Title: Contributions from tlie Botanical Laboratory and the Morris Arboretum of the University of Pennsylvania, vol.11 Place of Publication: Philadelphia Copyright Date: 1935 IVIaster Negative Storage Number: MNS# PSt SNPaAg243.3 <2252646> * *OCLC* Form.serial 2 Input: HHS Edit:FMD 008 ENT: 990118 TYP: d DT1: 1934 DT2: 19uu FRE: z LAN: eng 035 (OCoLC)7844146 037 PSt SNPaAg243.3-244.2 $bPreservation Office, The Pennsylvania State University, Pattee Library, University Parl<, PA 1 6802-1 805 090 10 580.8 $bP3c $cax $s+U1 1 XI 933/34-U 1 4X1 938 090 20 Microfilm D344 reel 243.3-244.2 $cmc+(service copy, print master, archival master) 245 00 Contributions from the Botanical Laboratory and the Morris Arboretum of the University of Pennsylvania 260 Philadelphia $b[s.n., 300 V. $bill. $c24 cm. 310 Irregular 362 0 Vol. 11 (1933/34)- 500 Some are reprints from various journals. 500 Some vols, issued in parts. 533 Microfilm $mv. 1 1 -v. 1 4 $bUniversity Park, Pa. : $cPennsylvania State University $d1999. $e2 microfilm reels ; 35 mm. $f(USAIN state and local literature preservation project. Pennsylvania) $f(Pennsylvania agricultural literature on microfilm). 650 0 Botany. 650 0 Botany $xBibliography. 710 2 University of Pennsylvania. $bBotanical Laboratory. 710 2 University of Pennsylvania. $bMorris Arboretum. 780 00 $tContributions from the Botanical Laboratory of the University of Pennsylvania 830 0 USAIN state and local literature preservation project. $pPennsylvania. 830 0 Pennsylvania agricultural literature on microfilm. L CONTRIBUTIONS FROM THE N i Botanical Laboratory and The Morris Arboretum OF THE UNIVERSITY OF PENNSYLVANIA VOLUME XI 1933-1934 4 PHILADELPHIA 19 3 5 « 1 » * I * V^ .« > The injection of certain salts into the protoplasm and vacuoles &c. 421 The microinjection technique has not, as yet, been apphed to plant cells It seemed profitable, therefore, to use it to investigate the effects of the salts of K, Na, Mg and Ca on a typical plant cell and to determme whether the effects noted by Chambers and Reznikoff can be duplicated upon plant material. The experiments described in this paper deal with the action of these salts upon the protoplasm and with the nature of the sap vacuole and the tonoplastic membrane. MATERIAL AND METHODS The plant cells here investigated were the young root hairs of the aquatic, LimnoUum Spongia (Bosc.) Rich. This plant is closely related and similar to Trianea hogotensis, known also as L. stolomferum, a form well known in physiological work because of the protoplasmic streammg of its large root hairs. These two plants have been confused by certain American horticulturists. L. Spongia is native to the southern United States, and is apparently quite variable. Its root hairs are 46-60 micra in diameter and the cells show rapid and regular protoplasmic streaming. A tvpical young specimen is shown in Fig. 1. The cytoplasm flows along the cell-wall from the base toward the tip, where it collects as it reverses its direction. In passing toward the base, the cytoplasm unites to form a central strand, which apparently traverses the sap-vacuole. The nucleus remains at the base of the root hair and can not be reached with a micro- needle The thin cell wall, protoplasmic streaming, low osmotic pressure lack of plastids and other cell inclusions, make these root hairs ideal material for micrurgical investigation. The moist chamber and the tech- nique of microinjection are those described by Chambers (2). The LimnoUum roots were cut off from the plant and pkced m a drop of pond water upon a large cover slip, which was nverted over a microdissecting moist chamber. In such a preparation the young root ha rs show d nomal streaming for at least 24 hours. The root haii. ex end from the root in all directions, so that in a liquid drop preparation a desired root hair was best reached by a micropipette of the vertical typ and the tip of the pipette generally punctured the lower surface of the cell As a rule, only the younger root hairs were used for injection, inasmuf as these a;e rilh in protoplasm and injury can, therefore, be ""' w£ Tmtoneedle or micropipette is brought up to the cell it easily pierces the cell wall, much as a needle would pierce an inflated •i 422 Kerr balloon. The cell can be most easily pierced at the base, but a finely pointed pipette can puncture the hair at any place along the length provided the cell has not aged too much as to be more than 500 micra long. Longer root hairs are too tough and flexible and can only be punctured at their basal end. When a microneedle is inserted into a cell, streaming continues and the cell appears to be normal. The pipette can remain inserted in- definitely without causing any apparent harm, as long as precautions are taken to prevent jarring. When the pipette is removed, a plug of coagulated protoplasm is formed where the cell was pierced. This plug is similar to those noted by Nichols (21), Scabth (26) and Martens and Chambers (20) for various plant cells. Upon the removal of the pipette, streaming may stop momentarily but is soon resumed if the puncture is not too severe. Operations were performed at all seasons of the year, and only healty root hairs were used. In these root hairs, protoplasmic streaming (i. e. cyclosis) is always seen under normal conditions, and the presence, type and rate of this cyclosis were taken as criteria for healthy cells. Cells which were injured as a result of injection either recover and regain their cyclosis at the end of an hour, or show positive evidence of death changes. These pathological conditions exhibited such symptoms as granulation and coagulation of the cytoplasm and nucleus, Brownian movement within the cytoplasm, and the formation of Ca oxalate crystals in the vacuole. When a micropipette is inserted in a root hair, the pipette tip usually penetrates the vacuole (especially if the insertion is at the base). With a micropipette within the vacuole the internal pressure tends to press the vacuolar sap back into the pipette. Therefore, in order to inject liquids into the cell, a pressure must be exerted on the syringe of the injection apparatus, to overcome the considerable osmotic pressure of the cell. The streaming protoplasm is usually in greater amounts at the tip of the cell, and, if the wall is punctured in this region, the cytoplasm may itself be injected. Injections here must be made almost at the mo- ment of piercing inasmuch as the cytoplasm tends to be pushed back into the pipette where it coagulates. The removal of the pipette from within the cytoplasm nearly always results in a cessation of the cytoplasmic movement away from the site of the plug. Thus, when the pipette is taken out of the cell tip, streaming in the direction of the base of the cell stops, but continues toward the tip, so that there is a heaping of the cytoplasm at the tip of the root hair. This massing of the cytoplasm lasts only a i> » *> • • f ^ ¥ t Sr V # V (^ i» 4> (* The injection of certain salts into the protoplasm and vacuoles &c. 423 few minutes. The amount injected from a micropipette can only be very approximately determined. Attempts were made to estimate the volume by comparing with the volume of a nucleus. On the average, twenty- five root hairs were injected with each concentration of salt studied. The fact that injections can be both cytoplasmic and vacuolar, may best be illustrated in the case of the pH indicator, phenol red (4). This indicator, when injected into the vacuole slowly diffuses from the pipette, coloring the sap yellow. The cytoplasm remains colorless. If the indicator is introduced into the cytoplasm, the color is yellow with an orange tinge, while the sap remains colorless. Protoplasmic injections were made only into streaming protoplasm. There is an ectoplasmic layer of appreciable thickness which could not be injected directly. LITERATURE Root hairs have been extensively used in physiological work and much of the literature has been reviewed by Farr (9), (10). It is necessary here to review certain phases of the work which are concerned with this investigation. Aquatic root hairs, such as are present in Limnohium, Trianea, and Hydrocharis are unusually large in diameter, in contrast to the root hairs of land plants, and the nucleus of these aquatics is always in the base of the epidermis. In land plants the nucleus is frequently behind the apex of the hair. Haberlandt (11) in 1887 first showed definitely that the root hair grows in length by the formation of additional material only at the apex, and this apical growth was later confirmed by Zacharias (30) and Reinhardt (27). Ziegenspeck (31) demonstrated that the cell wall at the tip is of a different chemical composition from that along the sides. The side walls give a cellulose reaction when treated chemically, whereas the tip stains blue with iodine and consists of a transitional carbo- hydrate, called amyloid. The apical growth of Limnohium root hairs can be easily demonstrated by making small punctures with a microneedle and following the position of the plug in the subsequent growth. This apical type of growth is in contrast to the type described by Ziegen- speck (32) in the closely related form, Hydrocharis, According to this author, Hydrocharis root hairs do not grow at the tip but deposit amyloid intercalarly near the base. Ursprung and Blum (29) reported that the root hairs of Vicia faba show the lowest osmotic pressure of any of the cells of this plant. They 424 Kerr also state the osmotic pressure of root hairs depends to some extent on the medium in which they are grown. Root hairs already developed do not adjust themselves to the osmotic pressure of a new medmm. Those however, which grow after the root has been placed m the new concentration are adjusted to the new osmotic pressure. I have found that UmnoUum root hairs in pond water possess a plasmolyzmg value equivalent to 0.22-0.26 M. sucrose solution, a value fairly constant over long periods. The low osmotic pressure of these root hau-s is an aid in microinjection. , , ,. ^ * v.„j«„ Pfepfer (23) was one of the first to note the hurstmg of root hairs at their tip, when they were placed in hypotonic solutions. Zachaeias (30) found that the cell wall might burst yet growth would continue with the formation of a new membrane. Klemm (18) found that low concen- trations of acid have an explosive action on Trianea root hairs. He also made a study of the differences in the coagulative action of nitric and oxalic acids on Trianea protoplasm. Cholodntj (6) worked on the effects of various mono- and bivalent cations on the protoplasm of Trianea and found that the alkali metals are toxic when the root hairs are immersed in single salt solutions. If the root hairs are placed in a solution of potas- sium salts, the movement of the protoplasm becomes slow and the proto- plasm masses at the tip, giving one the impression of an increase in vis- cosity. This change gradually results in the death of the cell but if the reaction has not gone too far, the cell may slowly recover on the addition of small amounts of calcium salts. Hansteen-Cranneb (12, 13), Coupin (7), KtJSTER (19), Trelease (27) and Parr (10) all found that calcium is apparently necessary for root hair formation in all the forms studied. Hansteen-Cranner (12) found no correlation between the growth of the roots of wheat seedhngs and root hair production; in fact the best growth of root hairs seamed to be in single salt calcium solutions. Kisser (17) demonstrated that root hairs would not grow in moist chambers if the seedlings were kept from coming in contact with calcium salts. In this case the roots grew, but no root hairs were formed. If the slightest trace of calcium was present, root hairs emerged. Both Hansteen-Cranner (12) and Farr (10) agree that the significance of calcium in the growth of root hau-s seems t^ be that it is utilized directly in cell wall formation. Calcium is m fact the only constituent of the cell wall which is not produced withm the plant. * ' » r y ^ s' •* i* (• 4 <* The injection of certain salts into the protoplasm and vacuoles &c. 425 EXPERIMENTAL I. Immersion Experiments It was found that Limnohium produces root hairs in solutions of calcium chloride in concentrations from 0.015 M. to 0.002 M. Mature plants were used in which root hair production was limited to the young adventitious roots. Growth, however, soon stopped and, after 6—8 days, the roots ceased to elongate and no further root hairs were formed. Root hairs were not produced when the plants were placed in distilled water or in solutions of either Na, K, or Mg chloride, and root hairs akeady present were dead at the end of 48 hours. Experiments in salt antagonism were not performed. Root hairs of Limnohium which had grown in pond water behaved toward Na and K chloride, when placed in these solutions, essentially like that of Trianea, reported by Cholodnyj (6). If a root hair was placed in 0.02 M. NaCl, it showed a slackening in the rate of protoplasmic streaming and within two minutes contained a large mass of protoplasm, at the tip. Fully 80 % of the root hairs thus affected, burst at the tip without showing any increase in length. In the remaining root hairs there was a slow streaming for 10—30 minutes in which very little proto- plasm took part in this streaming, as a large mass had accumulated at the tip. Gradually the protoplasm at the apex assumed the appearance of a dead coagulum. II. Protoplasmic Injections a) Injections of distilled water. A small or moderate amount may be injected into the streaming cytoplasm at the tip of the hair without stopping cyclosis. As the distilled water enters, it disperses the cytoplasmic . granules around the site of injection. The mass of cytoplasm at this site bulges into the vacuole but quickly sinks down again as the dispersed granules flow down the cytoplasmic stream. If larger amounts are in- jected, the effect is sufficient temporarily to stop streaming in the direction of the base of the cell. This causes the protoplasm to mass at the tip. Although the distilled water disperses the granules, the accumulation of normal protoplasm in this region, and the possible exosmosis of water, cause the apical mass to appear normal within 90 seconds. At the end of three minutes, streaming is again renewed and the cell quickly recovers. I 4 426 Ker I b) Injections of NaCl and of KCl. No differences could be noted between the effects of NaCl and of KCl. The effects of NaCI upon the protoplasm can best be illustrated by an account of a typical injection of 1.0 M. NaCl. A pipette filled with NaCl was inserted into the apex of the root hair, the tip of the pipette lying within the cytoplasm just at the point where the protoplasmic stream reverses its direction. When an amount of NaCI equivalent to the volume of the nucleus is introduced into the cytoplasm there are several immediate effects. The protoplasm and the salt mix freely and there is no evidence, whatsoever, of any coagulation of the cytoplasm. The vacuolar membrane bulges into the sap vacuole, the cytoplasmic volume increasing at the expense of the vacuole, for 10—20 seconds after injection has stopped. This increase is much more than can be accounted for by the amount injected and, there- fore, one must assume that it is due to the entrance of water to establish an osmotic equilibrium. The protoplasmic streaming toward the base of the cell stops almost immediately, whereas movement toward the cell tip continues. This results in a massing of protoplasm at the apex of the root hair, within two minutes of injection. The protoplasm which comes from the base of the cell appears normal until it reaches the injected region, but here it mixes with the affected cytoplasm until a large mass extending 150-200 micra from the tip, collects at the apical end of the cell (Fig. 2). At this time, there is a rapid Brownian movement of granules within this region, a condition never seen when massing is caused by a mechanical shock. The cytoplasm remains here without streaming for a period of 5—10 minutes. Toward the end of this period the volume of the cytoplasm gradually decreases without any apparent loss due to streaming away. Coincident with this loss in volume, Brownian movement IS no longer evident and the cytoplasm appears normal, while a churning movement now begins within the mass. Brownian movement continues longer in the protoplasm near the tip than in other parts of the cell. The decrease in volume is possibly due to the outward diffusion of the NaCl, either into the sap or to the outside and the attendant loss of water' About 10—15 minutes following injection, a thin stream of protoplasm gradually pushes out of the churning mass at the tip and flows to the base of the cell. Protoplasmic streaming is gradually renewed and the mass forms into strands, so that 30 minutes after injection the injected cells can not be differentiated from the controls. In NaCl and KCl injections, it is necessary to keep the micropipette within the root hair until the massed protoplasm has moved from the '1 The injection of certain salts into the protoplasm and vacuoles &c. 427 apex of the cell and the cell is recovering. If the pipette is removed im- mediately after the injection, most of the affected protoplasm flows out through the hole in the wall before plug formation occurs. If this is a small part of the total protoplasm, the hole becomes plugged, and the cell usually recovers. Should the pipette be removed just after the massing of the protoplasm at the tip, death invariably follows, since the protoplasm flows out without repairing the hole made by the pipette. Inasmuch as NaCl and KCl solutions adversely affect plug formation, the pipette must ,..'. I y-S.'.;-'.-, Fig. 1. Root hair of Limnobium showing normal strands of cytoplasm. Fig. 2. Latent effect of m/1 NaCl two minutes after injection. A cytoplasmic vacuole frequently develops at the site of pipette tip some time after injection. be kept in place until protoplasmic streaming has been renewed. It is plausible to assign the prevention of plug formation following Na and K injections primarily to their liquefying action. This lack of plug formation following Na and K injections was also shown in the tendency of the injected solution and protoplasm to run back into the pipette. Such a back-flow is caused by the osmotic pressure within the cell. When a pipette with an aperture as small as 1 micron in diameter is used, the tip will clog. If the opening of the pipette is greater than 2 micra, there is usually a flow back into the pipette and a successful injection becomes difficult. However, protoplasm, which 428 Kerr has entered the pipette, may be injected back into the cell and if the pressure in the pipette is maintained for some time the cell will recover as previously described. It has been possible, after an injection of NaCl or KCl, to take up the mixture of protoplasm and salt solution into the pipette, and mject this into the cvtoplasm of a second root hair. The second root hair will show a typical sodium effect, and after recovery, the injected protoplasm becomes incorporated with the cytoplasm of the second cell. The added volume of cytoplasm is quite apparent in the increased material m cyclosis. This experiment has been performed successfully more than a dozen times and it has been found that a maximum amount of cytoplasm can be exchanged after a preliminary injection of 0.2 M. NaCl. If an amount of cvtoplasm and salt mixture is injected directly into the vacuole instead of into the cytoplasm of a second cell the injected material becomes coagulated, gradually turns a dark brown color and persists as such in the vacuole. ^^^, , ^. When large amounts of concentrated NaCl or KCl solutions are injected the cell dies, but death occurs only after rupture of the plasma membrane either by the cell wall bursting at the tip of the cell, or around the region of the insertion of the pipette. The protoplasm then flows out as long as there is an internal pressure. The protoplasm that remains inside finally coagulates. In more than 200 injections of both NaCl and KCl, death never occurred unless the cell ruptured. ' A break has never been observed in the vacuolar membrane, except when excessive pressure was exerted, no matter what^ concentration of NaCl or KCl was injected into the cytoplasm. This was checked by in- jecting a cell, the sap vacuole of which was colored by a previous injection of a dye, either phenol red or acid fuchsin, which remains in the vacuole without coloring the cytoplasm. When the salts were injected, the contour of the vacuolar membrane could easily be seen against the colored sap. The vacuolar membrane never broke and the protoplasm never became colored unless the entire cell burst. Thus we have a marked difference in the action of the outer and inner protoplasmic membranes ; the outer membrane frequently breaking as a result of the NaCl or KCl injections, while the vacuolar membrane remains intact throughout. In the injections of NaCl and of KCl, the intensity of the reaction is conditioned by the concentration used and the amount injected. Concentrations ranging from 2.0 M. to 0.06 M. of both Na and KCl were used. Even in the highest concentrations there was never an internal A K #1** 4 V 1' 4 M n ^ 1 1 ■ . « The injection of certain salts into the protoplasm and vacuoles &c. 429 coagulation of the cytoplasm, but injections of a nuclear volume of 2.0 M. resulted in the death of the cell in approximately half the cases by the breaking of the external membrane. Typical effects occur with diminishing intensity down to a concentration of 0.1 M., while with 0.05 M. the effects were so shght as to be indistinguishable from those of a distilled water injection. Thus it can be seen that these salts only produce a noticeable effect when injected into the cytoplasm in concentrations which would cause plasmolysis, were the root hairs immersed in such solutions. It is a question whether the fluidity of the protoplasm seen upon NaCl and KCl injections is an osmotic effect or whether it is due to a specific action of these monovalent ions. The fact that this effect is only seen with concentrations greater than the osmotic pressure of the sap suggests that it is an osmotic phenomenon. Injections of concentrated dextrose and sucrose into the protoplasm produced a fluidity similar to that produced by K and Na chlorides. The protoplasm also increased in volume in a similar manner at the expense of the sap vacuole. Two essential differences were noted. First, during recovery after suger injec- tions, the protoplasm usually became vacuolated. These vacuoles formed within the cytoplasm, a fact that could easily be seen by contrast when the sap was colored. Secondly, the cell did not fully recover for over an hour following dextrose injections, whereas normal streaming was present in less than 30 minutes when NaCl was used. The effects of dextrose are thus in agreement with the assumption that the fluidity produced by Na and KCl injections is at least in part due to osmotic action. c) Injectio7is of CaC^. The effects observed upon the injection of CaClg into the protoplasm are almost the opposite to those produced with NaCl and KCl. Concentrations greater than 0.04 M. CaClg will coagulate the cytoplasm. The protoplasmic plug at the region of insertion of the pipette forms with great ease after calcium injections, and the pipette may be removed immediately after injection with practically no loss of protoplasm. If the pipette is inserted at the apex of the cell, a deep brown coagulated mass forms immediately upon injection. Move- ment toward the base of the cell is usually stopped, and the cytoplasm accumulates as a thick viscous mass behind the coagulum (Fig. 3). If the coagulum is less than the size of the nucleus, streaming is renewed within 15 minutes and the cell recovers. The coagulated mass remains unchanged at the point where it is formed as long as the root hair remains alive. Frequently, when recovery begins, a mass of apparently normal cytoplasm surrounds the coagulurh for a period of 10—30 minutes. This is usually ^ i 430 Kerr carried away ty the stream as viscid lumps, which have very little move- ment of their own, but later disappear in the main mass of the streammg protoplasm. In a normally streaming cell, only a localized region of cytoplasm around the pipette can be coagulated by the injection of calcium. Any further injection forces the calcium chloride through the tonoplast into the vacuole and not along the cytoplasmic stream. If the injection is made at the apex of the cell where the cytoplasm tends to collect, a maxi- mum amount of cytoplasm can be coagulated. Here, if the coagulum is greater than the volume of the nucleus, death of the cell invariably follows. Such a death is a slow reaction. If, for example, the calcium has coagulated half the apex of the cell, movement will stop throughout, but within 30 minutes a definite movement will be resumed at the base of the hair. The part of the cytoplasm which still appears normal at the tip gradually coagulates until the whole apical end of the cell is discolored, while sUght random movements may still be observed at the other end of the cell and may continue for as long as 3—4 hours. When a concentration of 0.04 M. is injected a distinct coagulum results only with amounts greater than a nuclear volume. With con- centrations from 0.04 M. to 0.005 M. the cytoplasm around the region of injection forms a jelly-hke mass. The vacuolar membrane breaks easily in this region in contrast to the pronounced plasticity noted in the sodium injections. The break is followed by repair so that very little, if any, cytoplasm escapes into the vacuole. The chief evidence of the break is the formation of Ca oxalate crystals in the sap. During recovery from the injection of dilute concentrations, the reaction is the same but less severe than in the higher concentrations. Concentrations of CaClj less than 0.005 M. only produce a distilled water reaction. d) Injections of MgCk- MgClj, in concentrations of 1.0 M. and above, coagulates the cytoplasm if more than nuclear volume is injected. The coagulation is not accompanied by the immediate discoloration of the injected region nor is it as locaUzed as the calcium effect. Concentra- tions less than M/1 were not found to produce coagulation, and below 0.04 M. there is no specific reaction to the salt. Solutions of M/1 MgClj cause the protoplasm to set firmly around the region of injection. If the solution is introduced into the cytoplasm at the tip of the cell a preliminary dilution effect spreads down the stream for about 100 micra and the vacuolar membrane bulges into the sap vacuole as protoplasmic streaming toward the base slows down and ceases. The cytoplasmic granules very The injection of certain salts into the protoplasm and vacuoles &c. 431 gradually move into the injected area, and blister like excrescences appear, protruding into the sap vacuole. The dispersal of the granules throughout the injected region is not completed until 60 — 90 seconds after injection, and during this time, there is no plug formation should the pipette be withdrawn. The affected protoplasm at the site of injection does not exhibit Brownian movement, but the turgor of the vacuole may force the material down the pipette in the same manner as described for sodium injections. The normal streaming cytoplasm coming from the base does r tt.-.wi-i«ij.,bl • 0^ ' I Fig. 3. The appearance of a root hair two minutes after injection of m/1 CaClj showing coagulated area around the pi- pette tip. Fig. 4. Effect of m/1 MgClg two minutes after injection, showing the cytoplasm from the base heaped up in masses distinct from the injected area. not reach the affected region but heaps up in irregular masses behind the injected area (Fig. 4). After 16—20 minutes all the cytoplasm at the tip mixes, and, following this, there is another period of 10—15 minutes before normal streaming is renewed. III. Vacuolar Injections Injections into the sap vacuole can be made easily and in much larger amounts than into the cytoplasm. The injection of a colored solution shows that diffusion in the vacuole is practically independent of proto- 432 Kerr i li .lasmic streaming With a root hair 400 micra long, it takes ff Y 2 "ji^^J^Jf plasmicstreaining J distributed throughout the cell. for an injected substance to oe e veiny . xtoPI ttpi and MeCL ThP effects produced by the injections of NaCl, KOI ana mgoi^ !-„r.likP Considerable quantities of the salts in concentrations into the vacuole with a finely pointed pipette and the f'^f^^''^^^^^ immedttdy, the protoplasm is not affected, and a plug forms at once. Fig. 5. Root hair showing apical growth 10 hour, af t.r it ^d burst at the tip following ^ a vacuolar injection of m/1 NaCl. ThP root hair be<^ins to increase in length as the vacuole enlarges with The root hail be ins t streaming. Frequently so much onitP nf the loss of considerable protoplasm. ' Occas onal V growth occurred after the cell had recovered from burstiS a the «p This is shown in Fig. 5 which had grown 53 micra L lengthlrstill s normal streaming ten hours after bursting a he t°? It is to be noted that the new growth took place to one side of the original tip of the cell. This demonstrates the apical growth of these "'* nthe cell does not burst, the increase in length due to ^.Uing is usually permanent and streaming is normal withm 5-10 minutes. "r i The injection of certain salts into the protoplasm and vacuoles &c. 433 A significant phenomenon relating to the viscosity of the protoplasm is the shrinkage of the axial protoplasmic strand in the vicinity of the injected hypertonic solution, probably due to the dehydrating action of the latter. After the salt solution has diffused throughout the vacuole the strand recovers its normal appearance. Calcium chloride injected into the sap vacuole, produces a shower of tetragonal crystals, indicating the presence within a normal vacuole of soluble oxalates which become precipitated as calcium oxalate. These crystals can be produced even with the mjection of 0.002 M. CaCla. Al- though frequently seen in injured cells, these crystals do not occur normally in the vacuole of these root hairs. When the crystals are formed upon injection they drop to the lower side of the cell where they eventually sink through the vacuolar membrane into the cytoplasm in which they are carried around by the protoplasmic stream. In order to observe the possible effect of immersion, Limnobiurn plants were placed in a 0.04 M. CaClg solution for 24 hours, after which only the younger root hairs of the roots were alive and showed cyclosis. Calcium oxalate crystals were present in the dead cells but none whatsoever in the streaming root hairs. On the other hand injections of the 0.04 M. CaCla into the vacuole, viz., the same solution in which the root hairs had been immersed for 24 hours, resulted in the usual shower of crystals. With calcium injections, there is always a danger of damaging the strands of cytoplasm in the vicinity of the injection. It was found that once the oxalate had been precipitated in the vacuole, a further injection of CaClg may coagulate the cytoplasm in the vicinity in much the same manner as if it were injected directly within the protoplasm. If a root hair is cautiously given successive vacuolar injections of M/10 CaClg, it is possible to precipitate all the soluble oxalate within the vacuole without immediately coagulating the cytoplasm. Usually death results but in every case protoplasmic streaming always ceases to be renewed again in some cases at the end of an hour. Usually, however, it takes two hours for the cell to regain normal movement. At the end of this time, if calcium is again introduced into the vacuole, more oxalate crystals are precipated, indicating that, in the meantime, additional soluble oxalates had been formed. This suggests that either the presence of soluble oxalates or the absence of calcium within the vacuole is necessary for cyclosis within these root hairs. Saturated Na oxalate, injected into the vacuole, never produces a precipitate and does not stop protoplasmic streaming. Protoplasma. XVm 28 434 Kerr CaCIa was also injected into the root hairs of Dianthus larbatus and Avena saliva, both of which normally show a certain amount of protoplasmic streaming. This injection also was followed by the precipita- tion of calcium salt crystals. Osterhout (22) reported that crystals of Ca oxalate are formed when the root-hairs of Dianthus are placed in calcium solutions. He concluded from this that these root hairs absorb calcium ions. The writer has grown Dianthus barbatus seedlings in a chamber similar to that described by Farr (9) in which the root hairs were constantly immersed in an ever-changing solution of 0.01 M. CaCla. When undisturbed, these root-hairs grow in a manner similar to those of collards reported by Farr, and no crystals appeared within the vacuoles. However, a slight disturbance, such as transferring the seedling to a slide, resulted in the precipitation of Ca oxalate crystals within the vacuoles. This, apparently, is not toxic since the root hairs may continue growing! LOCAL EXTERIOR APPLICATIONS The apical bursting of cells showing apical growth has been described frequently in literature for root hairs, pollen tubes, and fungal hyphae immersed in hypotonic solutions exposed to substances which damage the delicate growing apex. After a concentrated solution of NaCl has been injected into the vacuole, the root hair progressively elongates as the vacuole increases m volume until the root hair bursts at the tip. When the cells were immersed in 0.1 M. NaCl, most of the root hairs burst at the tip within 2 minutes without any increase in length. If sodium oxalate is blown on the outer surface of a root hair within 15—20 micra of the apex the tip bursts quickly and violently. On the other hand, if the oxalate is applied locally against the side wall no bursting occurs but crystals appear within the pectin layer which forms the outer part of the cell wall. Frequently the pectic layer rises in the form of blisters. The crystals presumably are of calcium oxalate produced by interaction of the sodium oxalate with calcium pectate which Roberts (25) and Howe (16) found to be the composition of the outer layer of root hairs. At the same time the pectic layer becomes converted