



. n 



U the same plant in more advanced state of growth, four luch 

 columoi on one side and one on the other hare acquired considerable 

 lisa, and rach radiate* toward* the cironmferenee of the iitem. As in 

 the Calytmiii**, so in these and all the other instance* of the some 

 kind, which these cnU represent (see K, K, and S), the woody columns 

 of the bark are destitute of pith. 



lYrhaps what we have called the separation of cones of wood at 

 N, M, O, P, Q, Jl, by thick layer* of cellular tissue, are rather to Iw 

 considered as other instances of wood formed in bark, but in a regular 

 and uniform manner. We are however uncertain how this may be, 

 and prefer allowing the statement to stand in its present form until 

 some one shall have examined such plants in their native forest* at 

 Singapore. 



In addition to such anomalous kinds of structure a* those now 

 described, Ezogens, like Endogens, contain species, the organisation 

 of whose stem is so imperfect a* to be reducible within no certain 

 rale*. Not to speak of CaUitridic, CcratopliyUum, or ifyriopkytlum, 

 wherein vessel* are scarcely developed, and the woody matter merely 

 forms a simple central axis of growth, we have in this class an exact' 

 parallel with Lemna among Endogens ; some Podostemaceous plants 

 have their leaves and stem completely fused together so as to resemble 

 a Ma reliant ia or an Alya. Such plants are to' be regarded rather as 

 instances of imperfect organisation than as deviations front a typical 

 form ; and it is by no means a violent supposition -to conclude that if 

 their organisation were more complete it would then become such as 

 is characteristic of the clora to which they belong. 



Si hleiden, in his ' Principles of Scientific Botany,' has treated tlie 

 structure of the stems of Exogeus in his usual lucid manner, and we 

 subjoin an analysis of his views upon the subject. He divides the 

 Dicotyledonous axial organs into two groups, the stalks and stems. Of 

 the stalks he says they frequently exhibit no essential differences from 

 those appertaining to Monocotyledonous plants, since the distinction 

 of the unlimited or indefinite vascular bundles is often imperceptible 

 in the growth of one year. But the vascular bundles generally close 

 in the first year into a simple circle, and the external parts in several 

 circles to form a ring, so that the pareuchyinatous masses separating 

 the individual bundles are compressed together into medullary rays. 

 In most cases the vascular bundles run from below upward in straight 

 parallel lines. They form a loop where the leaf begins, the edges of 

 which furnish vascular bundles for the leaf and the axillary bud, and 

 the pith of the bud is thus brought in connection with that of the 

 stem by means of their opening, as in the case of Troptrolum. The 

 vascular bundles supplying the leaves and buds generally separate 

 from this loop exactly at the point where they enter the leaf. Some- 

 times however they first pass through a longer portion of the paren- 

 chyma of the pith or the bark (as in the A maranthacea and C'/tfno- 

 pwlaeetr). In perfect nodes loops of vascular bundles are seldom 

 found passing across the stem ; in general the parenchyma merely 

 appears to be tougher and closer at these points. We are here, on the 

 whole, very destitute of accurate investigations, more especially 

 with regard to the first year's stalk with undeveloped in tommies. 



Steins differ in structure according as the interuodes are developed 

 or undeveloped. Those with developed interuodes are divided into 

 several sections : 



.1. With a Simple Ring of Vascular Bundles. Hero the vascular 

 bundles very seldom run parallel, but generally in serpentine lines, 

 alternately approximating and retreating from each other ; the meshes 

 thus formed are filled by the medullary rays. Where liber-bundles 

 lie in front of the vascular bundles they follow the same course. 

 Large and small medullary rays and annual rings are formed in the 

 manner indicated. Wherever there i* a leaf, one large or several 

 smaller loops are formed, from whose circumference the vascular 

 luiiidli's are given off fur the leaf and axillary bud, while the openings 

 furnish parenchyma for tin- tninti<in of the bud. The vascular 

 bundles of every newly developed inteniodn stand in iinmrdi.it>> 

 connection with, and are immediate prolongations of, that portion of 

 the vascular bundle of the preceding Intel-node still capable of develop- 

 ment, and thus the cambium of the vascular bundles forms a 

 < -.I'tiiiiMin ii--t through the stem and branrln-x of the whole plant. 

 I >ut ing the development* of the vascular bundles of the stem, and 

 those connected with them, and llonging to an axillary bud that 

 grow* into a branch, the base of this branch become* more and more 

 covered with newly formed wood. We thus see the same condition 

 established a* in the Monocotyledons : an under lateral branch crosses 

 nil the layers of wood pawing to the upper part*. The difference is 

 merely, that in the Dicotyledons they are portions of the continuous 

 mass of the progressively developing vascular bundle*; whil in 

 the case of Monocotyledon* they are discrete part*, new vascular 

 bnndli '. 



B. With Several Concentric Kings of Vascular Bundle*. This 

 condition appear* to be only met with in Piper (?) and J'irmia; and, 

 perhapi>, in a few of the Cramlaeta, a* in the C'rauiUa. The separata 

 vascular bundle* continue to grow, and finally close into a firm 

 woody mau ; each however retain* it* own cambium, and likewise a 

 mall portion of parenchyma, not perfectly dislodged : such, at any 

 rate, i* certainly the case in Pitmia. 



