain cut at different stages of maturity and cured in the shade, in the sun, and with the heads separated from the straw; on wheat sown during 6 years at the rate of 3 to 9 pecks per acre; on date of sowing between September 10 and October 8-15; on coöperative tests of fertilizers in southern Illinois, and on a test of 73 varieties.

The authors give the following summary of results:

"There is a steady increase in dry matter up to the stage of full maturity, and curing in the shade has a slight advantage over drying in the sun.

"Between 5 and 8 pecks of seed an acre there is little or no difference in yield.

"Sowings from September 10 to October 1 yield about equally, with a slight tendency to favor September 20. In the best seasons the later sowings give the highest yields.

"Barnyard manure is of great benefit upon the white clay soils of southern Illinois, and in most instances notably better than superphosphate.

"The variations in yield of a single variety on separate but neighboring plats are nearly as great as between different varieties.

"Trustworthy determinations of yield are difficult to secure and can be had only after a long series of years, or from duplicate plats upon land whose previous history is known."

Experiments with wheat and oats at Grignon in 1895, P. P. DEHÉRAIN (Ann. Agron., 21 (1895), No. 12, pp. 545-565).-The author says in the introduction that the culture of wheat in France covers each year nearly 7,000,000 hectares (17,297,000 acres). The average production is only 16 to 17 hectolitres (18.4 to 19.55 bu. per acre). Outside of the hot southern region the returns can be increased onethird.

A meteorological summary is given for the year ending October, 1894.

Wheat, size of kernels for seed.-During several years F. Desprez has selected the largest and the smallest grains of 5 varieties and sown them separately. From the grain produced, the largest kernels from the large ones were grown, and the smallest kernels from the small ones, and so on. The tabulated results for the 3 years, 1893-95, show a large increase from sowing the large kernels.

Wheat, test of varieties.—At Grignon the square-head varieties are sown and found to be well adapted. Two new square-heads, a white and a red, and a number of other varieties were tried. The yields of plats are tabulated and discussed. Australian wheat gave the largest yield, 57.99 bu. per acre, followed by Dattell, 49.37 bu., and the squarehead varieties: Porion, 46.24 bu., Scholley, 45.80 bu.; red, 43.87 bu., and white, 39.85 bu.

Manuring and place of wheat in the rotation.-Square-head Scholley sown after beets, manured, sown without manure, and without manure but with supplementary working of the soil, gave a yield of 11.04 bu. per acre more than when grown after fodder corn with 22 tons of manure per acre, or without manure with a supplementary working of the soil, or with a very moderate application of manure direct. The difference is attributed to the effect of the manure applied to the previous crop and to the place of wheat in the rotation; and preference is given to sowing wheat after beets that have been manured; after clover. with the additional application of manure, or after potatoes that have been manured.

Oats, test of varieties.-Of the 2 varieties tried, Ligowo yielded 115.7 bu. to 135.2 bu. per acre of grain and 2.54 to 2.85 tons of straw per

acre; Salines yielded 111.4 bu. of grain and 7.58 to 7.80 tons of straw per acre.

Variety tests of wheat, H. J. WATERS and E. H. HESS (Pennsyl vania Sta. Rpt. 1894, pp. 282–290).—This is a continuation of work published in the Annual Report of the Station for 1893 (E. S. R., 6, pp. 716, 719). On duplicate thirtieth-acre plats 44 varieties of winter wheat were tested in 1894. Tabulated data are given embracing yields of grain and straw for all varieties, and weight per struck bushel of many varieties, for the years 1887 to 1894, inclusive. The highest yields for 1894 were given by Reliable, Ontario Wonder, Rudy, Canada Wonder, Fulcaster, Valley, German Emperor, Extra Early Oakley, Royal Australian, Theiss Pool, Deitz, Fultz, Nigger, Finley, Deitz Longberry Red, Egyptian, Tuscan Island, and Mediterranean, all of which yielded over 30 bu. per acre.

During the past 5 years 7 varieties have returned an average of more than 30 bu. per acre, viz: Reliable, Valley, Fulcaster, Ontario Wonder, Deitz Longberry Red, Wyandotte Red, and Currell Prolific.

Experiments with foreign wheats, R. C. KEDZIE (Address read before the Michigan State Millers' Association, Jan. 14, 1896).-The author discusses the questions how to raise the grades of Michigan wheats, wheat breeding, cross-breeding wheats in Australia, and how to determine the milling qualities of wheat, and quotes the report of a committee of the Millers' Association on Buda Pesth winter wheat. Ten varieties of foreign wheats obtained through Mr. M. A. Carleton, of this Department, and 2 procured from Germany by Director Smith, of the Michigan Station, have been sown for trial.

The Buda Pesth winter wheat is spoken of as harder than the hard Fife of North Dakota; 20 bu. yielded 330 lbs. of patent flour, 440 lbs. clear, 8 lbs. of low grade, 36 lbs. fine middlings, 168 lbs. middlings, 177 lbs. bran, and 40 lbs. screenings. As a result of a baking test the flour was reported to be of a higher standard than spring wheat patent.