into soluble sodium pectate but the presence of cellulose maintains the required rigidity. At the apex of the cell the bursting is probably due to a similar demaging of the pectic layer. Evidently the amyloid apex described by ZiEGENSPECK (29) is not Sufficient to hold up against the internal pressure **^ ii M f \i^' 'I Ml V I u « I .0 The injection of certain salts into the protoplasm and vacuoles &c. 435 of the cell without the help of the pectic layer. It seems logical to assume that the bursting of the root hairs immersed in solutions of NaCl, is due to a similar action on the apex of the cell. The bursting reaction of concentrated sodium salts into the vac- uole is probably due in part at least to a different cause. The in- jection brings the salt into the vacuole where presumably it acts by increasing the internal osmotic pressure beyond the tensile strength of the apex of the cell. DISCUSSION It is not difficult to compare the effects of injecting various salts into Limnobium root hairs recorded here with the effects of injecting similar salts in amoebae and other animal cells described by Chambers and his co-workers. Chambers and Reznikoff (5) observed that the monovalent NaCl and KCl salts produce a liquefying and quiescent effect upon amoebae. The same action was noted both when the amoebae were immersed in NaCl and KCl and also when the amoebae were injected with these salts. On the other hand Heilbrunn (14) claimed that sodium and potassium ions cause an increase in viscosity of several types of protoplasm and initiate coagulative changes while the ions of magnesium and calcium have the opposite effect. Heilbrunn quotes Cholodnyj's work (6) on Trianea as support for his conclusions. The effects which Cholodnyj observed in Trianea, I have also noted when Limnobium root hairs are immersed in NaCl solutions, viz., a decrease in the rate of streaming and the accumulation of the protoplasm at the apex of the cell. The accumulation of the protoplasm at the apex can be accounted for by the ready penetration of the salt at this region here exerting its specific quiescent effect. The streaming from the base continues for a time so that the protoplasm becomes massed at the tip. In the majority of the cells, bursting soon occurrs at the tip, while in the others, the material at the tip finally forms a dead coagulum. The fact that it is the protoplasm at the tip which is first affected agrees with the work of Chambers and Kerr (4) in which it was found that the cell wall at the apex is more readily permeable than along the sides of the root hair. The protoplasm was never observed to coagulate from injections of either weak or strong NaCl or KCl and recovery occurred as long as the outer membrane of the cell remained intact. After bursting coagulation does follow but this 28* 436 Kerr The injection of certain salts into the protoplasm and vacuoles &c. 437 surely can not be interpreted as the immediate action of NaCl or KCI. Chambers and Reznikoff also describe coagulation of amoeba dying when immersed in NaCl. Several complicating factors must be considered in interpreting immersion experiments. It is well known that root hairs are normal only when kept in solutions containing calcium ions. True (28) quotes unpublished work of Miss Eckerson on the effect of placing wheat seedlings in solutions of K salts. Potassium rapidly penetrates and reacts with calcium pectate to form potassium pectate which dissolves in the water. At this stage, sugars, amino acids and salts, chiefly magnesium, diffuse rapidly out of the roots. A similar reaction occurs when the root hairs are placed in solutions of sodium salts in which case soluble sodium pectate is formed. In an attempt to study viscosity changes by immersion methods one must take into account the complicating factors of the effect of salts on the cell wall and the exosmosis of various chemical substances. On the other hand an injection of NaCl and of KCI always produces the reverse of an increase in viscosity. However, it is still possible to explain the increased liquefaction to the osmotic action of the added salts which might more than counter-balance any possible coagulation effects. At any rate, there is no visible irreversible coagulation. NaCl and KCI are relatively non-toxic when injected into cytoplasm and exhibit definite liquefying effects only when in high concentrations. The fact that dextrose and sucrose show a similar liquefying effect when injected, supports this view. The local coagulum produced by injections of CaClj into the cyto- plasm, is similar to that noted by Chambers and Reznikoff in amoebae. The amoebae are able to rid themselves of the coagulated mass by a pinching-off process, whereas the root hairs retain the coagulum within the rigid cell wall. The reaction of MgCl^ is also similar since in both amoebae and root hairs, magnesium ions produce a general quiescence of the protoplasm and the affected cytoplasm apparently sets in the form of a gel. The concentration of the solutions injected to produce the typical salt effects are greater in the case of UmnoUum root hairs than m amoebae, and are about of the same order as that in Aclinos- phaenum, (3). However, this may be largely due to the relative size of the organisms, the amoebae being much smaller than either the root-hairs or the Actinosphaerium. The formation of a protoplasmic plug to close a puncture in the cell wall takes place with ease after calcium injections but with the greatest *» difficulty after injections of NaCl, KCI and MgClg, and then only after the loss of considerable protoplasm. The plug consists of coagulated protoplasm and the difficulty of its formation after NaCl, KCI and MgClg injections may be due not only to the lack of coagulation but also to an increase in volume of the cell following the imbibition of water. At the same time protoplasmic injections of NaCl and KCI frequently resulted in the rupture of the outer membrane while this rarely happened with MgCl2 injections. SUMMARY By means of microinjection, the effects of the chlorides of K, Na, Ca and Mg were investigated upon the protoplasm and the vacuole of the root hairs of the aquatic plant, Limnohium spongia. This plant is closely related to the well-known Trianea hogotensis {Limnohium stoloni- ferum). 1. Root hairs develop in .015— .002 M. CaClg but not in Na, K, or Mg chlorides which are toxic. In .02 M. NaCl the massing of the cyto- plasm at the tip of the root hair is due to retardation of the streaming movement. After some time the majority of the root hairs burst at their tip. This does not occur when CaCl2 is present. 2. Sodium chloride and potassium chloride are relatively nontoxic when injected into the protoplasm. These salts increase the fluidity of the protoplasm, but only when used in high concentrations (2.0 M. to 0.1 M.). This effect is interpreted as essentially an osmotic phenomenon. Dextrose and sucrose produce a similar state of fluidity of the protoplasm when injected in high concentrations. In lower concentrations (below 0.1 M.) injections of these salts produce no visible effect except to slow temporarily the protoplasmic movement. Death results from injections of NaCl and KCI only when the cell wall breaks. 3. Injections of CaClg into the protoplasm in concentrations greater than 0.04 M. result in a local coagulation which persists, although the protoplast may recover its normal streaming with the coagulum remaining adherent to the cell-wall at the site of injection. Injections of CaClg less than 0.04 M. and greater than 0.005 M. produce a local solidification of the protoplasm which is later incorporated. 4. MgCla, in concentrations greater than 0.1 M., when injected into the protoplasm, produces a widespread reversible solidification of the cytoplasm. !l|l I 438 Ke rr 5. Injections of NaCl, KCl and MgCIg into the vacuole show no effect upon the cells except in sufficiently high concentrations to produce osmotic changes. Bursting of the cell apex may take place but this is frequently followed by recovery. 6. CaClg, when injected into 'the vacuole produces a shower of calcium oxalate crystals, indicating the presence of soluble oxalates within the sap. Cyclosis seems to be correlated with the lack of calcium ions in the sap of the vacuole in these root hairs. 7. Injections of high concentrations of sodium or potassium chloride into the cytoplasm causes the root hair to burst at the tip. The vacuolar membrane simply bulges into the vacuole. With calcium chloride the cell does not burst but the internal membrane breaks. This may be due mainly to their relative positions in the cell and to the respective weakening or stiffening action of the salts on the investing cell wall. 8. NaCl, KCl and MgClg injections retard the formation of a proto- plasmic plug at the site of puncture, while CaClg apparently aids in its formation. 9. The increased fluidity of the protoplasm after an injection of NaCl or KCl makes it possible to withdraw some of the mixture of proto- plasm and salt solution and to inject it into another root hair. When injected into the cytoplasm of a second cell, the injected material mixes with the protoplasm and becomes indistinguishable in the streaming. When injected into the vacuole, it persists as a coagulum. ACKNOWLEDGMENTS I wish to make grateful acknowledgments to Dr. William Seifriz of the University of Pennsylvania for the helpful cooperation which he has given me in carrying out this work. I also wish to thank Dr. Robert Chambers of New York University, under whose direction this problem was begun, and for the aid he has given me in preparing this manuscript. BIBLIOGRAPHY 1. Chambers, R. Microdissection and injection studies on the antagonistic action of salts upon protoplasm. Amer. Jour. Physiol. 72, 210 (1925). 2. — . Methods for the study of fresh material. Physical agents: microdissection and microinjection. McClungs Microscopal Technique, Chap. 2, 39 (1929). 3. — and Rowland, R. B. Micrurgical studies in cell physiology. VII. The action of the chlorides of Na, K, Ca and Mg on vacuolated protoplasm. Protoplasma 11, 1—18 (1930). The injection of certain salts into the protoplasm and vacuoles &c. 439 4. Chambers, R. and Kerr, T. Intracellular Hydrion concentration studies VIII. Cyto- plasmic and vacuolar pH of Limnohium root hair cells. Jour. Cell, and Comp. Physiol. 2, 105—119 (1932). 5. — and Reznikoff, P. Micrurgical studies in cell physiology. 1. The action of the chlorides of Na, K, Ca and Mg on the protoplasm of Amoeba proieus. Jour. Gen. Physiol. 8, 369—401 (1926). 6. Cholodnyj, N. Zur Frage iiber die Beeinflussung des Protoplasmas durch mono- und bivalente Metallionen. Beih. Bot. Centralbl. 39 Pt., 1, 231—238 (1923). 7. CouPiN, H. Influence des sels de calcium sur les poils absorbants des racines. Compt. Red. Acad. Sci. Paris 164, 641—643 (1917). 8. Ephrussi, B. et Rapkine, L. Action des differents sels sur le Spirostum. Proto- plasma 6, 35—40 (1928). 9. Farr, C. H. Studies on the growth of root hairs in solutions. 1. The problem, previous work and procedure. Amer. Jour. Bot. 14, 446 — 456 (1927). 10. — . Root hairs and growth. Quart. Rev. Biol. 3, 343—376 (1928). 11. Haberlandt, G. tJber die Lage des Kernes in sich entwickelnden Zellen. Ber. Deutsch. Bot. Ges. 5, 205—212 (1887). 12. Hansteen-Cranner, B. Uber das Verhalten der Kulturpflanzen zu den Boden- salzen. I. II. Jahrb. f. wiss. Bot. 47, 289—376 (1910). 13. — . t)ber das Verhalten der Kulturpflanzen zu den Bodensalzen. III. Jahrb. f. wiss. Bot. 53, 536—599 (1914). 14. Heilbrunn, L. V. The colloid chemistry of protoplasm. Protoplasma-Monogra- phien 1 (1928). 16. HOAGLAND, D. R. and Davis, A. R. The composition of the cell sap of the plant in relation to the absorption of ions. Jour. Gen. Physiol. 5, 629 — 646 (1923). 16. Howe, G. Pectic material in root hairs. Bot. Gaz. 72, 313—320 (1921). 17. Kjsser, J. tJber das Verhalten von Wurzeln in feuchter Luft. Jahrb. f. wiss. Bot. 64, 416—439 (1925). 18. Klemm, p. Die Organisationserscheinungen der Zelle. Jahrb. f . wiss. Bot. 28, 626 bis 700 (1895). 19. KiJSTER, E. t)ber die Beziehungen der Lage des Zellkerns zum Zellenwachstum und zur Membranbildung. Flora 97, 1—23 (1907). 20. Martens, P. and Chambers, R. fitudes de microdissection. V. Les poils staminaux des Tradescantia, La Cellule 41, 131—145 (1932). 21. Nichols, S. P. Methods of healing in some algal cells. Amer. Jour. Bot. 9, 18—27 (1922). 22. OSTERHOUT, W. J. V. On the penetration of inorganic salts into living protoplasm. Zeitschr. f. Physik. Chem. 70, 408-^13 (1910). 23. Pfeffer, W. Zur Kenntnis der Plasmahaut und der Vakuolen nebst Bemerkungen liber den Aggregatzustand des Protoplasmas und iiber osmotische Vorgange. Abh. Sachs. Ges. Wiss. 16, 185—197 (1889). 24. Reinhardt, M. O. Das Wachstum der Pilzenhyphen. Ein Beitrag zur Kenntnis des Flachenwachstums vegetabilischer Zellmembran. Jahrb. f. wiss. Bot. 23, 479—566 (1892). 26. Roberts, E. A. The epidermal cells of roots. Bot. Gaz. 62, 488—506 (1916). 26. ScARTH, G. W. The structural organization of plant protoplasm in the light of micrurgy. Protoplasma 2, 189—205 (1927). m 440 Kerr, The injection of certain salts into the protoplasm and vacuoles &o. 27. Trelease, S. F. and Trelease, H. M. Growth of wheat roots in salt solutions containing essential ions. Bot. Gaz. 80, 74^-83 (1926). 28. True, R.H. The significance of calcium for higher green plants. Science N.S. 66, 1—6 (1922). 29. Ursprung, a. and Blum, G. Zur Kenntnis der Saugkraft. V. Ber. Deutsch. Bot. Ges. 39, 139—148 (1921). 30. Zacharias, E. tJher das Wachstum der Zellhaut bei Wurzelhaaren. Flora 74, 466—491 (1891). 31 . ZiEGENSPECK, H. Das Amyloid jugendlicher Organe. Das Amyloid in den wachsenden Wurzelhaaren und seine Beziehungen zum Zellwachstum. Ber. Deutsch. Bot, Ges. 38, 328—333 (1920). 32. — . Die Lage des ZeUkemes in den Wurzelhaaren von Hydrocharis marsusranae wahrend des Wachsens. Bot. Arch. (Konigsberg) 20, 476 (1927). t »'i %i 4 Reprinted from Kew Bulletin of Miscellaneous Information, No. 4, 1933, by permission of the Controller of H.M.S.O, XX— SARRACENIA PURPUREA VAR. STOLONIFERA. A Noteworthy Morphological and Ecological Type. John M. Macfarlane and D. Walter Steckbeck. Description of Locality. In response to a request by the late Dr. Maxwell Masters, the senior author pubUshed a short paper in the '' Gardeners' Chronicle " for 1895 (series 3, xvii. 643) entitled " Sarracenias at Home.'' The article mainly treated of an excursion into the Southern United States undertaken, along with one of his students, nearly a year before, chiefly for field study of the native pitcher plants or Sarra- cenias. In a few concluding sentences the author wrote as follows : '* You will scarcely credit the show that our native S. purpurea at times makes. About three weeks after my southern tour I was botanizing in New Jersey, and was carried by train past a lake bordered by the plants. I determined to make an inspection and did not regret it. The lake was about a quarter of a mile long. Its margin was covered by a continuous sheet of large specimens mostly of deep crimson hue. " To look down on a piled up mass of pitchers, seven to twelve feet in width and mostly floating on the water, was a sight indeed. I hope some day to photograph the view and you will then be able to share the pleasure. ''In the sheltered bays of this lake floating beds of the purple bladderwort [Utricularia purpurea) in full bloom, alternated with sunken logs that were carpeted with Drosera rotundifolia and D. intermedia. A real carnivorous garden you will say and such it was." Many readers of the above, who are familiar with S. purpurea in its native haunts, may incline to query the possible correctness of such a description. For in the numerous and extensive expedi- tions made by the senior author of this article during the past forty-one years, from eastern Canada and Maine south to northern Floiida and the coast of Louisiana, he has encountered no localitv which would compare with that now to be described, nor any growths of pitchers and of stems comparable in size to those now figured. First acquaintance was made with the locality in mid July 1894, when the senior author along with one of his little sons — then four years old — was rapidly carried by train along the railroad embank- ment that rose about ten feet above the edge of the pond or lake. His eye quickly and with delight took in the scene, while the exclama- tion came from the small boy at his side " Oh, Papa, see the Sarra- cenias ! " For, over almost til entire shore margin that stretched away obliquely for about a quarter of a mile from the point of view, dense growths of Sarracenia purpurea extended. These varied in colour from rich crimson in sunny places to greenish-crimson or ahnost green in the shade, while their large pitchers bordered the shore often in continuous masses to a width of from two to frequently ten or even twelve feet. They produced a beautiful colour effect against the sparkling water and the tall cedar woods beyond, which anchored their roots among those of the Sarracenias. The timetable indicated that a "flag station/' Davenport, had just been passed, and that Whitings was five and a quarter miles ahead in the direction of Camden and Philadelphia. Needless to say, the writer resolved to visit so attractive a locality and to explore its riches. This was done about three weeks later when staying at Island Heights, a village about three miles east from Toms River. Around the lake on every side stretched undulating, sandy expanses of typical pine barrens soil on which grew a flora eminently characteristic of that region. This will be described later by the second author. But, looking westward along the railroad for about three hundred yards, the desired locality seemed to be in view, for both sides of the track were bordered by tall growths of the white swamp cedar or cypress {Chamaecyparis thyoides or C. sphaeroidea) which suggested swamp or even lake expanses. On nearer approach it was found that the railway had been carried over a bridge, beneath which flowed a stream whose waters were of the typical pine barrens yellow-brown tint. The stream issued from a pond or small lake on the left, which, after the bridge was crossed, gradually came into full view as an obliquely extended expanse of water. This was almost wholly encircled by tall growths of the white cedar whose trunks grew in the shallow water or in the shaded bordering sphagnum margin. This shaded area reached from ten to fifteen feet from the edge of the pond. A wide recess or bay had formed for about twenty feet to the left of the outlet, and here a striking scene was visible from our high vantage ground on the railroad. A few snags had drifted down and accumulated over the waters of this bay, in part submerged, in part exposed. These had entangled great beds of the purple bladderwort {Utricularia purpurea), whose floating masses formed a sheet of lavender-purple bloom. Along the moist sides of the logs or spreading over the sandy bay-shore were abundant patches of Drosera rotundifolia and D. intermedia. As the sketch-plan indicates, the lake, fully a third of a mile long, was of irregular and sinuous outline, its median line being a curve that extended from the out-flowing stream upward to where the lake again contracted into a gently moving and largely sphagnum-bestrewn current continuing for miles to the westward. From a study of the entire lake, as well as of its inlet and outlet areas, its history was evidently somewhat as follows. Originally a small continuous stream that arose in the higher moist lands near Whitings, it flowed for about two miles in a south-easterly direction, < t ¥ Y y 4 p ^i f 1 t % m « \- V 4) k 4' ', ^ ^. ♦ • r k then turned north-eastward for nearly nine miles as Davenport Branch when it joined two lesser streams, Wrangel Branch and Sunken Branch. These together formed a somewhat enlarged pond which in turn narrowed into a stream emptying into the Toms River at the county seat of that name. Many years ago, however, the Pennsylvania Railroad was first cut through, and continued from Whitings to Toms River and onward. In the process a railroad embankment was formed that cut across the depression in the midst of which Davenport Branch flowed. In constructing the embankment a trestle bridge about fifteen feet long was built across the actual stream. During spring freshets this embankment formed a slight temporary dam and restricted the flood water to the one outlet channel, which in the course of time was excavated so as to produce a pond that steadily increased in width and depth. Into this expanding lake, under normal environal conditions, a gentle precipitation of decomposing animal and vegetable remains was carried downward. These remains together with the inorganic mud gave rise to a fine, silty deposit or muck over the sandy lake floor. This evidently formed an excep- tionally rich and loose soil in which plants of Sarracenia could readily root while the protruding pitchers above could capture an ample supply of insects. Such clearly must have been the conditions which favoured an abundant growth of luxuriant Sarracenia plants. The soft, silty muck now varies in depth from three to twelve inches along the shore line, to as much as two or three feet in deeper water. The deepest parts of the lake, along its median curved line, are from four to six feet deep. At the time of the first visit in 1894 many of the white cedars were tall, thick trees which, along with those of lesser growth, formed an imposing cincture to the pond margin. About ten years later, however, the larger trees were cut out, and this seemed to admit abundant sunlight which gave crimson colour to the pitcher plants beneath, as well as to those farther out in the open. The surface of the pond margin that had been invaded by the Sarracenias was on an average three feet in width, but in places as much as ten to twelve feet. So solidly were the plants grown together, and so closely rooted amid the sphagum and silty mud, that one inclined almost to step out amongst them as if on firm ground. Mention has been made of the varying colour of the leaves of the Sarracenia plants. These showed a transition from uniform green in the deeper shade of the cedars, swamp magnolias and swamp blueberries, to a dark crimson-purple hue over the entire surface of the pitcher where fully exposed to strong sunlight. A similar variability is observable also from northern Maine southward to northern Florida, or for a distance of 1300 miles, and is equally characteristic for such southern species as S. Drummondii and S. psittacina. The colour of the upper leaf surface in Venus' flytrap I) i> V ^1 r / 1 4 ► \ 7 i. ^ i 'I' '/ y •J o o X be c a. u to o o o CO c o o u C/5 O J) > ^ ^ [To face page 4. then turned north-eastward for nearly nine miles as Davenport Branch when it joined two lesser streams, Wrangel Branch and Sunken Branch. These together formed a somewhat enlarged pond which in turn narrowed into a stream emptying into the Toms River at the county seat of that name. Many years ago, however, the Pennsylvania Railroad was first cut through, and continued from Whitings to Toms River and onward. In the process a railroad embankment was formed that cut across the depression in the midst of which Davenport Branch flowed. In constructing the embankment a trestle bridge about fifteen feet long was built across the actual stream. During spring freshets this embankment formed a slight temporary dam and restricted the flood water to the one outlet channel, which in the course of time was excavated so as to produce a pond that steadily increased in width and depth. Into this expanding lake, under normal environal conditions, a gentle precipitation of decomposing animal and vegetable remains was carried downward. These remains together with the inorganic mud gave rise to a fine, silty deposit or muck over the sandy lake floor. This evidently formed an excep- tionally rich and loose soil in which plants of Sarracenia could readily root while the protruding jntchers above could capture an ample supply of insects. Such clearly must have been the conditions which favoured an abundant growth of luxuriant Sarracenia plants. The soft, silty muck now varies in depth from three to twelve inches along the shore line, to as much as two or three feet in deeper water. The deepest parts of the lake, along its median curved line, are from four to six feet deep. At the time of the first visit in 1894 many of the white cedars were tall, thick trees which, along with those of lesser growth, formed an imposing cincture to the pond margin. About ten years later, however, the larger trees were cut out, and this seemed to admit abundant sunlight which gave crimson colour to the pitcher plants beneath, as well as to those farther out in the open. The surface of the pond margin that had been invaded by the Sarracenias was on an average three feet in width, but in places as much as ten to twelve feet. So solidly were the plants grown together, and so closely rooted amid the sphagum and silty mud, that one inclined almost to step out amongst them as if on firm ground. Mention has been made of the varying colour of the leaves of the Sarracenia plants. These showed a transition from uniform green in the deeper shade of the cedars, swamp magnolias and swamp blueberries, to a dark crimson-purple hue over the entire surface of the pitcher where fully exposed to strong sunlight. A similar variability is observable also from northern Maine southward to northern Florida, or for a distance of 1300 miles, and is equally characteristic for such southern species as S. Drummondii and S. psittacina. The colour of the upper leaf surface in Venus' flytrap •J 7 0 <) I 1 "^ .)i > tf. rX CO w" U u J'. T. r. A [To face pai^c 4. INTENTIONAL SECOND EXPOSURE V ^ 4 2. General view of widest part of lake. ^' . 0 'i - •' », •V > 3. Sinj^le })lant of Savvaceuia purpurea var. stohmifera (j ' 3. Single plant of Sarracenia purpurea var. stolonifera (greatly reduced) in - late May, showing 7 new leaves of the current year surmounting 7 still fresh or slightly withered leaves of previous season, and a few withered remnants of still older leaves. Widest diameter of plant from pitcher tip to pitcher tip 29 ins. To face page 5.] 4 4 and the tint of the tentacular hairs in leaves of the native sundews vary in the same way. When seen at this time (August) the plants showed thousands of strong, upright fruiting stalks, rising conspicuously above the dense masses of pitchers which made up each huge patch. The stalks varied in height from 14 to 19 ins. and in thickness from J to J in. Occasionally withered remnants of the petals could be traced, but the five large spreading and persistent green to purplish sepals were conspicuous, surrounding the large maturing fruit. While not a few of the fruits were maturing normally or had recently burst their carpels, many showed the serious havoc caused by the depredations of the larval stage of Exyra Rolandiana. This small moth attacks the carpels and soft young seeds, feeding on them and so reducing their tissues to a pulpy condition. They unquestionably cause great destruction to, and reduction of, the total quantity of seeds matured. This subject has been very thoroughly investigated by Dr. Frank M. Jones in a series of valuable papers* which have been published during the past few years. At irregularly succeeding periods, the senior author conducted various excursions with his students and with members of scientific societies to the region. He therefore had opportunities for studying the area and its local flora. Several of these observers have secured photographs illustrating both the plants and their surroundings. Special mention should be made of the views secured by Dr. Herbert Kribs, by Dr. Walter Steckbeck the junior author of this paper, and by Dr. W. Randolph Taylor. The first of these observers obtained some pictures of great detail and to him the present authors are grateful for permission to have one of them reproduced. Comparative study, then, throughout these years has shown that the plants of this limited area deserve to be regarded as a distinct variety. They differ strikingly in many structural details from the normal form, S. purpurea var. typica (Macfarl. in Pflanzenreich, Sarraceniaceae, 33). Accordingly specimens have been deposited in the Kew Herbarium, in that of the University of Pennsylvania and in various other herbaria of America and Europe. The following is a description of this variety. S. purpurea, var stolonifera Macfarl. et Steckb. Rhizoma elongatum usque ad 51 cm. longum, incrementa annua 1.25-3-2 cm. longa formans, ramulos laterales spatiis interiectis gerens rhizomati parenti similes atque interiectis 5-8 annis putrefacti ab eo separantes. Ascidia 10-45 cm. longa, parte tertia inferiore solida, partibus duabus superioribus cavis; ala usque ad 2-5 cm. lata; os i •5-2-8 cm. diametro; pedunculis elongatus usque ad 47-5 cm. longus. Flos quam in var. typica tertia parte vel dimidio major. ♦The most important of these appeared in Natural History, xxi. no. 3, 296-316 (1921). INTENTIONAL SECOND EXPOSURE Period of Blooming. During the last thirteen years observations have been made by the junior author on the blooming periods of this plant at Davenport and Crossley. Some sixteen trips were taken to the region ; most of these were made in the latter part of May when the pitcher-plant is in various stages of flowering, and when many other plants of the pine-barrens are at their climax of blooming. The results of these studies indicate that the average climax of flower expansion, that is when practically all flowers in a given area are fully open, is about May 28th. The earliest season was that of 192 1 when spring in this latitude was unusually early. By May 2ist of that year Sarracenia was past its best for flowering. The latest of the twelve seasons was that of 1924 when on May 31st barely J of the flowers had lengthened their petals. The spring of 1928 was about as late as that of 1924, and the climax of blooming was delayed to June 4th. Flower buds are noted on the Sarracenia plants early in May or even late in April. Gradually the stout pedicels elongate, the buds swell, and in an early season one may look for the first open flowers shortly after the middle of May, but the beginning of flower-opening obviously depends on the seasonal conditions. After complete expansion of the purplish-red petals, a period of six to nine days elapses before the showy corolla withers and drops, its work in attracting insect visitors having been accomplished. With the petals go the numerous stamens, and even in late seasons by June 15th only persistent bracts, sepals and pistil remain, the last consisting of developing ovary, the elongated lower half and the umbrella-like upper half of the style that bears five small stigmas between the tips of its pentagonal lobes. Environal Plant Associates. While Sarracenia holds first place in importance for the visitor to Crossley during May, yet a considerable number of other plants then in bloom adds to the interest. Along the water's edge one finds lingering clusters of Orontium aquaticum. Somewhat out from the banks of the stream Eriocaulon is just opening its capitula of flowers. One of the showiest pine-barrens companions of the pitcher plant is Xerophyllum asphodeloides, sending up long racemes of numerous flowers. At this time one may look for such other plants in bloom as Arethusa bulbosa, Cypripedium acaule, Leucothoe racemosa, Leiophyllum {Ledum) buxifolium, with flowers well past their climax, Arctostaphylos Uva-ursi, in full bloom a month earlier but at times showing a lingering flower, Euphorbia Ipecacuanhae, Hudsonia ericoideSy Pyxidanthera barbulata, nearly over, Pyrus arbutifolia, Helianthemum canadense, Linaria canadensis, Arenaria caroliniana, in large bud and well on to full bloom before Sarracenia passes, species of Vaccinium and Gaylussacia, Kalmia latifolia and K. angustifolia, both in large bud or just opened, etc. Of special interest are the three species oiDrosera — D. rotundifolia, D. intermedia, and D. filiformis — all very common but blooming after the pitcher plants. -• (^ < if ^ 4 i > 2 > 4. Chapman (1929) has been able to separate dif- ferent strains of Bacillus coli by comparing their potentials at a constant pH value, while Reed and Gardiner (1932) have compared complete mobility curves for both R and S types of Mycobacteriuin leprae. The differentiation of Haemophilus pertussis strains has been done by Shibley (1932) using these methods. Erythrocytes have also been shown by Abramson (1929) to pos- sess a specific P.D. dependent on species. Not only the isoelectric points but also the shapes of the velocity curves plotted against pH can be used to study an amphoteric substance. The difficulty of distinguishing between two pro- teins of the same isoelectric point but different characteristics is obviated when the curves show differences in shape. The curves give a complete picture of the changes which ensue with change of pH. It has been shown by many workers, following Loeb (1922) and Abramson (1928), that inert colloidal particles in protein solutions assume the isoelectric point and electrokinetic properties of the protein itself. The nature of the particle is immaterial ; only the adsorbed surface influences the reaction. The use of electrophoresis as a tool in this work has been summarized from a practical standpoint by Seifriz (1928) and by Prausnitz and Reitstotter (1931). For the theory, the reader is referred to Pauli and Valko (1929) and to Smoluchowski (1921). The work in relation to bacteria is discussed by Mudd, Nugent, and Bullock (1932). Particularly suitable subjects for investigation in this connection are latex particles. These colloidal particles occur naturally in living cells of many plants (Molisch, 1901 ; Bobilioff, 1919; Frey-Wyssling, 1932). The struc- ture of Hcvea latex particles has been studied by Freundlich and Hauser (1925) and by Hauser (1930) by means of microdissection. The particles were found to be pear-shaped with a liquid center surrounded by a shell of denser rubber. Hauser pictures a partial coating of adsorbed protein. Weber r -Vf ^> h > i . V f > 1* \ ^K I ¥ V > HI m •>i^ >' ^^ June, 1934] MOVER EUPHORBIA 295 (1903) seems to have been one of the first to suggest a protein-coated par- ticle, but most