As has been shown, no induced resin passages were present and no wood formation occurred until the latter part of April. Consequently this increased early yield came from the induced responses of the resin passages already present in the round timber. These apparently were stimulated to greater activity by the advance streak than would have been produced by ordinary chipping later in the season."

19

Observant operators were often aware of the fact that men who started operations with a little chipping very early in the season generally obtained high yields. The following statement by a practical man, whose family has been in the turpentine business for four generations, bears directly upon this point:

It has been a matter of common experience with everyone familiar with turpentining that with a box cut late in the season, followed by chipping immediately thereafter, the yield of gum is very little for the first two or three months of the season. * * * If you chip any face at one time 3 inches up, the face will not run well for two or three months following.20

Apparently the region just above the wound is profoundly influenced by it, but the response may not be apparent at once. This idea is incorporated in the standard Forest Service practice on the Florida National Forest, where it is the custom to cut one streak when the aprons (metal strips used instead of gutters) are placed; that is, the streak is cut some weeks in advance of the regular chipping. (See Pl. I, fig. 3.)

The influence of the advance streak is not only a factor in the case of the response of the ordinary resin passages, formed in the round timber, but is also probably manifested in the amount and responses of the resiniferous tissue found in the new wood which is formed as the season progresses. Similar responses may underlie the following observation, namely, that in the second year of the turpentine operation resiniferous tissue occurred somewhat earlier in the new growth ring than it had in the first year of the operation. Some experiments on obtaining the gum storax from red gum trees also appear to show a somewhat analogous situation.21 These trees, which normally have no resin passages, develop them as a result of wounding. In the case of red gum trees which were wounded as late in the spring as May 30, the first response in the developing annual ring was the formation of normal wood. It was from four to six weeks later that the formation of functioning resin passages, as indicated by the presence of any considerable amount of gum exudation,

19 Since this was written it has been learned that the advance streak is recognized as essential in turpentine practice in India.

20 Courtesy of Mr. A. Sessoms, of Bonifay, Fla.

1 Gerry, E., "American Storax Production," Jour. For. 19: January, 1921.

occurred. Here, as in the case of the pines, and even in that of the conifers studied by Tschirch in midsummer condition, considerable time was required to obtain the major response induced by the wound.

PART PLAYED BY RESIN PASSAGES FORMED AFTER WOUNDING.

When the new wood formation begins after wounding, one of the most striking features of the developing ring is the early differentiation of more resin passages than are normally found. As soon as these resin passages, induced by wounding, are formed, they add materially to the flow of oleoresin, since they are generally very numerous (Pl. II, fig. 3) and are often arranged in rather continuous tangential series (Pl. II, figs. 3 and 5). Although a considerable amount of gum comes from this newly formed resiniferous tissue, it seems doubtful if, as has been stated, it constitutes “the chief source of commercial crude turpentine." 22 The term "secondary resin passages" has been used to describe the resiniferous tissue formed in the annual rings after turpentining. In contrast to this the term "primary resin passages" has been applied to the resin passages present in the round timber. Both these terms seem undesirable, inasmuch as the distinction between the resin passages of the two regions is not clearly defined, and since a certain number of the so-called primary passages would have been present under any circumstances in the new wood. Which these are, it would be impossible to determine. The types of resin passages found will be discussed with reference to length and number in the following pages.

RESULTS OBTAINED BY DIFFERENT METHODS.

EXPERIMENTS AT COLUMBIA, MISS.

STANDARD PRACTICE.

The operating methods termed "standard" in this report are the regular methods practiced by the company with which the cooperative experiment here described was carried on. The chipping was

somewhat heavier than that known as the standard Forest Service method. In the Standard chipping as practiced at Columbia, Miss., the streaks were cut with a No. 2 hack or chipping tool. (See Pl. IV, fig. 2.) The weekly streak varied in depth from one-half to three-fourths inch and was about one-half inch in height. The type of forest, as indicated in Plate I, figure 2, consisted of large mature timber, but included also many old but relatively small suppressed trees. The chipping began early in the spring of 1916. The first material for microscopic study was collected in April, after four or five streaks had already been cut. The yield from

For. Serv. Bul. 90, p. 27.

about 5,300 cups, operated by this method, was separately recorded, so that it could be compared with the yield obtained from the trees on similar stands of timber adjacent, which were operated by the two special methods under experiment.

