bound into sheaves, several of which are put together, forming bundles, each containing stalks of equal length. The tops of the sheaves are then cut off, and only the portion preserved that will make good fiber. When the hemp is grown for use in spinning, that is, for fabrics, the stalks are not cut, but are pulled like flax. The operator first removes the leaves by passing his hand from top to bottom of the stalk, it being important to return the leaves to the soil where they were grown. Six to fifteen stalks are pulled at one operation, according to the ease with which they can be drawn out of the ground, and the earth shaken off. These handfuls are made into bundles about six inches in diameter; after bundling the roots and tops are cut off by means of an ax and chopping-block. The clipped stalks are then made up into larger bundles a foot or more in diameter, and are sent to be retted at once, as it is claimed that the hemp is not so white if it is dried before retting. When the seed is saved the method of procedure is as follows:

In some localities the gathering of the hemp is so managed as to secure the greatest quantity of seed possible of good quality. To this purpose the male stalks are first collected, which ripen six weeks earlier than the female stalks, the latter being given plenty of time to mature and not being gathered until their leaves and stems begin to turn yellow and the seeds to grow dark. They are tied in bunches, and of these there are made large bundles, which are placed upright, that the seed may complete its opening. The seeds are extracted by beating the stalks. This manner of operating produces less fiber, and these female plants yield fiber of inferior quality from those collected at the time of maturing of the male plants; but the harvest of seed compensates for the difference. If you take into account the expense occasioned by the double harvesting and double retting, we find that there is greater advantage in having but one harvest without reference to the seed. Dried in the air the male hemp contains an average of 26 per cent. of stripped hemp, and the female plants from 16 to 22 per cent. The stripped hemp dried in the air does not yield more than 60 to 75 per cent. of textile fiber, the remainder being foreign matter soluble in leached alkali, so that 100 parts of green hemp do not produce more than 5 to 8 parts of textile fiber.

There are two systems of retting practiced in western France, the retting in the open field, where the stalks are allowed to lie about a month, and similar to the plan followed in Kentucky, in our own country, and the water retting, which produces the best fiber. The water retting (rouissage) is accomplished both in pools and in running streams. The river retting seems to accomplish the best results, although taking a little longer time than the pool retting, the duration of immersion varying from five to eight days. If the weather is cold it retards the operation two or three days longer than if warm. This accounts, too, for the shorter time occupied when the immersion takes place in pools. This work is usually done in the latter part of August. The bundles of hemp are floated in the water, secured if in a running stream, and are covered with boards kept in place by stones or any weight that will keep them under. From all I could learn there is little pool retting in the Sarthe district, although public opinion is generally against river retting, on the score of its rendering the waters of the streams foul and detrimental to health as well as destroying all animal life with which

they should abound. I understand there are very stringent police regulations against the use of streams for this purpose, and as long ago as 1886, in a brochure published by M. Bary, a hemp-spinner of Le Mans, attention is called to the desirability of introducing an improved method of retting, which would accomplish all the beneficial results of retting in running water artificially, and therefore render unnecessary the polluting of streams. From M. Janvier (of the hemp-spinning establishment of Janvier, Pere et Fils et Cie, at Le Mans, successors to M. Bary) I learn that while many attempts have been made to bring about a better system, none have been successful, and, police regulations to the contrary notwithstanding, the best hemp fiber produced in the Sarthe district is still retted in the running streams. Where pool retting is followed the pools are specially constructed, dug out of the earth to the depth of a yard or more, walled up or the sides made solid, and lined and floored with cement usually, in order that the water shall remain clean and the hemp retain its color. The stalks are watched very closely after the third or fourth days, the farmer breaking and examining a few at intervals to guard against over retting, which weakens the fiber.