later evidence indicates that the protein coating, although pres- ent, is incomplete. Freundlich and Hauser (1925) and Beumee-Nieuwland (1929) have offered most convincing proof that this is the case. The work has been summarized by Whitby (1920), Hauser (1930), Fisher (1930), van Harpen (1931), and Morris and Greenup (1932). Concerning the structure of latex particles in other genera, little is known beyond their microscopic pictures and, in some cases, chemical analysis (Hauser, 1930). The present investigation was undertaken to see if resemblances between closely related species would be shown by the electrophoretic mobility curves of their latex particles. The electrophoresis of latex was first investigated by Henri (1906), who found it moved to the anode and hence its particles possessed a negative charge. This has been utilized by industries to plate substances with latex (Sheppard, 1927; Prausnitz and Reitstotter, 1931). Belgrave (1923) appears to have been the first one to report a positive charge on latex after addition of acid. Belgrave (1923), Sheppard and Eberlin (1925), Rowland (cited by Dins- more, 1926), and Twiss (1931) found varied isoelectric points for ammonia latex, depending on its degree of preservation. All of the values lie within the range of pH 3.0-pH 5.0 which includes the isoelectric points of most proteins (Mudd, 1925a; Pfeiffer, 1929). Fresh latex from Hevea has a more definite isoelectric point at pH 4.8 as determined by its acid coagulation optimum (van Harpen, 1931). Unfortunately, fresh Hevea latex can be obtained only in the tropics ; con- sequently, for this problem latex from various species of Euphorbia was stud- ied. The genus Euphorbia is a large one and includes many species, all of which contain latex. Hence, it is particularly suited for work of this char- acter. As far as known, there appears to be no literature on the electro- phoresis of latex from this genus. MATERIALS AND METHODS Some of the plants in this work were grown from seed obtained from the IVIuseum National d'Histoire Naturelle in Paris and from the University of Warsaw, some were from the botanical gardens of the University of Penn- sylvania, and others were collected locally. The latex was obtained by sever- ing a leaf or nicking the stem with a clean razor blade and suspending the exuding drops in M/50 acetic acid-sodium acetate buffer mixtures in a dilu- tion of approximately one drop of latex to 25 cc. of buffer. Both Mudd (1925b) and Abramson (1932) have shown that acetate buffers are preferable for this type of work, and since this buffer system has been used by other investigators in electrophoresis, all curves are confined to its pH range. Pauli (1920) has shown that strong acids yield anomalous results and cannot give a true electrophoretic isoelectric point, hence they have been avoided. Every buffer dilution was made immediately before each test and its pH measured 296 AMERICAN JOURNAL OF BOTANY [Vol. 21, (with the latex in it) by the quinhydrone method (Michaehs, 1926; van Harpen, 1929, 1930- A "type-K" potentiometer and a saturated calomel cell were used in each case. The standards proposed by Clark (1928) were used to convert voltage to pH. Measurements of electrophoretic velocity were made by the microscopic method with a modified Northrup-Kunitz (1925) apparatus after the design used by Mudd, Lucke, McCutcheon, and Strumia (1928) . Three radio B batteries, giving approximately 135 volts, were connected at each end of the cell to non-polarizable electrodes of zinc in saturated ZnSO,. The electro- phoresis cell was mounted with oil immersion contact over a Zeiss Wechsel- condensor. A single cell was used throughout. A 28 X Zeiss ocular and a Bausch & Lomb 8 mm. objective combined working distance with sufficient magnification. The apparatus is shown in figure i. \ Fig. I. The Northrup-Kunitz electrophoresis apparatus. Since velocities are given in ^/sec./volt/cm., the potential drop per cm. must be found. Following Ohm's Law, where E is the potential across the chamber itself ; L, the length of the cham- ber- Q, its cross section area; R, the specific resistivity of the fluid filling it; / tiie current ; and K, a constant dependent upon shape. For calibration,^ the cell is filled with mercury and placed in series with a dry cell, a rheostat, and a Weston standard ammeter. A potentiometer is connected across the cell by small platinum electrodes led into it. The P.D. and current are read 1 Thanks are due to Professor Charles Weyl for suggesting this method of caHbration. »' J .... If X 4 ^IV June, 1934I MOVER — EUPHORBIA 297 simultaneously and the resistivity of the mercury at room temperature is found from tables. Having found K, it is used, in practice, in the equation H = KRI where H is the P.D. per cm. R is measured by a Wheatstone Bridge; /. by a milliammeter, which can be introduced into the circuit ; and L is meas- ured directly. The cell was flushed several times with part of the buffer before the latex suspension was introduced. Care was taken to prevent the mclusion of air bubbles. The current and resistivity were measured at the start. Velocity was measured with a stop-watch by determining the time required for a latex particle in sharp focus to travel between two lines of the ocular micrometer and back again ; a Pohl mercury commutator was used to change the direction of migration when the particles reached the second line. Due to the charge which the glass wall assumes against the water and the consequent electroendosmotic streaming when the circuit is closed, it is neces- sary to measure the particle velocity at definite depths to eliminate this source of error. Smoluchowski (1921) has formulated these levels for flat cells. These " stationary " levels lie at 0.21 and 0.79 of the total depth of the cell. At least five readings were made at each level, and the mean of these was taken to calculate the velocity in />t/sec./volt/cm. Measurements were made at temperatures from 21° to 28° C. The tem- perature coefficient of velocity was found to be approximately 2 per cent per degree centigrade. This same coefficient has also been used by Abramson (1929, 1932) for protein-coated particles. To aid in comparison, all data were recalculated to a temperature of 25° C. Variations in the coefficient ap- pear to lie outside its experimental error, so that the same value was used throughout. Velocities were plotted against pH, all curves being drawn to the same scale. As shown by Abramson (1931), Freundlich and Abramson (1928), Henry (1931), and Sumner and Henry (1931), the Helmholtz-Lamb equa- tion, C HD (where H is P.D. per cm., D, dielectric constant, f, the electrokinetic poten- tial, rj, coefficient of viscosity, and V, the velocity — all units being electro- static), is valid for insulating particles. Using this equation and making certain assumptions for r; and D, the zeta potential can be calculated in milli- volts by multiplying the observed velocity by 12.6 (Northrup and Cullen, 1922). It was thought better to express the results simply in terms of mobility. Near the isoelectric point measurements of time are not very accurate, since the time needed to travel the required distance approaches infinity as 296 AMERICAN JOURNAL OF BOTANY [Vol. 21, (with the latex in it) by the quinhydrone method (Michaehs, 1926; van Harpen, 1929, i930- A "type-K" potentiometer and a saturated calomel cell were used in each case. The standards proposed by Clark (192b) were used to convert voltage to pH. Measurements of clectrophoretic velocity were made by the nncroscopic metho hi 0^ 9 i^ -L5 -2.0 0 0 .-2^ • » 3jO AO tA 3.0 4.0 5.0 §.0 pH. pH Fig. 2, 3. Fig. 2 (left). Mobility curves of latex particles from the section Trich- erostigma. Fig. 3 (right). From the section Tirucalli. shape set it apart from the other species just as it is set apart by taxonomy. Its protein content is relatively high, as shown by the ninhydrin reaction. Section XV, Poinsettia Three species of Poinsettia were investigated. The first species, E. puU cherrima Willd., was represented by three varieties: (i) the type, with ovate leaf and entire margin, and by two varietal forms; (2) red oak, named from its leaf shape; and (3) alba, named from its white bracts and petioles. All these forms are shown in figure 4 to be closely allied by the shape of their 2 Numbers and symbols represent the classification of Boissier. AMERICAN JOURNAL OF BOTANY ^,^^ A ■Mr-c-DJr' \-\J irillWIMAI. I Ih Klil/\-iNl |.vGl. 2I» curves. The I.P/s lie at pH 3.8 for E. pulcherrima and pH 3.9 for red oak and the white form. The curve of another species, E. dcntata Mich., is next to that of pulcherrhna. This is shown more clearly in figure 5, where the curves have been moved apart so that details may be observed. The LP. of June, 1934 J MOVER EUPHORBIA 301 + 1.5 +1.0 - +0.5 - • E. PULCHCRRMA o E. •■ MKR. OAK w C. » MMR. ALBA * C. DENTATA + E. HETEROPHYLUA -03 - • E PULCHERRIMA o E. " VAR.OAK w E. " VAR. ALBA + E. HETEROPHYLLA * E. DENTATA -1.0 - •2j0 Fig 45 Fig 4 (left). Mobility curves of latex particles from the section Poin- settia; all species plotted to the same scale. Fig. 5 (right). Curves spread out to show shapes. E. dentata lies at pH 3.9, close to alba. Although the curve of E, hetero- phylla L. has somewhat the shape of the others, its LP. is at pH 5.1, showing that its latex differs greatly from that of its supposedly closest relatives. It will be noted that as the LP. increases the curves become smoother. Protein reactions are high throughout this group. Section XIX, Diacanthiurn § I. Biaculeatae a. Splcndentes.—E. splendens Bojer is in some particulars similar to the other members, at least on the basic side of the LP., although it is not in the same taxonomic group (fig. 6). Its protein reactions are high and its LP. lies at pH 3.85. e, Trigonae.—E. lactea Haw. has an LP. (pH 3.8) close to that of E. splendens and a high protein reaction, but the characteristics of its curve are not very close to those of the other members of this group, E, grandicornis <^\ Goebel and E. grandidens Haw. Its curve on the basic side, however, does show some similarity to the others. E. grandicornis and E. grandidens are closely related taxonomically and have similar curves both in shape and in +1.0 •^0.5 + E. SPLENDENS A E. LACTEA o E. GRANDIDENS • E. GRANDICORNIS -03 •1.0 >l.5 -2.0 -2.5 ♦1.0 - -♦0.5 - -0.5 - 3^ 4.0 5i> e.0 pH -2.0 - -2.5 - Fig. 6, 7. Fig. 6 (left). Mobility curves of latex particles from the section Diacan- thium. Fig. 7 (right). From the section Tithymalus, sub-section Decussatae. I.P.'s (pH 3.45 and pH 3.5 respectively). E, grandicornis gives a medium protein reaction while the reaction of E. grandidens is low. Section XXII. Tinicalli E. tirucalli L., the sole member investigated in this group, gave a curve peculiar to itself. It has a relatively low LP. at pH 3.2 (fig. 3). Its protein content was low. Section XXVI . Tithymalus § I. Decussatae E. lathyris L. gives a strong protein reaction and also strong color reac- tions with iron. Molisch (1901) and Douin (1930) mention the presence of tannins in this form. In this respect it was unique among all the plants in- vestigated. Its LP. lies at pH 3.85. See figure 7. 302 AMERICAN JOURNAL OF BOTANY [Vol. 21, 9. Galarrhaei * Seeds smooth. t Capsule smooth or obscurely and minutely tuberculate, not verrucose E, pilosa L.3 represents a perennial species, while E. lagascae Spreng. is an annual. The two are close in respect to LP., which lies at pH 4.0 for E pi^ ♦1.0 - ♦ O.d o C. LAGASCAE • e. PILOSA -0.5 -1.0 -1.5 -2.0 - + 1.0 +0.5 • e.POLYCHROMA ♦ t.PLATYPHYLLA 2 u u •0.5 -1.0 -2.5 . 1 3.0 pH 4.0 -1.5 -2.0 - ■2.5 9U> 6.0 SO 4.0 pH SO 5.0 Fig. 8, 9. Fig 8 (left). Mobility curves of latex particles from the section Tithy- malus, sub-section Galarrhaei, with smooth capsules. Fig. 9 (right). From the group With warty capsules. ^ ^ losa and pH 4.1 for E. lagascae. On the basic side of the range both curves show a hump, but E. pilosa reaches a plateau on both sides before E. lagascae (hg. 8) . Protein reactions were low in this group. ttt Capsule covered by hemispherical, cylindrical or filamentous, elongated warts. ° Here, E. polychroma Kern is perennial while E. platyphyllos L. is annual rhe curves are of the same shape on the basic side (the acid side of the curve for E. platyphyllos could not be obtained, since it lies below pH 3 2) The magnitudes of the velocities of the two species are very different, however for the latex particles of E. platyphyllos move very slowly, while those of E polychroma flow faster than any other latex investigated (fig. 9). The I.P.'s 3 The name E pilosa L. was taken from the seed envelope ; unfortunately it could not be substantiated because the plants did not flower. Some uncertainty exists as to its specific Identity, although from all characters available it seems undoubtedly a member of this group. "^ iiiv-muci June. 1934] MOVER — EUPHORBIA 303 lie at pH 3.3 and pH 3.4 respectively. The ninhydrin tests gave a fairly strong protein reaction in the latex of both species. 10. Esulae **** Seeds irregularly pitted, marked or reticulate-rugose. E. segetalis L. and E. pinea L. are closely related taxonomically (E. pinea is a perennial, while E. segetalis is an annual). Their latex also behaves alike 1.5 ♦1.0 - ♦0.5 ■♦• E CYMkRISSIAS • E. ESULA 0 E.VIRCATA A E. SALICIFOUA X E. PINEA If E.SECETAUS -0.5 • 1.0 -1.5 -2.0 E MYRSINITES «-l.0 - ♦0.5 0 -0.5 • . V -1.0 - t o\ -1.5 -2.0 I 1 /sec/volt/cm \o x.^^ 0 -2.5 0 > U. so 4.0 5.0 6.0 SO 40 SO SO pH< pH Fig. ID, II. Fig. 10 (left). Mobility curves of latex particles from the section Tithymalus, sub-section Esulae. Fig. 11 (right). From the sub-section Myrsiniteae. (fig. 10), and their I.P.'s both lie at pH 3.3. Both species give a moderate protein reaction. ***** Seeds smooth. f Floral leaves free. In this group are E. virgata W. & Kit. (LP. at pH 4.0), E. cyparissias L. (ph 4.2), E. esula L. (pH 3.9), and £. salicifolia Host (pH 3.8).^ All these species have similar curve shapes with the same bends occurring in all four. E. salicifolia is slightly divergent, however. Protein reactions are low to me- dium in all four. Douin (1930) has shown by analysis that E. cyparissias has low protein content. II. Myrsiniteae * Seeds vermiculate-rugose. £. myrsinites L. (fig. 11) is not very closely related to the others, and although at first glance its curve might seem to be similar to the preceding 304 AMERICAN JOURNAL OF BOTANY [Vol. 21, June, 1934I MOVER EUPHORBIA 305 group, the position of the LP. at pH 3.4 does not indicate similarity. It gives a medium protein reaction. A table of the reactions is appended here for reference. Table i. Reactions of latex from different species N um- ber of Ninhydrin plants Species LP. test Xanth( Dproteic stud- Velocity HNO3 NH.OH ied at pH 5.9 pH i/sec/v/cm Tricherostigiua E. fulgens 4.3 + + + + -i-4- + 4-4-4-4- 2 — 1.06 Poinsettia E. pulcherrima 3.8 +4- + + -1-4-4-4- -f4--f-f 3 — 1.41 E. pulcherriiria var. oak 3.9 4-4-4-4- +++ -h4-4-4- 2 — 1.35 E. pulcherrima var. alba 3.9 4-4-4-4- +++ 4-4-f4- 2 — 1.25 E. dentata 3-9 4-4-4-4- — 1 — 4--i-4- 3 — 1.55 E. heterophylla 5.1 4-+4-4- +++ 4-4-4-4- 5 — 0.39 Diacanthium Splendentes E. splendens 3.85 4-4-4-4- +++ -f4-4- 2 — 0.54 Trigonae E. lactea 3.8 4-4-4-4- ++++ 4-4-4-4- I — 1.30 E. grandidens 3.5 4- 0 0 2 — 0.91 E. grandicornis 3.45 4-4- 4- + 4-4- I — 1. 10 Tirucalli E. tirucalli 3-2 4- 4- 4-4-4- 2 — 1.21 Tithymalus Decussatae E. lathyris * 3.85 4-h4-4- +4-4-4- 4-4-^4- 3 — 1.69 Galarrhaei Capsule smooth J- E. pilosa 4.0 4-4- 4-4- 4-4-4-4- 2 — 1.62 E. lagascae * 4.1 4- 4-4- 4-+4- 2 — 1.69 Capsule warty E. polychroiua 3.4 4-f4- -h4-4- 4-H-4-4- I — 273 • E. platyphyllos 3-3 4-h4- 4- 4--f- 2 — 1. 13 Esulae Seeds pitted ^^ E. segetalis * 3.3 4- 4- 4-4- 3 — 1.85 E. pinea 3.3 4-4-h 4-4- 4-4- I — 2.03 Seeds smooth E. virgata * 4.0 4-4-4- 4-4- 4--f4- 3 — 1.41 E. cyparissias 4.2 ochre -f-h4- 4-4-4-4- 6 — 1.38 E. esula 3-9 4--f4- 4-4- 4-4-4- 4 — 1. 21 E. salicifolia 3.8 4-4- 4-4-4- 4-4-4-4- I — 1.73 Myrsiniteae E. myrsinites 3.4 4-4-4- 4- 4- 2 — 2.25 * Identified by the Royal Botanic Gardens, Kew. o No reaction. 4- Very weak. 4-4- Pale color. 4-4-4- Good color. 4-4-4-4- Deep color. DISCUSSION The preceding results show that closely related plants have latex particles w^hose electrophoretic behavior and isoelectric points are similar or identical, while plants which are not members of the same taxonomic group have latices which differ in respect to these physico-chemical properties. As differences in electrophoretic behavior depend upon changes in the adsorbed surface of the particle concerned and not upon the size, shape, or composition of the particle itself, this indicates that the specific peculiarities of plants may extend even to the surfaces of their latex globules. This recalls the investigations of Reichert (1919), who found that plants could be grouped by the structure and chemical activity of their starch grains. More recently Bobilioff (1931) has formulated a key to Hevea clones by means of the color reactions of their latices. Calcium and magnesium salts are used to catalyze the activity of the natural oxidases present in Hezwa latex. The enzymes then bring on a re- sultant discoloration differing for each clone. Attempts to repeat these ex- periments with Euphorbia species were unsuccessful. They indicate, how- ever, the presence of other properties of latex which are specific in nature, in Hevea at least. In the earlier use of electrophoresis as a taxonomic tool, the prevalent prac- tice has been to group species of bacteria by a comparison of their mobilities in either an unbuffered medium, such as a physiological salt solution or dis- tilled water, or in buffers at a pH near neutrality. Such methods would give inadequate results in such an investigation as is here described, for a com- parison of the velocities at constant pH (as may be seen from the table) does not yield conclusive results concerning relationships of the latices. The iso- electric point affords a far better and more constant criterion, especially when accompanied by a complete curve of electrophoretic mobility with variation in pH. An examination of the curves shows that even within the smallest taxo- nomic groups, differences sometimes occur in the behavior of the latex par- ticles. A similarity is often shown between geographical distribution and these differences in electric charge. Considering first the poinsettias, it is shown in figure 5 that E. dentata has a curve which most closely follows E. pulcherrima and that the two varietal forms, red oak and alba, are more divergent. This indicates that E, dentata is more closely related to the type species, E, pidcherrima, than to the latter's varieties, which was to have been expected. The geographical range of E. pulcherrima is throughout Mexico and Central America. £. dentata is spread through the eastern United States, west to Ohio and Missouri, and south to Mexico. A third species, E. hetero- phylla, extends, however, from Illinois to Peru and Brazil (Boissier, 1866). Its curves and LP. show that in this respect, at least, it is not very closely related to the other two species in the poinsettia group which have been in- vestigated. 3o6 AMERICAN JOURNAL OF BOTANY IVoI. 21. June, 1934] MOVER — EUPHORBIA 307 In the section Diacanthium, the species investigated fall into two groups — the Splendentes (£. splendens, a native of Madagascar [Berger, 1907J ) and the Trigonae, three species of v^hich are shown in figure 6. Of these three, E, lactea, which is different both in shape of curve and in LP., is a native of the East Indies (Berger, 1907), while both E. grandidens and E. grandicornis, with similar curves and I.P/s, come from South Africa (Brown, 1925). Although E. pilosa and lagascae have nearly identical isoelectric points and curves which are to some extent alike, there are certain marked differences (fig: 8) which would not be expected in view of their close taxonomic rela- tionship. However, their geographical distribution differs markedly. E, pilosa spreads through North Spain, South France, South Germany, Silesia, Austria, Hungary, Italy, Rumania, Middle and South Russia to Siberia, while E, lagascae is restricted to the Mediterranean region. Central and South Spain, Sardinia, and the Canary Islands (Hegi, Beger, and Zimmermann, 1930). In the second group of the Galarrhaei (fig. 9) E. polychroma and E. platy- phyllos differ widely in the position of their mobility curves but not in their curve shapes or I.P.^s. It is interesting to note that these two taxonomically closely related forms have different distributions. E. polychroma extends through east and south-east Europe from South Poland through the Balkans, while E. platyphyllos is spread throughout South and Middle Europe from Great Britain and North Spain eastward to Middle and South Russia, the eastern Balkans, Asia Minor, and North Africa (Hegi, Beger, and Zimmer- man, 1930). The two m.embers investigated from the rough-seeded group of the Esulae {E. segetalis and E. pinea) show curves and I.P.'s (fig. 10) which are very much alike. Their geographical distributions are almost identical, ranging through the Mediterranean region and North Africa (Douin, 1930; Pax and Hoffmann, 1931). When this investigation was started, E. segetalis was called E, amygdaloides from the name on the seed envelope. E, amygdaloides belongs to a different group of the Esulae from E. pinea, and it seemed strange that the mobility curves and I.P.'s, when they were determined, should show a close relationship and not indicate this supposed difference. Specimens of '' E. amygdaloides " were identified by Kew Botanic Gardens as E, segetalis and the difficulty was removed. Of the smooth-seeded Esulae, E. cyparissias extends throughout Middle and South Europe and eastward to Siberia. Its northern limit is Cumber- land (England), Denmark, South and Middle Sweden, Lithuania, Latvia, Esthonia, and Middle Russia, while it is bounded on the south by Middle Spain, South Italy, Albania, Macedonia, and South Russia. E, esula ranges over virtually the same area. On the north it reaches Scotland, Denmark, South Sweden, Lake Ladoga in Finland, Novgorod, and the Onega Valleyi and on the south. North Spain, Middle Italy, the North Balkans, Rumania,' and Middle Russia. It also extends to western Asia. E, virgata is found from Czechoslovakia, Poland, Latvia, Lithuania, and Esthonia stretching n south-eastward through the North Balkans in the south and through Middle and South Russia on the north to Siberia, up to Dsungarei and western Asia. E. salicifolia extends from Bavaria eastward through the steppe region from Austria and Hungary, Galicia, and the North and Middle Balkans to Bulgaria and South Russia (Hegi, Beger, and Zimmermann, 1930). AH these four species have similar curves and I.P.'s. Of the four, E. salicifolia diverges most widely from the rest in geographical distribution and in curve shape. With few exceptions, species closely related and having similar geograph- ical distribution are marked by similar latex behavior ; between species having different distribution there is much less correlation. Furthermore, between members of different taxonomic groups there is little, if any, similarity in curve shapes or isoelectric points. These peculiarities can only be accounted for by differences in the surfaces of the latex particles. These surfaces all contain an ampholyte to some extent, for they all reverse their sign of charge as the pH is lowered. There are at least two groups of substances generally present in Euphorbia latex which are known to be amphoteric. These are the sterols and proteins. Cohen (1908) and Klein and Pirschle (1923) ^^ave investigated many species of Euphorbia and report the presence of sterols in the latex of all forms tested. Of the plants they investigated E. pidcherrima, E. cyparissias, E. lathyris, E. splendens, and E, myrsinites were found to contain these com- pounds. Klein and Pirschle also claimed that euphorbon, a resin constituent found throughout the entire genus, is a sterol, but neither Bauer and Schenkel (1928) nor Miiller (1929) could confirm this. Miiller, however, believes that sterols are present. Many authors, following Belgrave (1923)^ I'lave found sterols in Hevea latex and rubber (Whitby, Dolid, and Yorston, 1926; Bruson, Sebrell, and Vogt, 1927; Frey-Wyssling, 1929; van Harpen, 1931). There are but few references to the electrophoresis of sterols. Eagle (1930) finds the LP. of cholesterol to lie between pH 2.1 and pH 3.6, while Remesow (1930) gives pH 3.2 as the value of the LP. of the pure substance. Floccula- tion experiments by Rona and Deutsch (1926) on cholesterol show a maxi- mum flocculation at pH 2.4-pH 3.2. Van Harpen (1931) finds the floccula- tion optimum of the natural sterols in the acetone extract of Hcvca latex to be at pH 3.02. The I.P.'s of the proteins generally lie between pH 3 and pH 6 (Mudd, 1925; Pfeiffer, 1929; Tiselius, 1930), the range within which latex I.P.'s are found. As far as our present knowledge is concerned, there are several possibilities as to the structure of the particle surface. The surface for each species may be composed of a single specific protein. Any differences in curve shape would then be due to specific differences in the chemical or physical arrange- ment of these proteins. Abramson (1932) has shown that adsorption upon inert surfaces does not change or tie up any radicals which function in the electrophoresis of the free protein. Hence, if specific differences are ex- pressed by electrophoretic behavior, they must be due to a change in number 308 AMERICAN JOURNAL OF BOTANY tively, or all discharged. The at^na HH ^ ^^''•'" J^°^*^'^^'^' -" "ega- The I.P. could only be detern... '^ CtltTl'T ''' '"* ^° ^'- of the particles in the field move] tooZS.'TJu"^ approximately half tremely slow rate. This is an in icatirthf .. ''^ *° *^ "'^'^ ^^ -^ -- not composed of one homogen^^urproTein '^"'^'"^ '^ '"'^°'"P'^^^ ^ ^Ise In shape and in position of the T P r.h tt r. come closest to the typi al cJfve fT' . T ' '"^ ""• ''^'''^'^'^ ^PP^^^ to (Abramson. .9.8, ./.l Ster a d s e ^ ^^^^^^ ^ -.>e protein speces gave a high reaction for protein ' ' '93o)- All these seve^; protei'rorof ' (^ p'otl?" Tf"' T ""^ '^^^^ '"-^"^ of (0 yet known to be amphote ic' p ^tlTe urf?" T f'" ^°'"P°""^^' -^ a t'ons on the latex of Herea how thlt t, ,'^ *''" ^^'''''^''- ^"^estiga- Plete coating of protein (WhZ g o "h " ' ""'^' '^ ^^^^' '^^ ^ -- Stearn and Stearn (1031) and vf ; ?':' '9^°' ^^" ^^^P^"- mD Phoretic behavior of L?tls of two o/r^ '-e considered the ele'ctro- -ixture has an electrophoretic ^r be tw ^ t^ p"-^^^ ^ '"' '''' ^^'' The n,d.v,dual I.P.'s are not shown on tT. f' °^ "' constituents, interesting to notice that E. denZ\Z eT,T' °"'^ ''"" ^^^"'^^"^- '' - h>gh protein reactions and curves wWcb ^^ r""' ^'''^ '^^ ^^"^^ies yield ical curve for a mixture of two amphoivti: H ""'' "■'"''^'- *° ^'^^ ^'^'-t- g'ven by Stearn and Stearn. It T^s fwe It .." ""'"'' -"bination, as presence of approximately equa aC^ts f .,"'"''"' ''^"^ ^'^°" ^'^^ present in abundance in £ J/w,! , ''°"' ''"'"'''«' known to be P--t it is a constant^ot^:^'::;";!^ ^"""^- " '-' ^ ""-^ ^^ Ii the species with I P '« h^i^ u exception of £. myrsinitesandEl^ f' ''"" .'?'^' '' '' "°*^^ that, with the line with I.P.'s at pH 34) ",1 lff''''T" ^^°*^ '"^^ *''^ ^^itrary border This is seen in E. ..a«SL /l ^ ^- ' t ""^ '""^'^ ^^-^^^^ ^•■J<- ^-^^^o//., and E. pinea. AH ' these ""''""\f- ''''"^"^^'' ^- Platyphyllos, E. proteins. Their I'p.'s arf "l at ow^ ''Tf ' ""''"'" °^ '"^ test for te.ns and in that found for s ero s It f" K k? f " "^"^' ^^^^ ^^ P- or unknown non-protein amphols are T IV'^'' '" '''''' ^^^^ sterols In the case of E. polychroma to^nrT, ''^ '°"''"^ °^ ^'^^ Particles, these cause an altered reacTion '"' ''' '^'■^'^^'>' •^''^h^'-. and possibly The curves of E rnvr^inh^. ^u and £. /.,«.,,, ,„d £. £' ;t ffi^t' to " f • ^'^ ^'"-^'--ded Esulae, are extremely varied and their cur el: are "" •^'^^■7™*- -actions Shapes are very irregular, a feature not June, 1934I MOVER — EUPHORBIA 309 '>' ^ characteristic of pure protein. It is again possible that a mixture is here present. In general the results here obtained can be explained best by assuming that the latex particles of Euphorbia vary greatly in their surface structure as we pass from species to species but that the composition of the particles obtained from any one species is constant and speci ic, and resembles, in constitution, those of its nearest taxonomic neighbors. . SUMMARY 1. The isoelectric points and electrophoretic mobility curves for the latex particles (suspended in buffers) of twenty-one species of the genus Euphorbia have been determined by means of a Northrup-Kunitz electrophoresis ap- paratus. 2. The species investigated are: E. fidgens, E. piikhcrrima, E. pulcher- rima var. red oak, E. piilchcrrima var. alba, E. dcntata, E. hetcrophylla, E. splendens, E. lactea, E. grandicornis, E. grandidens, E. tirucalli, E. lafhyris, E. pilosa, E. lagascac, E. polychroma, E. platyphyllos, E. scgctalis, E. pinea, E. virgata, E. cyparissias, E. esida, E. salicifolia, and E. myrsinitcs. 3. These values of isoelectric points and mobility curves appear to be con- stant and specific for each species and do not depend upon environmental factors so long as specimens are in good, healthy condition. 4. These curves, representing the variation of the electrokinetic potential (plotted as velocity) with change of pH, can be grouped into families whose members have similar positions and shapes. 5. These families of mobility curves of latex particles correspond to the taxonomic groups already established for the genus. 6. If the isoelectric points of the latex particles of the several species are grouped according to the natural taxonomic arrangement, closely related plants are found to have isoelectric points at close or identical pH values. 7. A marked correlation was found between similarities in geographical distribution and curve shape. 8. Relations between species are not clearly shown when electrophoretic mobility values are taken at constant pH, since this method affords no indica- tion of the rest of the mobility curve or of the LP. 9. All plants investigated showed the presence of oxidases by the guaiacum reaction. 10. All plants studied gave an isoelectric point in the normal range on the pH scale in which protein isoelectric points occur. This is an indication that the surface of the latex particle is at least partially coated with protein in most cases. IT. Latex particles of Euphorbia behave as though some species are almost completely coated with a single protein, some with several proteins or mixtures of proteins with other ampholytes, possibly sterols, while other species seem to have an almost completely non-protein surface. 310 AMERICAN JOURNAL OF BOTANY [Vol. 21, June, 1934 J MOVER — EUPHORBIA 311 12. In spite of these variations between species, each species acts as though its latex particle surface is constant in composition and little influenced by individual variations. 13. The method of electrophoresis is suggested as a useful tool in study- ing the taxonomic relationships and latex particle structure of laticiferous plants. The author wishes to acknowledge his indebtedness to Professor Stuart Mudd for his most valuable advice and assistance, and to Professor Conway Zirkle for his criticism and suggestions regarding the manuscript. His thanks are especially due to Professor William Seifriz, under whose direction the present work was undertaken and whose interest and continued assistance made its completion possible. Botanical Laboratory, University of Pennsylvania, Philadelphia, Pennsylvania LITERATURE CITED Abramson, H. a. 1928. A new method for the study of cataphoretic protein mobility. Jour. Amer. Chem. Soc. 50: 390-393. . 1929. The cataphoretic velocity of mammalian red blood cells. Jour. Gen. Physiol. 12: 711-725. . 1931- The influence of size, shape and conductivity on cataphoretic mobility, and its biological significance. A review. Jour. Phys. Chem. 35: 289-308. 1932. Electrokinetic phenomena. VI. Relationship between electric mobility, ''^^\ ( N-. charge and titration of proteins. Jour. Gen. Physiol. 15: 575-603. Avery, O. T., W. F. Goebel, and F. H. Babers. 1932. Chemo-immunological studies on conjugated carbohydrate proteins. VII. Immunological specificity of antigens prepared by combining a- and i3-glucosides of glucose with proteins. Jour. Exper. ' Med. 55: 769-780. Eauer, K. H., and p. Schenkel. 1928. Zur Kenntnis des Euphorbiumharzes. Arch. Pharm. 266: 633-638. Belgrave, W. N. C. 1923. Studies on Hcvca latex. I. Coagulation. Malayan Agric. Jour. 11; 348-362. Beniasch, M. 1912. Die Saureagglutination der Bakterien. Ztschr. Immunitatsforsch. Orig. 12: 268-315. Berger, a. 1907. Sukkulente Euphorbien. Stuttgart. Beumee-Nieuwland, N. 1929. De coagulatie van Hevea latex. Arch. Rubbercult. Nederlandsch-Indie 13: 555-567 (Eng. summary). Bobilioff, W. 1919. Onderzoekingen over de vorming van latex bij Hcvca hrasiliensis. Arch. Rubbercult. Nederlandsch-Indie 3; 374-404. . 1931- Kleurreacties van latex als identificatiekenmerken van Hcvca-oXoono-n. Arch. Rubbercult. Nederlandsch-Indie 15: 289-308 (Eng. summary). Boissier, E. 1866. The tribe Euphorbieae. Prodromus systematis naturalis regni vege- tabilis, edited by A. de Candolle, 152: 3-188. Boom, B. K. 1930. Botanisch-serologische onderzoekingen. Diss. Wageningen. Brown, N. E. 1925. The genus Euphorbia. Flora Capensis, edited by Thiselton-Dyer, 52: 222-375. Bruson, H. a., L. B. Sebrell, and W. W. Vogt. 1927. Isolation of the natural oxida- tion inhibitors of crude Hcvea rubber. Ind. Eng. Chem. 19: 1187-1191. '^' Chapman, G. H. 1929. Electrophoretic potential as an aid in identifying strains of the B. coli group. Jour. Bacteriol. 18: 339-342. Clark, W. M. 1928. The determination of hydrogen ions. Third edition, Baltimore. Cohen, N. H. 1908. tJber Phytosterine aus afrikanischen Rubber. Arch. Pharm. 246: 515-520. De Kruif, p. H. 1922. Change of acid agglutination optimum as index of bacterial mutation. Jour. Gen. Physiol. 4: 387-393. DiNSMORE, R. P. 1926. Composition and structure of Hcvca rubber. Ind. Eng. Chem 18: 1140-1145. DouiN, R. 1930. The family Euphorbiaceae. In Flore complete illustree en couleurs, de France, Suisse et Belgique, edited by G. Bonnier. 