| REVISITON: 3: A OF THE I AND a ak OF PINUS ELLIOTTII. By DR. GEORGE ENGELMANN. ¢ From the Tr @sactions of the Academy of Science of St. Lonis, Vol.1V., No.1. : & ae a : | é : ST; LOUIS, MO, (S50 = R. P. Sruptey & Co., Printers and Binders. _ 9 161] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. I Revision of the Genus Pinus, and Description of Pinus ELLIoTTII. By Dr. GrorGE ENGELMANN. No difficulty exists in the circumscription of the genus Pinus ; floral unite with vegetative characters to establish it so firmly and so plainly, that nobody fails to recognize the species belonging to it. But when we come to analyze and to group the 60 or 70 spe- cies of pines which are known to us, we find that they appear so similar, that all attempts to arrange them satisfactorily have failed. The most obvious distinctive character was found in the number of leaves in each bundle, and thus the sections of 2-leaved, 3- leaved and 5-leaved pines were the only ones known to the older botanists. Spach (Syst. veg. 1834) separated Ceméra on account of its “‘wingless” seeds; Link (Linn:ea, 1841) relied on the num- ber of leaves only, adding two sections, one with 2 or 3, the other with 3 or 4 leaves in a sheath; Endlicher (Synopsis, 1847) was the first to point out the form of the cone-scales as an impor- tant character, and his first two sections Ceméra and Strodus, were by the form of this scale distinguished from the other Pines ; he retained the character of the “‘ large wingless” seeds, to sepa- rate by it Cemdra from Strodus, and Pinea from the other two- leaved pines which constituted his section Prmaster. Later writers did not add anything to our knowledge of the systematic relations of pines: Carritre (Coniféres, 1855) copied Endlicher, and Gordon (Pinetum, 1858) went back to the mere number of ‘leaves to characterize the sections. Ten years later Parlatore (DeCand. Prod. 162, 1868) followed Endlicher in adopting the differences in the form of cone-scales as the most valuable char- acter, and advanced a step further by discarding the proportion- ate size of the seeds as of sectional value. He divided his subgenus Pixus in two sections, Przea with pyramidate and Cemébra with dimidiate apophyses. The subsections of his Prxea were again based on the number of leaves, in twos, in threes, or in fives in each bundle; those with single leaves, with 2 or 3, and those with 3, 4 or 5 leaves had to find their place as best they could. 2 TRANS. ST. LOUIS ACAD. SCIENCE. [162 Not satisfied with such superficial knowledge of this interest- ing and important genus of trees, I have for a number of years devoted my leisure hours to the careful study of the different spe- cies accessible to me. In the following pages I give the principal results of my investigations. — Sizz. Almost all the pine species grow up to be trees ; the only shrubby one known to me is Pinus montana, heretofore known as P. Pumilio; a few make small, insignificant trees, such as 7. tuberculata ; the nut-pines, or cembroid pines, never grow large, but several others attain the greatest dimensions; P. Laméer- tiana grows to the height of 300 feet, with 20 feet in diameter, and P. ponderosa (at least in California) comes very near it; these two are probably the largest pines known. The aGe of pines varies between 15 to 25 years (P. tuberculata and perhaps P. montana), 300 years (P. mitis, P. ponderosa, P. Balfouriana), and 500 to 600 years (observed in P. montz- cola and P. Lambertiana). The Bark in some species is thin, only a few lines thick, flaky and detached in scales (P. contorta, P. restnosa) ; in others (e.g. P. ponderosa) it is several inches thick, persistent, rough, and deeply cracked. It is gray in some species, e.g. in the nut-pines, but most commonly of a brown red or cinnamon color, or some- times deeper brown; in P. australis and P. Lilliottiz, especially in the latter, it is laminated, the external layers peeling off in thin plates. The woop grows rapidly, especially during the first (often the first 50) years of their age, so that annual rings are sometimes 2 or 3 lines thick; in P. glabra I have seen them even GO, i 7. insignis 5 and in P. rigida var. serotina 4 lines thick ; in old . age or in the short seasons of high altitudes the wood grows_so slow that sometimes ten annual rings make not more than the thickness of one line. The sapwood is always white, and it takes many years before it turns into perfect or heartwood: in P. ponderosa, Lamber- téana, and mitis sometimes roo or even 150 years ; in others, e.g. P. flexilis and Sabiana not more than 20 or 30 years; but the majority of pines which I have examined may require 50 or 60’ years to mature their heartwood. In many other trees this pro- cess takes about 20 or 30 years, in most oaks on an average 163] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 3 about 20 years, in Cata/fa not more than 2 or 3 years. The thickness of the sapwood in pines is usually 2-4 inches, and rarely under 1 inch; in P. ponderosa I have found it sometimes even Io inches. The wood cells and especially those more compact ones of the late summer growth, the outer part of the layers, are often strongly impregnated with resin and thereby darker colored, yel- lowish or brown, and become in thin sections semi-transparent ; _ this is much more the case in those of the section Pinaster than of Strobus. The former have mostly heavier and harder wood than the latter, though we find exceptions, such as P. contorta, which has soft wood similar to that of the white pine or spruce. Spirally marked cells, such as abound in. Pseudotsuga and in Taxus, have not been found in the pines. The LEAVES, in the wider sense, are of seven different forms: the cotyledonous or seed-leaves, the primary leaves, the ordinary bracts, the secondary leaves, the bracts constituting the sheath of these, the bracts forming the involucrum of the male flowers, and the bracts supporting the carpellary scale. The CoTYLEDONOUS LEAVES form a whorl of 4 to 18 in num- ber, are triangular, flat on the back, keeled above, higher than broad and mostly entire; in P. Strodus I find the keel slightly. spinulose -dentate. Stomata are found only on the inner and upper sides, as is the case in the cotyledonous leaves of most conifers; those of Scéadopitys are, as far as I know, the only ones that have stomata merely on the under and none on the upper side. ; . The PRIMARY LEAVES succeed the cotyledons on the main axis ; in some species (P. zxops, P. rigida, P. Canariensis, etc.) they are also found on the sprouts. They are always subulate from a broader base, flat, keeled on both surfaces, always serru- late, even in those species whose secondary leaves are entire (P. edulis), with stomata in rows on both surfaces, more on the lower than on the upper face. The primary leaves not rarely produce in their axils buds with secondary leaves, but they are most generally reduced to BRACTs (Hochéblaetter) before their axils become productive. These bracts are triangular-lanceolate, membranaceous or coriaceous, 4 TRANS. ST. LOUIS ACAD, SCIENCE. [164 entire or mostly fringed on the edges, more or less persistent. or mostly deciduous, sometimes articulated above their base. The sECONDARY LEAVES constitute the foliage of the tree: they are borne on an undeveloped branchlet in the axils of pri- mary leaves or mostly of bracts, and are surrounded at base by a sheath of bud-scales (Wieder blactter). These consist of 2 short, rigid, strongly-keeled, lateral bracts, anda number (6-10 or more) of longer, thinner inner ones, which generally are woven together by the delicate fringes of their edges, and are then persistent with the leaves, though in time worn off at the ends; or they are loose, at last spreading and deciduous at the end of the first season. This is the case in all the species of the section S¢rodus, in the nut-pines, and in a few others: ?. Balfouriana, Gerardiana, Bungeana, Chthuahuana, and usually also in P. decophylla. The secondary leaves generally occur in definite numbers, 1 to 5, in a bunch, or their number is slightly variable: some species have regularly 2 and 3 leaves (P. mztis, P. Eiiivttic), ties vary with 3 to 5 leaves (P. Montezume) ; species with regularly 3 leaves have occasionally 2 or 4, such with 5 leaves are sometimes found with 6 and even 7 leaves. Where we have one (only in P. monophylia), the leaf is terete ; where there are two, the leaves are semi-terete, convex on the lower surface and flat on the upper one when fresh, or channelled when dry. ‘Those leaves that grow in bundles of 3 or more are triangular, the upper surface being more or less elevated and keeled ; ternate leaves are gener- ally somewhat flatter, and quinate ones higher and regularly triangular. Thus the shape