A table is given of the composition of 18 varieties of Michigan wheat and 15 of foreign wheat.

The Poona experimental farm, J. W. MOLLISON (Indian Agr., 21 (1896) No. 1, pp. 11–15).—Variety tests were made with Indian, American, and African sorghums, and teosinte. The largest yield was given by American sugar sorghum (Sorghum saccharatum), followed by Nilva Jowari (S. vulgare), Imphee (S. saccharatum), and Sundia Jowari (S. cernuus). The last is regarded as one of the best fodder crops for that country. The author says that teosinte yields well in a season of good rainfall, but the fodder is coarse and the stalks thick and woody.

The wild forms of cotton (Gossypium stocksii), (G. arboretum), and (G. wightianum) have been under cultivation for 2 years with a view to improve them and to secure better cultivated varieties by crossing or by sports from the seed of the wild forms. The results so far are of no practical value.

In a trial of plows the turn-wrist plow was thought by the author to be best suited to the conditions of Indian agriculture. By this plow the furrows are all turned in the same direction while plowing a field. Cawnpore experimental farm, F. E. TAYLOR (Indian Agr., 21 (1896), No. 3, pp. 76-79).-This is a report on a fertilizer experiment with Indian corn, a test of different varieties of cotton, an experiment with mixed crops, a test of the vitality and productive power of indigo seeds of different ages, and a fertilizer experiment with wheat. The American varieties of cotton, Upland, Georgian, S. B. Maxey, Jones Improved, and Shine Early Prolific were grown with good results.

Some culture experiments with winter barley, J. ERIKSSON (Kgl. Landt. Ikad. Handl. Tidskr., 35 (1896), pp. 87-94).

In what stage of ripeness shall barley for brewing be harvested? (Deut. landw. Presse, 23 (1896), No. 51, pp. 449, 450).—The results are given, with tabulated data, of experiments carried on with 4 varieties of barley by the experiment station and school for brewing at Berlin. The barley was harvested in the milk stage, yellow ripe, fully ripe, and dead ripe. The milk ripe barley was uniformly considered unsuitable for brewing, and the yellow ripe stood far behind the fully ripe and the dead ripe. Between the latter two there was little difference, showing that the brewing value of barley increases with advancing ripeness. The varieties stood as follows in the order of yield of extract: Chevalier, Hanna, Goldthorpe, and Selchower.

Crimson clover, D. O. NOURSE (Virginia Sta. Bul. 44, pp. 113-117).—A popular article on crimson clover (Trifolium incarnatum), treating of the season for sowing, habits of growth, and value of crimson clover for silage and hay.

Cotton culture, R. J. REDDING (Georgia Sta. Bul. 30, pp. 379–382).—This is in part a reprint from Bulletin 27 of the station (E. S. R., 6, p. 884), followed by some general cultural directions.

The fiber bearing plants of Florida, C. W. PARSONS (Associated Railway Land Department, Savannah, 1895, pp. 62, figs. 15, pls. 5).—Popular descriptions are given of Florida hemp (Agave sisalana), bow-string hemp (Sansevieria spp.), wild pineapple (Bromelia sylvestris), pineapple (Anana sativa), Florida jute (Frena lobata), and ramie (Boehmeria nivea), and their adaptability to fiber manufacture.

Practical manual for the culture of fodder plants, C. and H. DENAIFFE ( Manuel pratique de culture fourragère. Paris: J. B. Baillère et fils, 1896, pp. 369, figs. 108).—The principal topics treated of are the laying down and maintenance of meadows; harvesting, preservation, use, and food value of forage plants; grasses, legumes, and other forage plants; plants injurious to meadows; silage; and a series of tables of interest in connection with fodder plants.

The authors differ from recognized authorities in claiming that maize originated in the Levant instead of on the Western Continent.

Composition of certain forage plants-spurry, Lathyrus sylvestris, mint hay, R. C. KEDZIE (Michigan Sta. Rpt. 1894, pp. 131-135).—A reprint from Bulletin 101 of the station (E. S. R., 5, p. 782).

Chemical-botanical analyses of hay from Ottenby (Sweden), A. ATTERBERG (Tidskr. Landtmän, 17 (1896), pp. 129–132).

Manuring, especially of hops, J. SCHNEIDER (Jahresber. Hopfenbaurereins Tišice, 1896; abs. in Chem. Ztg., 20 (1896), No. 42, Repert., p. 157).

Millet, A. A. CROZIER (Michigan Sta. Rpt. 1894, pp. 422-429).—A special bulletin compiled from Bulletin No. 117 of the station (E. S. R., 6, p. 713).

Experiments with pulses, G. VALDER (Agl. Jour. N. S. Wales, 7 (1896), No. 1, pp. 41-43). Notes are given on chick pea or gram (Cicer arietinum), horse gram (Dolichos biflorus), square podded pea (Lotus tetragonalobus), French beans, lentils, Dolichos, tares or vetches, and Wagner's flat pea (Lathyrus sylvestris).