10: 5-22. Paris. Eagle, H. 1930. Studies in the serology of syphilis. IIL Explanation of the fortifying effect of cholesterin upon the antigen as used in the Wassermann flocculation tests. Jour. Exper. Med. 52: 747-768. Falk, I. S. 1928. A theory of microbic virulence. In The Newer Knowledge of Bac- teriology and Immunology, edited by E. Jordan and I. S. Falk. Chicago. Fisher, H. L. 1930. The chemistry of rubber. Chem. Reviews 7: 51-138. Freundlich, H. 1922. Kapillarchemie. Leipzig. , AND H. A. Abramson. 1928. liber die kataphorctische Wanderungsge- schwindigkeit grober Teilchen in Solen und Gelen. II. Ztschr. Physik. Chem 133: 51-68. , AND E. A. Hauser. 1925. Zur Kolloidchemie der Kautschukmilchsiifte. Kolloid Ztschr. 36: 15-36. Frey-Wys SLING, A. 1929. Microscopisch onderzoek naar het voorkomen van harsen in de latex van Hci'ca. Arch. Rubbercult. Nederlandsch-Indie 13: 371-392 (Orig.); Eng. transl., 393-413. . 1932. Onderzoekingen over de verdunningsreactie en de bewegung der latex tijdens het tappen van Hcvca hrasiliensis. Arch. Rubbercult. Nederlandsch-Indie 16: 241-284 (Orig.) ; Eng. transl., 285-327. VAN Harpen, N. H. 1929. Voorloopige mededeeling over de coagulatieverschijnselen en waterstofionenconcentratie in de latex van Hevea hrasiliensis. Arch. Rubbercult. Nederlandsch-Indie 13: 44-60 (Orig.) ; Eng. transl. 61-77. . 1931. The electrometric determination of the hydrogen ion concentration in the latex of Hcvca hrasiliensis and its applicability to technical problems. Medan, Sumatra. Hauser, E. A. 1930. Latex. New York. Hegi, G., H. Beger, and W. Zimmermann. 1930. The family Euphorbiaceae. In Illustrierte Flora von Mitteleuropa, by G. Hegi. 5: 1 13-190. Miinchen. Henri, V. 1906. £tude de la coagulation du latex de caoutchouc. Compt. Rend. Soc. Biol. Paris 60: 700-703. Henry, D. C. 1931. The cataphoresis of suspended particles. I. The equation of cata- phoresis. Proc. Roy. Soc. London A, 133: 106-129. HoeffgeNj F. 1922. Serodiagnostische Untersuchungen ueber die Verwandschaftsver- haltnisse innerhalb des Columniferen-Astes der Dicotylen. Bot. Arch, i: 81-99. Klein, G., and K. Pirschle. 1923. Nachweis und Verbreitung der Phytosterine im Milchsaft. Biochem. Ztschr. 143: 457-472. Landsteiner, K., and J. VAN der Scheer. 1929. Serological differentiation of steric isomers (antigens containing tartaric acids). Jour. Exper. Med. 50: 407-417. Levene, p. a., and H. S. Sim MS. 1923. Calculation of isoelectric points. Jour. Biol. Chem. 55: 801-813. LoEB, J. 1922. The influence of electrolytes on the cataphoretic charge of colloidal particles and the stability of their suspensions. II. Experiments with particles of gelatin, casein and denatured egg albumin. Jour. Gen. Physiol. 5: 395-413. 312 AMERICAN JOURNAL OF BOTANY [Vol. 21. June, 1934] MOVER EUPHORBIA Mez, Carl. 1922. Anleitung zu sero-diagnostischen Untersuchungen fiir Botaniker. Bot. Arch, i: 177-200. , AND H. ZiEGENSPECK. 1926. Der Konigsberger serodiagnostische Stammbaum. Bot. Arch. 13: 483-485. MicHAELis, L. 191 1. Die Saureagglutination der Bakterien, insbesondere der Typhus- bazillen. Deut. Med. Wochenschr. 37: 969-971. . 1926. Hydrogen ion concentration. I. Translated by W. A. Perlzweig. Baltimore. MoLiscH, H. 1901. Studien iiber den Milchsaft und Schleimsaft der Pflanzen. Jena. MoRiTZ, O. 1928. Zur Kritik der Phytoserologie. Biol. Zentralbl. 48: 431-443. . 1929. Weitere Beitrage zur Kritik und zum Ausbau phytoserologischer Methodik. Planta 7: 758-814. 1932. Prinzipien und Beispiele der Anwendung phytoserologischer Methodik. 313 Planta 15: 647-697. — , AND H. voM Berg. 1931. Serologische Studien iiber das Linswickenproblem. Biol. Zentralbl. 51: 290-309. Morris, V. N., and H. W. Greenup. 1932. Rubber latex. Recent scientific and tech- nical developments. Ind. Eng. Chem. 24: 755-770. Morrow, C. A. 1927. Biochemical laboratory methods for students of the biological sciences. New York. MuDD, S. 1925a. Electroendosmosis through mammalian serous membranes. I. The hydrogen ion reversal point with buffers containing polyvalent ions. Jour. Gen. Physiol. 7: 389-413. . 1925b. Electroendosmosis through mammalian serous membranes. II. Com- parison of hydrogen ion reversal points with acetate and with citrate-phosphate buffers. Jour. Gen. Physiol. 9: 73-79. , B. Lucre, M. McCutcheon, and M. Strum ia. 1928. Methods of studying the surfaces of living cells, with especial reference to the relation between the surface properties and the phagocytosis of bacteria. Colloid Symp. Monographs 6: 131-138. , R. L. Nugent, and L. T. Bullock. 1932. The physical chemistry of bacterial agglutination and its relation to colloidal theory. Jour. Phys. Chem. 36: 229-258. MuLLER, J. A. 1929. Zur Kenntnis des Euphorbons aus Euphorbiumharz. Jour. Prakt. • Chem. 121: 97-112. NoRTHRUP, J. H., AND G. E. CuLLEN. 1922. An apparatus for macroscopic cataphoresis experiments. Jour. Gen. Physiol. 4: 635-638. , AND M. KuNiTZ. 1925. An improved type of microscopic electro-cataphoresis cell. Jour. Gen. Physiol. 7: y2<^y2J^. Pauli, Wo. 1920. Kolloidchemie der Eiweisskc.rper I. Dresden and Leipzig. , AND E. Valko. 1929. Electrochemie der Kolloide. Vienna. Pax, F., and K. Hoffmann. 1931. The family Euphorbiaceae. Die natiirlichen Pflanzenfamilien, edited by A. Engler and H. Harms, 2d Ed., 19c: 11-233. Leipzig Pfeiffer, H. 1929. Elektrizitat und Eiweisse. Dresden and Leipzig. Prausnitz, p. H., AND J. Reitstotter. 1931. Elektrophorese, Elektroosmose, Elek- trodialyse in Fliissigkeiten. Dresden and Leipzig. Reed, G. B., and B. G. Gardiner. 1932. Studies in the variability of tubercle bacilli. V. Acid agglutination and electrophoretic potential in Mycoh, leprae. Canadian Jour. Research 6: (i22-(y2,\. Reichert, E. T. 1919. A biochemic basis for the study of problems of taxonomy, heredity, evolution, etc., with special reference to the starches and tissues of parent- stocks and hybrid-stocks and the starches and hemoglobins of various species and genera. Carnegie Inst. Washington Publ. 270. M Remesow, I. 1930. Physikalisch-chemische Untersuchungen iiber den kolloidalen Zus- tand des Cholesterins, Cholesterinesters and Lecithins. II. Electrokinetische Mes- sungen. Das ^-Potential und die kataphoretische Wanderungsgeschwindigkeit der Cholesterinsole. Biochem. Ztschr. 218: 134-146. RoNA, P., AND W. Deutsch. 1926. Untcrsuchungen uber Cholesterin- und Lecithin- suspensionen. Biochem. Ztschr. 171: 89-118. Seifriz, W. 1928. The physical properties of protoplasm. In Colloid Chemistry, edited by J. A. Alexander 2: 403-450. New York. Sheppard, S. E. 1927. The electrical deposition of rubber. Trans. Amer. Electro- chem. Soc. 52: 47-82. , AND L. W. Eberlin. 1925. The electrodeposition of rubber. Ind. Eng. Chem. 17: 711-714. Shibley, G. S. 1932. Differentiation by cataphoretic velocity of fresh and old labora- tory strains of H. pertussis. Proc. Soc. Exper. Biol, and Med. 30: 31-32. Smoluchowski, M. von. 1921. Elektrische Endosmose und Stromungstrome. In Handbuch der Elektrizitat und des Magnetismus, edited by L. Graetz 2: 366-428. Leipzig. Stearn, a. E., and E. W. Stearn. 1931. Metathetic staining reactions with special reference to bacterial systems. Protoplasma 12: 435-464. Sumner, C. G., and D. C. Henry. 193 i. Cataphoresis. 11. A new experimental method and a confirmation of Smoluchowski's equation. Proc. Roy. Soc. London A, 133: 130-140. Tiselius, a. 1930. The moving boundary method of studying the electrophoresis of proteins. Nova Acta Reg. Soc. Sci. Upsaliensis 7, No. 4, 107 pp. Twiss, D. F. 1 93 1. Some considerations of latex processes. Trans. Inst. Rubber Indust. 6: 419-430. Vles, F. 1924. Considerations theoriques sur le point isoelectrique des ampholytes: leur application a la formation de complexes. Arch. Phys. Biol. 4: 228-254. Weber, C. O. 1903. The chemistry of india rubber. London and Philadelphia. Wells, H. G. 1929. The chemical aspects of immunity. 2d Ed. New York. Whitby, G. S. 1920. Plantation rubber. London. , J. DoLiD, and F. H. Yorston. 1926. The resin of Hevea rubber. Jour. Chem. Soc. London 129: 1448-1457. ; ' THE GURWITSCH RAYS BY WILLIAM SEIFRIZ Reprinted from The Science of Radiology 1933 THE GURWITSCH RAYS 413 CHAPTER XXV THE GURWITSCH RAYS WILLIAM SEIFRIZ TEN years ago, the Russian histologist, Alexander Gurwitsch, discovered that living cells give off radiant energy which will stimulate other tissue to more active growth. The existence of such a form of energy has been ques- tioned although it is supported by experiments of considerable diversity, many of which have been carried out in other than Russian laboratories. The original experiment of Gurwitsch (14), though apparently a simple one, is difficult to reproduce. For this and other reasons it has been the chief Fig. 102.— The original experiment of Gurwitsch with onion roots. (After Gurwitsch.) point of attack by critics. Historically, it deserves mention, though later ex periments are more convincing and hiore readily duplicated. The first experi- ment which led directly to the discovery of Gurwitsch rays was done with the roots of onions. One root was held in a horizontal position close to and pointing directly toward another held in a vertical position (fig. 102). After some hours the tip of the vertical root, the so-called receptor, was "fixed" (killed with chemicals), sectioned, and subjected to microscopic examination. It was found that in the receptor root there were more cells in the process of division on that side which had faced the sender than on the opposite side, indicating that the sender root had radiated some form of energy which accel- 412 My 'it 4^ ^k 4 ^ -t * ^ .4 crated cell division, or growth, in the receptor (fig. 103). The following table gives some results: Table 1 Mitotic cells Induced half of root Other half of root Difference in percentage Without intervening plate 292 770 464 2455 1422 231 634 312 2190 1200 +26 +21 +48 +12 +18 With quartz plate 934 476 733 395 +27 +20 In order to eliminate all effects possibly due to volatile oils given off by the onion which may have an influence on tissue, and further to ascertain the nature of the emanation, Gurwitsch placed glass and quartz plates between direction of na 'f Fig. 103. — Section of a root tip which has been radiated, show- ing greater cell division on the treated (right hand) side. (From Reiter and Gabor.) the sender and the receptor roots. In the case of quartz, the increase in cell multiplication in the receptor still took place (see table 1). Ultra-violet rays pass through quartz but not through glass. For this reason Gurwitsch thought the "vital" rays comparable to, if not identical with, ultra-violet rays. Subse- quent work (9) has proved this to be true, certainly as far as stimulation to growth is concerned and possibly in all other respects. The fact that the source of the "vital" rays is exceptional and distinct justi- fies giving them a distinguishing name, and we can do no better than honor 414 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 415 the discoverer by using his name to characterize the new rays. Gurwitsch himself called the rays "mitogenetic," because they increase cell division or "mitosis." It is quite possible that later experiments will show the Gurwitsch rays to have a much wider influence than merely a stimulation of cell division. Gurwitsch rays are by no means limited to the onion; they appear to be characteristic of all living matter under certain (metabolically active) con- ditions. That this is true is indicated by the long and varied list of tissues so far known to radiate this form of emanation. The following cells, tissues and organisms have been shown to be sources of radiant energy: bacteria (57, 30, 31), yeast (3), Hydra, the eggs of lower animals (13), plant seedlings (12), potatoes (26), beets, blood (51) of man, frog, and rat, cancerous tissue (29, 17, 27), muscle, nerve (1), the brain of young axolotls, a mash of Drosophila (fruit fly) larvae, a mash of tadpoles, and regeneration processes after amputation in tadpoles. Cells from all these varied sources radiate energy which increases the rate of cell division in other tissues. But not all tis- sues of an organism radiate equally well, nor do the tissues necessarily radiate at all times and under all conditions. The first work definitely locating the ex- act center of radiation was that done by Gurwitsch. He located the radiation center in the onion, and found it to be in the base of the bulb. The energy orig- inates there, not in the root, and travels down the root, where it is given off at the tip. Reiter and Gabor (43, 44) also found that in the tadpole the radia- tion center is located in definite tissues. Only a mash made from the head radiates, and not from any other part of the body. Crushing the tissue has in some cases no ill effect upon it as a source of radiant energy. (This is true only where the chemical reaction responsible for radiation is proteolytic.) Malignant tumors emit rays in the strongest fashion ; benign tumors yield less radiation. Dorfman and Sarafanow (unpublished work) found that unfertilized sea- urchin eggs radiate weakly, but when sperm cells are added, much more in- tense radiation results. An increase in radiation takes place also in artificial "fertilization" (parthenogenesis) with butyric acid. A very important feature of the radiation, which answers a question that must inevitably arise in connection with the transmission of the energy through tissue, is that of secondary radiation. It is the key to the whole problem. Without it there can be no induction effect, as surrounding tissue would rapidly absorb the radiation from within. Without secondary radiation the primary form would never reach the outside except where it is superficial in origin. Also, a cell may not in itself be capable of radiation but may be aroused to emanation activity by a neighboring radioactive cell. The sec- ondary radiation may be greater than the primary. This is not speculation but an experimentally proved fact. Gurwitsch was able to shut out the sec- ondary radiation, leaving the primary, and the induction effect was not present. Besides Gurwitsch, Potozky and Zoglina (37) have established the presence of secondary radiation, and also proved that the secondary radiation from yeast may be greater than the primary from a beating frog's heart (the heart was the primary source; a yeast culture radiated by the heart was the i * > f * secondary source; and a second yeast culture was the detector of the sec- ondary radiation). The number of known kinds of receptors has not, strange to say, in- creased with that of the senders. Yeast is the most satisfactory receptor. It has now fully replaced the onion root as a detector of the rays. An excellent detector is the epithelium of cornea (19, 34). Bacteria also respond well (53, 57). Zirpolo (59, 60) has found yet another, but less convenient form of receptor in seeds. He observed an increase in the rate of growth (germina- tion) of the seeds of plants (Brassica, Lactuca, etc.) which had been radiated with Gurwitsch rays from bacillus cultures. An average length of 14.4 mm. of the treated as compared with 10 mm. of the control seedlings was obtained. Zirpolo also added the fungus Penicillium to the list of sources of Gurwitsch rays. He found that the rate of segmentation of sea-urchin eggs is greatly increased when the eggs are subjected to radiation from Penicillium. Gurwitsch's discovery has been questioned not only by those who have done no experimental work in the field, but also by others who have performed some experiments in an endeavor to confirm or disprove his work. Rossmann (46) and Guttenberg (20) repeated the onion root experiment and obtained numerous negative results. Gurwitsch (15) admits the difficulty in technic in the onion experiment and prefers that investigators repeat the later experi- ments on yeast which are much more readily controlled. Richards and Taylor (45) find objection to the yeast experiment on the basis of the variations in making the cultures and in mathematical treatment. The latter criticism seems to have been met by subsequent work of Gurwitsch. To offset the opposition, there is a great amount of research in support of Gurwitsch. In addition to the Russian investigators, some of whom have worked in collaboration with Gurwitsch and some independently, there are Reiter and Gabor (43), Siebert (48), Loos (28), Rajewsky (40) and Heinemann (22) in five separate Ger- man laboratories, Magrou (29) in France, Maxia (32), Protti (39) and Zirpolo (59) in Italy, Wolff (57) in Holland, and a score or more of others. Efforts have been made in several American laboratories to duplicate and augment the work of Gurwitsch, but so far with little success. All work up to 1932 is given in a monograph on the subject by Gurwitsch (16) . A somewhat earlier but excellent review is also by Gurwitsch (15) . The work of the foreign investigators stands as an unequivocal substantia- tion of Gurwitsch's discovery. The negative results which have been published (21, 33, 45, 56) may be explained in several ways. It is much easier, espe- cially in work involving delicate technique, to obtain negative results than positive ones. An investigator who does not believe the thing he is trying to prove greatly lessens, at the outset, the likelihood of confirmatory results. Lack of familiarity with material, and the performing of but one experiment in a totally new field of research, leave an investigator in a rather weak posi- tion when he denies a decade of work by another worker. In nearly all cases where negative results have been obtained, the experimenter devised his own technic. Few have followed precisely the methods of Gurwitsch. The new factors introduced are probably one of the causes of the negative outcome of the experiments. 416 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 417 Among the Russian workers, other than Gurwitsch, Baron (2), and Frank (6, 7, 8) stand first. The chief contribution of the former is the discovery that yeast is an efficient sender and receptor of the vital rays. Baron succeeded in obtaining a sufficiently great increase in the growth of yeast cultures to be distinguished with the naked eye. Proof is, therefore, macroscopic as well as microscopic. The area of the treated culture is visibly larger than that of the control (the controls run very uniform), and the number of cells per unit area in the field when microscopically viewed, is considerably greater in the treated cultures. Baron made the addi- tional discovery that the yeast cultures must be used at a definite time in their life cycle. Salkind (47) has likewise demonstrated a rhythmic nature in the emanation from developing sea-urchin eggs, and Gurwitsch has stated that yeast cultures, after being transferred, must run for 12 to 15 hours at 25-28°C. for successful radiation. Twenty hours is too long. Only cultures which double or triple in cell number in four hours are suitable as detectors. More sluggish, or more active cultures do not respond. The situation is similar with the fruit-fly, Drosophila, which when used as a source of Gurwitsch rays must be taken at a definite time in its life cycle, namely, just before and during pupation of the lar- vae. Such facts as these, when unknown, become pitfalls for less experienced workers. Later work has revealed what is rather a general rule in biologic processes, namely, that frequent, successive, short exposures are more effective than one long continuous exposure. Intermittent exposure of yeast cultures to a ray source from which they are separated by a re- yolving, notched (fractioning) disc, increases the effect so that the time of ex- posure has been lowered in some cases to twenty seconds (16). The earlier method for ascertaining J- . . 1 , , , r whether or not acceleration of cell division had resulted from induction was the laborious one of counting the cells which show mitotic (dividing) figures (in onion tissue), buddW (t yeast) or an increase in cell number (in bacteria). This method has, in the case of yeast and bacteria, been replaced by a very neat technic. For yeast the Fig. 104. — The centrifuge-capillary test: left treated, right control. (From Gurwitsch.) f/) ' i4) procedure is as follows: the cells are radiated in open tubes or chambers, then 0.2 cc. of the radiated and the control cultures are accurately pipetted off into small test tubes which contain 1 cc. of fresh sterile beer-wort. The cultures are put into an oven at 25-28°C. for three or four hours, then killed with sulphuric acid and thoroughly shaken. Fine capillary ( "Myzetokrit" or hematocrit) tubes are filled, one containing the treated and one the control culture. The capillaries are then centrifuged, precipitating the yeast cells which pile up in the bottom and reveal the results of the experiment in a graphic manner (fig. 104). Bacterial cultures present a more difficult problem because of the minute- ness of the organisms. The following method, which may also be used for i|i|i|i|i|i|i|i|i|i|t Fic. 105. — The nephelometer. (From Gurwitsch.) yeast, accurately determines the slightest increase in growth. The method in- volves measuring the light effect of a turbid medium. The apparatus is known as a nephelometer; the form used was devised by Frank (8). It consists of two compensating photo-electric cells (C, C, fig. 105), con- nected to a "Null" electrometer (E) and housed in metal boxes (B, B'), with openings (0, 00, regulated by adjustable diaphragms (D, D'), between which two glass vessels (V, V), rest, the one containing the treated and the other the control culture. The bacterial cultures are suspensions of a definite and determinable degree of turbidity. The two cultures are illuminated lat- erally (L). The scattered light coming from them affects the two photo-elec- tric cells. Difference in turbidity, due to unlike concentrations of the two suspensions resulting from different growth rates of the treated and untreated cultures, is revealed by the photo-electric cells, or rather by movement of the electrometer which records whether or not a current is passing through the light-sensitive electric cells. This inequality in the light effects of the cultures is compensated for by adjustment of the vernier scales of the two 416 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 417 Among the Russian workers, other than Gurwitsch, Baron (2), and Frank (6, 7, 8) stand first. The chief contribution of the former is the discovery that yeast is an efficient sender and receptor of the vital rays. Baron succeeded m obtaining a sufficiently great increase in the growth of yeast cultures to be distinguished with the naked eye. Proof is, therefore, macroscopic as well as microscopic. The area of the treated culture is visibly larger than that of the control (the controls run very uniform), and the number of cells per unit area in the field when microscopically viewed, is considerably greater in the treated cultures. Baron made the addi- tional discovery that the yeast cultures must be used at a definite time in their life cycle. Salkind (47) has likewise demonstrated a rhythmic nature in the emanation from developing sea-urchin eggs, and Gurwitsch has stated that yeast cultures, after being transferred, must run for 12 to 15 hours at 25-28°C. for successful radiation. Twenty hours is too long. Only cultures which double or triple in cell number in four hours are suitable as detectors. More sluggish, or more active cultures do not respond. The situation is similar with the fruit-fly, Drosophila, which when used as a source of Gurwitsch rays must be taken at a definite time in its life cycle, namely, just before and during pupation of the lar- vae. Such facts as these, when unknown, become pitfalls for less experienced workers. Later work has revealed what is rather a general rule in biologic processes, namely, that frequent, successive, short exposures are more effective than one long continuous exposure. Intermittent exposure of yeast cultures to a ray source from which they are separated by a re- yolving, notched (fractioning) disc, increases the effect so that the time of ex- posure has been lowered in some cases to twenty seconds (16). The earlier method for ascertaining , whether or not acceleration of cell had resulted from mduction was the laborious one of counting the cells which show mitotic (dividing) figures (in onion tissue), buddinf (in yeast) or an increase in cell number (in bacteria). This method has, in the case of yeast and bacteria, been replaced by a very neat technic. For yeast the Fig. 104. — The centrifuge-capillary test: left treated, right control. (From Gurwitsch.) di ivision f/) ,i-i «» procedure is as follows: the cells are radiated in open tubes or chambers, then 0.2 cc. of the radiated and the control cultures are accurately pipetted off into small test tubes which contain 1 cc. of fresh sterile beer-wort. The cultures are put into an oven at 25-28°C. for three or four hours, then killed with sulphuric acid and thoroughly shaken. Fine capillary ("Myzetokrit" or hematocrit) tubes are filled, one containing the treated and one the control culture. The capillaries are then centrifuged, precipitating the yeast cells which pile up in the bottom and reveal the results of the experiment in a graphic manner (fig. 104). Bacterial cultures present a more difficult problem because of the minute- ness of the organisms. The following method, which may also be used for III III I I I 11 Fig. 105. — The nephelometer. (From Gurwitsch.) yeast, accurately determines the slightest increase in growth. The method in- volves measuring the light effect of a turbid medium. The apparatus is known as a nephelometer; the form used was devised by Frank (8). It consists of two compensating photo-electric cells (C, C, fig. 105), con- nected to a "Null" electrometer (E) and housed in metal boxes (B, B'), with openings (0, 0'), regulated by adjustable diaphragms (D, D'), between which two glass vessels (V, V), rest, the one containing the treated and the other the control culture. The bacterial cultures are suspensions of a definite and determinable degree of turbidity. The two cultures are illuminated lat- erally (L). The scattered light coming from them affects the two photo-elec- tric cells. Difference in turbidity, due to unlike concentrations of the two suspensions resulting from different growth rates of the treated and untreated cultures, is revealed by the photo-electric cells, or rather by movement of the electrometer which records whether or not a current is passing through the light-sensitive electric cells. This inequality in the light effects of the cultures is compensated for by adjustment of the vernier scales of the two i INTENTIONAL SECOND EXPOSURE 418 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 419 diaphragms, which are between the photo-cells and the cultures. The differ- ence in adjustment of the two diaphragms, and therefore in light-intensity of the two cultures, is determined from the vernier readings. The two ves- sels containing the cultures may be rotated, and rebalanced, thus checking one photo-cell against the other. The criticism, whether justified or not, that one cannot count and compare with precision the number of dividing cells in a tissue, or budding ones in a culture, is met by these indirect methods of measuring which wholly elimi- nate the personal element. Also, whatever objection there may be to the mathematical handling of the results, is eliminated as a sweeping criticism TO VACUUM PUMP AND GAS PREbbURC CONTROL QUARTZ RtCEPTACLL FOR MITOOENCTIC RAWATOR TOAMPUriER.LOUD SPEAKER AND TAPE RECORDER TRANbWVRENT CXJARTZ WINCIOV5 CONDENSING COLLECTOR 5URFACE. PHOTOCLECTRIC ACTIVE SURfACE TO ACCELERATING POTENTIAL Fig. 106.- -Two types of Geiger-Mueller counter-tubes for measuring Gurwitsch rays. (From Glasser and Seitz.) of the work by the simple fact that the two yeast cultures, treated and con- trol, in the measuring (hematocrit) capillaries, pile up to different heights. That an increase in cell number results from exposure to Gurwitsch rays can, in the face of such evidence, hardly be questioned. The experiments so far cited have all had to do with living material, both sender and receptor being growing cells. We can now proceed to a considera- tion of physical means of detecting, and physical and chemical means of producing the rays. The earliest attempts to produce a non-living detector in- volved the photographic plate. If the Gurwitsch rays are of the nature of ultra-violet light, they should readily affect a photographic plate. Reiter and Gabor (43) thought that they had observed such an effect, but Gurwitsch believes that the photographic plate is not a sufficiently sensitive detector of the rays. The photographic plate is particularly sensitive to ultra-violet light and is accumulative in its action, as attested to by the long astronomic ex- posures made of invisible stars. The photographic plate should, therefore, be especially well suited as a detector of the Gurwitsch rays, but so far this has not proved to be true. The difficulty possibly lies in the fact that tissues radiate only at definite periods in their life rhythm. If light is emitted periodically from a very weak source, the accumulative action on the photo- graphic plate may be insufficient to record it, while a more sensitive detector may catch one flash. There is also the fact that certain sources of rays, in particular yeast, require light in order to radiate, while the photographic plate requires darkness to record (36). Again, such difi&culties lead the in- experienced experimenter astray. When a problem involves the determination of weak light effects, the in- strument ultimately resorted to for measuring them is the photo-electric cell. The name, photo-electric cell, suggests that the instrument is one for meas- uring light by means of the electrical properties which light acquires when considered in terms of its ultimate unit, the quantum. For the determination of Gurwitsch radiation the photo-electric cell alone however is not sufficiently sensitive. It is therefore combined with a Geiger-Mueller counter tube. This counter is a glass tube in which an electric field of high potential is estab- lished (fig. 106). The voltage across the partial vacuum of the tube is raised to near the point of discharge. The tube has a quartz window under which is a photo-electric sensitive metal (aluminum or cadmium). Through the axis of the tube is a metal wire which is connected with an electrometer for measuring feeble currents. Quanta or units of energy in the form of, for example, ^ particles or roentgen rays will, on entering the tube, disrupt elec- trons from the metallic surface. These cause momentary ionization of the gas within the tube which results in sudden electric discharges. The latter may be amplified so as to produce sudden deflections of an electrometer, the number of which is proportional to the quantum number of the radiation. The extreme sensitivity of the Geiger-counter makes it susceptible to meas- uring the cosmic radiation and the radioactivity of the earth which may be sufficient to overshadow the effect of the Gurwitsch rays, though according to Frank, Gurwitsch and Siebert, the latter rays exceed cosmic radiation in intensity. The first work on a photo-electric means of detecting the Gurwitsch rays was done by Rajewsky (40, 41) in Frankfurt and shortly thereafter by Frank and Rodionov (10, 11) in Leningrad. Glasser and Seitz at about the same time used the Geiger-counter hoping to amplify the effects of the radia- tion to such a degree that they could be registered on an electrometer or a tape recorder and heard from a loud speaker. Their results are not yet sufficiently conclusive to justify publication. The following table from the work of Frank gives the number of deflections per unit of time caused by Gurwitsch rays. Table 2 Radiation source Frog sartorius Four frog hearts Six frog hearts Muscle (macerated) Time of ex- posure in minutes Number of deflections Without induction 6 8 5 11 6 5 4 With induction Effect in percentage 12 12 16 19 35 34 13 40 20 27 23 42 46 20 231 66 59 21 23 39 54 H 418 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 419 diaphragms, which are between the photo-cells and the cultures. The differ- ence in adjustment of the two diaphragms, and therefore in light-intensity of the two cultures, is determined from the vernier readings. The two ves- sels containing the cultures may be rotated, and rebalanced, thus checking one photo-cell against the other. The criticism, whether justified or not, that one cannot count and compare with precision the number of dividing cells in a tissue, or budding ones in a culture, is met by these indirect methods of measuring which wholly elimi- nate the personal element. Also, whatever objection there may be to the mathematical handling of the results, is eliminated as a sweeping criticism TO VACUUM PUMP AND GAS PREbbURC CONTPOL QUARTZ MICEPTACLL FOR MITOGENLTIC RADIATOR TOAMPUFIER.LOUD SPEAKER AND TAPE RECORDER TRANbPARENT QUARTZ \VIN0OV5 CONDENSING COLLECTOR SURFACE, PHOTOCLECTRIC ACTIVE. SURrACE. TOACCELERATING POTENTIAU Fig. 106.