of the leaf and especially its trans- verse section is mostly sufficient to determine the number i in which they occur. he leaves are in most species minutely but sharply serrulate on tRe edges, and mostly also on the keel of the upper surface. These serratures are closer together, or more distant, coarser or more delicate, but are absent only in a very few West-American species: the ceméroid or nut-pines, P. Balfouriana, and most forms of P. flexélés. The tips of the leaves are generally entire, acute, or acuminate, rarly obtusish; but in all the species of the section Strobus they are in the young and fresh leaves finely denticulate. The stomata are usually distributed in longitudinal rows over 165] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 5 both surfaces in the Pixaster section; only P. Balfourtiana has none on the back, and thus approaches S¢rvdus in this as it does in many other respects. In S¢rodus we find on the back no or but few stomata, or sometimes a single or an interrupted line of them. P. Lambertiana only has numerous stomata on the back, thus approaching Pinaster. I will have to dwell somewhat extensively on the internal struc- ture of the leaves, as it provesto be of the greatest impor- tance for the classification of the species. We distinguish in a transverse section the thick epidermis, the chlorophyll- bearing parenchyma cells, and in the centre the fibro-vascular bundle. This latter is single in the terete and mostly in the quinate leaves; it is double in the broader triangular or ternate, and ‘in the semi-terete or binate leaves. This difference, however, is of very little diagnostic importance, as we find occasionally single or double bundles in the same species. The fibro-vascular bundles always show wood cells on the upper or ventral, and bast cells on the lower or dorsal side, traversed by delicate medul- lary rays, usually obliquely diverging from the lower to the upper side. The bundles are imbedded in a mass of small (medullary?) cells, free of chlorophyll, and are together with those surrounded and separated from the parenchyma by a sheath of larger cells, also destitute of chlorophyll. Within the parenchyma of the leaf a smaller or larger number of longitudinal tubes or ducts are found, the RESIN DUCTS, nor- mally probably two, but very often more, even as many as a dozen or more. These ducts occupy a certain definite position within the leaf. They lie (1) close to the epidermis, peripheral ducts, in some species more on the ventral, in others more on the dorsal side of the leaf; or (2) they occupy a place within the parenchyma and surrrounded by it on all sides, parenchymatous ducts; or (3) they lie close to the sheath which surrounds the . vascular bundles, zzternal ducts. This position of the ducts is so constant, and seems to be so intimately connected with the essential character of the plant, that I venture to adopt it as one of the principal characters for the subdivision of the genus. I must add, however, that in some few species smaller, accessory ducts do sometimes occupy an abnormal position. Thus I find occasionally in some Strodz, especially in P. exce/sa, where there 6 TRANS. ST. LOUIS ACAD. SCIENCE. [166 are mostly two peripheral dorsal ducts, a third upper parenchy- matous one; in P. Bungeana, which generally has numerous peripheral ducts all around, occasionally a single lateral paren- chymatous duct is observed; P. Canarzensés has regularly paren- chymatous ducts, but sometimes they are connected with the epidermis by a very thick bundle of strengthening cells (of which presently) ; 7. Zariczo has normally parenchymatous ducts, but in a specimen of var. Pyrenazca from the Pyrenees in Herb. Cos- son I find them sometimes almost touching the epidermis cells, and therefore easily mistaken for peripheral. In P. rigida and Teda, and also in P. pungens and filifolia, which all have normally parenchymatous ducts, I occasionally have observed a number of smaller accessory ducts close to the sheath of the ves- sels. In pines with very slender leaves it is sometimes difficult to discover the ducts, and in some forms they are, I believe, really absent, especially in cultivated specimens. Such may give us some trouble in their classification. A peculiar element in the structure of the pine leaves are cer- tain cells which had been formerly named ** hypoderm cells” ; but as they also occur in other parts of the leaf apart from the epidermis, they more appropriately receive the name of sTRENGTH- ENING CELLS. They are thick-walled, elongated, colorless cells, much larger than the bast and wood cells, generally of the diame- _ter of the epidermis cells, rarely a little larger, often smaller, and always smaller than the cells of the parenchyma. They give to the leaf its rigidity, and are most abundant in the most rigid pine leaves; in the softer more flaccid ones they are almost entirely wanting. Thus they are scarce or entirely absent in some species of the Strodus section; in P. Pseudostrobus and P. Jilifolia they are very imperfectly developed. The strengthening cells are prin- cipally found under and close to the epidermis (whence the name hypoderm cells) either ina continuous layer or mostly in bun- dles, interrupted by the lines of stomata; they are generally most abundant within the angles of the leaves. Sometimes they surround the ducts, and in all the species allied to P. resz- nosa and P. sylvestris they are found only there, and not or scarcely at any other places. In some species these cells also occur within the sheath, above and below the fibro-vascular bun- dies. Their presence and position is not absolutely constant, but 167] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 7 may be relied on to some extent for diagnostic purposes. Thus the quinate smooth-edged leaves of P. flexilis and P. Balfour- tana, which would be difficult to distinguish without their cones, may be readily recognized by the strengthening cells, which in the latter surround the more closely approximating ducts, while in the former the ducts, widely apart from one another, are desti- tute of these cells. € PERSISTENCE of the leaves is very different in different species; in P. Strodus and others they fall in the autumn of the second year; more commonly they last to the end of the third year; in some species, e.g. P. Banksiana, they do not fall before they are 4, 5 or even 6 years old; in P. Balfourtana, or at least in var. arzstata, I have seen them persist 12 to 14 years. en the leaves persist only a short time and are long, and the annual growth of the axis is short, they form brushes or tassels (P. aus- tralis) at the end of the branchlets, but where they are short and persist long (P. Balfouriana) they give the branchlets that “‘fox- tail or bottle-brush” appearance of which travellers speak. In young and vigorous trees the leaves are apt to persist longer than in old ones. In exceptional cases and as a monstrosity the leaf-bundles be- come proliferous, the branchlet which bears the secondary leaves elongating and forming a regular branch. The pines are moneecious trees which bear their male and fe- male flowers generally on different branchlets, the male commonly on the lower, the female frequently on the upper part of the tree ; sometimes both are found on the same axis, the male below, the female above. The MALE FLOWERS are borne on the lowest part of the year’s shoot, in the axil of bracts, either crowded together in a kind of a head or elongated in a spike; the axis usually continues to elongate during or after flowering and makes a leafy branch, which in its continuation in succeeding years often again bears flowers. Male flowers sometimes abnormally make their appear- ance higher up.on the axis mixed with leaf-bundles and occupy- ing the place of such. The male flowers consist of an indefinite number of anthers sessile on a more or less elongated column, and have the form of an oval or a cylindrical ament, for which they used to be taken. They are surrounded by a somewhat defi- 8 ; TRANS. S1. LOUIS ACAD. SCIENCE. [168 nite number of bracts, which assume the functions of a calyx and have been sometimes designated as such. Linnzus, in his Syst. Nat. ed. 1, speaks of a calyx 4-phyllus. Their number varies in the different species from 3 to 15 or 16, but is fairly constant in the same species. The two exterior lateral bracts are strongly keeled, like those of the sheath of the leaves, and stouter and. mostly shorter than the others; the third is placed on the upper side, towards the axis of the inflorescence ; the fourth on the lower or dorsal side, opposite the supporting bract, and so forth. The innermost ones not rarely exhibit a transition to the anthers, bearing small or incomplete anther cells on the lower part of their back. In P. resimosa and Canariensis 1 find the involucral bracts articulated in the middle. - A table exhibiting the numerical proportion of involucral bracts in the different species, the male flowers of which I could exam- ine, may not be without value. 