the older cells tend to become more or less impregnated with resin as they are transformed into heartwood, which serves primarily for mechanical support. The resin in the heartwood flows out much less easily than that in the sapwood, as may be seen on any freshly cut log or stump. Although the heartwood contains more resin by chemical analysis, it will not, if tapped, produce a sustained yield of resin, as the sapwood, if properly treated, will. Gum turpentine is obtained from a number of the outermost sapwood rings (Pl. III, fig. 2). The depth of the cut in relation to the width of the sapwood is a very significant factor with reference to both the yields obtained and the injurious effect of turpentining upon the tree. Successful turpentine operations depend upon keeping the sapwood in a healthy condition, capable of responding not only to normal stimuli but to the special wound stimuli as well.

TRACHEIDS.

During the growing season, April to November, the individual cells of the outermost annual ring of the sapwood, as they are derived from the cambium, become differentiated and develop into specialized structures. These are the tracheids. (T, Pl. II, figs. 1 to 5.) They are vertically extended cells which complete their formation and lose their protoplasmic content early. They then serve for sap conduction and finally for the mechanical support of the tree. They are not directly connected with oleoresin production, although they constitute a significant part of the organic mechanism of the tree. They differ somewhat in appearance, as was stated in the discussion of spring wood and summer wood (p. 5). The character of the summer wood tracheids and also the total number of tracheids formed often serve as indicators of the vitality and responsive power of the trees under observation.

PARENCHYMA.

The parenchyma tissue found in pines may be considered under two main heads: First, the horizontally extended parenchyma cells constituting the rays (R, Pl. II), including the fusiform rays (FR, Pl. II, fig. 4) with their horizontal resin passages; second, the vertically extended parenchymatous elements, which grouped together (GP, Pls. II, V, and VI), form the resin-producing centers, the so-called resin passages, ducts, or canals. The parenchyma cells are characterized by the fact that they "remain alive," that is, retain their protoplasmic content and function actively for a number of years or until approximately the time when the inner rings of the sapwood change into the heartwood.

It is both from the horizontally extended fusiform rays, as exposed on a freshly cut tangential surface, and the vertically extended parenchyma aggregates, as exposed on the cross section, at the streak, that the droplets of oleoresin may be seen to exude. Large nuclei are often to be seen in these cells, as well as accumulations of starch grains. Tannin is also reported to be associated with resin formation.12 The vertical and horizontal systems of resin-producing parenchyma are more or less united, since they frequently cross each other. The method of turpentining which keeps these cells active and which provides suitable stimulation to insure their greatest productivity without undue injury will give the highest and best yield of oleoresin over a period of years. In the discussion in the following pages of the results from the experiments, the methods used were judged, not alone by the yield obtained, but also by the amount and type of the wood tissue produced, as indicative of the vitality and responsive power of the trees. In each case the wood tissue produced by neighboring, comparable, unturpentined trees growing under similar environment was studied and used as a check upon the judgments formed.

METHODS OF STUDY.

COLLECTION AND TREATMENT OF MATERIAL.

The first material to be studied was in the form of fresh chips or pieces collected by the writer from the living trees and put in corked bottles. This material was examined within a few hours after cutting. Specimens shipped to the laboratory at Madison periodically during the season were handled in the same way, with the exception that sometimes moistened cotton was put in the bottles to prevent the drying out of the specimens. Later, fixing solutions were used, and the chips or increment borings were placed in the solution selected as soon as cut. These were kept for a length of time which necessarily varied with circumstances, washed with water, and stored in glycerine and alcohol. The fixatives used were: (1) Mercuric chloride, a saturated solution in 90 per cent alcohol; (2) mixture of a saturated solution in 90 per cent alcohol of mercuric chloride (3 parts) and of picric acid (1 part); (3) chrome-acetic fixative, consisting of a mixture, in 100 cubic centimeters of water, of 1 per cent glacial acetic acid and 0.7 per cent chromic acid.

13 Haas and Hill: "Introduction to the Chemistry of Plant Products," Vol. I, 3d Ed. 1921. pp. 195, 217. "In Pinus it is stated that the amount of tannin varies with that of the resin; thus in the spring it was found that as the tannin decreased in amount so the resin increased. * * * That starch frequently is contained in the same cells with tannin suggests a connection between the two and it is not impossible that the starch may contribute glucose for the constitution of the tannin. The cells surrounding the epithelium of resin ducts contain tannin and starch." Wiesner" Die Rohstoffe des Pflanzenreiches " concluded that tannin was an intermediate product in resin formation.

SECTIONING AND MOUNTING.

Relatively thin sections (10 to 48 microns) were made from the greater number of the specimens studied. For this purpose a sliding microtome of the Jung type with the Thomson modification was used. Some observations were made with a hand lens on smoothly cut surfaces. The sections to be preserved or photographed were stained with safranin (water-alcohol mixture), dehydrated with alcohol, cleared with xylol, and mounted in Canada balsam. Microscopic examinations were also made of microtome or hand sections temporarily mounted in glycerine alcohol (50-50 mixture).

AMOUNT OF MATERIAL.

Samples were obtained each month during the 1916 and 1917 seasons, from 15 trees, 5 from each of the three experimental plots.18 Specimens were also examined at the end of each season from 50 trees on each plot and from 50 unturpentined trees from the same locality. In addition to this, considerable material was collected from Bogalusa, La.; Kokomo and Gulfport, Miss.; Daytona, Bonifay, and the Forest Service experiments and leases near Camp Pinchot, Fla. More than 4,500 microscopic mounts were prepared and examined.