— Two types of Geiger-Mueller counter-tubes for measuring Gurwitsch rays. (From Glasser and Seitz.) of the work by the simple fact that the two yeast cultures, treated and con- trol, in the measuring (hematocrit) capillaries, pile up to different heights. That an increase in cell number results from exposure to Gurwitsch rays can, in the face of such evidence, hardly be questioned. The experiments so far cited have all had to do with living material, both sender and receptor being growing cells. We can now proceed to a considera- tion of physical means of detecting, and physical and chemical means of producing the rays. The earliest attempts to produce a non-living detector in- volved the photographic plate. If the Gurwitsch rays are of the nature of ultra-violet light, they should readily affect a photographic plate. Reiter and Gabor (43) thought that they had observed such an effect, but Gurwitsch believes that the photographic plate is not a sufficiently sensitive detector of the rays. The photographic plate is particularly sensitive to ultra-violet light and is accumulative in its action, as attested to by the long astronomic ex- posures made of invisible stars. The photographic plate should, therefore, be especially well suited as a detector of the Gurwitsch rays, but so far this has not proved to be true. The difficulty possibly lies in the fact that tissues radiate only at definite periods in their life rhythm. If light is emitted periodically from a very weak source, the accumulative action on the photo- i graphic plate may be insufficient to record it, while a more sensitive detector may catch one flash. There is also the fact that certain sources of rays, in particular yeast, require light in order to radiate, while the photographic plate requires darkness to record (36). Again, such difficulties lead the in- experienced experimenter astray. When a problem involves the determination of weak light effects, the in- strument ultimately resorted to for measuring them is the photo-electric cell. The name, photo-electric cell, suggests that the instrument is one for meas- uring light by means of the electrical properties which light acquires when considered in terms of its ultimate unit, the quantum. For the determination of Gurwitsch radiation the photo-electric cell alone however is not sufficiently sensitive. It is therefore combined with a Geiger-Mueller counter tube. This counter is a glass tube in which an electric field of high potential is estab- lished (fig. 106). The voltage across the partial vacuum of the tube is raised to near the point of discharge. The tube has a quartz window under which is a photo-electric sensitive metal (aluminum or cadmium). Through the axis of the tube is a metal wire which is connected with an electrometer for measuring feeble currents. Quanta or units of energy in the form of, for example, ^ particles or roentgen rays will, on entering the tube, disrupt elec- trons from the metallic surface. These cause momentary ionization of the gas within the tube which results in sudden electric discharges. The latter may be amplified so as to produce sudden deflections of an electrometer, the number of which is proportional to the quantum number of the radiation. The extreme sensitivity of the Geiger-counter makes it susceptible to meas- uring the cosmic radiation and the radioactivity of the earth which may be sufficient to overshadow the effect of the Gurwitsch rays, though according to Frank, Gurwitsch and Siebert, the latter rays exceed cosmic radiation in intensity. The first work on a photo-electric means of detecting the Gurwitsch rays was done by Rajewsky (40, 41) in Frankfurt and shortly thereafter by Frank and Rodionov (10, 11) in Leningrad. Glasser and Seitz at about the same time used the Geiger-counter hoping to amplify the effects of the radia- tion to such a degree that they could be registered on an electrometer or a tape recorder and heard from a loud speaker. Their results are not yet sufficiently conclusive to justify publication. The following table from the work of Frank gives the number of deflections per unit of time caused by Gurwitsch rays. Table 2 Time of ex- posure in minutes Number of deflections Effect in Radiation source Without induction With induction percentage Frog sartorius Four frog hearts Six frog hearts Muscle (macerated) 6 8 5 11 6 5 4 12 12 16 19 35 34 13 40 20 27 23 42 46 20 231 66 59 21 23 39 54 INTENTIONAL SECOND EXPOSURE 420 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 421 A rough average indicates that the intensity of the Gurwitsch radiation is about 1000 quanta per second per square centimeter, from which follows that the sensitivity of living cells, as measured by their susceptibility to Gurwitsch rays, is about equivalent to that of the human retina for light. The most recent work with photo-electric cells is that of Siebert and Seffert (50) who report positive results with an electron counter-tube to the extent of 30 per cent over the average spontaneous (terrestrial and other) radia- tion. The sources of the emanation were blood, carcinoma, urine and artifi- cial oxidation processes. Two counter cells were used, one being subjected to a mitogenetic source, and the other to an indifferent liquid. The effect of the radiation reached a maximum of 100 per cent with an average of 40 to 60 per cent, under favorable conditions, over the control. Rajewsky (42), using a counter sensitive to 9 X 10'^^ erg/cmVseconds gives 10"^^ — 10-* erg/cmVseconds as the strength of the Gurwitsch rays. A novel type of detector is that discovered by Stempell (53) who ob- served disturbances in Liesegang ring formation (in chromgelatin) due to radiation from cultures of Daphnia. His work met with criticism (49) but he later (54, 55) proved, by using cultures hermetically sealed in a quartz- covered chamber, that only radiation and not volatile oils or other chemical substances, produces the disturbances in periodic precipitation. While American scientists are still wondering whether or not the Gurwitsch rays really exist, Russian workers are busy determining the chemical nature of the reaction which produces them. The most natural assumption as to the nature of the source of the emanation is that it results from one of the chemical processes which is essentially peculiar to life. There is also the pos- sibility that some radioactive metal in protoplasm is responsible. Earlier experiments indicated that oxidation might be the source of the rays because anesthetics suppress their production; thus, onion bulb mash anesthetized with chloral hydrate emits no rays; potassium cyanide likewise stops emana- tion. Apparently, all oxidation processes, whether in living or in non-living material, produce the rays. The oxidation of ferrous to ferric iron, of stan- nous chloride, and of potassium dichromate, emanates rays. (The last re- action radiates more strongly and the second less strongly, than does living tissue.) It is of importance to realize that oxidation processes in both living and non-living systems are sources of the rays, which indicates, as Gurwitsch early realized, the probable identity of the "vital" rays with emanations of a kind from non-living material. The freshly cut surfaces of onion and potato are active sources of radiation due to contact of the exposed tissue with the oxygen of the air. It appears certain, therefore, that oxidation, including fermentation, is a source of the rays in both living and non-living material. Gurwitsch has interpreted the oxidation processes which give rise to emana- tions from the onion (only) as DuBois did the production of light in insects and molluscs. DuBois was able to isolate two substances, luciferin and lucif- erase, which, when in contact with each other, produce light. Luciferin, in contact with oxygen and the ferment luciferase, extracted from the light organ, goes over into oxyluciferin. In the same manner it was possible for Gurwitsch to extract from the onion two fractions which he has called, in analogy to the nomenclature of DuBois, "mitotin" and "mitotase." Only in the presence of mitotase does mitotin emit rays, probably by going over to oxymitotin. This has been confirmed by Loos (28). Oxidation is but one of the reactions which produces Gurwitsch rays. An- other, which has come more and more to the fore as a source of the rays, is glycolysis. Glycolysis, or the splitting of glucose into lactic acid, is a process common to many tissues. In muscle, it is, perhaps, more correctly referred to as glycogenolysis, where it involves the reduction of glycogen into lactic acid (through glucose) with further oxidation of a smaller part of the lactic acid into carbon dioxide and water (and conversion, by glyco- genesis, of the remainder of the lactic acid back into glycogen). There are three distinct proofs that glycolysis is a source of Gurwitsch rays. Pure lactic acid fermentation emanates the rays. Blood, protected from clotting, loses its power of radiation after standing one-half hour. If glucose is added, the blood will radiate for some ten minutes and then stop, only to start again if more glucose is added. It is possible to establish the loss of sugar in this re- action, which means that glycolysis has taken place. A carcinoma taken in- tact out of the body and put into Ringer solution, does not radiate. If glucose is added to the Ringer solution, intense radiation immediately begins (38). Marked glycolysis under these conditions has been established by Warburg. The emanation spectra of all three of these reactions in living tissues have been studied and found to agree in full. A third type of chemical process in tissues which gives rise to Gurwitsch rays is proteolysis, or the breaking down of proteins into their soluble de- composition products, notably peptones, as in the case of the digestion of albumin by natural stomach juices. L. Gurwitsch (18) found that emana- tion due to hunger inanition is of the proteolytic type. Considerable work has been done on digestion as a source of the rays. Karpas (25) was the first to obtain radiation from the digestion of albumin. L. Gurwitsch (18) then followed with a spectrum analysis of emanations from the digestion of serum albumin by pepsin. Billig, Kannegiesser, and Solowjew (5) deter- mined the emanation spectrum of the digestion of albumin by pepsin. This work indicates the chemical nature of one of the non-living sources of the rays. The final problem to be considered is the physical nature of the emanation. It has already been suggested that the rays are similar to, if not identical with, ultra-violet light. The first experiments of Gurwitsch indicated this, for he found that the emanation would penetrate quartz but not glass. That the rays lie in the ultra-violet region of the spectrum is now generally accepted, but the exact part of the spectrum has only very recently been accurately estab- lished by spectrographic analyses. The earlier and less exact spectrum studies of Frank (7) consisted in subjecting the sartorius of a frog to electrical or mechanical stimulation, for only muscle in tetanus emits Gurwitsch rays. The radiation so produced was directed through a quartz prism and the spectrum thus formed allowed to fall upon a series of three to six yeast cuhures in agar blocks (fig. 107). The spectrometer records which spectral lines each culture receives. Frank found that muscle in tetanus emanates rays within the wavelengths 2000- 2500 A. Other senders were similarly investigated and the spectral region of I I 'I 422 THE SCIENCE OF RADIOLOGY t ..A f« Aif^er with the type of chemical reaction which the emanations found to ditter witn me tyP" , ^ glycoly- produces them. Oxidation gives rays between 2240 and 2290 A., and g y y sis between 1900 and 1970 A. „ct=V,lUh the limits of In the earlier experiments, it was not possible to establish the limits THE GURWITSCH RAYS 423 \ F,c. I07.-Spectrographic analysis, (a) source of the rays (in tW«^~ ^scle^ (b) prism; (c) yeast cultures in agar blocks each exposed to a different pa emanation spectrum. (From Gurwitsch.) the spectra more accurately than within 100 A. Newer work (24, 35J^^^^^ has resulted in the development of a spectrum atlas ^^^ f;;f ^J^^^^^^ range of the emanations to within 10 A. An adjustable optical slit, which JioL«ma>i»aMX0 22M^^meisoiyioo»^ F,r 108 -Six spectra of various Gurwitsch-ray-producing reactions: (A) glycolysis (in lac^cacfd fementation, blood, and carcinoma); (B) oxidation; C) peptic digestion ind auto vsisMD) nerve; (E) muscle; (F) splitting of nucleic acid by nuclease it wn be noticed A) that the spectra from the three separate sources lactic acid, fermemation bl od and carcinoma, agree, and fonn one common ^r>eclr-moi ^j h.nd^ ifwiH also be noticed that the spectra obtained from nerve and muscle (D and E) are , nTo! identica" and that each contains the five bands of the glycolysis spectrum A) ^nU^^f the broad band of the oxidation spectrum (B), showing that the radiation :?VrUsltysrsilated nerve and muscle is ^ue Pnmarjy^ to glycolysis, but also in part to oxidation (and proteolytic reactions). (From Gurwitsch.) selects waves of a definite length, i. inserted in a spectrometer. The living material, which is to act as sender, is placed in front of the prism and the latter di perses the emanation of the former into its spectrum. A yeast cul ure, acring as detector, is then placed back of the slit. By exposmg one culture H after another to successive 10 A. regions, the spectrum limits of that particu- lar material acting as sender are determined. Thus have the ultra-violet spectra of numerous sources been monochromatically investigated. Table 3 Height of culture in "Wavelength in A. hematocrit capillaries Induction effect in percentage Control Induced 2320-2330 15 15 0 2330-2340 20 23 15 2340-2350 20 26 30 2350-2360 19 19 0 • 2360-2370 18 17 ? (within experimental error) 2380-2390 12 16 33 2390-2400 21 26 24 Billig and his collaborators (5) obtained the above results on yeast cul- tures radiated by serum albumin in the process of digestion by dog stomach juices (fig. 108). In the face of such experimental evidence it is extraordinary that the existence of the Gurwitsch rays should be questioned. The remarkable thing about them is not that they have been discovered but that their presence was not suspected long ago. In these days when knowledge of our radiation en- vironment increases in leaps and bounds, from a primitive recognition of the beneficial effect of the sun's rays, to the scientific study of roentgen rays, radium emanations, and cosmic rays, it is unreasonable to regard living matter, which is infinitely more complex than any of these other emanating substances, as less likely to be a source of radiant energy than they. The complexity of protoplasm is not in itself sufiScient evidence of radioactivity, but it does leave one more ready to suspect it of being thus active. Let us approach the hypothesis that living matter gives off radiant energy from another angle. One of the chemical constituents of protoplasm is potas- sium. Another metal which has been found in plants is rubidium. Both potas- sium and rubidium are slightly radioactive. Knowledge of the presence of these elements in living matter is of long standing. Recent work at the Rus- sian Radium Institute has shown that radium, in almost infinitesimal amounts, has been found in living plants and animals. Two species of duckweed scooped off the surface of a pond showed the presence of over fifty-six times as great a concentration of radium as was found in the water. Is protoplasm less radioactive than the radioactive elements which it contains? It may be, but a priori reasoning is against it. Every chemical reaction involves a transfer of energy which may manifest itself in the form of heat, light, or an electrical potential. The measurement of oxidation-reduction potentials is now common laboratory practice. The emanation of invisible light is not essentially different from the giving off of visible light. The chemist is quite familiar with emanations of a fluorescent nature when substances, such as sodium chlorid crystals, are exposed to 424 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 425 roentgen rays. Fluorescence also occurs when sodium is burned in the pres- ence of chlorine. Chemoluminescence is an ever growing chapter in present day chemistry. Emanations, therefore, are not uncommon in relatively sim- ple chemical reactions. Wolff and Ras (58) found that Gurwitsch emana- tions (i.e., ultra-violet light) are given off by such simple reactions as the neutralization of sodium hydroxid by acid. Opposition to the Gurwitsch rays appears to rest not so much on objection to the experimental work, though there is such objection, but rather on the faulty belief that admission of such a source of radiant energy is tacit ad- mission that protoplasm is possessed of extraordinary "vital" powers. While it must be admitted that protoplasm is alive, the Gurwitsch rays do not prove this. A thoroughly mechanistic interpretation of that most complex and baffling of all substances, as the simplest possible mixture conceivable, still allows for radioactivity. The emanation of radiant energy in the form of ultra-violet light by living matter is no more remarkable than the giving off of heat, light, and elec- tricity by living animals. These are of such common experience that we un- hesitatingly admit their existence, for we know of heat from our own bodies, light from the firefly and the glow-worm, and of electrical shocks from the ray-fish. It is but a step from heat, light, and electricity to such other forms of radiant energy as ultra-violet rays, roentgen rays, radium emanations, and cosmic rays. While certain of these manifestations of energy are less well understood than others, and in this sense are more mysterious, yet none is vitalistic except in so far as it is given off by living matter. From every point of view, each of them is as physical, as materialistic as are the others, and this includes the Gurwitsch rays which are but ultra-violet rays coming from living matter. The presence of radiant energy in protoplasm is one of the least mysterious things about it. We need only think of movement, growth, and reproduction, to realize this. This review of the work of Gurwitsch and his supporters is admittedly a favorable one, as are, in the main, the comments of a number of other physi- ologists, notably Beutner (4), Hollaender (23), and Spek (52). A favorable viewpoint is based on the fact that fully 90 per cent of the total number of experiments have yielded positive results, and the fact that the experimental results are justified by our present knowledge of protoplasm. Had this review been written by a less sympathetic writer, more doubt could have been thrown on the results. But one need only to recall the history of science and the basis of most criticisms to realize how easy it is to doubt experimental find- ings. Some of the doubts cast upon the work of Gurwitsch will have to be met in time, if the work is to hold. One such doubt deserving serious con- sideration is the following. A type of radiation so weak that it cannot be readily detected by a photo-electric cell, will, after dispersal (diffraction) by the prism of a spectograph, be so thinly spread out that detection of its component (spectral) parts would be beyond even the sensitivity of living matter. This criticism can be met with the simple statement that experiments by Gurwitsch, Frank, and others show that this is not true. We may theorize but we have always the experiment to consider. The criticism can also be tentatively answered by calling attention to a fact often not realized, namely, >f ) V Vfl .4,t 4(^ that living matter, while apparently very resistant at times, is at other times sensitive beyond all expectation. Plants deal with salt concentrations which the chemist cannot detect. As for the photo-electric cell, delicate as it is, it is not as sensitive to light as is the human retina. In theory it is, but methods of construction and extraneous disturbing influences keep it from being so ex- perimentally. Gurwitsch employed the experimental method and it is on this basis that his work must be judged. Note: I am indebted to Professor A. Gurwitsch, Dr. Otto Glasser, and Dr. L. S. Moyer for cooperation in the writing of this review. REFERENCES 1. Anikin, A., Das Nervensystem als Quelle mitogenetischer Strahlung. Roux' Arch. 108:609,1926. 2. Baron, M., Ueber mitogenetische Strahlen bei Protisten. Roux' Arch. 108 :617, 1926. 3. Baron, M., Bakterien als Quelle mitogenetischer Strahlung. Zbl. Bakter. 73:373, 1928. 4. Beutner, R., Physical chemistry of living tissues. Baltimore, 1933. 5. Billig, E., Kannegiesser, N., and Solowjew, L., Die Spektralanalyse der mito- genetischen Strahlung bei Pepsinverdauung und bei der Spaltung von Glycyl-Glycin durch Erepsin. Z. physiol. Chem. 210:220, 1932. 6. Frank, G., Das mitogenetische Reizminimum und -maximum und die Wellenlaenge mitogenetischer Strahlen. Biol. Zbl. 49:129, 1929. 7. Frank, G., and Popoff, M., Die mitogenetische Strahlung des Muskels und ihre Verwertung zur Analyse der Muskelkontraktion. Pfluegers Arch. 223:301, 1929. 8. Frank, G., Ueber die Erforschung mitogenetischer Strahlung mittels einer neuen nephelometrischen Methode. Biol. Zbl. 52:1, 1932. 9. Frank, G., and Gurwitsch, A., Zur Frage der Identitaet mitogenetischer und ultra- violetter Strahlen. Roux' Arch. 109:451, 1927. 10. Frank, G., and Rodionow, S., Ueber den physikalischen Nachweis mitogenetischer Strahlung und die Intensitaet der Muskel strahlung. Naturwiss. 19:659, 1931. 11. Frank G., and Rodionow, S., Physikalische Untersuchung mitogenetischer Strahlung der Muskeln und einiger Oxydationsmodelle. Biochem. Z. 249:323, 1932. 12. Frank, G., and Salkind, S., Die Quellen der mitogenetischen Strahlen im Pflanzen- keimling. Roux' Arch. 108:596, 1926. 13. Frank, G., and Salkind, S., Die mitogenetische Strahlung der Seeigeleier. Roux' Arch. 110:626, 1927. 14. Gurwitsch, A., Die Natur des spezifischen Erregers der Zellteilung. Roux' Arch. 100:11, 1923. 15. Gurwitsch, A., Ueber den derzeitigen Stand des Problems der mitogenetischen Strahlung. Protoplasma 6:449, 1929. 16. Gurwitsch, A., Die mitogenetische Strahlung. Berlin: Springer, 1932. 17. Gurwitsch, A., and Gurwitsch, L., Die mitogenetische Strahlung des Carcinoms II. Z. Krebsforsch. 29:220, 1929. 18. Gurwitsch, L., Die mitogenetische Spektralanalyse. IT. Die mitogenetischen Spektren des Carcinoms und des Cornealepithels. Biochem. Z. 236:425, 1931. 19. Gurwitsch, L., and Anikin, A., Das Cornealepithel als Detektor und Sender mito- genetischer Strahlung. Roux' Arch. 113:731, 1928. 20. von Guttenberg, H., Zur Theorie der mitogenetischen Strahlen. Biol. Zbl. 48:31, 1928. 21. Haberlandt, G., Ueber "mitogenetische Strahlung." Biol. Zbl. 49:226, 1929. 22. Heinemann, M., Cytagenin und "mitogenetische Strahlung" des Blutes. Klin. Wschr. 11:1375,1932. 23. Hollaender, A., and Schoeffel, E., Mitogenetic rays. Quart. Rev. Biol. 6:215, 1931. r 426 THE SCIENCE OF RADIOLOGY THE GURWITSCH RAYS 427 I 24. Kannegiesser, N., Die mitogenetische Spektralanalyse I Biochem. Z. 236:415, 1931. 25. Karpass, A., and Lanschina, M., Mitogenetische Strahlung bei Eiweissverdauung. (Dritte Quelle der mitogenetischen Strahlung). Biochem. Z. 215:^^7, l^z^. 26. Kisliak-Statkewitsch, M., Das mitogenetische Strahlungsvermoegen des Kartotfellep- toms.Roux' Arch. 109:283, 1927. , . ^, ^ r - i 7 ICr.h^ 27. Kisliak-Statkewitsch, M., Die mitogenetische Strahlung des Carcinoms. 1. Z. Krebs- forsch. 29:214, 1929. . , ^ ,, ., . t> , 7o./:n 28. Loos, W., Untersuchungen ueber mitogenetische Strahlen. Jb. wiss. Bot. /Z.Oll, 29. Magrou, J., Radiations mitogenetiques et genese des tumeurs. Comptes Rendus Acad. Sci., Paris 184:905, 1927. , r> • / • lo 30. Magrou, J., and Magrou, M., Action a distance du ^«^^^?"f^i"^^^^^^^^^ '"'^ ^^ developpement de I'oeuf d'Oursin. Comptes Rendus Acad. Sci., Paris 186:802 1928. 31. Magrou, J., and Magrou, M., and Choucroun, F., Action a distance du Bactenum tumefaciens sur le developpement de Toeuf d'Oursin. Comptes Rendus Acad. Sci., Pans 188*733 1929. 32. Maxia,' C, Conferma delF esistenza delle radiazioni mitogenetische del Gurwitsch demonstrata sui blostomeri di Paracentrotus lividus. Monit. zool. ital. -JO^llS, 1929. 33. Moissejewa, M., Zur Theorie der mitogenetischen Strahlung. Biochem. Z. Z4l:l, 1931. , , . r . 34. Naville, A., Action des rayons mitogenetiques a travers un ecran de quartz. Uomptes Rendus Soc. Phys. Genova 46:128, 1929. 35. Ponomarewa, J., Die mitogenetische Spektralanalyse. III. Das detaillierte glykoly- tische Spektrum. Biochem. Z. 239:424, 1931. , . , u 36. Potozky, A., Ueber Beeinflussung des mitogenetischen Effektes durch sichtbares Licht. Biol. Zbl. 50:712, 1930. 37. Potozky, A., and Zoglina, I., Ueber mitogenetische Sekundaerstrahlung aus abge- schnittenen Zwiebelwurzeln. Roux' Arch. 114:1, 1928. 38. Potozky, A., and Zoglina, I., Untersuchungen ueber die mitogenetische Strahlung des Blutes. Biochem. Z. 211 :352, 1929. 39. Protti, G., Impressione fotografiche di radiazione ematiche ottenute attraverso il quarzo. Comm. Sed. Sci. Ospedale Civile Venezia, 1930. 40. Rajewsky, B., Eine neue Messanordnung fuer kleinste Lichtintensitaeten. Strahlen- therap. 39:194, 1930. 41. Rajewsky, B., Ueber einen empfindlichen Lichtzaehler. Phys. Z. 32:121, 1931. 42. Rajewsky, B., Zur Frage des physikalischen Nachweises der Gurwitsch-Strahlung. In Zehn Jahre Fors'chung auf dem physikalisch-medizinischen Grenzgebiet. Ber. Inst. Phys. Grundlagen Med. (F. Dessauer) , Frankfurt, Leipzig, 403, 1931. 43. Reiter, T., and Gabor, D., Zellteilung und Strahlung. Berlin, 1928. 44. Reiter, T., and Gabor, D., Der heutige Stand des Problems der Gurwitsch-Strahlen. Arch, exper. Zellforsch. 11:21, 87, 1931. 45. Richards. O., and Taylor, G., Mitogenetic rays— a critique of the yeast detector method. Biol. Bull. 63:113, 1932. 46. Rossmann, B., Untersuchungen ueber die Theorie der mitogenetischen Strahlen. Roux' Arch. 113:346, 1928. 47. Salkind, S., Ueber den Rhythmus der mitogenetischen Strahlung bei der Entwick- lung des Seeigeleis. Roux' Arch. 115:360, 1929. 48. Siebert, W. W., Ueber die Ursachen der mitogenetischen Strahlung. Biochem. Z. 202:123, 1928.' . 49. Siebert, W. W., Das Stempell-Phaenomen an den Liesegangschen Rmgen. Biochem. Z. 220 :487, 1930. 50. Siebert, W. W., and Seffert, H., Physikalischer Nachweis der Gurwitsch Strahlung mit Hilfe eines Differenzverfahrens. Naturwissenschaften 21:193, 1933. 51. Sorin, A., Zur Analyse der mitogenetischen Induktion des Blutes. Roux' Arch. 108:634, 1926. 52. Spek, J., Allgemeine Physiologic der Entwicklung und Formbildung. In Lehrbuch der allgemeinen Physiologic, edited by E. Gellhorn. Leipzig, 508, 1931. 53. Stempell, W., Nachweis der von frischem Zwiebelsohlenbrei ausgesandten Strahlen durch Stoerung der Liesegailgschen Ringbildung. Biol. Zbl. 49:607, 1929. 54. Stempell, W., Ueber Organismenstrahlung. Strahlentherap. 40:777, 1931. 55. Stempell, W., and Romberg, D., Weitere Versuche ueber die Wirkung der Organis- menstrahlung und Gasung, besonders derjenigen des Karizinoms und des ermuedeten Muskels auf die Liesegangschen Ringe. Biol. Zbl. 52:413, 1932. 56. Taylor, G., and Harvey, E., The theory of mitogenetic radiation. Biol. Bull. 61 :280, 1931. 57. Wolff, L., and Ras, G., Einige Untersuchungen ueber die mitogenetischen Strahlen von Gurwitsch. Zbl. Bakt. I Abt. Orig. 123:257, 1932. 58. Wolff, and Ras, G., Ueber Gurwitschstrahlen bei einfachen chemischen Reaktionen. Biochem. Z. 250:306, 1932. 59. Zirpolo, G., Ricerche sulle radiazione mitogenetiche. Boll. Soc. Nat. Napoli 42:169, 1930. 60. Zirpolo, G., Nuove ricerche sulle radiazioni mitogenetiche. Comm. XI Cong. Internal. Zool. Padova, 4, Sept., 1930. Reprinted from Science, October 20, 1933, Vol. 78, No. 2025, pages 361-363. MORE ABOUT THE SPIRAL HABIT Under the title, "Twisted Trees and the Spiral Habit," I recently published^ evidence of considerable variety indicating that spiral movement and develop- ment among organisms are expressions of a wide- j spread tendency which is protoplasmic in origin. Barely had the manuscript left my hands than I 1 yciENCE, January 13, 1933. realized that I had failed to carry that part of m: discussion dealing with twisted trees to the individual wood cell rather than stopping at the cotton fiber. Il had in mind at the time the work of Scarth.^ Befon taking this up, I should like to turn for a moment t( other examples of the spiral habit which have beei brought to my attention as a result of the first account. 2 Trans. Roy, Soc. Can., Sec. V, 269, 1929. 362 SCIENCE Vol. 78, No. 2025 Boctober 20, 1933 SCIENCE 363 In listing the articles which have appeared in Science on the twisting of tree trunks, I overlooked the one by Koehler,^ in which he states with emphasis that twisted grain is not due to prevailing winds act- ing on asymmetrical crowns, because there is no evi- dence within the tree trunk that actual twisting took place after the wood was formed. The twisting of vines and tendrils around their sup- ports is common knowledge to all, but possibly it is not generally known that both may change their direction of twist several times between the base and the tip. The tropical liane Bauhinia may change its direction of twist six times within four feet, or within nine complete revolutions. Mr. L. F. Brady, of Mesa, Arizona, has been kind enough to send me photographs showing twisting in the stems of the cactus, Chamaecereus sylvestrii. Out of twelve stems on a single plant, four show a clock- wise twist, six a counter-clockwise twist and two are straight. In general, Brady says, the twist is to the right. Echinocereus also shows twisting of the stem. The spines of Neomammillaria are arranged in perfect spirals. It is reasonable to see in the wedge shape of long cambium cells an explanation of the twisting of tree trunks. The slippage or sliding growth of cambium cells would bring about a spiral twist. But such development can not serve as an explanation of the twisting of individual cotton fibers or of the walls of a single bast cell. Also, as the spiral habit is equally characteristic of animals from the lowest to the high- est, the ultimate cause, if there is a universal one, must be protoplasmic in character. A brief reference^ was made to the spiral nature of certain body organs (the gall duct) in man. My attention has since been called to the work of F. T. Lewis,* who, in an article on symmetry in plants and animals, presents evidence of a pronounced tendency among body organs in mammals to show a marked right or left twist. The cardiac loop in man rotates dextrally. Human viscera do likewise. (The primary loop of intestines may rotate sinistrally, but rarely so.) The coiled colon of the pig is dextrally wound through three or four complete revolutions. The trachea and esophagus show dextral rotation. There is a dextral spiral trend of muscle fibers throughout the digestive tube. Returning now to the spiral character of wood ele- ments, there is the work of Scarth,^ already referred to and illustrated in Fig. 1. The figure is of a por- tion of a single wood cell showing part of the wall. The wall is built up of concentric layers (20 in num- ber and about 0.5 m- thick). Each comprises a parallel 3 Science, May 1, 1931. ^American Naturalist, 57, 1923. Fig. 1 series of fibers, the orientation of which varies from layer to layer. In the outer layers the fibrils are in- clined at nearly 90° to the long axis of the fiber, in others at 0°-30°. (Fig. 1 is by Dr. Scarth.) Herzog,^ in a very interesting article on the. struc- ture of cellulose, gives evidence of the spiral wrapping of bast fibers. I had hoped to find in this article of Herzog^s some suggestion of a molecular interpreta- tion of the spiral orientation of structural units in natural cellulose, but no such suggestion is given. There is only a very cautious reference to crystals, which, from a solution that is slightly polluted, crys- tallize with a spiral structure. While we can not yet iind in the molecule an ultimate explanation of the spiral structure of animate and inanimate things, we have at least some indication that molecules are, at* times, also subjects of the same habit. There are molecules which show spiral (axial) symmetry (in the same way as do crystals), and molecules in which the atoms are arranged on a helical curve. The cellulose chain has a spiral axis of symmetry in that along the chain the rings are alternately right- and left-handed. The cellulose molecule, as ordinarily pictured, is linear. However, spirally wound mole- cules have been proposed for cellulose as fitting in well with some of their chemical characteristics, although this has been offered only as a speculation. The same can be said of the spiral molecule suggested for rubber. Returning to less speculative and grosser, though still microscopic examples of the spiral habit, we have the newly discovered spiral structure of plant chromo- somes. A spiral twist appears to be universally characteristic of plant chromosomes. It was first well established by Kaufmann.^ Such a structure of plant ^Koll. Zeitschr., 61: 280, 1932. ^Amer, Jour, Bot., 13: 59, 1926. jhromosomes is of particular interest in connection nth the statement that the spiral habit is a heritable )ne. Chromosomes are thus carriers of a trait which they themselves possess. Right- or lef t-handedness may be a more fundamen- tal character than the spiral tendency and possibly Iresponsible for the latter habit. Mirror writing, in which some children are adept, is an extreme form of left-handedness. Perhaps right- and left-handed- ness and the spiral habit are both expressions of a common, deep-seated and heritable protoplasmic I quality. There has just appeared an article by Haskins and JMoore^ in which they report the spiral twisting of two young citrus plants grown from irradiated (x-rayed) seed. Both plants showed marked twisting in a counter-clockwise direction during early life. I After six months, the habit was abandoned and sub- I sequent growth was normal. The plants gave other evidence of x-ray injury when young. Haskins and I Moore conclude that the experimental conditions in- Idicate that twisting was the result of a physiological i rather than an environmental condition — possibly I x-ray induced abnormal mitoses. Crampton adds a fuitther note to that already Igiven^ on the spiral coil of marine snails, which is usually dextral. It appears ihsut where the one or the [other mode of coil predominates, dextrality is a Men- delian dominant with reference to sinistrality, al- though the case is complicated by the fact that the mode of coil in snails is one of maternal inheritance. Again we come to the conclusion that the spiral habit among organisms is of wide-spread occurrence jand protoplasmic in origin. This statement does not preclude the possibility of the characteristic being suppressed, accentuated or otherwise modified by en- vironmental influences. After the manuscript to the preceding account had ^left my hands there appeared in Science two articles on the spiral habit, one by M. Copisarow^ and one by E. J. Kohl.» The latter author takes up in de- tail the suggestion made above that the spiral grain in trees is due to slippage between long, wedge- shaped cambium cells, a hypothesis first brought to my attention by I. W. Bailey. I wish merely to add here that there can be no question as to the possi- bility from the point of view of structural mechanics, that spiral grain in trees is due to the gliding growth of cambium cells with oblique transverse walls. Cer- tainly this type of structure must be a contributing factor to spiral growth in trees. But the explana- tion does not take care of the experiments of Haskins 7 Science, March 7, 1933. 