3 or 4 involucral bracts I find in P. sylvestris and Pinaster; 3 to 6 in P. densifiora; 4in P. Palosriond, Canariensis, and Greggit; 4to5in P. edulis and Parryana; 4to6in P. Pinea ane PP. hed oleserion’ 6 in P. letophylia, Liste: and contorta; 6 to 7 in P. fos 9 om montana, and Masso 6to8in P. Strobus. excelsa, aoe Sua pirie tuberculata, muricata, aa Banksia 8 to 10 in P. monticola aes oe Chihuahuana, pees te Laricio var. Pyrenaica, and Austriaca, Coulteri, inops; i eda; g to 12 in P. eetgricnnd and wd 10 in P. insi, tt7; 12 to14in P. Khasta, glabra. and australis; 14 to 16 in P. Lambertiana and Cubensis. The ANTHERS consist of two parallel extrorse cells, which open longitudinally on their back; their connective, heretofore often called a bract, spreads out into a transverse semi-orbicular or almost orbicular, entire or denticulate (in most species of Pin- » 169] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 9 aster) or lacerate (P. Laméertiana) crest, or it terminates in a knob or a few teeth (in most S¢rodz and a few Pinasters such as P. Balfouriana and sylvestris). The POLLEN has the well known bilobed form, consisting of an elliptic central portion, which emits the pollen tubes, and two lateral sacs which are said to contain air. The longer diameter measures 0.025 to 0.045 lines, mostly between 0.030 and 0.040 lines, while the pollen grains of Aédzes and Picea are much larger and in many instances twice as large, viz. 0.045 to 0.070 lines long. Thus by the pollen alone Pénus can generally be distinguished from those allied genera. The different species of pines are pretty constant in the size of their pollen. Without going into minute detail, I will only state that I find pollen grains of 0.025-0.030 lines in P. edulis and P. Banksiana ; 0.030-0.035 ‘* in P. Balfouriana, peer ees grrr fe resinosa, Chihuahuana, Laricto, inops, contor 0.035-0.040 “ in P. Strobus, excelsa, Pinea, rigida, T sion mitts; 0.040-0.045 ‘* in P. Lambertiana, flexilis, cue terumea, Pinaster ponderosa, Sabiniana, Elliot The property of the pine-pollen to float for a long time in the air, and to be carried by storms to very distant localities, is well known. I have found in streets of St. Louis after a rainstorm from the south, in March when no pines north of Louisiana were in bloom, pine-pollen which must have come from the forests of P. australis on Red river, a distance of about 6} degrees of lati- tude or 400 miles in a direct line. The FEMALE FLOWER consists, as in all Adée¢inea@, of a car- pellary scale in the axil of a smaller, concealed, bract, bearing two pendulous ovules on the lower part of the upper side. A number of such scales in the axils of their supporting bracts, and spirally arranged, form the female ament. The question of the nature of the scales, and of the ovules they bear, is not to be discussed here, but it may be stated that the best lights force the view on us that the carpellary scale consists morphologically of two leaf- organs, lateral to an undeveloped axis and united at their poste- rior edges (those turned towards the axis of the ament), and thus bearing their naked ovules on their morphologically outer but now reversed and apparently upper side. Io TRANS. ST. LOUIS ACAD. SCIENCE. [1 70 The carpellary scales, which in the flower as well as in the fruit we call, in short, scales, are either rounded, obtuse, and appressed (in Strodus, etc.), or they have a short (P. resinosa, sylvestris, etc.) or a longer (P. ponderosa, Twda) or an elon- gated subulate, often squarrose, point (P. contorta, inops, pungens). The aments are globose, oval or elongated, subsessile or pedun- cled, single or several together, always erect, each borne in the axil of a bract, its base invested by sterile bracts which gradually or suddenly give place to the carpel-bearing bracts, just as the involucral scales of the male flowers give place to stamens. They make their appearance on the upper part of the year’s shoot, often just below the terminal bud, when we call them sudéer- minal; or they become /ateral, when the axis elongates beyond them, and sometimes more aments form above them in the same season. The axis above the aments continues covered with leaf- bundles in some, while in others it is naked for some distance, or rather destitute of leaves, bearing only bracts; a second stage of aments or the terminal bud is always preceded by a number of leaf-bundles. The position of the female ament, whether subterminal or late- ral, seems to be connected with an essential difference in the spe- cies of pines, secondary in importance only to the leaf structure as described above, and both of these together will enable us to ar- range the species in something like a natural order. It ought to be understood, however, that the relative position of the ament on the axis is not absolute and that variations do éccur. Species with or- dinarily subterminal aments may in young and vigorous shoots sometimes bear lateral aments; this occurs, though very rarely, in P. ponderosa and australis, and perhaps in others, but I have never seen it in any of the S¢rodus section, nor in P. sylvestris, restnosa, Laricio or its allies. More frequently subterminal aments are found in species which normally bear lateral ones, probably when with the formation of the aments the vigor of the axial growth has been exhausted; thus sometimes a second stage of aments is.subterminal, while the first is of course lateral; or subterminal and lateral ones are occasionally found on differ- ent branches of the same tree; or, very rarely, a tree bears al- most entirely subterminal aments. This last case I have seen in 171] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. II the Californian P. muricata and in the Mediterranean P. Pyre- maica. ‘This character has to be studied intelligently among the native trees in their homes. So long as only a few herbarium specimens can be consulted it must remain doubtful, and errors may creep in, especially as collectors have heretofore paid so lit- tle attention to the necessity of obtaining instructive specimens, which, however, are easily procured in any season of the year, provided the tree bears at all; for always either flowers or young cones, or in spring both together, can be obtained. The compound Fruir resulting from these aments, known as the cone or strodile, matures at the end of the second, or in a single species, P. Finea, of the third season; during the first twelve months it does not enlarge much; in most species it re- tains its erect position during that period, but in a few it becomes reversed soon after flowering and before the leaves are developed (P. sylvestris and Eliiottii) ; in the allies of P. Strobus the slender peduncle bends downwards in the second summer apparently by the weight of the swelling cone; but in the majority of the species the cones in that period assume a horizontal or somewhat de- clined, rarely a strictly recurved, position. Only in P. Banksiana it is as often curved upwards as horizontal. We continue to speak of subterminal and of lateral cones in regard to that part of the axis which bore the flowers, though the branch elongates in the next year, and the maturing cone, strictly speaking, thus always becomes lateral. The cones are, as the name might indicate, conical, from sub- globose to oval or subcylindrical, mostly more or less symmetri- cal, often slightly oblique, and in some Californian and Mexican species (P. insignis, tuberculata, muricata, patula) 30 much so, that the scales on the inner and the outer side become very unequal ; in the first named spedies especially we find the scales on the outer, convex, side much larger and tumid; on the inner, more flat, side smaller and depressed, but singularly enough more fer- tile than the big outer ones. The color of the cones is from gray to light leather-brown, reddish, or deep brown, with a dull ora glossy or almost varnished surface. They vary in length from 13 or 2 to 12 or even, in 2. Laméertiana, to 18 inches. The phyllotactic arrangement of the scales is quite interesting, 12 TRANS. ST. LOUIS ACAD. SCIENCE. [172 but not of much diagnostic importance; nevertheless it will be necessary in the description of the different species to mention it, and also to state the number of the more prominent secondary spirals, two of which, inclining in opposite directions, are always the most conspicuous. e long cones of P. Strobus, excelsa and S: donemired and the short ones of P. edulis and monophylla, show