When sufficiently retted, whether the work is done in streams or pools, the hemp bundles are removed from the water, but first agitated to remove all waste matter that may be adhering to the stalks. They are then drained, and the bundles, opened at the bottom, are set up in conical sheaves to dry, this operation being accomplished in two or three days. Considerable of the hemp grown in the Sarthe district (I can not speak for other sections) is further dried in brick-kilns. One of these examined on a large hemp farm visited near Le Mans, and at that time in operation, may be described as a circular brick structure some 10 or 12 feet in height, resembling a smoke-house in our country. It was built on a side hill, the door opening into the chamber where the hemp was drying being on one level, the higher, while the floor to the fire-pit, at the back of the building, was on the lower level. As no evidence of a fire was observed, I infer that the fire is drawn when the right temperature has been reached, and the hemp introduced upon the grated floor to dry slowly by moderate heat. I witnessed the process of breaking hemp in the Sarthe district and brought away samples of both stalks and cleaned fiber as sent to market, as well as samples of scutched, softened, and dressed fiber prepared both for cordage manufacture and for weaving into "linen." The stalks are of creamy whiteness, as brittle as pipe-stems, and the filasse, particularly next the wood, so bright in color that no tinge of yellow is observable. A farm operator questioned told me he was able to break out 30 to 35 kilograms of fiber per day (say 60 to 75 pounds). A brake similar in principle to the old fashioned Kentucky hemp-brake is used, though lighter and smaller in the first place, produced with seven instead of five breaking-slats (arranged three opposite to four), both wood and metal being used in its construction. Double this quantity of hemp is cleaned in a day by the

negro operators in Kentucky, but it should be explained that the French operator is nicer in his manipulation of the fiber, running through a smaller quantity at one time, skillfully twisting the product into a very loose rope or "streak" of fiber, these as produced being laid most carefully side by side so that when the larger bundle of fiber is made up each · has its place and can be detached from its fellows by the scutcher with hardly the disarrangement of a filament.

At a scutch-mill, where, by the way, only hand-scutching was practiced, I was shown some bales of softened fiber, and afterwards visited the establishment of a hemp-softener (Batteur de Chanvre), near Le Mans, to observe the process. The mill was run by water-power, the fiber being manipulated on a circular platform a couple of feet in height and perhaps eight in diameter, made of solid oak timber, the end-wood forming the surface. To a heavy spindle in the center was attached a short conical cylinder of iron weighing some 2,400 pounds. The "streaks" or ropes of fiber as received from the farmer are made up into bundles weighing perhaps 64 pounds each, and these to the amount of 130 pounds are arranged over the surface of the circular bed or platform. The heavy iron cone is then made to revolve or travel around in a circle at a rate of speed equal to thirty-five times a minute, the softening process requiring from half an hour to one hour and a half, dependent upon the condition of the hemp under treatment. Only the finest fiber is softened, the product going to the spinning mills for the manufacture of coarse sheeting, shirting, canvas, and similar fabrics, the peasantry of Brittany, for the most part, employing hemp instead of flax in the domestic economy.

Although these details relate to the manufacturing side of the industry, rather than the agricultural, they are interesting as showing by what careful means a fiber is produced in this country (France) that will take the place of linen. While on this subject I would add that the softened hemp is not used in its whole length, but is broken (pulled apart) into three pieces on a mechanical device for the purpose found in all hemp-mills (and even in our own country). The bottom third is the best, and is kept separate for use in the finest numbers of yarn.

How much of the French methods of hemp culture and manipulation might be adopted in America, with advantage, remains to be determined. Two points however may be noted: That a more careful practice with more thorough methods of handling throughout will be necessary to improve the fiber to that point that will make it available for the higher grades of manufacture, and also that a better system of retting must be followed, though the contamination of streams in the rural districts of the United States will hardly be allowed by the residents of any section of the country. American ingenuity must devise a plan which will be distinctly American, and both practical and economical, and one that will not at the same time tend to make the cultivation of this crop exhaustive to the soil.

THE RAMIE MACHINE TRIALS.

Probably no one fiber interest represented in the Paris Exposition of 1889 attracted more attention than ramie, nearly every country of any prominence which took part in the exposition either sending specimens of fiber to show the result of experiments or progress of its own culture, or commissioning representatives to ascertain the latest facts regarding it. The United States Department of Agriculture made a small display of ramie illustrating the simple fact that the plant can be grown successfully in the United States, and produce a filasse of good quality. During my residence in Paris, and while connected with the American Commission to the Exposition of 1889, I studied as far as possible the recent progress that has been made towards establishing the ramie industry, and especially in relation to machines or processes for the decortication or cleaning of the fiber. I can only place on record here, however, the result of the official tests of the ramie machines, which took place September 23, 24, and 25, on the grounds of the exposition, with brief descriptions of these machines and such other general information as may be deemed important.