8 Science, June 16, 1933. 9 Science, July 21, 1933. and Moore cited above nor of the spiral twist in ca<;ti, and of course not of the many other examples of spiral development and movement in numerous and varied forms of organisms. The purpose of my first account^ was to look further than the one in- stance of twisted tree trunks, and to recognize that there is throughout nature a very marked tendency toward spiral form and motion. Slippage of wedge- shaped cambium cells may be a correct explanation of twisting in trees (it may also be only the means by which a tendency toward spiral growth is able to manifest itself), but it does not take care of the several other forms of spiral structure in plants, the coiled thickenings of the walls of xylem vessels, the twist in bast and cotton fibers, etc. Each may have its own ultimate cause, but the habit is too wide- spread to preclude the strong possibility of a gen- eral tendency toward spiral form and movement in plants and animals. It seems, therefore, that the spiral habit, whether in trees, snails or chromosomes, is a fundamental heritable protoplasmic quality. William Seifriz University of Pennsylvania Reprinted from Ecology, Vol. XV, No. 3, July, 1934 Printed in U. S. A. THE PLANT LIFE OF RUSSIAN LAPLAND William Seifriz University of Pcnnsylvama The Kola Peninsula or Russian Lapland lies within the Arctic Circle some 750 miles north of Leningrad (fig. 1). It is bounded by the White Sea on the south and the Arctic Ocean on the north, thus extending from lati- tude 66° N. to about latitude 69° 5' N., the north-western corner nearly reaching latitude 70° N. From west to east the Peninsula extends from Finland, at longitude 30° E., to the strait of the W^hite Sea, at longitude 41° E. The northern part of the Peninsula may be reached by a forty-two hour rail journey from Leningrad to Murmansk, the northern terminal of the road, passing through the much older town of Kola. Fourteen miles north of Murmansk lies Alexandrovsk, the most northern village of the Peninsula, and six miles more bring us to the open shores of the Arctic Ocean near which lies the Island of Kildin. In the center of the Peninsula are the Hibini Moun- tains, extending eastward from Lake Imandra. The Mountains have now become known to all Russians because of the recent discovery of valuable deposits of apatite (phosphorus ore). This article has to do with the plant life of the two regions of the Hibini Mountains known as Tachtarvumchorr and Kukisvumchorr, and of the Is- land of Kildin. On the east shore of Lake Imandra is the small settlement of Hibini ^ with a government agricultural experiment station (fig. 2). Less than a mile from the lake shore rise the Tachtarvumchorr Mountains. Ten miles south from Hibini is the station Apatiti from which a small railroad goes eastward to the new and rapidly growing town of Hibinogorsk on Lake Vudyavr. Four miles to the north, beside Small Lake Vudyavr, is the experimental station of the National Academy of Science of Leningrad. The mountains of this region are more majestic than those of Hibini. The panorama, though bleak, is truly a fine one. Glaciers are not present as the mountains are not suf- ficiently high, the maximum altitude being under 4000 feet. The climate of Lake Imandra and the Hibini Mountains is not as severe as one might expect from their arctic situation. The usual winter temperature is about — 10° C. (14° F.) though —40° C. may be reached. Snow attains an average max- imum depth of two meters. Summer temperature is about 15° C. Rains iHibini is usually written with a K or C as the initial letter, Khibinl, probably in an attempt to obtain an English spelling which will give a pronunciation similar to the Russian. 306 307 WILLIAM SEIFRIZ Ecology, Vol. 15. No. 3 July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 308 J\lORTHW£6T£RM RUS5IA. STATUTE MILES, Fig. 1. Map of northwestern Russia. and chilling winds are frequent, interrupted by a month of uncomfortably warm weather in summer. Daylight, or twilight, is perpetual from April to September. Three Russian botanists, temporarily or permanently stationed in these regions, were of great help in the planning of trips and in the final identifica- tion of the plants collected ; they are Johann Eichf eld. Director of the Hibini Station, George M. Kreps, Keeper of the Imandra Game Reserve, and Vlad- HlSini MOUtiTAITiS, /^USSIAn LAFUliP, iXemtvons artinftet aJbovtthcJIrctic Oc^aK. yfmi;s Fig. 2. Map of the Hibini Mountain region of Russian Lapland. imir Fridoline of the Leningrad National Academy, specialist in entomolog- ical biology. To these gentlemen are my thanks due for their courteous as- sistance. I am also indebted to IVIr. W. R. Williams of the New York Botan- ical Gardens for identifying the mosses, to the late Professor C. C. Plitt of the University of Maryland, for the identification of the lichens, and to Miss Emily Rock for listing and arranging the plants names. Lake Imandra is 430 feet above sea level (fig. 2). Its shores, where recent foresting has left them untouched, are covered with pine, spruce, and birch The pine, Finns silvcstris var. femiica, usually predominates. The spruce is chiefly the Siberian species, Picea ohovata, with P. cxcclsa also pres- ent and according to some authorities, equally abundant. There is difhculty 309 WILLIAM SEIFRIZ Ecology, Vol. 15, No. 3 in distinguishing the two. Juniperus communis occurs near the lake shore but is more frequent at higher altitudes. Its maximum height is rarely above five feet. Birches, as bush and prostrate forms, are abundant, especially at higher altitudes, but the lowland tree forms are not numerous. Among the latter are Betula pubescens, D. odorata (B. pubescens var. odorata) and B. ver- rucosa, of infrequent occurrence. B. kusmischeffii occurs higher up as a tree- shrub. It and B, odorata are hybrids. Two other birches of the Kola Pen- insula are B. tortuosa and B. nana, both typical of higher altitudes. AInus borealis and Popidus tremula occur as scattered specimens, the lat- ter being usually found as a small tree at higher altitudes. Sorbus aucuparia forms low bushes in moderate quantity. Neither Populus tremula nor Sorbus aucuparia characterize the arctic vegetation as strikingly as they do the moun- tain flora of the southern Soviet Provinces. The willows are abundant, but only in low forms being more typical of the higher slopes where they exist as prostrate plants. Among the nine or more species which are said to occur, Salix glauca is most abundant as a tree-shrub along the lake shore. With it occurs 5*. lapponica. Kreps and Kegel distinguish several types of forest associations at the lake shore, with pine and spruce as the dominant members. Pine is the more abundant and spruce exhibits greater variety in its associates. Kegel enumer- ates ten spruce associations distinguished by their undergrowth among which are Picetum myrtillosum, Picetum sphagnosum, and Picetum microbetulosa- empetrosum (fig. 4). Kegel mentions five pine associations but Kreps re- duces these to three: Pinetum sphagnosum, with Riibus chamaemorus, on wet ground; Pinetum hylocosum, with Plypnaceae, chiefly Hylocomium proliferum and Hypnum, and Vaccinium, on moist ground; and Pinetum cladoniosum, consisting of pine and numerous lichens chiefly Cladonia al- pestris, on dry ground. The spruce gets into all three of these associations and may replace the pine, thus forming the similar associations of Picetum sphagnosum, etc. The struggle between pine and spruce in the arctic is similar to that in the Caucasus.^ Pine predominates on dry, sandy, and rocky soil, and spruce on wet sand, but better on clay. There is little pure clay soil in this region and therefore few pure stands of spruce. The clay with its spruce is mixed more or less abundantly with the sand and its pine. Few if any areas exist which have not been repeatedly fire swept. In a fire, the spruce dies first because its large lower branches often sweep the ground and take root while the pine, with its high branches, more easily survives. The temperature fac- tor is also in favor of the pine for spruce seedlings freeze unless well pro- tected by a covering of larger growth, consequently, the pines come in first on ^Seifriz, W. 1932. Sketches of the vegetation of some southern provinces of Soviet Russia. III. Plant life in the Bakuriani Basin, Minor Caucasus. Jour. Ecology 20: 53-68. July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 310 u n3 c C u O o u o CO C C S u I H o H-l 309 WILLIAM SEIFRIZ Ecology, Vol. 15, No. 3 July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 310 in distinguishing the two. Junipcms commiims occurs near the lake shore l)Ut is more frequent at higher altitudes. Its maximum height is rarely ahove five feet. Birches, as bush and prostrate forms, are abundant, especially at higher altitudes, but the lowland tree forms are not numerous. Among the latter are Bctiila pubcsccns, B. odorata (B. puhcsccns var. odorata) and B. ver- rucosa, of infrequent occurrence. B. kitsmiscJicffii occurs higher up as a tree- shrub. It and B. odorata are hyl^rids. Two other birches of the Kola Pen- insula are B. tortuosa and B. nana, l)oth typical of higher altitudes. AIuus horcalis and Populus trcmida occur as scattered specimens, the lat- ter being usually found as a small tree at higher altitudes. Sorhus aucnparia forms low Inushes in moderate quantity. Neither Populus trcmula nor Sorhus aucuparia characterize the arctic vegetation as strikingly as they do the moun- tain flora of the southern Soviet Provinces. The willows are abundant, Init only in low forms being more typical of the higher slopes where they exist as prostrate plants. Among the nine or more species which are said to occur, Salix glauca is most abundant as a tree-shrub along the lake shore. With it occurs 5'. lapponica. Kreps and Kegel distinguish several types of forest associations at the lake shore, with pine and spruce as the dominant members. Pine is the more al)undant and spruce exhibits greater variety in its associates. Kegel enumer- ates ten spruce associations distinguished by their undergrowth among which are Picetum myrtillosum, Picetum sphagnosum, and Picetum microbetulosa- empetrosum (fig. 4). Kegel mentions five pine associations but Kreps re- duces these to three: Pinetum sphagnosum, with Ruhus chamacmorus, on w^et ground; Pinetum hylocosum, with 1 lypnaceae, chiefly Hylocomium prolifcrum and Hypnum, and Vacciniuin, on moist ground; and Pinetum cladoniosum, consisting of pine and numerous lichens chiefly Cladoma al- pcstris, on dry ground. The spruce gets into all three of these associations and may replace the pine, thus forming the similar associations of Picetum sphagnosum, etc. The struggle between i)ine and spruce in the arctic is similar to that in the Caucasus.- Pine predominates on dry, sandy, and rocky soil, and spruce on wet sand, but better on clay. There is little pure clay soil in this region and therefore few pure stands of spruce. The clay with its spruce is mixed more or less abundantly with the sand and its pine. Few if any areas exist which have not been repeatedly fire swept. In a fire, the spruce dies first l)ecause its large lower branches often sweep the ground and take root while the pine, with its high branches, more easily survives. The temperature fac- tor is also in favor of the pine for spruce seedlings freeze unless well pro- tected by a covering of larger growth, consequently, the pines come in first on -Seifriz, W. 1932. Sketches of the vegetation of some southern provinces of Soviet Russia. III. Plant life in the Bakuriani Basin, Minor Caucasus. Jour. Ecology 20: 53-68. i u c £ C O c u O cn C CO ■*-» c § c u I H 1> x: 05 < u: o -o . *— •t-t rt in o ' '^ u O <-• j:: c5 H E rt 'c5 3 O -1^ -y> o X r3 > OJ INTENTIONAL SECOND EXPOSURE 4i 311 WILLIAM SEIFRIZ Ecology, Vol. 15, No. 3 July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 312 'il new, fire-swept soil, and the spruce seedlings follow. The roots of the pine go deep while those of the spruce are superficial, which gives the pine the advantage, particularly on stony ground where it predominates. The spruce holds better in clay soil. Fortunately the spruce is not wholly outdone by the pine; it has one substantial evolutionary factor in its favor, that of vigor, so that in time the spruce should and occasionally does win out over the pine, unless other factors interfere. A small island in Lake Imandra just off the Hibini Station (fig. 4) is a rich collecting ground, although not a pure association. It presents a typical spruce-pine-moss-Ericaceae complex of the following composition: Picea cxcelsa {P. ohovata), Pinns silvestris var. fcnnica, Betula pubescciis, B. nana, Salix glauca, S. lapponica, Vaccinium vitis-idaea, V. uliginosum, V. myrtilliis, Ledum paliistrc, Empetrum nigrum, Calluna vulgaris, Rubus charnaemorus, Cornus suecica, Eqiiisctum spp., Lycopodium clavatmn. Neph- roma arcticum. Sphagnum spp., Hylocomiuni proliferum, Hypnum spp., Polytrichum juniperinum, Ceratodon purpureus, Bryum caespiticium, Cladonia alpestris, and C. rangiferina. In such an association complex occur many other plants, scattered or in restricted areas where the ground is more open and drier or moister, but as these plants are typical of other regions none need be enumerated here, except the gem of all the arctic flowers, Linmea borealis, which has a very wide distribution preferring, however, the wooded moors. The herbarium at Hibini records Cladonia alpestris and C. rangiferina as the predominating terrestrial lichens. My own collections contained the following additional species for the lowlands at Hibini: Cladonia coccifera stemmatina, C. deformis gonecha, C. uncialis, Stereocaulon paschale, Neph- roma arcticum, Parmelia centrifuga, and Cetraria islandica. Within the needle forests, or bordering them, are wet moors, bogs, swamps, and open water, each of which harbors a typical vegetation, though there is much intermingling. To distinguish sharply between moor, bog, swamp and tundra is impossible in the present state of confusion of these terms. Moor might better be retained for the dry heaths, yet wet moors, which are not un- common, as in the North German Heide, practically become bogs. Kreps translates the Russian word for " bog " into the German word " moor.'* Tundra is just as loosely used. The consensus of Russian opinion is that it should be retained for the high (mountain) moors yet it is a convenient term for the low northern moors which duplicate the high moors of the mountains. The wet moors or bogs of Hibini are usually association complexes, but Krebs tells of pure stands of Molinia caerulea forming one of the typical sedge moors of the Peninsula. More common are the sphagnum moors with scattered pines. These are ideal habitats for a number of plants which find their home here without characterizing the association. The first half dozen plants of the following list are typical of the moor; the rest are frequent inhabitants of it : Sphagnum acutifolium, Molinia caendea, Carex gracilis, Betula nana, Empetrum nigrum. Ledum palustre, Cladonia spp., Marchantia polymorpha, Equisetum palustre, Lycopodium selago, Eriophorum alpinum. Orchis macu- lata, Drosera rotundifolia, Rubus charnaemorus, Loiseleuria procumbens, Pirola rotundifolia, Phyllodoce taxifolia, Arctostaphylos uva-ursi, Arctous alpina, and Pinguicula vulgaris. Where the moor becomes swamp or the lake forms inland pools, the fol- lowing aquatic and semi-aquatic plants occur : Potamogeton spp., Phragmites communis, Carex aquatilis, and Menyanthes trifoliata. The borders of the moors, the lake shore, and the larger semi-moist burned-over areas, all harbor a considerable variety of plants which cannot be ascribed to any one association, being widely scattered. The more abun- dant of the flowering species add color to the otherwise somber tone of the arctic flora. The most prolific of these is Epilobium angustifolium which in great profusion covers the burned areas from Leningrad to the Arctic Ocean. Among these widely scattered plants are : Stereocaulon denudatum Peltigera spp. Ochrolechia tart area Cetraria nivalis Alectoria ochroleuca A. diver gens Sphagnum lindbcrgii S. fuscum S. medium S. balticum (20 spp. occur) Mnium cincidioides Nephrodium dryopteris Equisetum silvaticum Deschampsia spp. Carex rotund at a C, alpina (20 spp. occur) Luzula multifiora Maianthemum bifolium Orchis lanceolata A In us (incana) Stellaria holostea Lychnis flos-cucuH L. alpina L. vis carta Vise aria alpina Trollius europaeus Ranunculus spp. Drosera anglica Rubus saxatilis Rosa acicularis Alchemilla sp. Comarum palustre Geiim rival e Trifolium re pens Astragalus alpinus Geranium sylvaticum Viola canina V, bi flora V. tricolor Angelica archangelica {A. sylvestris) Pirola secunda P. media Trientalis europaea Myosotis sp. Veronica sp. Bartschia alpina Euphrasia minima Rhinanthus sp. Linnaea borealis Campanula rotundifolia Solidago virga-aurea Antennaria dioica Achillea millefolium Tussilago farfara Matricaria inodora Cirsium heterophyllum Saussurea alpina Taraxacum ceratophorum T. lapponicum Hicracium silvaticu m H. alpinum 313 WILLIAM SEIFRIZ Ecology, Vol. 15, No. 3 July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 314 I I The sand deltas of the Lutnjermajok River, where the latter flows into Lake Imandra near Hibini, illustrate one of the most interesting of the eco- logical features of the region. The majority of the species which form the sparse vegetation on the river delta are typical high altitude plants which do not normally occur on the lowlands except where, as here, they have been brought down with the sand. The deltas are formed of nepheline sand (nephelin-syenite) which is brought down from the high regions by the river waters. The most abundant and conspicuous plants in this rather meager delta flora are the typically alpine forms Saxifraga aisoides and Silcnc acaulis. They actually thrive better at the lake level than in their natural habitat at a higher elevation. Their presence and healthy condition on the lake shore indicate that altitude, with its characteristic qualities of temperature and light, is after all not always an important factor. Rather does competition deter- mine the successful growth of these plants. At the lower altitude there are no competitors on the deltas, yet the high altitude forms find there the same soil that they would have near the mountain tops. Silcnc occurs in two varie- ties, one with yellow and one with orange flowers ; both may form mounds 18 inches in diameter. There also occur Silcnc alpina, Oxytropis sordida, Papavcr radicatnm var. lapponicuni, Oxyria digyna, the rare aljMue Sagina linnaea, and the thoroughly alpine Carcx rigida, all high altitude forms trans- ported with their soil to the lowlands. Poa alpina may be added to the list although it is a relatively widely scattered grass. Kreps was among the first to call attention to this interesting ecological oddity. An altitude of 1800 and 2500 feet can be conveniently reached near Hibini in a two or three hour walk from the lake. It is desirable to climb the lower Hibini Mountains near the lake, even though they present a bleak appearance (fig. 3) because the more picturesque Hibinogorsk Mountains (fig. 5) lack the lower altitude plants. Three lichens were collected on the 1800 foot summit of one of the Hibini (Tachtarvumchorr) Mountains which were not seen else- where; they are Cctraria tcnuifolia, Gyrophora proboscidca, and Cladonia sylvatica. The following is an account of the altitudinal distribution of plants on both the Tachtarvumchorr and the Kukisvumchorr Mountains. Hibinogorsk is the new city on Lake Vudyavr, three miles from the re- cently discovered phosphorus mines. The settlement may be reached by train from Apatiti, or by a twenty hour walk through the mountains from Hibini. The house of the Leningrad National Academy on Small Lake Vudyavr (fig. 5) 1200 feet altitude, serves as a base for the visiting naturalist. At the lake, moors predominate, moors which more closely approach tundra than do the bogs of the lowland. With Ledum palusfrc in abundance, the association is definitely moor and not tundra. A few plants not so far found, were gathered here: Gymnadcnia conopsca, Dianthus supcrhus, Saxifraga stcllaris, Cassiopc hypnoidcs, and Gnaphalium supinuni. Where forest creeps up into the almost treeless Vudyavr Valley, the spruce •^1^ *m p ^ is the sole representative of the larger arborescent forms. Pines are few and exist only as dwarf specimens seldom exceeding three feet in height as com- pared with the large trees found at Hibini. The treeline on the Kukisvumchorr Alountains is formed by Picca obovata, in upright form, and Bctula pubcsccns. With these occur Bctula kitsmis- chcffii, a controversial species of birch which may be a variety, if not merely a synonym, of B. tortiiosa. The birches here, as throughout Russia, are taxonomically indefinite. If we are to distinguish them B, kiismischcffii is the tree-shrub form while B. tortuosa is prostrate, much resembling B. nana but with larger leaves, y. inch in length as compared with the B. nana leaves which are less than 14 inch. B. nana, a small prostrate shrub, is the most widely distributed species, growing at all altitudes. The willows offer an even more diflicult problem. Ten species occur in the Hibini Mountains : Salix glauca, S. lapponica, S. phylicifolia, S. reticulata, S. hcrbacca, S. rotundi folia, S. polaris, S. alpina {S. polaris) , S. myrsinitcs, and S. lanata. None form large trees. 5*. lanata is a bush form, growing abundantly at the tree line. ^. reticulata and 5. hcrbacca are small and pros- trate, the latter becoming more abundant higher up. Sorbus aiiciiparia is not infrequently met with. Populus trcmula is rare and seldom attains a height of more than 18 inches. It is found mostly on southern slopes. At the tree line a number of plants find their upper limit and others make their first appearance, while still others take on ecological modifications. This last is true of the spruce which at higher altitudes is prostrate in form. An- other prostrate gymnosperm is the juniper which could have been mentioned among the lowland plants, where it is moderately abundant and grows to a maximum height of 6 feet. Here, at 1700 feet, it lies closer to the ground and most authorities refer to this prostrate arctic juniper as /. nana. Linnaea borcalis finds its upper limit near the tree-line. This is also true of Dianthus supcrbus. A plant typical of the willow scrub at this altitude is the small (15 inch) Rubus saxatilis which here replaces the related R. chamacniorus of the lowland. Another and quite rare tree line plant growing always in isolated wet rocky exposures on southern slopes, is Cotoncastcr sib erica. With it occur a Rosa, Rubus saxatilis, and the two grasses, Melica nutans and Molinia caerulea. While ferns grow in the lowlands, it is in the wet ravines and gorges above the lake shore that they are most abundant. They climb to near the tree line, but are never profuse. The species collected were Phegoptcris dryoptcris, Dryoptcris linneana, and Aspidium lonchitis. This modest list indicates the paucity of ferns in the Hibini Mountains, but it does not convey quite a fair impression, as other species have been reported, namely, Woodsia alpina, Cys- topteris fragilis, C. montana, Polystichum lonchitis, Athyrium felix-femina, A, alpestre, Allososus crispus, and Polypodium vulgare. Botrychium simplex was also found, the only species listed for the Hibini Mountains. 315 WILLIAM SEIFRIZ Ecology, Vol. 15, No. 3 July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 31^ I I But one moss, Dicranoiveisia crispiila, was collected at Hibinogorsk. Five hundred feet above the tree line, at an altitude of 2200 feet, there grows a true subalpine flora (pseudo-tundra) intermixed with species from below and above. All plants collected in this subalpine zone between 1800 and 2500 feet, i.e., at an average altitude of 2200 feet, are given in the fol- lowing three lists, the first of which includes the plants typical of lower alti- tudes, the second, the plants which characterize the subalpine zone, and the third, the plants typical of the higher alpine flora. Such widespread forms as Betula nana and the Vacciniums are classified with difficulty; they char- acterize the lowland swamps as much as they do the subalpine slopes. Typical of Forest (below 1300 feet) Hylocomium proliferum Lycopodium aurantia J uncus spp. Polygonum viviparum Alchcmilla sp. Geranium sp. Pirola minor P. uniflora Loiscleuria procumbens Andromeda poly folia Colluna vulgaris Pinguicula alpina Campanula Solid ago Antennaria dioica Saussurea alpina Alpine (above 2500 feet) Rhacomitrium hypnoides Salix herhacea (S. alpina) Silenc acaulis Sax if rag a a isoides S. oppositifolia Dryas octopctala Cassiope hypnoides Veronica alpina Pedicularis lapponica Subalpine (1800-2500 feet) Juniperus spp. Tofieldia alpina T. horealis Coeloglossum z'iride Salix reticulata S. myrsinites S. phylicifolia Betula natui B. kusmischeffii Oxyria digyna Ranunculus glacialis Pa paver radicatum Sihhaldia procumbens Potcntilla alpestris Oxytropis sordida Viola montana Diapensia lapponica Phyllodoce caerulea Cassiope tetragona Vaccinium vitis idaea V. uliginosum V. myrtillus Among the above plants a number make their first appearance at this altitude (2500 ft.), these are Veronica alpina, Pedicularis lapponica, and the two Cassiope species. Both of the last mentioned prostrate Ericaceae are distinctly high altitude forms, the smaller of which, Cassiope hypnoides, climbs the higher. C. tetragona is a rare plant, the mountains at Hibinogorsk being among its few habitats. The willows are especially interesting because of their small size and ecological change in form, thus S. myrsinites occurs at lower altitudes where it attains a maximum height of 3 feet, but here in the subalpine belt it becomes a low plant with very small leaves, the old ones of which persist through the second summer. The smallest of the willows, which lies very close to the ground like a creeping vine, is 5^. herhacea (S, alpina). S, reticulata also is prostrate here, but has larger leaves. Betula nana is as abundant at 2200 feet as it is in the bogs below at 430 feet. This small-leaved prostrate birch therefore extends from the lowlands up to the edge of the high alpine tundra. The preceding list of plants growing in their natural habitat at 2500 feet .•s-.M. !-i J l^^ « A im ' contains the names of those specimens found growing on the Lutnjermajok River delta, at Hibini, including Saxifraga aizoidcs and Silene acaulis, which grow equally well at both altitudes. No marked altitudinal distribution of lichens was observed, although lichens (and mosses) may show as striking an altitudinal distribution as do flowering plants. Those collected on the Kukisvumchorr Mountains at Hibinogorsk are: Stcreocaulon alpinmn, Cetraria islandica, Solorina crocea, Gyrophora proboscidea, and Rhizocarpon gcographicum astrovirens. The summit of the ridge, at 3250 feet, is a rocky alpine tundra. The soil is badly weathered and the vegetation sparse. Climbing still higher along the ridge, one reaches the maximum altitude of the mountain, 3400 feet. The following plants were observed at the top: Cladonia (and other, lichens), Dicranum spp., Polytrichum spp., Lycopodium alpinum, Deschampsia sp., Carex spp., Juncus spp., Salix alpina (S. herhacea), Silene acaulis, Saxifraga aicoides, Dryas octopctala, Empetrum nigrum, Loiscleuria sp., Phyllodoce caerulea, Vaccinium vitis-idaea, V. uliginosum, Cassiope hypnoides, Bartschia sp.. Campanula rotundifolia, and Solidago znrgaurca. Dryas octopctala is credited with being the flowering plant which climbs highest ; Silene acaulis and Saxifraga aizoides are close behind. The journey north to the Arctic Ocean and the Island of Kildin was made in order to follow the change in type of vegetation with change in latitude. If the successive altitudinal zones through which one passes in climbing the Hibini Mountains were brought down to sea level and placed one after the other as one travels north, a fairly accurate picture of the arctic lowland flora extending from Hibini to the Arctic Ocean, a distance of 125 miles, would be had. Only two features of this northernmost vegetation of western Russia will be added to that already given, namely, a delineation of the tree line, and a description of the flora of the Island of Kildin. The tree line on the Kola Peninsula has been established by Kreps. It is better given by a line on the map {^g. 1) rather than in words. It represents the northern limit of the needle forest. Deciduous trees (birch, willow, and poplar) get down to the ocean shore along all river banks. The northern limit of the needle forest does not run parallel to the coast nor to the lines of latitude, but approaches parallelism with the ice barrier. This is due to the Gulf Stream, which, north of Norway (between latitude 7^ and 75° N.), di- vides into four branches. The southern one sweeps the coast of the Kola Peninsula until it disappears at longitude 40° E. Here the ice barrier com- mences, closing the White Sea in winter but leaving the north Murman coast open the year round. Alexandrovsk (its name was recently changed to Polarnoje) is the only Russian seaport which is never closed to shipping be- cause of freezing. One-half mile off the Murman coast, 25 miles from Alexandrovsk, is Kildin (fig. 6). It is a treeless island, one and a half to two miles across and three or four times this in length. It rises to a maximum height of about • A 317 WILLIAM SEIFRIZ Ecology, Vol. 15, No. 3 1000 feet. Its shore is precipitous and its surface a relatively flat plateau wh,ch ,s aIn,ost pure and typical tundra, if we may use this conventnt tern" for arct.c moors. Here and there on the island are wet depressions wh.l harbor a bog vegetation. One small lake exists. It is separatTf oTthe ' by a narrow rocky, natural causeway. This lake is one of the few U not the only one. of ,ts kind. Its surface is fresh water to a depth of seve altche supp bed by surface water from the island, while its deeper water! altv The lake was probably cut off from the ocean in early times leavin. asuZl' ranean connection with the sea. Its surface water is' ufficientTf Ih e -' lit dnnk„.g, and .ts deeper water contains marine organisms of great variety such as codfish and echinoderms. variety, Were the alpine, and part of the subalpine. flora of the mountains at tundra eii tin. th J' "' '''°"'^ '^'"^ ' ^^^^ '^°'' reproduction of the KildL ot nf H ""' '"'"'^ '""'''' "°^ y^' "^"^^d ^-e collected on o^L s are cll' ""? '"*"T""^ '' '""''^ '^ ^"-"Z-^" — Among the others are . CalamagrosNs neglccta, Cemstitan sxlvaticum, Caltha Palustrh July, 1934 PLANT LIFE OF RUSSIAN LAPLAND 318 Stercocaulon alpinum Hacmatomma ventosum Rhizocarpon geographicum R. geographicum atrovirens Lccidea confluens Gyrophora polyphylla Nephroma arcficum Rhacomitriuni hypnoides Mnium cincidioides Aulacomnium tiirgidum Drcpanocladus exannulatus Dryopteris sp. Equisctiim sp. Lycopodium amwtinum L. alp mum Junipenis couununis J. nana Calamagrosfis neglccta Eriophorum polysfachion h. vaginatum Carcx tetragona C. hypnoides C. rotundata Juncus polyfonuis Orchis maculafa Salix lanata Bctida nana Rumex acetosella Polygonum znmparum Saginu linnaca Ccrastiuni sylvaticuiu Dianfhus super bus Ranunculus pygmaeus Caltha palustris Seduin roseum Saxifraga rosea Parnassia palustris Rubus chamacmorus Alchcmilla sjx Dryas octopctala Filipendida ulmaria Comarum palustre Geum rivale Oxytropis sordida Geranium pratense Vicia cracca Empetrum nigrum Epilobium palustre Angelica archangelica Diapensia lapponica Cornus suecica Pirola secunda P. rotundi folia Loiscleuria procumbens Arctous alpina Vaccinium vitis-idaea V. myrtillus V. uliginosum Andromeda polifolia Trientalis europaea Bartschia alpina Euphrasia officinalis Pedicularis sceptrum Melampyrum pratense Rhinanthus minor Pinguicula palustris Linnaea borealis Campamda rotundifolia Solidago virgaurea Antennaria alpina (A. dioica) Achillea millefolium Saussurea alpina Cirsium heterophyllum Hieracium sp. EXTRAIT DE LA REVUE GBnERALE DE BOTANIQUE Tome 46, Page 200 (1934) / PROPRlfiTfiS PHYSIQUES DU PROTOPLASMA DES MYXOMYCETES It par M. W. SEIFRIZ Certains caracteres s'appliquent en general a tout protoplasma. Ainsi tout protoplasma est elastique et glutineux ; il est d*une vis- cosite moderee ; il est sensible a la blessure ; il possede une structure continue et une organisation. Mais ces caracteres generaux, si vrais soient-ils, varient dans de telles proportions qu'on ne pent leur don- ner une valeur exacte sans specifiei^ le protoplasma particulier au- quel ils s'appliquent ej: aussi son etat actuel. Considerons, par exemple, deux des proprietes du protoplasma : la viscosite et la sensibilite a la blessure. La premiere varie d'une valeur de 0,02 X (ou de dix a vingt fois celle de I'eau) a la consistance d'une ferme gelee. II n'y a done pas une valeur de viscosite du pro- toplasma, mais un nombre infini de valeurs. Que le protoplasma soit sensible a la blessure, cela est evident du fait que c'est un systeme vivant. Sa sensibilite est parfois extr^- mement grande. Un oeuf de Fums ou un protozoaire pourra eclater . a la plus legere atteinte d*une aiguille ; la configuration mitotique pourra s'affaisser immediatement et completement si la cellule (oeuf d*Echinoderme) est perforce par I'aiguille du micromanipulateur. D'autre part, le protoplasma pent etre extr^mement resistant aux actions mecaniques (3). Une Amibe tolerera quelquefois la perte de son noyau ; de meme si I'on coupe un lobe de son corps, elle vivra plusieurs jours. Elle est susceptible de supporter trois operations successives, chacune entrainant la separation d'une assez grande partie de son corps, jusqu'a etre reduite de moitie ; mais ^s ^ 4 REVUE GENERAL^ DE BOTANIQUE dans ce dernier cas, elle ne pent vivre longtemps. Le protoplasma en mouvement pent s'arreter instantanement au contact de I'aiguille, et ensulte recommencer a couler comme si de rien n'etait. On ne pent done pas dire que le protoplasma soit tres sensible ou trfes resistant ; il pent etre I'un et