the z's order of scales, and the 3 and 5 spirals are the most prominent ones. P. Lambertiana and Sabiniana have the 3+ arrangement with the 8 and 13 or the 13 and 21 spirals radat’ Aehapietions. The intermediate orders of $3 and 3} are the most common ones; abnormal orders are extremely rare. The cone scales furnish us the most valuable characters for the classification of the species. Their exposed part, not covered by adjoining scales and more or less thickened, has been called the apophysis ; it is rather depressed and terminates in a blunt point in the section S/rodus ; in FPinaster it bears its point on the usually more thickened back, the wméo, mostly armed with a prickle, weak or strong, early deciduous (in 2. Balfouriana, insignis, Bank- siana) or stout and persistent (in P. rigida, Teda, inops, pungens ) ; in some species (P. Sabiniana, Coultert) it becomes a thick,-long, and often curved or twisted spur. The bracts which support the scales remain concealed, but become greatly enlarged and mostly thickened and corky, and help to form lodges for the — which are enclosed between them and the scales. ‘The cones generally open their scales soon after maturity, drop their seeds, and fall off soon afterwards ; in most cases they sepa- rate at the insertion of the peduncle, but in a few instances (P. ehadge P. australis) the peduncle and the lowest part of the xis toge with a number of scales remain on the branch. In some cone (P. Sabiniana, Coulter?) the open cones persist for several years on the tree, and in others they remain almost indefi- nitely, so that they are apt at last to be partially enclosed in later layers of wood. Such are P. Banksiana, inops, pungens, insignis, muricata, rigida, and some Mexican species. ' Most specimens of Pinus contorta retain their cones in this manner, while those of the higher sierras of California are early decidu- ous, proving that this character is not of great specific impor- tance. The persistence of the cones may be connected with the 173] ENGELMANN— REVISION OF THE GENUS PINUS, ETC. 13 peculiarity of some species to retain their seeds in temporarily or permanently closed cones, when they are called serotinous. Such are southern forms of P. rigida (var. serotina) and P. inops (var. clausa), rarely P. Teda ; in some Californian (P. insignis, tuberculata,muricata) and Mexican species (P. patula, Teocote and Gregg?) this is still more conspicuous. The seeds of such serotinous cones seem to retain their germinating power for many years longer than loose pine seeds, which are known soon to lose their vitality.* - The sEEDs are obovate, or often more or less Sail ccite trian- gular, rarely (in P. Sedénéana and Gerardiana) nearly cylin- drical, generally somewhat compressed, 2 to 12 lines in length, smooth or often on the lower surface ridged or slightly tubercu- lated, always destitute of balsam vesicles, pale gray or yellow- ish, or spotted, or brown, and often black. A wing is always present, and is generally several times longer than the seed; in some large-seeded species (P. flexilis, Cembra, edulis and the other nut-pines, and Pézea) it is reduced to a narrow rim, which is apt to remain attached to the scale when the seed is liberated ; in P. parviflora, Bungeana, Gerardiana, Torreyana, and Sabiniana, itis more conspicuous, but shorter than the seed itself; in P. Coulterd it is about as long as the seed, and in P. Lambertiana longer. The size of the seed and the proportion of the wing to it has been considered to furnish valuable sec- tional characters, but it proves to be only of specific importance. The wing is always more or less oblique and widest in some species upwards, in others near the base. The base of the wing forms a rim which surrounds the seed, leaving its under side free and with its edge covering part of, or rarely the greater part (P. £lliottiz) or the entire upper side (only seen in P. Bankstana). Generally the wing and its rim is completely separable from the mature seed, but in a few species (P. Strodus and allies) it ad- heres to it closely, and is at last broken off irregularly. The coTyLEpons, 4 or 5 to 15 or 18 in number, are mostly several times shorter than the caulicle, usually not longer than its eeds from closed cones of P. contorta, two to eight years old when I collected them in pa a and then kept four Spee in a hot garret, germinated freely with Prof. Sargent of the Arnold Arboretum, Mas / 14 TRANS. ST. LOUIS ACAD. SCIENCE. [174 diameter, and rarely as long or a little longer than it; this seems to be the case especially where the leaves also are of unusual length, e.g. in P. australis. It is easy enough and very satisfactory to ascertain the number of cotyledons where a large quantity of seedling pines is at one’s disposal. With me this was unfortunately not the case; hence I had to examine the seeds themselves, quite a laborious process, rarely extending over more than six or eight specimens, and often less. In examining greater numbers more variation will probably be discovered. As it is, the different species show a tolerable constancy in the number of their cotyledons. I give here only the result of my own observations, leaving out those found in the oks. I have observe about 5 cotyledons in P. Balfouriana, montana (3-6), Laricio, rigida, inops (4-6), muricata (4-5), glabra (5-6), Banksiana (4-5); about 6 cotyledons in P. Balfour. var. aristata (6-8), resinosa (6-7), sylvestris (6-8), insignis (5-8), tuberculata (5-8), Teda (5-8), pungens (7), Pinaster (5-8), mztis (4-7); about 8 cotyledons in P. S¢robus (7-11), monticola (6-9), parvifiora (8-10), flexilis (8-9), monophylla (7-10), edulis (7-10), Parryana, falepensis (6-9), ponderosa (6-11), Canariensis, australis (7-10), Eliiottit (6-9) ; about to cotyledons in P. excelsa (8-12), Peuce (9-10), Cembra (9-12), cembroides (9-12), Bungeana (11); about 12-15 cotyledons in P. Ayacahurte (12-14), Lambertiana (12-15), Pinea (10-14), Torreyana (13-14), Sabiniana (12-18), Coulter? (10-14). In germination the seed-shell is raised like a hood on the tip of the cotyledons, mostly after the wing has come off, but some- times the wing is raised high above the plantlet (P. australis). The axis generally soon elongates, bearing the primary leaves, but the species just mentioned behaves peculiarly in this period, almost as do many moncotoyledonous trees, For six or eight years it grows not in length but only in thickness, and bears in the axils of the short primary leaves numerous tufts of long and slender secondary leaves, which give the plantlet the appearance of a coarse grass or a rush; only after it has acquired sufficient vigor the thick axis rapidly shoots up. 175] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 15 I now propose an arrangement of the species of Pinus based upon the more essential characters above analyzed, and, though I by no means claim it to be a faultless one, I expect that it will deserve the character of a natural one as much as any that can be devised. I find with Endlicher the most valuable character in the fruit scale, or rather, to speak more correctly, I find that the form of the fruit scale in this genus corresponds with a series of other characters which constitute two very natural sections of tlge genus. My section S¢rodus in a wider sense includes his Strobus and Cembra, and my Pinaster, also enlarged, comprises all his other sections, viz. Pseudo-strobus, Teda, Pinaster,and Pinea. The subsections are distinguished by the position of the ducts within the leaf, whether peripheral, parenchymatous, or internal. Subordinate to this character is the subterminal or lateral position of the female ament and the cone. Only after this may the num- ber of leaves ina sheath be taken into consideration, and perhaps the presence or absence of strengthening cells around the ducts. It will be found that thus not only natural but to some extent even geographical alliances are best preserved. I enumerate only such species or subspecies (these in brackets) which I have been able to examine myself; the list, however, will be found nearly complete. The nomenclature of Parlatore in DC. Prod. xvi.2 is adopted unless otherwise stated. , Sect. I. STROBUS. Apophysis with a marginal unarmed umbo, — ge nerally thinner; cones subterminal ; leaves in fives, their sheaths loose and deciduous; anthers terminating in a knob, or a few teeth, or ina short incomplete crest; wood softer, lighter, less resinous. § 1. Eustrosr. Ducts peripheral.