Six machines and one chemical process were entered, as follows: The Armand Barbier machine (E. Armand, 46 Boulevard RichardLenoir, Paris); two forms of the Favier machine, one for green and one for dry stalks, exhibited by the "Société la Ramie Française," (P. A. Favier, 14 Rue St. Fiacre, Paris); two forms, a large and small, of the De Landtsheer machine (Norbert de Landtsheer, 2 Place des Batignolles, Paris), and the machine exhibited by Felicien Michotte (of 43 Rue de Saintonge, Paris). The process was that of Ch. Crozat de Fleury et A. Moriceau (7 Rue de Londres and 4 Faubourg Poissonniére, Paris), and was for treatment of the stems in green condition.

THE FAVIER MACHINE.

The first trial was that of the smaller of the machines exhibited by the French Ramie Association, adapted for work upon green stalks. Ten kilograms of stripped stalks (equal to about 22 pounds) were put through the machine in four minutes and thirty seconds, which included one or two brief stoppages. The net product of well cleaned wet ribbons weighed 2.82 kilograms, equal to almost 6 pounds.

In the second test 60 kilograms of stalks with leaves were used (about 123 pounds), divided into lots of 10 and 50 kilograms respectively.

The first lot ran through in two and a half minutes, the second in fif teen minutes and a half, the difference in time being due to some of the ribbons fouling the last pair of rollers, necessitating a stoppage. The product of the decortication was 18.1 kilograms, equal to about 40 pounds of wet ribbons nicely cleaned.

On the afternoon of the 24th of September a test was made with the Favier machine on dry ramie with the following results: Thirty kilograms of stalks ran through the machine in thirty-three minutes, there being several stops. (The actual time, that is, deducting time lost in stoppages, was twenty-seven minutes forty seconds.) The yield of dry ribbons was 7.70 kilograms, or very nearly 17 pounds. A later trial of this machine on five kilograms of ramie stalks dried in a furnace at 30 centigrade (which makes softer fiber) ran through in three minutes and sixteen seconds actual time. The product was 21 per cent. of fiber, as claimed by M. Favier, though the record of actual weight can not be given. At these tests two men were employed, although a feeder and an assistant and a receiver and an assistant make up the usual complement of attendants or operators required.

Some days before the official trials I had an opportunity of examining the larger machine privately, and seeing its work for dry ramie stalks grown in several countries, including some secured by myself from two localities in Texas and sent to Paris with the United States agricultural exhibits. Without considering the capacity of the machine, that is, the amount of fiber it will turn out in a given time, it must be admitted that on dry ramie it does its work more perfectly than any decorticating machine I have ever seen. Some of the filasse from Spanishgrown stalks was almost nice enough to work up into twine or cordage or similar coarse manufactures without further manipulation. It may be remarked, however, that ramie is too valuable a fiber to be employed in cordage or the coarser manufactures. There is little or no waste by this machine; the chief objection that may be urged against it is its very complicated mechanism, adapting it more for use in a central factory, where it would be attended by experienced operators, than for employment on the farm to be run by ordinary farm hands. Its cost, too, makes it at once a machine for the central factory and not for the farm. The smaller machine, adapted for work on green stalks, costs 2,500 francs, and the dry stalk-machine, making a complete decortication, 5,000 francs; practically $500 and $1,000 respectively. Either machine requires a force of three fourths horse-power, the refuse of decortication supplying more fuel than the boiler requires. In fact the refuse of one machine is sufficient to furnish power for four. As has been stated, while two operators can run the machine, it is adapted for four persons; but as the work is light, it is claimed that it can be performed by two women and two children. It should be remarked that women are frequently employed in such occupation in Europe where only men would be employed 20789-No. 1—3