I'autre. La m6me variabilite pent se trouver dans toutes les reactions protoplasmiques. La vie elle-mfime est dans un etat constant de changement. Ce fait est particulierement evi- dent dans les reactions protoplasmiques : la viscosite, I'elasticite, la perm^abilite, la sensibilite, etc. changent perpetuellement.' Negliger d'apprecier ce fait, c'est negliger d'apprecier le caractere le plus fondamental de la substance vivante. L'introduction de la technique de la culture des Myxomycetes a mis, a la disposition des biologistes, le materiel de choix qui convient a une grande variete d'experiences, et notamment a celles de micro- dissection. Un milieu trfes favorable a la culture des Myxomycetes a ete indique par Howard (1). C'est de la souche qu'il nous a fournie qu'ont ete obtenues les cultures ayant servi au prdsent travail. Le milieu est simplement du gruau d'avoine avec de I'agar dans les proportions suivantes : Gruau d'avoine 15gr. Agar.... 30gr. Eau l.OOOcc. Le melange est chauffe a Tautoclave et verse ensuite dans les fla- cons de cultures. Des fragments de plasmodes frais ou de sclerotium sees sont places sur le milieu gelifie. Apres un jour ou deux, le dt ve- loppement commence; les jours suivants, la plus grande partiedela surface des verres de cultures pent etre recouverte de plasmodes. Pour travailler avec le micromanipulateur, il est necessaire de transporter le materiel sur une mince lamelle de verre qui servira ensuite de couvercle a la chambre humide. Dans ce but, il est commo- de de faire ramper le plasmode sur une lamelle placee sur son chemin. II semble etrange que le plasmode paraisse « preferer >> a la lamelle qui lui est offerte la surface d'autres verres. Ceci est pent etre du a une substance toxique contenue dans le verre recemment introduit et qui n'existerait pas dans le verre usage. Mais, avec plusieurs lamelles piquees verticalement dans le milieu gelose sur lequel se PROTOPLASMA DES MYXOMYCETES 5 deplace un fragment de plasmode, on obtient generalement, au bout de deux jours, une lamelle recouverte du plasmode. Les reci- pients de culture doivent etre soigneusement fermes pour b|en con- server rhumidite. Une lamelle avec du plasmode est renversee sur la chambre humide. Les aiguilles fines de verre employees pour la dissection entrent dans la chambre par les cotes. Elle , sont actionnees par un micromanipulateur. Les recherches relatees ici ont ete faites a Paris, au Laboratoire de Cryptogamie du Museum National d'Histoire Naturelle. Je reste tres oblige a son directeur, M. le professeur Allorge et a ses colla- borateurs, le Sous-Directeur M. Heim, M. R. Lami et M. Lefevre de leur aide precieuse et courtoise. Le plasmode d'un Myxomycete pent exister sous deux formes. II pent etre constitue soit par un filament unique, soit par une sur- face de protoplasma perforce et sillonnee par plusieurs arteres anas- tomosees. Dans le filament unique, comme dans les arteres, le pro- toplasma coule rapidement, dans une seule direction, pendant un cer- tain temps. Apres quelques secondes, ce mouvement s'inverse. A la peripheric du plasmode, des prolongements protoplasmiques (pseu- dopodes) se forment. La force du protoplasma fluant fait avancer le plasmode. Viscosite. — On pent considerer, dans le plasmode d'un Myxo- mycete, trois degres majeurs de viscosite, avec tons les degres intermediaires. Le degre le plus bas de viscosite est celui du proto- plasma qui coule le plus rapidement. Mais la rapidite du courant a elle seule ne pent fournir une idee exacte de la viscosite. Le protoplas- ma fluant n'est pas tres fluide, sa consistarice est sensiblement celle de rhuile d'olive. II n'est pas cependant possible d'en donner une valeur exacte, car, lorsqu'il s'arrete, sa viscosite augmente. II existe des lames de protoplasma interposees entre les arte- res ; leur viscosite a Tetat immobile est analogue a celle d'une gelee molle offrant peu de resistance au mouvement d'une aiguille. (Comparable a peu pres a la pate a pain). Recouvrant le plasmode entier, c'est-a-dire les arteres et le protoplasma intermediaire, se trouve une membrane relativement m^ \< I PI ii»»— e REVUE g^n6rale de botanique I dure de plasma ferme et dastique, appelee la membrane protoplas- mique. Elle poss^de, a I'ctat de repos, la rigidite d*une gel^e ferme. L'existence de cette membrane morphologiquement definie sur le pfotoplasma appele « nu », est facilement demontree lorsqu'on dechire un plasmode a Taide d'une micro-aiguille. Presque toujours c'est la couche externe, e'est-a-dire la membrane protoplasmique, qui est la derniere rompue par suite de sa plus grande resistance. Cette membrane pent etre souvent etiree sur une assez grande lon- gueur : et, si elle est rompue par cette manoeuvre, elle se retracte ensuite comme un ruban de caoutchouc. Ces trois valeurs principales de viscosite, si elles caracterisent le plasmode dans son ensemble, sont cependant susceptibles de changements. Quand le protoplasma fluant s'arrete un moment, sa viscosite augmente ; par contre, si le protoplasma situe entre les arteres commence h couler, sa viscosite diminue. La variation la plus marquee de la viscosite se trouve a la surface. Evidemment, la membrane d'un pseudopode doit etre fluide pour que le protoplasma puisse avancer. Normalement la membrane n'est jamais trestendue parce que sa surface se trouve constamment accrue par la matiere venant de Tinterieur. Pour que cet accroissement en surface et le mouvement qui en resulte puissent se produire vers I'avant, il faut que la membrane rigide devienne fluide. Une telle difference de vis- cosite entre la membrane d'un pseudopode en mouvement et la mem- brane du protoplasma immobile, pent etre facilement demontree sur une amibe. Si Ton f^it penetrer une aiguille dans Textremite d'un pseudopode en mouvement en I'enfon^ant lentement, le proto- plasma se reparera comme le ferait une goutte d'eau ; mais, quand ce meme pseudopode devient inactif, et que Ton continue k enfoncer I'aiguille, la membrane se dechire et laisse la marque d'une bles- sure. Un examen des travaux publics (6) indique combien sont varies les resultats obtenus sur la viscosite du protoplasma et montre la grande influence des facteurs externes. Elasticite. — Quand le protoplasma poss^de une viscosite suffi- sante pour permettre de le saisir k I'aide d'aiguilles, il montre tou- jours des qualites d'^lasticite. II n'y a pas d'exception a cette rfegle, sauf parfois quand le protoplasma est mort. Si le protoplasma est en mouvement, il n'est pas possible de demontrer aisement ces quah- PROTOPLASMA DES MYXOMYCETES 7 tes elastiques. On note cependant certains indices d'elasticite. Ainsi ScARTH (5) pense que I'aspect du protoplasma fluant se manifeste meme dans le cas ou le filament de protoplasma est librement sus- pendu. Aucune raison n'autorise a penser qu'une augmentation de vis- cosite serait, a elle seule, capable de conferer 1' elasticite au protoplas- ma si cette propriete n'etait pour lui une caracteristique constante. On doit faire une distinction entre I'elasticite, au sens stricte- ment physique, et I'extensibilite. Le protoplasma pent presenter une assez forte elasticite et une certaine rigidite, en meme temps qu'une faible extensibilite, ou le contraire. Les excellents films cinematographiques de Comandon et de FoNBRUNE donnent un exemple tres convaincant de I'elasticite du protoplasma. Quand les leucocytes touchent, en passant, les globules rouges, ils y adherent souvent, et avec une telle tenacite qu'ils peu- vent obliger le protoplasma a s'etirer sensiblement. Quand le fila- ment protoplasmique ainsi forme se rompt, il revient sur lui-meme, comme un ruban de caoutchouc. Ce fait indique egalement la qualite glutineuse du protoplasma, qualite tres importante dans la vie des cellules animales. Ces cellules sont capables de se joindre et de for- mer des tissus, grace a la glutinosite de leur protoplasma. L'elasticite du protoplasma est considerablement modifiee par les sels, comme I'ont si bien montre Faure-Fremiet, et d'autres. L'allongement maximum d'un filament de protoplasma que Ton etire est considerablement diminue par le sodium ; le magnesium n'a aucun effet ; le calcium I'augmente fortement. L'elasticite du protoplasma est une propriete d'une importance fondamentale, et la preuve la plus significative que nous ayons de sa structure sub-microscopique. SemibiliU, — Quand une aiguille entre dans le protoplasma vivant, mais immobile, il n'y a ordinairement aucune reaction visible. Mais, quand I'aiguille penetre dans le protoplasma fluant, on observe ordinairement une reaction immediate. Lorsqu'une Amibe en mou- vement est percee par une aiguille, en general, elle s'arrete, contracte ses pseudopodes et reste au repos, puis elle commence un mouvement actif et donne tons les indices d'un « effort » pour s'echapper. Elle peut meme aller jusqu'a se separer de la partie de son corps retenue '; ' 'J" "' 8 REVUE GENERALE DE BOTANIQUE > ) * ^1 par Taiguille. Un plasmode ne montre pas, dans le meme cas, des reactions si complexes. Apres que le plasmode fluant s'est arrete, par suite d'une blessure, il reprend generalement aussitot son mouvement. Bien que Taiguille demeure dans le protoplasma, celui-ci la debordera tout simplement. Dans tons mes travaux de dissection des Myxo- mycetes, je n'ai pas trouve une seule exception a la regie suivant laquelle le protoplasma fluant donne une reaction immediate sous Taction de I'aiguille. La piqure d'une aiguille diminue invariablement la vitesse du mouvement, et, habituellement, I'arrete temporairement, mais jamais il n'y a acceleration. Sans vouloir attribuer a cette observation une valeur definitive, on pent remarquer cependant, en passant, qu'elle Concorde avec celle de Nichols (2) mais elle differe de celles d'autres auteurs. Que certains stimulants augmentent la vitesse, cela a ete demontre par Martens (4) pour la temperature, et par d'autres pour les sels. II se pent que la reaction ne soit que momentanee ; mais aussi, elle pent etre assez severe pour causer la mort. Ce dernier fait se produit parfois pour les ceufs et les Proto- zoaires qui peuvent reellement eclater a la moindre piqure de Tai- guille. Ce cas est le plus extreme de la sensibilite du protoplasma. La plus grande resistance du protoplasma est montree par Texperience sur un plasmode qui continue son mouvement actif apres plusieurs serieuses dechirures. D'apres cela, on voit combien il serait futile d'avancer une assertion generale au sujet de la sensibilite du protoplasma. II n'y a pas de raisons de croire que, dans tons les cas, le proto- plasma ait les memes reactions sous I'influence d'un meme facteur exterieur. A certains egards, le protoplasma d'un Myxomyxete est semblable a celui d'une feuille ; mais, a beaucoup d'autres egards, il est fondamentalement different. Immixtion du protoplasma. — La description originale du protoplasma par Dujardin est d'une extraordinaire exactitude. II dit : « Je propose de nommer ainsi ce que d'autres observateurs ont appele une gelee vivante, cette substance glutineuse, diaphane, M^OTOPLASMA DES MYXOMYCETES 9 insoluble dans I'eau, se contractant en masses globuleuses, s'atta- chant aux aiguilles de dissection, et se laissant etirer comme du mucus, enfin se trouvant, dans tous les animaux inferieurs, interpo- see aux autres elements de structure ». Aujourd'hui, apres un siecle, cette description de la substance vivante est exacte, j usque dans ses moindres details. La partie de la description qui nous interesse en ce moment est celle de I'immix- tion du protoplasma dans I'eau. Cette immixtion est, non seulement, int^ressante par elle-meme, mais aussi parce qu'elle a une importante relation avec la structure protoplasmique et I'organisation cellulaire. Quand le plasmode est dechire, le protoplasma ne montre au- cune tendance a se meler a I'eau environnante. Ceci est vrai du pro- toplasma qu'il soit liquide et deborde librement, ou non, mais seule- ment a la condition que ce protoplasma garde son identite : c'est a dire qu'il reste vivant. La membrane protoplasmique empeche normalement I'immixtion du protoplasma ; cependant I'immixtion du protoplasma ne depend pas uniquement de la membrane, mais de sa structure continue. II faut soigneusement distinguer I'immixtion, au sens de solu- * bilite, de I'immixtion au sens de I'imbibition. Un sel est soluble dans I'eau mais une eponge ou un buvard s'imbibent d'eau. Le proto- plasma absorbe I'eau par imbibition, ce qui est la consequence de sa structure. Le protoplasma n'est pas soluble dans I'eau. Structure, — Le protoplasma, vu au microscope, presente I'apparence d'une emulsion (7). L'emulsion vivante pent etre tres fine ou grossiere. Spek a observe que, dans le protoplasma, les plus petites particules se fusionnent quand elles viennent au contact les unes des autres. Ce fait met en evidence la nature liquide de ces par- ticules qu'on a appelees granules. Quand les globules sont gros et disposes regulierement et sous pression, on a la structure alveolaire bien connue depuis Butchli. Ainsi toutes les hypotheses connues sous les noms de structures granulaires, alveolaires, vacuolaires, etc. se rapportent toutes aux formes d'une « emulsion ». Mais cette emulsion est une donnee purement superficielle ; il y a cependant des savants qui considerent une emulsion ultra-microscopique comme la struc- ture fondaflientale du protoplasma. Mais alors si les theories basees sur ce concept etaient vraies, il serait necessaire, pour l'emulsion vi- 1 »* k I s I ! < i *' u 10 REVUE GENERALE DE BOTANIQUE vante, de pouvoir inverser sa structure comme celle d'huile dans I'eau en celle d'eau dans I'huile: or,il n'est pas demontre que Temulsion protoplasmique s'inverse. De plus, les propriet^s protoplasmiques telles que Telasticite, Timbibition et la coagulation indiquent que la substance vivante est un colloide hydrophile analogue a la gela- tine. La structure dite « emulsoide » des gels a ete entierement aban- donnee par les chimistes. Prenons, par exemple, le lait. On a coutume de le considerer comme une emulsion, mais, quand le lait se coagule, ce n'est pas I'emulsion d'un corps gras (le beurre) qui se transforme, mais la pro- teine (caseinogene). II en est de meme pour le protoplasma. II est generalement admis que la structure des gelees elastiques du type gelatine et protoplasma est un enchev^trement ou un lacis de fibres longues, tenues, cristallines, de dimensions moleculaires ou colloidales. Cette structure n'est pas visible, mais elle est necessitee par les propietes des gelees elastiques. Un t^moignage indirect de sa presence dans le protoplasma est fourni par la structure fibrillaire qu'on observe quand le protoplasma vivant est fixe. Cette structure est bien connue dans les preparations histologiques. On pent aussi la constater dans le protoplasma vivant, examine dans des conditions favorables. Quand le plasmode est dechire, le protoplasma montre souvent une structure fibrillaire distincte. L'enchev^trement des fibres ultra-microscopiques donne au protoplasma une structure en treillis qui est la condition mecanique de ses proprietes d'elasticite. Evidemment, ce treillis est tr^s mobile quand le protoplasma est en mouvement : et il est rigide si le protoplas- ma est d'une grande viscosite (gelatinise). Le pouvoir de se trans- former d'un etat liquide en un etat rigide est bien connu maintenant comme propre a certaines gelees. Ce phenomene se nomme « thixo- tropie )) et il est caracteristique du protoplasma. La structure que nous avons mentionnee ici est, sous une forme ou une autre, largement re- pandue dans la nature. Un enchevetrement d'elements lineaires constitue non seulement un systeme mecanique satisfaisant Tin- terpretation des proprietes physiques du protoplasma, mais encore il aide a la comprehension de Torganisation du protoplasma. Il est difficile de concevoir un systeme vivant sans une struc- ture continue. PROTOPLASMA DES MYXOMYCETES 11 Je reste trfes oblige a MM. R. Lami et Comandon de s'etre interesses a cet article. BIBLIOGRAPHIE 1. Howard, F. L. — Laboratory cultivation of Myxomycete Plas- modia. Amer. Journ. Bot. 18, 624 (1931). 2. Nichols, S. P. — Methods of healing in some algal cells. Amer. Jour. Bot, 9, 18 (1922). 3. Nichols, S. P. — The effect of wound upon the rotation of the protoplasm in the internodes of Nitella. Bull. Torrey Bot. Club, 52, 351 (1925). 4. Martens, P. — Nouvelles recherches experimentales sur la cinese dans la cellule vivante. La Cellule, 39, 167 (1929). 5. ScARTH, G. M. — The structural organization of plant protoplasm in the light of micrurgy. Protoplasma, 2, 189 (1927). 6. Seifriz, W: — Viscosity values of protoplasm as determined by microdissection. Bot. Gaz., 70, 360 (1920). 7. Seifriz, W. The alveolar structure of protoplasm. Protoplasma, 9, 177 (1930). NEMOURS. IMPRTMERIK ANDRE LESOT. ( I 9 THE SIERRA NEVADA DE SANTA MARTA* AN ASCENT FROM THE NORTH William Seifriz University of Pennsylvania THE Sierra Nevada de Santa Marta has been singularly neglected. Practically all the accurate information we have about this mountain massif has been assembled in the last decade. Even so fundamental a fact as its height has not been definitely established: the figures given range between 18,000 and 21,000 feet. Until now no plants had been collected above 14.000 feet. Yet the snow-capped summits of the Sierra are only 25 miles distant from a readily accessible coast. In b b. 74- e a Rio Hach S Cabo de la A^uja 5 Santa Marta SahJuandeCi^na -r\-^r Goajira ^' Oi H •'(- 1,11 '^^^^'^'^ GEOOR BEVltW, JULV, 1934- ./ /'' I'^J^i i v-'X ^--^ Fundacidn *>— 7^ <^ X' ,3000 10 S 20 MILES / 73 10 20 KILOMETERS Heights in meters Fig, I — Map of the Sierra Nevada de Santa Marta showing the author's route. 1 93 1 Griffith Taylor made an ascent of the western slopes: he described his observa- tions under the title of ''Settlement Zones of the Sierra Nevada de Santa Marta, Colombia."^ In 1932 the writer made a geobotanical excursion into the massif from the north: he here gives an account of the zonal distributions on the northern slopes. The western route has also been used by Mason^ and Cabot,' the northern trail by Carriker,"* Wollaston,^ and. in part, by De Brettes.^ The best time of year for a visit to the Sierra Nevada is determined by one's interest. For alpine climbing and distant views the dry and clear season, from January to March, is the better. For biological study the early part of the rainy season, when plant and animal life are at their best, is preferable. On the northern slope, east of Santa Marta, there is a lull in the rainy season during late June and July, the "verano de San Juan," of which it is well to take advantage. *I wish to express my gratitude for the courteous hospitality and material assistance of their Excellencies, Don Miguel A. Zuniga. Acting Governor of the Province of Magdalena, and Don Manuel Julian de Mier, Mayor of Santa Marta. I am also indebted to Mr. Ellsworth Killip, of the United States National Museum, for the identification of many of the plants collected. » Geogr, Rev., Vol. 21, 1931. PP. 539-558. 'J. A. Mason: Archaeology of Santa Marta, Colombia: The Tairona Culture, Field Museum of Nat. Hist. Publ. 304 {Anthropol. Ser., Vol. 20, No. i), Chicago, 1931. ' T. D. Cabot: Mountains of the Caribbean, Appalachia, Vol. 18, 1930-1931. pp. 17-22, * W. E. C. Todd and M. A. Carriker, Jr.: The Birds of the Santa Marta Region of Colombia: A Study in Altitudinal Distribution, Annals Carnegie Museum, Vol. 14, 1922, pp. 3-61 1. ' A. F. R. Wollaston: The Sierra Nevada of Santa Marta, Colombia, Geogr. Journ., Vol. 66, I925t pp. 97-1 1 1. • Joseph de Brettes: Chez les Indiens du Nord de la Colombie, Le Tour du Monde, Vol. 4 (N.S.), 1898, pp. 61-96 and 433-480. 478 fX. * ^St^ VI- 4) ^ >4 ^ k ^T -^ ^l^ SIERRA NEVADA DE SANTA MARTA 479 Fig. 2 — A street in Dibulla, with coconut palms. The narrative of our ascent of the mountains begins at Dibulla (Fig. 2), a hot, malaria-infested negro settlement 60 miles east of Santa Marta. The trail follows the shore for some five miles westward, passing one of the few coconut plantations of the region, the coastal vegetation otherwise consisting almost wholly of the low tree Coccoloha (sea grape). The trail then turns inland across the plain where all of the discomforts of travel in the tropics seem to be combined — insects, heat, dust, marshes, and the absence of drinking water. It is, however, by no means without interest for the naturalist, for here, in the xerophytic coastal zone, is the home of the palmiche palm {Euterpe), the giant cactus {Cephalocereus), the sprawling calabash tree {Crescentia), which is so typical of Caribbean shores, and the bottle tree (Ster- culia rupestris), with its bulging trunk. Taylor refers to this last-mentioned tree as very characteristic of the somewhat higher xerophytic zone on the west slope. The Lowland Forest Ten to twelve miles from the shore bring one to the edge of the lowland forest. Here, in clearings near the site of the ancient village of Bonga, the natives have their garden patches— plantain, corn, and yuca; banana, papaya, and pineapple. The Fig. 3 — A street in Pueblo Viejo. with broad-leaved plaintain "trees SIERRA NEVADA DE SANTA MARTA 479 THE SIERRA NEVADA DE SANTA MARTA* AN ASCENT FROM THE NORTH William Seifriz University of Pennsylvania THE Sierra Nevada de Santa Marta has been singularly neglected. Practically all the accurate information we have about this mountain massif has been assembled in the last decade. Even so fundamental a fact as its height has not been definitely established: the figures given range between i8,o(^)o and 21,000 feet. Until now no plants had been collected above 14,000 feet. Yet the snow-capped summits of the Sierra are only 25 miles distant from a readily accessible coast. In C r b b 'e- a n s Cabo de la A|uja jr Santa Marta ■■^-^Vv. ,Dibu Riq Hach^--.., ^\-f^ Goajira 1 \ 0- \ \ '% '7 1 «r. -i^San Juan deCi^naga^ GEOCR REVtt»v, JULY, 1934. V, tr\ ,\jbrvniO'^ /: . J ►#— ?\ ^ agriculturally developed with success. Almost any type of crop that flourishes in the tropics or subtropics could be grown. The cultivated plants now to be found there include oranges, bananas, plantains, sugar cane, cotton, tobacco, onions, cab- bages, arracacha (Conium arracacha), potatoes, beans, and yams. The oranges gathered at San Francisco near by proved to be sweet and juicy. Apples and toma- toes could undoubtedly be grown with equal success. Coffee is not raised but is grown in quantity in identical regions elsewhere in the Sierra Nevada. Life at Pueblo Viejo would, under good living con- ditions, be comparable to that at the haciendas above Santa Marta at the same altitude. Furthermore, the northern slopes of the Sierra Nevada are watered by permanent streams — a natural and easily applied source for power and irrigation. Several trails lead out of Pueblo Viejo. one of them to San Francisco, the first of the few Arhuaco Indian villages. The facial characteristics of the inhabitants here and also the square houses indicate Spanish influence. Another trail leads west to the Indian village of Palomino, and yet another south to San Miguel, the highest of the Indian settlements. It is the last-named route that we shall travel. The trail rises rapidly, crosses a steep divide, and then descends to the Rio Ancho. Simons, on his map.^ shows San Miguel west and somewhat north of Pueblo Viejo, but this is an error: it lies to the southwest. San Miguel and the Arhuacos About two hours out of Pueblo \'iejo a group of several Indian huts, known as Santa Cruz, is passed. Another four or five hours, through open and mostly treeless country, bring one to San IVIiguel. at an altitude of 5260 feet. The first in the series of impressive approaches to the village is a bridge built of poles and woven cane stalks, spanning the deep ravine above which lies San Miguel, well protected from attack. This gorge proved to be unusually rich in ferns. The bridge leads directly to a pre- Columbian megalithic stone stairway, now much disarranged. Beyond the stairway lies the temple yard.- It reveals a feeling for order— a trait lacking in the coastal negroes. On the grounds are several circular huts, two with the characteristic tops that indicate the religious function of the buildings. The whole is surrounded by a 7 F. A. A. Simons: On the Sierra Nevada of Santa Marta and Its Watershed. Proc. Royal Geogr. Soc., Vol. 3 (N.S.), 1881, pp. 705-723. FiG. 5 — A cansatnana, or native teinyie, ctt Aiucutama. The Indians are chewing coca leaves, handling their suigis or gourds containing lime. g.TTwi:i3aa»a 480 THE GEOGRAPHICAL REVIEW SIERRA NEVADA DE SANTA MARTA 481 forest begins at an altitude of about 300 feet. It represents the finest plant associ- ation that the tropics have to offer. The trunks of the giant caracoli (Anacar- dium), higueron (Ficus), and guayabo (Eugenia) rise in the air to 150 feet. From their uppermost limbs swing the larger lianas, among them the sinuous air roots of young "strangling" figs. Every limb has its quota of epiphytes, mostly bromeliads. and with them orchids and pendent ferns. The forest floor is covered with the brilliantly flowered Heliconia (the so-called wild plantain), aroids. es- pecially the perforate-leaved Mon- stera and the renowned South American genus Anthurium, small palms (Chamaedorea). and the fan- tastic, white, spider-like flower of Hymenocallis, Through this luxu- riant tropical v^egetation the Indians of pre-Columbian times cut a broad roadway, which still exists, giving a superb vista into the forest. As the trail leads upward, there occur in abundance the fruits of the avocado. At an altitude of 2000 feet a marked change in plant life takes place. The huge lowland trees are now rare — they continue to greater heights on the western slopes of the Sierra — and the vegetation in gen- eral is sparser. New forms appear, among them the tree fern and the climbing bamboo, certain species of which have the extraordinary habit of flowering but once in their life- time of 32 years. Terrestrial orchids of great beauty, rivaling the epi- phytic ones, are numerous in the grassy fields here and at higher altitudes. Fig. 4 — The patriarch of the Arhuacos. The Negro Village of Pueblo Viejo The trail proceeds to the negro village of Pueblo Viejo (Fig. 3). at an altitude of 2800 feet. In spite of its name the hamlet is relatively new. replacing the now extinct village of San Antonio, the disappearance of which is variously ascribed to destruction by Colombian soldiers during the last revolution or by civilizados who drove out the inhabitants or by fires set by the Indians in retaliation against the blacks. Living conditions at Pueblo X'iejo are no better than on the coast from the point of view of accommodations, but climate and surroundings are more agreeable. The village, though in itself unattractive, nestles among grass-covered hills domi- nated by the precipitous Cerro Nanu. The higher altitude, with its cool nights and abundant rainfall, adds greatly to the health of the people and to the luxuriance of plant life. Bananas and plantains reach a huge size, with stalks 15 feet high and a foot in diameter. Sugar cane is a prolific grower. The natives are less diseased, and the pigs are fatter. Two valuable additions to the food supply are to be found here — milk and oranges. As far as climate and soil are concerned. Pueblo X'iejo is a region that could be k>\^ ♦) \ f ♦) ' > .1. r \ I ' 4k lillMM 1 .iijI^.^M '^^- ._..., V^H, :\^t agriculturally developed with success. Almost any type of crop that flourishes in the tropics or subtropics could be grown. The cultivated plants now to be found there include oranges, bananas, plantains, sugar cane, cotton, tobacco, onions, cab- bages, arracacha {Conium arracacha), potatoes, beans, and yams. The oranges gathered at San Francisco near by proved to be sweet and juicy. Apples and toma- toes could undoubtedly be grown with equal success. Coffee is not raised but is grown in quantity in identical regions elsewhere in the Sierra Nevada. Life at Pueblo X'iejo would, under good living con- ditions, be comparable to that at the haciendas above Santa Marta at the same altitude. Furthermore, the northern slopes of the Sierra Nevada are watered by permanent streams — a natural and easily applied source for power and irrigation. Several trails lead out of Pueblo X'iejo. one of them to San Francisco, the first of the few Arhuaco Indian villages. The facial characteristics of the inhabitants here and also the square houses indicate Spanish influence. Another trail leads west to the Indian village of Palomino, and yet another south to San Miguel, the highest of the Indian settlements. It is the last-named route that we shall travel. The trail rises rapidly, crosses a steep divide, and then descends to the Rio Ancho. Simons, on his map.^ shows San Miguel west and somewhat north of Pueblo X'iejo, but this is an error; it lies to the southwest. fMiib^J^ '^ -idif ^..m.- Fig. 5 —A cansamaria, or native leniple. ai Macotaiad. The Indians are chewing coca leaves, handling their siiigis or gourds containing lime. A ^ Sax Miguel and the Arhu.\cos About two hours out of Pueblo X'iejo a group of several Indian huts, known as Santa Cruz, is passed. Another four or five hours, through open and mostly treeless country, bring one to San Miguel, at an altitude of 5260 feet. The first in the series of impressive approaches to the village is a bridge built of poles and woven cane stalks, spanning the deep ravine above which lies San Miguel, well protected from attack. This gorge proved to be unusually rich in ferns. The bridge leads directly to a pre- Columbian megalithic stone stairway, now much disarranged. Beyond the stairway lies the temple yard. It reveals a feeling for order— a trait lacking in the coastal negroes. On the grounds are several circular huts, two with the characteristic tops that indicate the religious function of the buildings. The whole is surrounded by a 1 F. A. A. Simons: On the Sierra Nevada of Santa Marta and Its Watershed, Proc. Royal Geogr. Soc, Vol. 3 (N.S.). 1881, pp. 705-723. intentional second exposure 482 THE GEOGRAPHICAL REVIEW SIERRA NEVADA DE SANTA MARTA 483 planting of Fourcroya, a stiff-leaved plant closely related to and very much resembling the agave, which serves the Indian as a source of fiber for rope, bags, and the like. A few minutes from the temple grounds bring one to the village proper, with its small and artistic portal, a kind of lich gate (Fig. 6)— further evidence of the Indian's love of seclusion and beauty. Immediately beyond lies the village, hemmed in by high mountain ridges and resting on a small plateau just a mile above the sea. Some sixty or seventy huts grouped closely together with neat open spaces between them constitute the village of San Miguel (Fig. 7)— room enough to house fully a hundred Fig. 6 — The entrance gate to San Miguel. (Drawn from a photograph by E. W. Emmart.) people, yet when we entered not a person was in sight. The village is used only as a gathering place. Every Indian has his one, two, or three small farms some distance away, where he has built himself a hut as comfortable as the one he owns in town. Here he goes with his wife, children, pigs, chickens, and dog to care for first one crop and then the next. In the higher altitudes he grazes his cattle and plants potatoes and arracacha. Near the village, or slightly lower, plaintains, sugar cane, corn, cotton, coca, and tobacco are grown. In the hot lowlands he raises yuca, bananas, and some few other fruits, though it is the negro who enjoys the pine- apple and papaya more. The houses of San Miguel are of several types, each serving a different purpose. The commonest are the small circular ones of mud. When a man is young, he builds himself a hut. On marrying, he builds a second one opposite and close by, for his wife and children. The family eats outdoors between the two huts, or the husband takes his meal to the men's hut. This latter is a large structure with walls of plaited cane instead of mud. The religious houses possess a specially constructed apical ornamentation the character of which indicates the precise use to which the temple or cansamaria is put — for novitiate, priest, or high priest. Finally, there are the square houses for foreigners or for purposes that are foreign to the life of the Indian, as is, for example, the Catholic church, opened once a year when the priest pays his annual visit. The Arhuacos differ markedly from the coastal Goajiras, their nearest aboriginal neighbors, both in appearance and in character. Mason* contrasts the two tribes as follows: the Goajiras are strong, progressive, and aggressive; the Arhuacos are weak, inveterate coca chewers, sedentary, and shun contact with the outside world. They have a curiously Asiatic look; indeed, they were formerly referred to by the Colombians as "Chinos" rather than as "Indios." Among the Arhuacos families 8 J. Alden Mason: Coast and Crest in Colombia: An Example of Contrast in American Indian Culture, Nat. Hist., Vol. 26, 1926. pp. 31-43. ■ are small, with two children as the usual maximum. Intermarriage with Colombians, either white or black, is rare, nor do the Arhuacos mingle with the adjoining Goajira tribes. The name "Arhuaco" is not used by the Indians themselves; it, or rather the original form of it, " Aurohuacos," seems to have first appeared in Jose Nicolas de la Rosa's "Floresta de la Santa Iglesia Catedral de la Ciudad de Santa Marta" (1739). Its similarity in sound to Arawak, one of the large primitive stocks of South American Indians, suggests a possible origin of the name, though anthropologically it is misleading. The tribe is not identified with the ancient inhabitants of the PiG, 7— San Miguel. The square houses include the Catholic church, bishop's house, and house for guests. (Drawn from a photograph by E. VV. Emmart.) Sierra Nevada. These were the Taironas. a warlike people as compared with the peaceful Arhuacos. The Taironas disappeared four hundred years ago. and the Arhuacos will probably not long survive. There are probably not more than 3000 Arhuacos all told in the Sierra Nevada. That branch of the Arhuacos living on the northern side of the mountains is known as Kagaba (or Koggaba). Those on the southern side are the Ijca (De Brettes refers to them as the Bintoukouas). There are also smaller groups— the Sanha, Busintana. Gouamakas. and Marocaso. Above San Miguel are several small settlements of three or four huts each where priests reside. The cansamarias are here at their best (Fig. 5). The priests also represent the finest specimens of the dying Arhuaco race. Two of these groups of huts are named— the first, Taquina, where lesser priests reside; and the second, Macotama, the home of the most renowned priest, or mama, in these parts, patriarch of the Arhuacos, famed far and wide for his medicinal powers. He was not at home on our upward journey, but on the return we met him (Fig. 4). and in the customary exchange of gifts I received a kerchief full of camomile, a plant cultivated in abun- dance around the huts of the priests and used medicinally. The tree line is passed shortly after leaving San Miguel, at about 6000 feet. On Mt. San Lorenzo, on the western side of the Sierra Nevada, at this altitude and higher— up to 7500 feet— there occurs the tall wax palm Ceroxylon (Fig. 8). Nec- tandra, which the natives know as laurel, climbs highest of the dicotyledonous trees. Four hours of hard going from San Miguel bring one to a group of mud huts at an altitude of 9200 feet. Fog and the usual afternoon rain make further climbing inadvisable. Again there is a pronounced change in plant life. Only shrubs, grasses, and alpine herbaceous forms now cover the few areas where there is soil. The Paramos Another half day brings one to the last group of shepherds' huts, at 12.600 feet. We are here at the edge of the paramo, which in this region is mostly bare rock u SIERRA NEVADA DE SANTA MARTA 485 Fig. 8 Fig. 9 ^\ V f K N. ^^0. "^*v ^ i p Fig. 10 Fig. 8 — The wax palm at 7500 feet on Mt. San Lorenzo. Fig. 9— Laurel trees forming the tree line at 7000 feet on Mt. San Lorenzo. Fig. 10 — The paramos at 12,600 feet. 484 ^iC> ^)v (Fig. 10). At 14,000 feet the glacial Lake Macotama is reached, and 1200 feet higher a second smaller body of water, named vSummit Lake by Carriker. Neither lake contains any fish. Just above the second lake on the floor of the paramo, at an alti- tude of 16,000 feet, a distinctive alpine vegetation flourishes. Among the flowers the elegant purple spike of the lupine is conspicuous, and the semi-arborescent frailejon (Espeletia), a plant like a gigantic mullein in appearance and limited to the paramos. Carriker states that the flowers of frailejon furnish a large portion of the food of hum- ming birds, which are very numerous in the Sierra Nevada. The queen of all the alpine flowers is Draba, and it has the added honor of climbing the highest. A speci- men was collected on the crest of the ridge, protected in the crevice of a rock, at an altitude of 17,640 feet. It proved to be a new species and has been named Draba sanctae-martae. Of the 600 plants gathered some thirty were heretofore unknown to science; they include among them new species of Weinmannia, Piper, Peperomia, and Tibouchina. The altitude of 17,640 feet is probably a record for flowering plant life in the Sierra Nevada de Santa Marta. Ferns persist on the paramos, chief among them the indigenous and extraordinary Jamesonia, its narrow, stiff" frond. 