--Northern or mountain species of the Old and New World. * Wings longer than the seeds; leaves sharply serrulate, denticu- late at tip. + Strengthening cells few, none around ducts. P. Strobus, monticola, excelsa, Peuce, parvifiora,? Bonapartea,® Ayaca- utte. ++ Strengthening cells abundant under the epidermis and surrounding ducts. P. Lambertiana. : ** Wings much shorter than seeds; leaves mostly entire, not denticulate at tip. ‘ » . P. flexilis, (albicaulis,) pygmea. * 16 TRANS. ST. LOUIS ACAD. SCIENCE. [176 2. Cum Ducts parenchymatous; leaves sparingly serrulate, scarcely denticaea at tip.—Europe and principally Asi P. Cembra,* Mandschurica, Koratensts. Sect. Il, PINASTER. Apophysis with a dorsal umbo, mostly armed, generally thicker; leaves 1 to 5 in a bundle, their sheaths usually rsistent; anthers mostly terminating in a semi-orbicular or al- most orbicular crest; wood generally harder, heavier, and more resinous. s A. Ducts inion al. a. Cones subterminal. 3. INTEGRIFOLL®. Leaves smooth-edged, their sheaths ea anthers nice in a knob or a few teeth.— Western Nort ic ic exic Cones short jabalovees: with thick scales, unarmed; seeds large with a minute wing; leaves brotdes. ‘ P. Parryana, cembrotdes, emi monophy * Cones oval or elongated, se armed with a deciduous or persistent prickle or an —. seeds much saints than the wing, leaves in fives. Balfour P. Balfouriana,® (aristata). § 4. SytvestTres. Leaves serrulate, their sheaths persistent; anthers crested or (onlyin P. aaa. merely knobbed.—Europe and Asia, one eee in America. * Leaves in a es; wings much Soar ee seeds.—East India and its islands. Judice P. Khasia," insularis,? ht eee , * Leaves in twos; strengthening cells abundant, hotaiaeed around ducts; fructification biennial, cones and s wings large. nse eae World, one phe in Genie eastern Americ Pi Ndi montana,® risinen densifiora,'° Massoniana,'! ? Mer- kustt.+? *** Leaves in twos; strengthening cells under the epidermis and around ducts; fructification triennial, cones and seeds large, wings rudimentary.—A sia Mediterranean species. P. Pinea. : 4. Cones lateral. § 5. HALEPENsEs.—Old World. * Leaves in threes, their sheaths deciduous; umbo very promi- nent; wings shorter than the large seeds. Gerardiane. Asia. P. Gerardiana,'* Bungeana. * Leaves in twos, their sheaths persistent; cones smoothish; wings much longer than the seeds. Euhalepenses.—Mediter- ranean regions. . 177] _ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 17 P Halepensis,\* Pyrenaica.*® B. Ducts Dela RHA a. Cones subterminal. § 6. sb rampviuer ~Mtosty America, with three Old World species. * Leav n fives, ducts usually free of strengthening cells. Pestdictvudh stant Priger and Mexico to Arizona and California. P. leiophylia,® tenuifolia, te" pig aidak ai Montezume™ ( Hart- wegit), Torreyana,” Arizonic ** Leaves in threes, sometimes in fours or fives, their sheaths persistent; strengthening cells under the epidermis, around ducts, i us ually alto 1 near the oS ee bundles. — Eup America, Mexico, and Canary Islande. r. pene ponderosa? (Feffreyi), Canariensis.?* aves in threes, their sheaths deciduous.—Mexico and hsidol ee Oe Tie *** TLeaves in twos, generally with some gear’ cll around ducts. Laricfones.—Europe to Asia an ca. P. Laricio?® (Austriaca), Thunbergii,*® contorta®’ cturayana) &. Cones lateral. § 7. Tap#.—Mostly American, only one Old World species. * Leaves in threes, ducts tle without strengthening cells. Eutede.—North America to Mexico. P. Sabiniana,?* Coulteri,?* inven ak paitaict Teda,*' hen eomiiinndig- Greggti, Teocote, pat ** Leaves in twos; cones with very stout prickles. Pungentes. + Ducts without strengthening cells.—North America. P. inops (clausa**), pungens,** muricata.*® ++ Ducts surrounded by strengthening cells. —Southern Eu- rope. P. Pinaster.*" *** Teaves in twos, or in the first often also in threes; cones with weak or deciduous prickles. M/tes. —Eastern North America. P. mitis,>*® glabra,*® Banksiana.*° C. Ducts roedenns - § 8 AusTRALES. Leaves in twos to fives; timber very heavy and resinous. Bis ai North America, West Indies, and one spe- cies in Mexico. / z 18 TRANS, ST. LOUIS ACAD. SCIENCE. [178 * Cones subterminal; leaves in threes to fives. uaustrales. P. oocarpa,** occidentalis, australis.*? * Cones lateral or mostly so; leaves in twos to threes. £/- liottie. P. Elliottit,+® Cubensis,+* Wrightit.+® NOTES. . Peuce, Griseb., may after all be distinct from P. excelsa; it has sigect shorter leaves and sheaths, and, if my specimen can be relied on, a short fruiting peduncle; the structure of the leaf is nearly the same in both. P. Peuce, excelsa, and monticola have a layer of strengthening aw) all , around under the epidermis, interrupted only by the stomata, and not in distinct bundles as in Lambertiana and csiagighin while ty oo and iit gt have scarcely any. all have regularly two sal duc nly. In P. excelsa T have ae pric found a third, upper, ps enh a parenchymatous one. 2. P. parvifiora, Sieb. & Z A branch in Herb. Haenke in the Prague Museum, marked ‘ P. deters Mitta: Bec Nutka Island,” seems to ‘belong to this species, which is distinguished by slender, ot and very ed serrulate leaves, and scarcely any strengthenin ga oe apartea, Roezl. Prof. E. Purkinje, of the Foresters’ Academ of welowaienr: Bohemia, who was probably the first to carefully study the microscopic anatomy of the pine-leaves with a view to the diagnosis of the lis extensive and copiously illustrated work, has directed my attention to the leaf-structure of this form. It deviates from all the other S¢rodz in having numerous, usually 7, ducts, 3 on the back and 2 on each of the upper sides, and having strengthening cells in numerous bundles all around and espe- cially in the angles. I find no stomata on the back. Roezl’s P. Don Pedri has exactly the same structure, but has 3 or 4 series of stomata on the back; both evidently belong together. Though I have not been able to study the flowers and fruit, I do not hesitate to pronounce it distinct from P. Ayacahuite, which, like P. Strobus, has scarcely any strengthening cells, and only 2 dorsal ducts. 2 P. — Lin. The ducts, generally in the middle of the paren- aiiyes metimes nearly approach the epidermis, but I have always found them ieee from it by at least one layer of parenchymatous cells. 5. P. monophylla, Torr. & Frem. The number of ducts is excessively variable; I have found from 3 to 14 in different leaves. The leaves are usually curved, and the upper side, proved to be such by the relative posi- on of the wood and bast cells (see p. igs is always directed towards the branch. Sometimes two-leaved bundles occur. It is an open question she the four species of the subsection Eeebroths may not properly ted into one, as the difference of flowers and fruit is very slight, and bak of the foliage only relative. 179] ENGEILMANN—REVISION OF THE GENUS PINUS, ETC. 19 6. P. Balfouriana, Jeffrey, and P. aristata, Engelm., of the Colorado Rocky Mountains, are identical in leaf-structure and in flowers and must be united, tho he cone of the former is elongated, often even cylindri- cal, the apophyses thicker and peculiarly spongy, and at maturity unarm- ed, ile th oval co h thinner scales and awnlike prickles. In Ut d Nevada a form ith es like the latter, con but with short, stout, recurved prickles. eine enumerates aristata, but does not mention Balfouriana. 7. P. Khasia, Royle, and its two allies, form a very natural little group. Leaves in this species with 2 dorsal ducts; strengthening cells very slight Ree anthers } to 1 line long; crest only 4 line wide, nearly en nsularis, Endl., has similar leaves, ducts often indistinct, sie about 8, outer pair more than half as long as the inner ones; anthers less than 1 line long, crest nearly entire. P. longifolia, Roxb. Ducts few (in Wallich’s rea ). or many (in Hooker and Thompson’s), or none at all er’s; Thuret’s cult.); strengthening cells strongly developed in wham all around leaf; bracts pits strongly fringed, deciduous; male flowers larger than in last, 1-14 nches long, thicker; anthers 1} aes long; crest } line wide, strongly Silo de nticclaes involucre not seen. The thick bundles of strengthen- ing cells and the larger male flowers readily distinguish it from the two others. ontana, Duroi, is so well characterized that it is inconceivable how it could have been taken for a variety of P. sylvestris, unless some hybrid forms, which are said to occur, have created the difficulty. The involucral bracts are always more numerous, usually about 6, the anthers crested, the female aments subsessile, and the young cone erect; in sylves- tris the involucral bracts rarely exceed 3, the crest of the anthers is reduced to a small ridge or a few teeth, the female ament is not longer than its a duncle and becomes recurved soon after flow ring. g. P. resinosa, Ait., is the only American representative this well enaracteried group. The 6 involucral scales are articulated in the middle, e upper part fling off early (p. 168); ducts almost always say 2 on the agate side of the I coir, Sieb. & Zucc. Leaves with numerous ducts, mostly sg eae by strengthening cells, also some of these within the sheath; in a few mar May in Japanese as well as in cultivated specimens, the strengthening cells are almost wanting; male flowers Apa seg 2 to 3 lines long, in an elongated spike; involucrum of 3 or 40 r 6 bra of equal length; anthers only 4 line long, or less, vies a small, slightly denticulate crest. Only in Japan. Sometimes cultivated under the name of the following. : 20 TRANS. ST. LOUIS ACAD. SCIENCE. [180 1. P. Massoniana, Lamb. Parlat., well distinguished from the tree thus named by Siebold & Zuccarini and by Endlicher, which was named by Parlatore P. Thunbergii (see note 26). It is similar to the last, but has longer and more slender leaves and is a native of the warmer climate of Southern China, and is not hardy where densifora and Thunbergii are. Ducts few or many, often with a few strengthening cells, these cells also in the corners, very few under the epidermis, rarely some with the vessels; male flowers slender, cylindric, 6-8 lines long, in a spike, involucrum of 6 or 7 bracts, the outer pair rather shorter than the inner ones. Griffith, ©. 