C. Stems of (.'limbing Plant*. The item* of many tropical climU-r.-. 

 {Liaucs, Llanos) exhibit a peculiar structure, which has long been 



misunderstood. Even in our own indigenous plant* we meet with 

 some indication* of it. In the first year, most of them exhibit 

 nothing striking, if we do not regard the generally square *talk a* 

 mich ; and we find that they have a simple ring of vascular bundles, 

 which close* towards the end of the first period of vegetation into an 

 ordinary wood cylinder. In the following year* however the pecu- 

 liarities are more and more strikingly manifested, consisting in tin- 

 wood not being uniformly developed towards the exterior throughout 

 its whole circumference, but ceasing to grow at definite parts, often 

 regularly, and as frequently in a fantastically irregular manner, 

 allowing the substance of the bark to replace it In this manner 

 stems are produced, which, in a transverse section, exhibit the most 

 varied distribution of the wood. We meet with the first indie. 

 of this peculiarity in- our indigenous species of Cltmatu forming stems ; 

 iu the strikingly broad and regularly arranged Urge medullary rays ; 

 and in the six narrower portions of wood which are not nearly so 

 fully developed towards the exterior as the six broader ones. To 

 these we may add the Biynoniaeete. After the wood has continued 

 for some time to be regularly developed, it ceases growing in four 

 different places, so that the bark is no longer pushed outward ; and 

 on the further development of the wood in the remaining places, the 

 bark forms, in the transverse section, four septa of variable thickness 

 between the four portions of wood. 



In some species these cortical masses become a definite degree 

 broader iu each succeeding annual ring, so that a sharply marked step 

 is formed on each side ; in another species all that is formed are four 

 very thin flat platen, wholly separated (in consequence of drying) 

 from the wood. Still more striking is the cross-section of many 

 climbers of the family of Hapintlnceir. A hasty glance would lead us 

 to imagine that we hod here a cylinder of wood surrounded with Kirk 

 in which other stems or branches with their bark had become blended 

 in their growth. A minute observation however refutes this view at 

 once from the absence of pith in the exterior woody masses. 



Finally, the most astonishing phenomena are seen in the families of 

 the Arwlolocltiacea", Atctepiadacea", Malpighiaccrr, Baukinift, in whieh, 

 in the transverse section, the woody moss appears divided in the 

 strangest ways by cortical substance, separated into various portions, 

 and often elegantly lobed. 



The great diameter of the porous tubes may apparently be regarded 

 as a general peculiarity in the ligneous structure of all climbing 

 plants. These have also strikingly large pores which (as I have never 

 yet seen in vessels) form even ramified canals, as is seen particularly 

 well in Bauliinia. 



Of Stems with Undeveloped Intel-nodes, Schleiden says they have 

 scarcely been investigated at all in the Dicotyledons. Most of them 

 remain very short, since they die below as they increase upward. 

 They belong principally to the subterraneous stems and rhizome*. 

 The leafless uphorliace<e, Carica, Thropknwta, Nympk<ta, and 

 .\iij,!i/ii; as well as many Qactaottt, afford excellent material. The 

 most important researches in reference to this point are those of 

 Schleiden, into the stems of Cactacea, especially Mammitlaria, h'chino- 

 cactiu, Melocactui. The vascular bundles at first make an arc of 

 considerable curvature ; by the gradual development of the pith the 

 curvature becomes almost effaced, and it only remains in the upper 

 part, where the vascular bundles pass off to the leaves. The first 

 succeeding layer developed in the vascular bundle is applied over and 

 up lieyond this, dividing at the point where the primary vascular 

 bundle goes off to the base of the leaf, and uniting again above to 

 pass up to the base of a leaf situated higher up. The next layer of 

 structure forms in the same way, by splitting and reuniting, two 

 meshes, one for the primary vascular bundle. ;.n.l one for the portion 

 of the first layer of increase, running to the iipp<-r 1,-al", then 

 this it runs up to the base of another leaf. This structure is continued 

 up throughout the \vliol, > nlem, which thus possesses a form of wood 

 exhibiting jierfeetly t<-:;iilur meshes, or areoho, which appear to be 

 formed by an alternating sni>erposition of vascular bundle-, and each 

 gives passage to a bundle coming from the innermost part of the wood, 

 (if coin-no there is here a perfect crossing of the vascular bundles 

 going to the lower leaves by all the subsequently formed portions of 

 vascular structure, and by a little care we may make jnvpai 

 not very unlike the ntnuturo of a Monocotyledonous stem with 

 undeveloped internodes. The whole structure bears great similarity 

 to that of the arborescent Ferns, allowing for the different nature 

 of the vascular bundles and the difference of dimension. 



Many interesting varieties in the structure of the wood occur here 

 also ; aud the wood of the Mammillanic and Miiocaeti, composed 

 entirely of peculiar spiral-fibrous cells, is particular worthy of notice. 



The stems of the Itliizanthacca (Blume) appear to be altogether 

 aberrant and irregular in their structure. As a Kent-nil rule, we may 

 f.iy that in the Monocotyledon; the vascular bundles ore simplest in 

 their lower part, often, for instance in the Palmn, composed at that 

 part solely of elongated parenchyma (liber); in the middle becoming 

 more complicated from within outward, exhibiting almost all the 

 forms corresponding to the varied expansion of the cell ; above they 

 become simpler again, particularly when- they pass off into a leaf or 

 branch, and consist frequently merely of such elements as correspond 

 to considerable expansion in the longitudinal direction after the 

 appearance of layers of thickening. In the Dicotyledons the vascular 