PROCESS OF TURPENTINING.

METHODS OF WOUNDING OR SCARIFYING THE TREES.

In ordinary methods of turpentining the oleoresin is made to flow from the tree by periodically cutting a wound or streak through the bark and for a variable distance into the sapwood. In the United States the cut or streak is generally made in the form of a somewhat flattened V, the point or peak of which is at the center of the face. (Pl. I, fig. 4.) The wound used by the French is different in shape (Pl. I, fig. 6), and smaller. The Germans recently have been practicing modifications of the American system.

Except for the general type of face used, the commercial practice of turpentining is not standardized in the United States. The size of the faces used, the amount of bark bars between faces, and the dimensions of the streak or wound cut each week, show marked differences in different operations. The depth of the weekly chipping may vary from one-half inch to 14 inches (bark not included), and the height of the chipping from one-half inch to 1 inch, or enough, as some operators say, to "keep ahead of the lightwood."14

13 When more than one face was on a tree each face was indicated by a letter descriptive of its general position with reference to points of the compass, as N., S., E., W. (See, for instance, tree 1, fig. 1.)

14 See p. 31.

It is customary to cut from 28 to 40 streaks each season. Often one-third of the trees in a commercial operation die, chiefly because of the undue severity of the methods used in turpentining. In some places the old method of cutting boxes or cavities at the butts of the trees (Pl. I, fig. 1) to hold the exuding gum, is still used, although, for the most part, cups of various types are employed (Pl. I, figs. 1, 3, 5, and 6).

After chipping, the gum or oleoresin exudes from the freshly cut surface. The most abundant exudation (88 per cent) has been observed to occur during the first three days after chipping. (Table 1.)

TABLE 1.-Rate of exudation of gum from chipped longleaf pine.1

SIGNIFICANCE OF RESIN PASSAGES PRESENT IN THE UNTURPENTINED TREES.

The gum which exudes during the first weeks of a virgin or firstyear operation comes from the normal resin passages which were already present in the round or unturpentined timber. Plate II, figure 1, shows the characteristic condition of the trees at Columbia, Miss., on April 13, 1916. At that time no new wood or resiniferous tissue for the 1916 ring had formed. Plate II, figure 2, shows a cross section cut from a tree on June 13, 1916. The development of new wood cells is apparent, and new resiniferous tissue may be seen to be differentiating in the region next the cambium. Plate II, figure 3, shows a cross section cut from a tree on July 5, 1916. Marked differences between individual trees occurred, however, and the range of development of the 1916 ring shown in the three figures might be encountered in material cut at the same time from different trees during either late May or early June. Often it is the end of May before the new and augmented development of the resiniferous tissue, formed after wounding in the new wood, is sufficiently advanced to yield appreciable amounts of resin. Indeed it frequently happens that in a virgin operation the normal resin passages present in the round timber may fill the cups not only for the first but also for the second and sometimes for the third time, or " dipping," before any new resin

passages are differentiated in the annual ring beginning to form for that year. An instance of a yield obtained by the end of April, ranging from 16 to 24 barrels of gum per month, per 10,000 cups, or " crop," was observed on a virgin operation where at the time no new resiniferous tissue had been formed. Therefore it is apparent that the resin passages already present in the outer sapwood of the round timber play a significant part in producing the gum obtained.

USE OF THE "ADVANCE STREAK."

15

A practice, the consequences of which are as yet not fully explained, but which appears to produce desirable results, is that of cutting a streak or wound some time before regular chipping begins.1 The results of such scarification were pointed out by Dr. C. H. Herty in 1911. In the early experiments, made to demonstrate the advantages of replacing the box system by the cup system, the boxes were cut in winter and cornered in late winter. The first streak was made somewhat later along the upper edge of the wound made by cornering. In the case of the cupped trees, on the other hand, no such severe wound was inflicted. The trees, however, were cornered-that is, the bark and some of the wood were removed in order to obtain a suitable surface for the gutters. The first streak was cut at a little distance above the curved rim of the cornered surface. The ends or corners of the streak were farther above the cut surface than the peak and consequently did not reach directly to this open wound made some time before. Until May the yield of gum was notably less in the cupped trees than in the boxed trees, which had received a more intense wound stimulus.

Wounding the outer sapwood, therefore, in this manner (or even less severely, as is the practice on the Florida National Forest and in India) appears to have a very definite effect on the early yield of gum. It was estimated by one operator, for instance, that this practice gained for him, on one operation, a total of $500,000 in one year.16 This practice, strongly advocated by Dr. Herty, was repeatedly employed with success. The theoretical explanation of it, advanced by him on the basis of the results of Tschirch's " investigations on other resin-yielding trees, was however not in accord with the facts. His deduction was as follows:

17

Immediately after cornering (late winter) the formation of secondary resin ducts begins at all points of the cut. Later when the tree is chipped, these secondary ducts are opened along the full length of the cut and a good yield is consequently at once obtained.18

15 For. Serv. Bul. 90, p. 28.

16 Herty, C. H., "The Turpentine Industry in the Southern States." Jour. Franklin Institute, March, 1916, p. 362.

17 See footnote p. 3.

18 For. Serv. Bul. 90, p. 28.