This operation was carried on for a period of two years. Figures 1 and 2 give the monthly observations on the five trees selected in 1916 and in 1917 (not the same five trees both years). The number of tracheids counted in a radial direction across the ring indicates. the diameter increase since the beginning of the year. The fluctua

Apr May June July Aug. Sept. Oct. Nov. Dec. Apr. May June July Aug. Sept. Oct. Nov. Dec: FIG. 1.-Standard trees, 1916.

Number of tracheids, observed April to December; in 1916, growth ring. °Summer wood present.

Number of resin centers per unit area (an arbitrary tangential extent; diameter of microscopic field by the width of the annual ring observed). Observed April to December, 1916; in 1916, growing ring; in 1915, completed ring.

tions in the number of resin passages present are also shown. In Tables 2 and 3 are given the relative increases and decreases in width of annual ring and percentage of summer wood as observed in December, 1916, in 50 trees chipped by each of the three methods under observation and in 20 unturpentined trees. In Table 4 are given similar observations for the years 1915, 1916, and 1917.

In 1916 considerable variation was found in the number of tracheids observed from month to month in the specimens from the five selected trees. (Fig. 1.) The width of the 1916 annual ring in the specimens cut near the end of the season was in several cases 87404°-22- -3

The num

less than that observed at lower levels earlier in the year. ber of resin passages per unit area 23 also was often smaller at the end of the season, but was throughout larger than that in the round timber. In 1916 this material was taken wherever a good chip chanced to be cut along the streak.

In 1917 care was taken to cut the specimens midway between the peak and the corner in all cases. By comparing figures 1 and 2, it is apparent that marked variations occurred, nevertheless, in the

1917 material. Consequently these variations may be considered as characteristic of the annual ring, whether studied in the same or in different parts of its circumference and at different heights in the tree. During 1917, furthermore, many more resin passages than were normally found were produced, but at the levels where

An arbitrary unit made up of the portion of the annual ring in question which could be included in a microscopic field of a given magnification, the diameter of which was moved so as to include the approximate rectangle bounded by the beginning and end of the annual ring.

the 1917 material was cut it was obvious (fig. 2) that the number of resin passages present by June or July in the 1917 ring was greater than in the 1916 ring; and at no time in 1917 was the maximum number of resin passages in the 1916 ring as great as the maximum number in the 1917 ring. If, on the other hand, the number of resin passages present in the July, 1916, material cut in 1916 (fig. 1), is compared with the 1917 material cut in 1917, it is apparent that approximately similar numbers of resin passages were present.

Apr. May June July Aug. Sept. Oct. Nov. Dec. Apr May June July Aug. Sept. Oct. Nov. Dec.

FIG. 3.-Double trees, 1916.

Number of tracheids, observed December; in 1916, growth ring. mer wood present.

April to
• Sum-

Number of resin centers per unit area (an arbitrary tangential extent; diameter of microscopic field by the width of the annual ring observed). Observed April to December, 1916; in 1916, growing ring; in 1915, completed ring.

The conclusion therefore seems justified from this and other data that some of the resin passages are shorter than others, and are entirely cut away as chipping progresses. The microscopic observations indicate that in the case of the standard tract, the trees, judged by their wood formation in the neighborhood of the faces, a region where the wound response is very pronounced, did not suffer seriously from the effects of turpentining by this method for a two-year period. The wood formation was reduced more than in the case of the narrow chipping, as is brought out in the comparisons given in Tables 2 to 4. From Table 2, for instance, it is apparent that in