4992, from Afghanistan, in Herb. Kew, with 2 ducts on the upper side of the broader leaves, may belong here, which would extend the geogra- phical area of this species. . P. Merkusii, Jangh. & De Vriese, seems to be closely allied to the last ‘and probably belongs here, or ought perhaps to be considered rather a two-leaved Jvdica. In the poor specimens at my disposal I could not dis- cells similarly disposed. The involucrum consists of 12 bracts, the outer pair not half as long as the inner o 13. P. Gerardiana, Wallich. pen crests semi-orbicular, laciniate- neti seeds nearly 1 inch long. . P. Halepensis, Mill. Cones with longer or shorter peduncles, lateral oa often low down on the axis, generally single, with flat or sometimes pene tumid scales P. Pyrenaica, peg fide Parlatore, P. Brutia, Ten., and with yay synonyms, not to be confounded with that other P. Pyrenxaica which is a form of P. Laricio. This species is so closely allied to the last that it is often considered a variety of it. But the leaves are stouter, the more numerous duc surrounded by strengthening cells, which are very scarce in the AGC of the other; in both e cells are found near th vessels; the male flowers are twice as large; the outer pair of involu- cral bracts is ae equal to the inner ones; the cones are nearly sessile and thicker, generally several — and often lateral and terminal on the same tree (see p. 171); the densely clustered cones in pean s Sasa specimen in the botanic atti n of Naples are the pag of disea 16. haa latone Schiede & Deppe, has often6a ven vies aves; th ducts are very s and often wanting; the media cells, usually - well developed in Saatiee under the epidermis, are, as well as the ducts, absent in Gregg’s No. 821 from Zamora; the sheaths are usually decidu- ous, but plan er) so in Hartweg’s No. 441. \ ; 17. P. filifolia, Lindl. In ods xa cultivated in Kew gardens the sar are sometimes intern . Montezuma, Lam , is, if I understand it correctly, a most vari- atis § species, the largest suit re different forms of which is preserved in the Berlin herbarium ; some forms have longer, others anaes leaves, or stouter *more slender ones, 3, 4or 5 in a bundle; cones long cylindrical or oval or conical; the scales in the typical form are depressed and regularly rhom- - boidal, in other forms they become strongly umbonate. It is quite diffi- 181] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 21 cult, a pose to “ig apid circumscribe the species; for the present I feel obliged to unite with iteven P. Hartwegii and a number of others already included by Sartetoes! ad a closer study on the Mexican mountains will decide whether or not several well characterized species may be hidden among them. All those that I could examine have numerous and strong bundles of strengthening cells under the epidermis and also near the ves- sels, but none around the ducts. 19. P. Torreyana; Parry, has the same structure of the leaves. The ame was published in the Botany of the Mexican Boundary, 1859, and is therefore older than P. lophosperma, Lindl. of 1860. Oo. rizontca, Engelm. in Bot. Wheeler, p. 260, has also this struc- ture and is thus distinguished from P. ponderosa, besides being five-leaved. . P. Engelmanni, Carritre, Conif. p. 356; P. macrophylla, Engelm- in Wislis, Mem. p. 103, note 25, is a tree only known from Wislizenu . single specimen gathered in 1846 on the mountains of Cosiquiriachi, 1 of Chihuahua, where it is said to be abundant. wees name was changed Carriére because it clashed with Lindley’s prior one; this, ho sami, tx is ieee by Parlatore to be a form of Montezuma, tet which I have not been ine. Our plant differs from this species by having its very stout leaves in threes and fours and very rarely in fives, in the strongly developed strengthening cells under the epidermis and also a round the ducts, and inthe form of the cone. Parlatore does not mention it. 22. P. ponderosa, Douglas, a variable and wide-spread species of West- ern North America, sg a forms of which have ‘tks described as distinct. The only one which m haps claim specific recognition is our var. Feffreyt (P. Feffreyt, Mu urr.), habeus rized by its darker more finely cleft bark, glaucous branchlets, paler foliage, and much larger cones, with rather slender sharp recurved prickles and larger seeds; but it seems that inter- notice is var. scofpulorum, of the Rocky Mountains, with shorter and often binate leaves and smaller cones (see Engelm. in Fl. Calif. 2, p. 125). 23. P. Canariensis, Ch. Smith, is perhaps more nearly related to P. Laricio epee to ponderosa. The articulation of the 4 involucral bracts is acurious feature which it has in common with our P. resinosa (see p. 168). 24. P. Chihuahuana, Engelm., first described from the mountains of Chihuahua, but now repeatedly found in wai is well distinguished from all its relatives by its deciduous sheat 25. P. Laricio, Poir. Strengthening peti avonnd. ducts and in bundles all around leaf; the ipicsl form has slender leaves and is tender in culti- ar. Mo leaf, and is more hardy than the species. Var. Austriaca or nigra is per- fectly hardy; it has the stoutest ehank of all the forms, with abundant strengthening cells. A specimen in Herb. Kew, Birmah, Griffi 4993» may belong here, thus a the range of the species far into Asia. 22 TRANS. ST. LOUIS ACAD. SCIENCE. [182 26. P. Thunbergii, Pariat. This is P. Massoniana, Sieb. & Zucc., and of many authors and many gardens, but is easily distinguished by its stout- er, shorter leaves with parenchymatous ducts. It seems peculiar to Japan, though cultivated in apres whence P. Australasiaca, Steud., an origi- nal specimen of which I have been able to examine in Herb. Cosson. Its parenchymatous duct distinguish this species at once from: any other Japa- nese pine and place it near P. Laricio. 27. P. contorta, a. is a little out of place here and evidently be- longs nearer to the next group, but it has the subterminal cones of this. ducts, which character distinguishes it at once from its Old World associates, and so do the subulate points of the female scales. The low- growing narrow-leaved coast form, which is found along the Pacific from Northern California to Alaska is the original P. contorta, Douglas (from cino, California); it is a regular seaside tree, an excellent screen against e Pacific storms and pes salt spray, just as PH alepensis is on the Mediterranean ; its leaves are often entirely destitute of ducts. The broader leaved mountain form is P. Murrayana, Murr., as Jeffrey’s original speci- mens prove, which come from the sierras; P. muricata, with which Parla- tore unites it, is very different and belongs tothe coast region; this broad- leaved form extends to Oregon and to the Rocky Mountains. While the forms of the coast and of the Rocky Mountains have very knobby, oblique, serotinous, and persistent cones (see p. 172), those of the sierras have occa- sionally more regular, less tuberculated, readily opening and deciduous cones, without being otherwise distinguishable (C..S. Sargent). The woo of this species is white and soft, and the tree is therefore often called white pine or spruce-pine. 28. P. Sabiniana, Dougl. and P. Coulteri, Don (macrocarpa, Lindl.) cannot be confounded by those that have been able to compare both grow- brown; the seeds larger, 9-12 lines long, almost cylindrical, with muc shorter wings; those of P. Coulter’ are more slender, of a paler leather color, the seeds shorter, 6-8 lines long, and their wings longer. P. Sadi- niana makes a round-topped tree with spreading branches and looser, more slender and lighter foliage on glaucous branchlets; P. Coulter? is 1 tree with rigid brown-green branches and denser, coarser and darker foliage. The seeds of P. eeee are or have been a most important cass of subsistence for the Indian 29. usignis, Dougl., distinguished by its fresh green foliage and closely and strongly serrulate leaves. Cones generally thick and very ob- de a equally flat. For the synonymy I refer to Flor. Calif. 