1. The fibro-vascular bundle of the leaf, single or double.

2. The base of the bract subtending the leaf-fascicle, non-decurrent or decurrent 3. The phyllotaxis of the cone, simple or complex.

4. The flower-bud, its less or greater development.

Some characters indicate the same distinction but are subject each to a few exceptions.

5. The fascicle-sheath, deciduous or persistent.

6. The walls of the ray-tracheids, smooth or dentate.

7. The connective of the pollen-sacs, large or small.

8. The formation of bark, late or early.

SUBSECTIONAL CHARACTERS

An exact subdivision of the Soft Pines is possible on the following characters.

1. The umbo of the cone-scales, terminal or dorsal.

2. The scales of the conelet, mutic or armed.

3. The pits of the ray-cells, large or small.

EVOLUTIONAL CHARACTERS

The progressive evolution of the fruit of Pinus, from a symmetrical cone of weak tissues, bearing a wingless seed, to an indurated oblique cone with an elaborate form of winged seed and an intermittent dissemination, appears among the species in various degrees of development as follows--

The seed

1. wingless.

2. with a rudimentary wing.

3. with an effective adnate wing.

4. with an ineffective articulate wing.

5. with an effective articulate wing.

6. with an articulate wing, thickened at the base of the blade.

The cone

1. indehiscent.

2. dehiscent and deciduous.

3. dehiscent and persistent.

4. persistent and serotinous.

and as to its form

5. symmetrical.

6. subsymmetrical.

7. oblique.

These different forms of the seed and, to some extent, of the cone, are available for segregating the species into groups of closely related members; while the gradual progression of the fruit, from a primitive to a highly specialized form of cone and method of dissemination, points to a veritable taxonomic evolution which is here utilized as the fundamental motive of the systematic cla.s.sification of the species.

SPECIFIC CHARACTERS

All aspects of vegetative and reproductive organs may contribute toward a determination of species, but the importance of each character is often relative, being conclusive with one group of species, useless with another. Characters considered by earlier authors to be invariable with species, such as the dimensions of leaf or cone, the number of leaves in the fascicle, the position of the resin-ducts, the presence of pruinose branchlets, etc., prove to be inconstant in some species. In fact, as the botanical horizon enlarges, the varietal limits of the species broaden and many restrictions imposed by earlier systems are gradually disappearing.

Variation is the preliminary step toward the creation of species, which come into being with the elimination of intermediate forms. Variation in a species may be the result of its partic.i.p.ation in the evolutionary processes culminating in the serotinous Pines, or it may result from the ability of the species to adapt itself to various environments by sympathetic modifications of growth, or it may arise from some peculiarity of the individual tree.

Evolutionary variation is a.s.sociated with the gradual appearance of the persistent, the oblique and the serotinous cone, and of the multinodal spring-shoot. For these conditions appear in less or greater prevalence among the species of the genus.

Variation induced by environment finds familiar ill.u.s.trations among the species that can survive at the limits of vegetation and can meet these inhospitable conditions by a radical change of all growing parts. Such variations are mainly of dimensions, but, with some species, the number of fascicle-leaves is affected and the shorter growing-season may modify the cone-tissues. In Mexico and Central America are found extremes of climate within small areas and easily within the range of dissemination from a single tree. The cause of the bewildering host of varietal forms, connecting widely contrasted extremes, seems to lie in the facile adaptability of those Pines, which are able to spread from the tropical base of a mountain to a less or greater distance toward its snow-capped summit.

The peculiarities of individual trees that induce abnormally short or long growths, the dwarf or other monstrous forms, the variegations in leaf-coloring, etc., etc., are not available for cla.s.sification, for they may appear in any species, in fact in any genus of Conifers. These variations are artificially multiplied for commercial and decorative purposes. But inasmuch as they are repeated in all species and genera of the Coniferae that have been long under the observation of skillful gardeners, their significance has a broader scope than that imposed by the study of a single genus.

PART II

CLa.s.sIFICATION OF THE SPECIES

The following cla.s.sification is based on the gradual evolution of the fruit from a cone symmetrical in form, parenchymatous in tissue, indehiscent and deciduous at maturity, releasing its wingless seed by disintegration--to a cone oblique in form, very strong and durable in tissue, persistent on the tree, intermittently dehiscent, releasing its winged seeds partly at maturity, partly at indefinite intervals during several years. This evolution embraces two extreme forms of fruit, one the most primitive, the other the most elaborate, among Conifers.

Two sections of the genus, Soft and Hard Pines, are distinguished by several correlated characters, and moreover are distinct by obvious differences in the tissues of their cones as well as in the quality and appearance of their wood.

With the Soft Pines the species group naturally under two subsections on the position of the umbo, the anatomy of the wood and the armature of the conelet. In one subsection (Cembra) are found three species, P.

cembra and its allies, with the cone-tissues so completely parenchymatous that the cones cannot release the seeds except by disintegration. In both subsections there is a gradual evolution from a wingless nut to one with an effective wing, adnate in one subsection, adnate and articulate in the other. The different stages of this evolution are so distinct that the Soft Pines are easily separated into definite groups.

Among the Hard Pines a few species show characters that are peculiar to the Soft Pines. These exceptional species form a subsection (Parapinaster) by themselves.

With the remaining species, the majority of the Pines, the distinctions that obtain among Soft Pines have disappeared. The dorsal umbo, the articulate seed-wing, the persistent fascicle-sheath, the dorsal and ventral stomata of the leaf and its serrate margins, the dentate walls of the ray-tracheids have become fixed and constant. But a new form of seed-wing appears, with a thickened blade, a.s.suming such proportions in P. Sabiniana and its two allies that these three const.i.tute a distinct group, remarkable also for the size of its cones.

Here also appear a new form of fruit, the oblique cone, and a new method of dissemination, the serotinous cone. a.s.sociated with the latter are the persistent cone and the multinodal spring-shoot. These characters do not develop in such perfect sequence and regularity that they can be employed for grouping the species without forcing some of them into unnatural a.s.sociation. The oblique cone first appears sporadically here and there and without obvious reason. The persistent cone, the first stage of the serotinous cone, is equally sporadic in the earlier stages of evolution. The same may be said of the multinodal shoot.

Nevertheless these characters show an obvious progression toward a definite goal, where they are all united in a small group of species remarkable for the form and texture of their cones, for a peculiar seed-release and for the vigor and rapidity of their growth. It is possible, with the a.s.sistance of other characters, to segregate these species in three groups in which the affinities are respected and the general trend of their evolution is preserved.

The first group, the Lariciones, contains species with large ray-pits, cones dehiscent at maturity, and uninodal spring-shoots. They are, with two exceptions, P. resinosa and P. tropicalis, Old World species.

The second group, the Australes, contains species with small ray-pits, cones dehiscent at maturity and spring-shoots gradually changing, among the species, from a uninodal to a multinodal form. They are, without exception, species of the New World.

The third group, the Insignes, contains the serotinous species. The ray-pits are small and the spring-shoots are, with two exceptions, multinodal. With two exceptions, P. halepensis and P. pinaster, they are New World species.

These three groups, being the progressive sequence of a lineal evolution, are not absolutely circ.u.mscribed, but are more or less connected through a few intermediate species of each group. The systematic position of these intermediate species is determined by their obvious affinities. It cannot be expected that the variations, which take an important part in the evolution of the species, progress with equal step or in perfect correlation with each other.

As to specific determinations, a little experience in the field discloses an amount of variation in species that does not always appear in the descriptions of authors; and species that are under the closest scrutiny of botanists, foresters or horticulturalists, attest by their multiple synonymy their wide variation. The possibilities of variation are indefinite and, with adaptable Pines, the range of variation is somewhat proportionate to change of climate. In mountainous countries, where there are warm sheltered valleys with rich soil below cold barren ledges, the most variable Pines are found. The western species of North America, for instance, are much more variable than the eastern species, while in Mexico, a tropical country with snow-capped mountains, the variation is greatest.

Therefore in the limitation of species undue importance should not be given to characters responsive to environment, such as the dimensions of leaf or cone, the number of leaves in the fascicle, etc. Moreover, there are familiar examples (P. sylvestris, etc.) that show the possibility of wide differences in the cone of the same species.

In the following cla.s.sification species only are considered without attempting to determine varietal or other subspecific forms. But varieties are often mentioned as one of the factors ill.u.s.trating the scope of species. Synonymy serves a like purpose, but synonyms not conveying useful information are omitted, Roezl's list of Mexican species, for instance, and variations in the orthography of specific names.

PINUS