2, 127, repeating here only that the original P. tuderculata, or is founded on an unusu- 183] ENGEILMANN—REVISION OF THE GENUS PINUS, ETC. 23 ally slender cone of this, and that P. Stnclatrii, Hook. & Arnott, is a fac- titious species compounded of a cone of this and a branch of P. Monte- @. The old and evidently sone described P. Californiana, Lois., is rashid: our species, ra cannot now be identified 30. P. tuberculata, Gordon Pin al es art, at Doni!» a name at first erroneously given to a species le ‘s effrey, is to be retained as now in general use and because Don’s ah ira ident is amere form of znsig- nis. P. Californica, Hartw., is thes It is the smallest pine known as a tree, fruiting often when only 2 te it high and rarely ever exceed- ing 15 or 18 feet. (See Engelm. in Flor. Calif. 2, 128. 1. P. Teda, Lin., and P. rigida, Mill., have sometimes, besides the regular hinted anattend: smaller, accessory internal ducts, thus approach- p- The co The most inland localities ree be the Stone Mountain near Atlanta, Georgia, and Camden in Arkan 32. P. rigida var. serotina, P paler Michx., I cannot distinguish specifically from rigida ; it is more apt to grow on wet places (whence the name Pond- fine) and has longer leaves (occasionally, on strong shoots, in fours), and the cones often do remain closed for several years, pical Dr. Mellichamp sends them; further inland the cones are more elongated, often twice as long as in the northern rigida (Wm. H. Ravenel, Aiken, S.C.) Prof. Sargent observed it on the mucin and East eae coast, but not in West Florida or in Alabama. Felled trees or posts se ground sometimes make sprouts bearing primary leaves. P. patula, Schiede & Deppe. The ead cells of the leaves protiodé so o that the surface appears minutely tuberculate 4 ops, var. clausa, was discovered and ads sy Dr. Chapman at Apalachicola, Florida, and Prof. Sargent finds it quite common o Cedar Keys. It is distinguished from the species by decidedly narrower leaves and by its cones being often serotinous, more in one tree than in ous; young branches green, in the northern form a gaa involucru of to to II, in émops of 8 to 9 bracts; cones larger, mostly s sbsenil, recurved ; in the other, mostly ahi peduncled and patulous; cotyledo fewer, 4 or rarely 5, in the other 5 0 35. PB. pungens, Michx. hae rasety’ in threes and sometimes with accessory internal ducts. The cones persist sometimes 20 years or longer. P. muricata, Don. Male flowers only 4 inch long in a spike of shdut 1 inch in length, similar to those of ¢uberculata and insignis; an- theral crest strongly denticulate, in the others nearly entire. Specimens 24 TRANS. ST. LOUIS ACAD. SCIENCE. [184 have been collected at Tomales Point with subterminal cones, not differ- ent in any other respect. The cones of the southern specimens (from Monterey, etc.) have usually very long, 14g inch, and stout curved spurs, especially on the outer side, and fully deserve their specific name, but oth- rs from farther north (Mendocino, etc.) are more regular, with short and thin, age very sharp, prickles inaster, Ait. The joule flowers form a large oval head ; invo- cally and pina this species is more nearly allied with P. Canar?- ensis, less so with Laricio, but is distinguished from both by the lateral (quite cele Casas female ament. 38. P. mitis, Michx. Wide-spread through the middle and partly the ; species of pine; it is found always on silicious soil; it fur- sists excellent ‘‘hard pine” lumber. The outer pair of the 9-12 involucral bracts is scarcely half as long as the inner ones. 39. P. glabra, Walt. Similar to the last, with slender foliage, smoother bark (in young trees and on the branches the grayish bark is quite smooth) and almost unarmed cones, JPorera gir by Walter 100 years ago, but W. Rav where it grows on the edge of or in swamps, and on the knolls in them, with Magnolia, Fagus, and pnsead! gees on sandy soil, and never in the so called Pine-barrens. He describes the branching of the tree as singu- larly characteristic, the spray u scaniie ts flattened somewhat ae biad of Cedrus. It probably — throu the lower parts of t e leaves are usually 2$ to 3 inches long, not half as thick as they are wide, while in P. mitzs their sate exceeds half their es the external in- volucral bracts are min so. PF. Banksigna, wad , published 1803 in Lambert’s first edition, a with erect or at least patulous cones; the small prickles of a ee SN cones soon disappear, so that the mature ones are unarm The b of the wing entirely covers the outer side of the seed and ates 0 it, just as it does in Picea, and which I have not seen in any other pine to this extent. The cones are often serotinous and persist for a long time. The seeds seem to germinate most readily, just like those of P. Teda, and 20 or 30 feet high, and 10-12 or very rarely 18 inches in diameter. Very 185] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 25 common in Northern Michigan and Wisconsin, it does not seem to extend farther westward than the Saskatchawan, where it is replaced by P. contorta 42. P. australis, Michx. Male flowers 24-3 inches long, the longest of any pine, of rose-purple color; lowest pair of involucral bracts minute. On a very vigorous shoot I have seen the female ament lateral (see p. 170), arare anomaly. In the germinating plantlet, the long wing remaining attached to the seed shell is raised up like a flag by the growing cotyled- onous leave 43. if Biot Engelm. For a full account see below Cubenszs, Griseb. Leaves in threes, only exceptionally in twos, 8-10 a re es long, rarely longer, stout, about § line wide, rigid, strength- ening cells largely developed under the epidermis (so that their bundles sometimes extend from the epidermis to the ducts) and also near the ves- si ; bracts 3-34 lines Tee. strongly fringed, reflexed, rather persistent; e flowers about aie hes long; involucral bracts 13-15, the outer pair aa as long as the inner ones; anther-crests scarcely denticulate; cones 24-3 inches long, bet elias scales depressed; seeds 3$ lines long, faintly ridged; wing nearly twice as long, widest at base, oe to an acutish point. The var. tertrocarpfa, Wright, in Gris. Cat. Cub. 217, is a very curious form but not a variety. It seems that in this case the growth I ha of growth in the biennially-maturing Quercus chrysolepis. It is found in different parts of Cuba, in the maritime districts as well as on the moun- tains, and is probably the same that gives the name to the Isle of Pines. A cone from the Bahama Islands, preserved in the Kew Museum under 45. P. Wrightit, Engelm. n. sp. Leaves in twos, very rarely in threes, slender, 5 to 8 inches long, }¢ line or less wide; sheaths 4 lines “iy with age a little see bracts ar (14 lines ies, very slightly fringed and rather deciduous; cones lateral, peduncled, recurved, oval, is to 24 inches long, Bae, radiately grooved, thickened on the crenulated edge, ophyses retused, umbo immersed, prickles short; seeds 24 lines long, v The cone-scales of both species are pis 3 in the 2} igen the 8 and 13 spirals being the most prominent. 26 TRANS. ST. LOUIS ACAD. SCIENCE. [186 Pinus Exuiorti, Engelm. n. sp. A large tree, 50-100, rarely to 110 feet high, 2-4 feet in diame- ter, with (7-15 lines thick) laminated, reddish-brown bark ; leaves in twos and threes, in the axils of lanceolate, long-fringed, some- what persistent bracts, 7 to 12 (mostly about 9) inches long, ¢ to nearly 1 line wide, rigid, closely serrulate, acutish ; sheaths at first about 3 inch long, later withering to one-half that length; resin ducts internal (adjacent to the sheath of the vascular bundles). Male flowers from the axils of similar, persistent bracts, cylindri- cal, elongated (13 to 2 inches long), in a short head (not more than t inch long), each one surrounded by an involucre, 4 lines in length, of about 12 bracts, the exterior pair strongly keeled, half the length of the inner ones; anthers with semicircular, denti- culate, rose-purple crests; pollen grains 0.037 to 0.045, on an average 0.04 lines in the longer diameter. Female aments pe- duncled, mostly 2 to 4, or rarely to 6 together, oval, purplish, at first erect, but soon assuming a horizontal and (a month later, and before the leaves are well developed) a recurved position, the _axis meanwhile elongating and in vigorous trees not rarely form- ing a second tier of aments several inches above the first ones ; the bracts above the aments bear the usual leaf-bundles, so that no naked space is left; carpellary scales broad, rounded, more or less abruptly cuspidate, their bracts half their length, transverse, retusé. Cones peduncled, recurved, oval to cylindrico-conical, 3 to 63, usually 4 to 5 inches long, 17 to 24 inches in diameter (when closed), of a rich brown color and almost. glossy ; bracts thickened, retuse, or emarginate; scales in 33 order, the 5 and 8 spirals most conspicuous; larger scales 2 inches long and 7 lines wide; apophyses marked with grooves, radiating from the slightly prominent umbo, transversely divided by a sharp ridge, armed with a short stout or rarely a slender sharp prickle. Seeds triangular, 24 to 3} lines long, dark, slightly ridged, and rough on the under side ; wing 4 or 5 times as long (13 to 16 lines long), _ somewhat oblique, obtuse, with nearly parallel sides, or usually somewhat broader below, its base covering the greater part of the outer or upper surface of the seed; cotyledons 6 to 9, usually 8.—P. Teda, var. heterophylla, Elliott, Sketch 2, p. 636. . Common, in light sandy damp soil, among the sandhills near the seabeach and along the marshes near the mouths of rivers; 187] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 27 also found in damp clayey pine lands and with P. rigitda, var. serotina, in pine-barren ponds, rarely exclusively covering small tracts, and only as a second growth in old fields. From South Carolina, on the sea islands near Charleston, to Georgia along the coast, and sparingly as far as 15 to 20 miles inland, but never very far from the influence of salt water, Dr. Mellz- champ ; to Georgia, Elliott ; and Florida, Canby, Curtiss ; form- ing forests on the St. John’s river, where it is often called Slash- pine, and is not cut for timber, Sargent; “ the most common pine in South Florida, the ‘short-straw pine’ of the wood-cutters, taller, more slender, and with harder wood than the ‘ long-straw. pine,’ P. australis, which is the principal forest tree of Eastern, Middle and Northern Florida,” Dr. A. P. Garber; extending westward to Alabama, ‘‘a common tree along the bay of Mobile,” Mohr, Sargent. Prof. Sargent observes that while the long-leaf pine rapidly disappears under the axe, Elliott’s pine becomes more and more common, the young second growth forests in Flo- rida almost entirely consisting of this species and of Tada. This is the earliest flowering pine of those regions, from 2 to 4 weeks in advance of any other pine, showing its rose-purple male flower-buds already in December, and in January or February, according to latitude and season, shedding its abundant pollen, which, wafted by the winds, is apt to cover roads and streets, and especially sheets and pools of water, far and wide, with its sulphur-looking powder. P. austra/is, also with rose-purple flow- ers, comes several weeks later, and then the others, P. Teda, next P. glabra and mitis, and lastly P. rigida var. serolina with greenish-yellow flowers. Our species bears abundantly: every year (at least in South Carolina), different from P. australis, which, like many other pines, is fruitful only every other season. The cones also mature and drop off earlier than those of the associated pines, and shed an abundance of seeds, which readily germinate about November, and develope their young stems in spring. -This tree Prof. Sargent considers by far the handsomest of all the southern pines, readily distinguished from those, with which it is associated, by its heavier, denser heads, darker foliage, and larger and heavier branches. The red-brown bark is very characteristic of this species; it is 28 TRANS. ST. LOUIS ACAD. SCIENCE. [188 regularly laminated, and the outer laminz exfoliate in rather rigid and brittle, often very thin, plates of a purplish color when fresh, whence the local name of Blue-pine ; the bark of P. australis is somewhat similar, but the plates are much larger. The timber is excellent, heavy, very tough, and more resinous even than that of P. australis, which it resembles; of a striped yellowish-brown and paler resin color (the inner portion of each ring, formed earlier in the season, being paler; the outer part, of later growth, brown); fibre coarser than in australis and more tenacious. It grows rapidly, at least in its youth: a tree of 22 inches diameter, and about 140 years old, had in the semi-diameter 8} inches of heartwood with 74 annual rings, and white sapwood 2} inches thick with 60 to 70 rings; another of over 3 feet diameter, 109 feet high, and 200 years old, had a radius of 17 inches heartwood with about 140 rings, and 3 inches of sapwood of 60 rings. Thus the average rings of the heartwood were over 1} lines and those of the sapwood (because of later growth) about 3 line wide. The leaves of young trees are more frequently in twos, in older ones as often in threes; those of trees from swampy soil are apt to be shorter than others ; the structure in all of them is the same. Our species is closely allied to P. Cubensis (see p- 185), and _ further study of the latter may possibly prove them to be nothing but geographical varieties. Meanwhile the constantly three-leaved foliage, the larger number of involucral bracts of the smaller male flowers, the smaller cones with smaller, shorter-winged seeds, distinguish P. Cudensis from our species. Of the bark, of the timber, or of the behaviour of the young cones in this species we know nothing. P. Eliiottii was imperfectly known to Elliott, and was consid- ered by hima form of P. Zieda. Later botanists ignored it till Dr. _ J. H. Mellichamp of Bluffton, S. Car., rediscovered it about ten years ago and directed my attention to it. Without his diligent investigations, ample information, and copious specimens, this paper could not have been written. At the same time I grate- fully acknowledge my obligations to many botanical friends in this country and in Europe, and especially to the directors of the 189] ENGELMANN—REVISION OF THE GENUS PINUS, ETC. 29 botanical gardens and the curators or possessors of the great her- baria, who most liberally furnished me with the material to carry on my investigations of the Pines and of the Conifers in general. I am particularly indebted to Messrs. Bolander, Brewer, Parry and Lemmon for their contributions of the Californian and Rocky Mountain Conifers, and to Messrs. Canby, Gilman, Ravenel and Mellichamp for those of the northern and eastern American ines. EXPLANATION OF FIGURES. Pl.I. fig.1. A branch, gathered in Sear showing two mature zavest of the pre- ceding years, floweri and tk g f the spring. One-half nat, size T + d th Nat. Le fig. 2. e, fig. re Their close serratures, est OF ao re: figs. 4-7. Sections of | 30 times, 4 and 5 of binate, leaves; the ducts are seen clo sely appressed to the sheath which encloses r es; these bundles are, as in most pine either paren, or closely approximate most united wide or small, few or many, in these specimens, varying from 4 to 9. PL Il. fig. $8. Male inorescece, capitate, with the elongated flowers in the axils of ringed bra Pi. fig. 9. One nora reste magn. 3 times, exhibiting the calycoid involucrum PL 7 fig. 10. A bract, an fig. 11. The Siivetinevsinn from the dorsal, and fig. 12 from t tral xhibit- sac e lowest lateral pair of bracts and the pe Nene inner ro upper agn, 4 times. : Fist. fig. 3 Diagram m of the involucrum with the supporting bract; the 2 outer scales re strongly, the 4 eas ones Pies: keeled. fig. 14 aia 10 te Bftcte an! nties fro: abov ve and the age, WE er the transverse nei Magn. 1otimes. hl th Pipe, § th } 4 } ef elongating Pic lie: figs ts: fig, 16. The same, a little more edvanced. fig. 17. An ament magnified twice fig. 18. A — flower (carpel sc iat e) in the axil of the broad retuse ool upper, cuspidate, half being visible, and below the bifid t of the ‘ata Pea cae vules—in February, Magn. 10 gas. fig. 19. Female ae Reet hehe ‘wo months later, recurved. Pl. I. fig. 20. One of th Pl. III. figs. 21-24. hed cones sof re sizes at = showing their variability- fig. 25. eadin. Base of an open cone with spr figs. 26, 27,28. Scales aay cone. Beta 26, a al view, showing the bract and the eager hie fig. 27, v rom above, soe ay ee reg ithe hooey by seed and a surface pies dere a ing h ; fig- fig. a9. Post from the lower, and fig. 30 from the 1 upper sd with pier shaped wings; t in fig. 29 the nrg under surface of the seed i is seen; fig. Album and embryo et different shapes. Saves 4 thes: mber. figs. 3 Fi. as fig. oe ee a nie ead in N fig. 34. A seedling che the flowing spring, ae ay 8 sing ease the pri- y Mr. us Roetter, late of S who eer ‘ai himself he favorably persis lees — went year $ ago, by the