18 inches tall, resembling a flattened stick stuck in the ground. The only animal seen at this altitude was a small, glistening, ebony-black frog {Antelopus carrikeri). The weather on the morning of our arrival at the last camp (12,600 feet) was not propitious; heavy fog had collected, and a cold wind, threatening snow, whistled across the paramos. Only two of our party continued beyond 16,500 feet. Over bare rock and through patches of snow we slowly climbed upwards, finally reaching the ridge which is the crest at that point. Over the edge is a sheer drop of some two or three thousand feet to the valley below. A sudden rift in the clouds revealed the peaks to the west, at least two thousand feet higher. The ridge increased in elevation to the east. We proceeded in this direction, climbing slightly higher, until we came against a rocky prominence that defied ascent. The setting in of a driving snowstorm was the final circumstance that turned us back. The Height of the Sierra Nevada Statements in regard to the maximum height of the Sierra Nevada de Santa Marta range between 18,000 and 21,000 feet. Only De Brettes has ascended the highest peaks, and he gives 5887 meters (19,310 feet) as the altitude of the summit. The crest of the ridge reached by us is 17.640 feet. The snowy peak just beyond, to the west, rises at least 1500. if not 2000, feet above. This peak is therefore cer- tainly not less than 19,000 feet. Wollaston's photographs show the snowfield and indicate it to be of substantial size. The central peaks of the range are still higher and rise enough above the others to dominate them when viewed at a distance. They too show snowfields. A conservative estimate of the altitude of the Sierra Nevada de Santa Marta places it at 20.000 feet. l'^ SIERRA NEVADA I)E SANTA MARTA 485 Fig. 8 Fig. 9 Fig. 10 F"'iG. 8 — The wax palm at 7500 feet on Mt. San Lorenzo. Fig. 9— Laurel trees forming the tree line at 7000 feet on Mt. San Lorenzo. Fig. id — The paramos at 12,600 feet. VI «V V #-<-. "^i. (Fig. 10). At 14,000 feet the glacial Lake Macotama is reached, and 1200 feet higher a second smaller body of water, named Summit Lake by Carriker. Neither lake contains any fish. Just above the second lake on the floor of the paramo, at an alti- tude of 16,000 feet, a distinctive alpine vegetation flourishes. Among the flowers the elegant purple spike of the lupine is conspicuous, and the semi-arborescent frailejon (Espeletia), a plant like a gigantic mullein in appearance and limited to the paramos. Carriker states that the flowers of frailejon furnish a large portion of the food of hum- ming birds, which are very numerous in the Sierra Nevada. The queen of all the alpine flowers is Draba, and it has the added honor of climbing the highest. A speci- men was collected on the crest of the ridge, protected in the crevice of a rock, at an altitude of 17,640 feet. It proved to be a new species and has been named Draba saftctae-martae. Of the 600 plants gathered some thirty were heretofore unknown to science; they include among them new species of Weinmannia, Piper, Pepcromia, and Tibouchina. The altitude of 17,640 feet is probably a record for flowering {)lant life in the Sierra Nevada de Santa Marta. Ferns persist on the paramos, chief among them the indigenous and extraordinary Jamesonia, its narrow, stiff frond, 18 inches tall, resembling a flattened stick stuck in the ground. The only animal seen at this altitude was a small, glistening, ebony-black frog {Avtelopus carrikeri). The weather on the morning of our arrival at the last camp f 12,600 feet) was not propitious; heavy fog had collected, and a cold wind, threatening snow, whistled across the paramos. Only two of our party continued beyond 16.500 feet. Over bare rock and through patches of snow we slow ly climbed upwards, finally reaching the ridge which is the crest at that point. Over the edge is a sheer drop of some two or three thousand feet to the valley below. A sudden rift in the clouds revealed the peaks to the west, at least two thousand feet higher. The ridge increased in elevation to the east. We proceeded in this direction, climbing slightly higher, until we came against a rocky prominence that defied ascent. The setting in of a driving snowstorm was the final circumstance that turned us back The H right of the Sierra Nevada Statements in regard to the maximum height of the Sierra Nevada de Santa Marta range between 18,000 and 21,000 feet. Only De Brettes has ascended the highest peaks, and he gives 5887 meters (19.310 ^eet) as the altitude of the summit. The crest of the ridge reached by us is 17,640 feet. The snowy peak just beyond, to the west, rises at least 1500. if not 2000. feet above. This peak is therefore cer- tainly not less than 19,000 feet. Wollaston's photographs show the snowfield and indicate it to be of substantial size. The central peaks of the range are still higher and rise enough above the others to dominate them when viewed at a distance. They too show snowfields. A conservative estimate of the altitude of the Sierra Nevada de Santa Marta places it at 20.000 feet. J V 484 INTENTIONAL SECOND EXPOSURE ««M«wi^ii*dayui 'c/ &I J . Reprinted from Proceedings or the Pennsylvania Academy of Science, Vol. VIII, 1934, pages 43-48. \i I Vi 4> A' ,»-r- \ i A <► MORPHOLOGY OF SCHIZANTHUS WISET0NEN8IS By Esther D. Still Department of Botany, University of Pennsylvania Introduction Species of Schizanthus have been in cultivation for at least 100 years, and have been known by such different popular names as Poor Man's Orchid, Butterfly Flower, and Fringe Flower. In addition to some seven species now cultivated, there are several well established hybrids, as well as a few fixed varieties. These half-hardy, annual herbs are na- tive to Chile. Hipolito Ruiz Lopez and Jose Pavon in their Flora of Peru and Chile (1794) described Schizanthus pinnatus, the first species to be recorded. At present Schizanthus is placed under Solanaceae, a family unique in many respects in that it presents many features which are important both to morphologists and taxonomists. This genus occupies an ad- vanced position in this, the Nightshade family, due to the fact that it has a markedly irregular corolla (zygomorphic) and only two fertile sta- mens, in contrast to the regular corolla (actinomorphic) and five fertile stamens of the great majority of the members of the family. The struc- ture of the flower is so different from the general type of Solanaceae that the earlier taxonomists placed Schizanthus in the closely allied family, Scrophulariaceae. The change from the latter family to the former is a comparatively recent one. The anatomy of a number of genera and species of Solanaceae has been studied, but members of the genus Schizanthus have received very little attention, and those studied were not the species now under con- sideration. All the members of the Solanaceae that have been investigated show bicollateral bundles in the stem. In this family, as in others^ with this anomalous condition, the nature and origin of internal phloem has attracted many investigators. The main purpose of this study is to consider the morphology of Schizanthus WisetonensiSj and especially 44 PENNSYLVANIA ACADEMY OF SCIENCE i the internal anatomy of the plant in an effort to discover the origin of the internal phloem, constituting a part of the bicollateral bundle. General Morphology of the Mature Plant Schizanthus Wisetonensis, like other species of this genus, is char- acterized by its finely-cut, compound leaves and variously colored flowers. It is preferred to some of the other species because of its rather smaller and more compact habit, its more profusely branching stems, and greater variation in flower color. This last-named character is prob- ably due to the heterozygous nature of this form, since it is undoubtedly a hybrid form, originating at Wiseton, England, from a cross of S. Grahami and S. pinnatus. The mature plant when in full bloom has a very extensive root system, consisting of a short thick tap-root fully 10 mm. in diameter and many long side roots which average 2 mm. in diameter and are often more than 15 cm. long. Even though the root system is unusually extensive when the plants are grown under conditions where they have plenty of space, yet the plants do very well in relatively small flower pots. The main axis is green and erect, and covered by small unicellular hairs. In the dwarf forms recently produced in this species by selection numerous branches arise from the main axis, approximating the central stem in thickness and length. Thus the mature plant is rounded and quite symmetrical. The inflorescence is cymose with numerous flowers borne on delicate, pubescent pedicels. The calyx is only slightly irregular, and like the corolla consists of five segments. The bilabiate corolla is white, tinged with pink. The mid-lobe of the upper lip is light brown at the base, flecked with brilliant yellow spots. The androecium consists of two exserted fertile stamens and two shorter slender staminodes; sometimes the fifth stamen is represented by a very small reduced staminode. The gynoecium follows the usual structure of that part of the flower in this family. . General Morphology of the Seedling Under ordinary conditions the seeds germinate within 10 days after planting, the time for germination depending on the season of the year. It is particularly difficult to obtain good seedlings during July and August. Germination is epigeal, the hypocotyl being well-developed at the end of 2 weeks. At this stage the cotyledons have been carried aloft by the developing hypocotyl, unfold and begin their temporary function of photosynthesis. The complete seedlings, at this time, vary in length Si PENNSYLVANIA ACADEMY OF SCIENCE 45 y * m' from 4 to 5 cm., with cotyledons which average 6 mm. in length and 2 mm. at their widest part. The cotyledonary stalk is only slightly developed. The primary root has developed only a few short side root- lets. At the end of 4 weeks the seedlings have developed many secondary rootlets which are slightly smaller and are much longer than the primary root. The cotyledons still persist at 15 to 20 mm. above the ground level, their length now averaging 10 mm. and width 4 mm. The petiole of the cotyledon is well developed, often attaining a length of 12 mm. Between the cotyledons the epicotyl has given rise to the first foliage leaves. Soon after the first foliage leaves appear the cotyledons turn yellow and drop. Anatomy of the Seedling The lower end of the primary root has a typical though immature root structure. It shows definitely the diarch condition, although only a few xylem elements are present with small phloem patches located one on each side of the xylem. The pericycle can be distinguished as a more or less regular layer of cells which defines the limits of the stele. It consists of small thin-walled parenchymatous cells which are crowded closely against the endodermis and are, therefore, variously shaped. The endodermis is a row of thin-walled, regularly shaped, large cells distin- guished by the presence of slight thickenings in its radial walls. The remainder of the cortex is made up of from three to four layers of parenchymatous cells which can be distinguished from the endodermis because the cells are larger and more irregularly arranged. The cells of the epidermis are rectangular and give rise to numerous root hairs. Serial sections of a seedling about fifteen days old, in which foliage leaves have not yet developed, reveal the internal conditions and serve to demonstrate the transition from root to stem. This transition takes place in the hypocotyl, beginning about 10 mm. below the cotyledonary level. The transition is completed just as the cotyledonary petioles arise from the main axis. The change from root to stem structure is, there- fore, accomplished in a relatively short time in marked contrast to most dicotyledons of this type in which the transition is relatively long. That the transition is complete before the petioles of the cotyledons arise, seems to contradict the findings of other workers in the family. Thiel, working with several solanaceous plants, found that the transition was completed in the cotyledonary petiole. He states that ''In the coty- ledonary petiole and midrib the protoxylem differentiates abaxially until the endarch condition is fully established." Avery, however, working with tobacco (Nicotiana tahacum) finds that the internal phloem is estab- lished at a level 0.7 mm. below the cotyledons. 1 Ill 46 PENNSYLVANIA ACADEMY OF SCIENCE PENNSYLVANIA ACADEMY OF SCIENCE 47 h I i At an early stage in the transition the patches of phloem elongate laterally toward the protoxylary ends thus forming approximately a concentric condition. Within a short distance this ^^ring'^ of phloem divides into distinct patches of unequal size. Some of these groups of phloem cells are destined to become the internal phloem of the coty- ledonary and cauline traces higher up in the hypocotyl. The primary xylem of the hypocotyl, consisting of a few protoxylem elements and a limited number of metaxylem vessels, is supplemented by the addition of a small amount of secondary xylem developed by the cambium as transition proceeds. The xylem area of the hypocotyl forms first an irregular mass. It then divides gradually into two equal, triangular strands separated by parenchyma cells. In each of these strands the protoxylem is at the apex of the triangle. The protoxylem of each of the two areas now begins to swing in toward the center and inversion from the exarch of the root to the endarch of the stem takes place rapidly as suc Fig. 1. Road through the red-cedar forest, with Bowmans Hill in the distance. 8 there is an extensive growth of red-cedar, in places forming nearly pure stands and reaching forest-tree stature. Toward the northeast there are more recently abandoned fields, and these are being invaded by pioneer plants, such as sedge-grass {Andropogon) and sumac (Rhus) in addition to the red-cedars. Mr. W. W. Heinitsh, of Chestnut Hill, Philadelphia, has been desig- nated warden of the preserve, and is to have charge of all development on the ground. Under his direction a number of signs are being made, by smoothing oif surfaces of variously shaped sections of tree trunks and Fig. 2. Conservation signs in course of construction. Stumps, and burning letters into them with heated irons. One set marks entrances to trails in the preserve, another carries conservation mottoes, and a third indicates what is believed to be a novel feature, namely an extensive field in which the public is invited to pick flowers. It is hoped that this provision will lessen the danger of destruction of the rarer species planted in the preserve proper. Later on it is hoped to have nature trails, labels marking colonies of noteworthy species, and other educational features. Teachers throughout the region are invited to make use of the tract for class instruction, and to send in suggestions for increasing its value in this direction. it 3 laurel (Kalmia) and trailing-arbutus (Epigaea). In the valley to the north the underlying rock is a sediment of Triassic age, which has been metamorphosed by heated waters emanating from the igneous mass, and changed from its original character of soft red shale to a hard gray argillite. This outcrops along the banks of the creek, and is especially well exposed at the copper mine, w^hich lies in the hill to the north, and was famous in colonial times, though apparently never produced much ore. North of the creek valley the rock is deeply covered by a pinkish- gray, sandy loam soil, which also tends to become acid wiierever leached by the rain. The water of Pidcock Creek is slightly alkaline, but several small acid-water springs or seeps are present on the tract. Since it is much easier to develop neutral soil, for plants which need it, in regions of natural acidity than to carry out the reverse operation, conditions would seem to be favorable here for the establishment of a considerable proportion of the plants native to the State. Much of the ground is heavily covered with oak woods, mingled with hemlock along the north slope of Bowmans Hill and occasionally else- where. On an area which apparently represents a long-abandoned field there is an extensive growth of red-cedar, in places forming nearly pure stands and reaching forest-tree stature. Toward the northeast there are more recently abandoned fields, and these are being invaded by pioneer plants, such as sedge-grass {Andropogon) and sumac {Rhus) in addition to the red-cedars. Mr. W. W. Heinitsh, of Chestnut Hill, Philadelphia, lias been desig- nated warden of the preserve, and is to have charge of all development on the ground. Under his direction a number of signs are being made, by smoothing off surfaces of variously shaped sections of tree trunks and Fig. 1. Eoad through the red-cedar forest, with Bowmans Hill in the distance. Fig. 2. Conservation signs in course of construction. stumps, and burning letters into tliem with heated irons. One set marks entrances to trails in the preserve, another carries conservation mottoes, and a third indicates what is believed to be a novel feature, namely an extensive field in which the public is invited to pick flowers. It is hoped that this provision will lessen the danger of destruction of the rarer species planted in the preserve proper. Later on it is hoped to have nature trails, labels marking colonies of noteworthy species, and other educational features. Teachers throughout the region are invited to make u.se of the tract for class instruction, and to send in suggestions for increasing its value in this direction. INTENTIONAL SECOND EXPOSURE Separately printed, without change of paging, from Bulletin of the Torrey Botanical Club 61 : 81-84. 1 February, 1934 ( ; I ' \ j<^ i;^<^\ ..> Fig. 3. The field where the public is invited to pick flowers. Introduction of plants from various parts of the State will be begun this season, and will go on as rapidly as funds permit. It is hoped that members of the Pennsylvania Academy of Science will aid the project by advising us as to places in the State where colonies of interesting plants are threatened with destruction through road-building, draining or flooding of swamps, or vandalism on the part of the public, to the end that clumps may be transplanted and given an opportunity to survive in a permanently protected place. A series of lantern slides was shown at the meeting to illustrate the present aspect of the tract. Three of these views are reproduced here. '4 Ai»< ^ The box huckleberry as an illustration of the need for field work^ Edgar T. Wherry The plant here discussed is a low evergreen ground-covering shrub with leaves resembling those of the common box, but with the technical char- acters of the blueberry family (Vacciniaceae), Michaux, in his *^ Flora Bor- eali-Americana/' named it Vaccinium brachycerum, and gave the locality as "Virginia circa Winchester," although the specimen in his herbarium is labelled "Warm Springs," perhaps referring to what is now known as Berkeley Springs, in West Virginia. It was again found by Matthias Kinn, about 1800, in the Greenbrier VaUey, and by Pursh in 1805 near Sweet Springs. Upon the death of these early collectors, the locaUties from which they had obtained the box huckleberry were lost to science, and for many years Asa Gray was unable to obtain material for study in connection with his monograph of the family. In 1845, however, a colony of it was discovered by Spencer F. Baird near New Bloomfield, Perry County, Pennsylvania, enabling Gray to prepare an adequate description of the species and to refer it to the correct genus, its name becoming Gaylussacia brachycera. The friendship which sprang up between Gray and Baird ultimately led to the latter's becoming Secretary of the Smithsonian Institution, so this plant may be said to have played an important part in the development of science in America. In 1918 Dr. Frederick V. Coville recognized that the isolated colony of the plant discovered by Baird consisted of a single individual, which had spread over 8 acres by means of rootstocks. As these grow on the average but 6 inches per year, the colony must have sprung from a seed which had germinated about 1200 years previously, and was thus to be numbered among the oldest of living things. Being nearly self-sterile, the seeds failed to produce young plants with sufficient vitality to reach maturity, and the species appeared to be in real danger of extermination through clearing of the land and vandaHsm on the part of the public.^ If, however, another colony could be found and cross-poUination be carried out, vigorous seed- lings might conceivably be obtained, permitting the introduction of the 1 Presented in a Symposium on Objectives and Methods in Field Work, System- atic Section, Botanical Society of America, at Atlantic City on December 28, 1932. Contribution from the Botanical Laboratory and Morris Arboretum of the University of Pennsylvania. « This area has subsequently been made a state preserve. 81 I Separately printed, without change of paging, from Bulletin of the Torrey Botanical Club 61: 81-84. I February, 1034 Fig. 3. The field where the public is invited to pick flowers. Introduction of plants from various parts of the State will be begun this season, and will ^0 on as rapidly as funds permit. It is hoped that members of the Pennsylvania Academy of Science will aid the project by advising us as to places in the State where colonies of interesting plants are threatened with destruction through road-building, draining or flooding of swamps, or vandalism on the part of the public, to the end that clumps may be transplanted and given an opportunity to survive in a permanently protected place. A series of lantern slides was shown at the meeting to illustrate the present aspect of the tract. Three of these views are reproduced here. 'I i«l ^ ^ The box huckleberry as an illustration of the need for field work^ Edgar T. Wherry The plant here discussed is a low evergreen ground-covering shrub with leaves resembling those of the common box, but with the technical char- acters of the blueberry family {V acciniaceae) . Michaux, in his ^^ Flora Bor- eali-Americana/' named it Vaccinium hrachycerum, and gave the locality as "Virginia circa Winchester/' although the specimen in his herbarium is labelled "Warm Springs," perhaps referring to what is now known as Berkeley Springs, in West Virginia. It was again found by Matthias Kinn, about 1800, in the Greenbrier Valley, and by Pursh in 1805 near Sweet Springs. Upon the death of these early collectors, the locaUties from which they had obtained the box huckleberry were lost to science, and for many years Asa Gray was unable to obtain material for study in connection with his monograph of the family. In 1845, however, a colony of it was discovered by Spencer F. Baird near New Bloomfield, Perry County, Pennsylvania, enabUng Gray to prepare an adequate description of the species and to refer it to the correct genus, its name becoming Gaylussacia brachycera. The friendship which sprang up between Gray and Baird ultimately led to the latter's becoming Secretary of the Smithsonian Institution, so this plant may be said to have played an important part in the development of science in America. In 1918 Dr. Frederick V. Coville recognized that the isolated colony of the plant discovered by Baird consisted of a single individual, which had spread over 8 acres by means of rootstocks. As these grow on the average but 6 inches per year, the colony must have sprung from a seed which had germinated about 1200 years previously, and was thus to be numbered among the oldest of Hving things. Being nearly self-sterile, the seeds failed to produce young plants with sufficient vitaUty to reach maturity, and the species appeared to be in real danger of extermination through clearing of the land and vandaUsm on the part of the pubUc.^ If, however, another colony could be found and cross-pollination be carried out, vigorous seed- Ungs might conceivably be obtained, permitting the introduction of the 1 Presented in a Symposium on Objectives aad Methods in Field Work, System- atic Section, Botanical Society of America, at Atlantic City on December 28, 1932. Contribution from the Botanical Laboratory and Morris Arboretum of the University of Pennsylvania. 2 This area has subsequently been made a state preserve. 81 INTENTIONAL SECOND EXPOSURE 82 BULLETIN OF THE TORREY CLUB [VOL. 61 species into horticulture as well as into protected areas, and its consequent perpetuation. At that time, however, not a single other occurrence was known to any botanist consulted by Dr. CoviUe, nor was there any infor- mation in the Uterature which would lead to the finding of one. Had a hst of American reUc-endemics been compiled at that time, it would certamly have included the box huckleberry. Fig 1. Supposed distribution of Gaylussacia brachycera in 1918. 1. "Warm Springs," Michaux about 1790 = "Near Winchester," 1803; presumably Berkeley Sprmgs, Mor- gan County, West Virginia. 2. "Krien Preyer," Kinn, 1800; = Greenbrier VaUey east of Lewisburg, Greenbrier County, West Virginia. 3. Sweet Sprmgs, Pursh, 1805, m Monroe County, West Virginia. 4. New Bloomfield, Perry County Peonsylvama, Baird, 1845. 5. Millsboro, Sussex County, Delaware, Commons, 1876. 6. I'amvuie, Polk County, Tennessee, Gattinger, 1901 . AU but No. 4 were lost to science m 1918. A colony of the species had been found by A. Commons of Wihnington, Delaware, in the southern part of that state about 1875, and although it was reported to have been destroyed, the writer succeeded in rediscovering it in 1919. Cross-poUination between this and a clump from the Pennsyl- vania colony was carried out, and an article on the plant was pubUshed by Dr. Coville.s Then one after another additional occurrences were brought to notice, —in Pennsylvania a few miles east of Baird's locaUty, in Maryland less than an hour's ride from Washington, in southwestern Virginia, and so on. •Science, SO: 30. 1919, X • 1 i V r* 1934] wherry: box huckleberry 83 In 1921 Rev. Fred W. Gray observed it near Dorr, West Virginia, and learning that it was locally known as "juniper-berry" and was used for food by the people of that region, he published in a local newspaper an inquiry as to where the plant so-named could be found. Notwithstanding the fact that the species had not been mentioned in any of the compila- tions deaUng with the flora of the state, he received reports of over 75 Fig. 2. Known distribution of Gaylussacia brachycera in 1932. Pennsylvania: New Bloomfield; opposite Losh Run Station, also in ^^e^ry a)unty, H. A. Ward, 1920; 15 miles northwest of Lebanon, Lebanon County, H. J. Roddy, Delaware: Millsboro, rediscovered by the writer, 1919; west of Bethel, also in Sussex County, W. S. Taber, 1932. . mm u . f Maryland: Pasadena, Anne Arundel County, C. C. PUtt, found about 1910 but not reported until 1920. West Virginia: Michaux's locaUty never rediscovered; numerous localities found through efforts of Rev. Fred W. Gray, 1921, in Greenbrier, Mercer, Monroe, Pocahontas, and Summers counties. ^ ,„ <-, i? • Virginia: Several locaUties in Alleghany, Bath, and Craig counties, F W. Gray; Fnes Junction, CarroU County, W. K. La Bar, found about 1912, but not reported Kent'Sy: Just west of Cumberland FaUs, PenneU and Wherry, 1927; two miles fur- ther west. Miss E. L. Braun, 1932, both in McCreary County. Tennessee: Gattinger's locaUty never rediscovered; 3 miles east of Allardt, Fentress County, S. H. Essary, 1920; along White Oak River near Rugby, boundary be- tween Fentress and Morgan counties, S. A. Cain, 1930; along Obed River near Rugby, Morgan County, Essary, 1931 84 BULLETIN OF THE TORREY CLUB rvOL. 61 stations distributed through five counties, and representing hundreds of acres of ground-cover.* Still later it turned out that in Kentucky the plant is called "ground-huckleberry" and in Tennessee "bear-huckleberry," and no doubt colonies additional to the few now known could be located in those states by similar inquiry under these names. Instead of being a relic-endemic, limited to one or two isolated colonies, as believed for a time, the species actually occurs in at least 7 states, from sea-level on the Coastal Plain to 3000 feet up on the Appalachian Plateau. The fact that the ranges of native plants are stated with a considerable degree of definiteness in manuals and local floras tends to lead the student and amateur botanist to feel that we are already well informed as to the distribution of species in this country. Circumstances such as those above discussed indicate, however, that such is not the case. When a plant which forms a cover over many acres in at least five counties, and is known to na- tives as a source of food, fails to be so much as mentioned in a pretentious state flora, the need for more intensive work on plant geography is evi- dent. Teachers will do well to encourage their students who are so fortu- nate as to live within reach of areas where man has not yet destroyed the native vegetation, to compile local floras with full field notes, including the names used for the plants by laymen. Before theorizing as to the principles of plant distribution, the relation of area to age, etc., let us first find out more as to where our species of native plants really grow. Department of Botany University of Pennsylvania *Torreya22:17. 1922. I h rr y. 4 k ♦ « 'I 4 i A V Eeprinted from Bartonia, No. 15, 1933 The Eastern Veiny-leaved Phloxes^ Edgar T. "Wherry For the fourth section of the eastern Phloxes the name pro- posed by Peter^ is acceptable. Its distinctive features com- prise wholly deciduous foliage, relatively large leaves with prominent areolate veins and minutely hispid-serrulate mar- gins, a compound corymbose-paniculate inflorescence, whitish anthers, and styles equalling or exceeding the corolla-tube. Although numerous names have been proposed for plants which belong here, the majority of these were applied to vari- ants without taxonomic significance or to horticultural forms of unknown origin, and only two species are clearly distin- guishable, as indicated by the following tabulation. PHLOX, SECTION PANICULATAE: KEY TO SPECIES Mature plant 50 to 150 cm. tall, with few nodes; leaves opposite, rela- tively broad, their surfaces often beset with coarse bristles; inflores- cence densely glandular-pubescent; corolla-tube glabrous; anthers usually all included 14. P. amplifolia Mature plant 75 to 200 cm. tall, with numerous nodes ; leaves tending to be subopposite, narrow to moderately broad, their surfaces glabrous to pubescent but rarely coarse-bristly; inflorescence more or less pubes- cent but infrequently glandular ; corolla-tube often pubescent ; one or more anthers exserted 15. P. pamculata While the large and often coarse-bristly leaves of P. ampli- folia seem more specialized than those of P. paniculata, most of the features of the former indicate it to be the more primi- tive. The ancestor of both is presumably to be sought in a member of the section Ovatae, and P. Carolina heterophylla, discussed in a preceding article in this series, does indeed bear some resemblance to them. The differences are too marked, however, to regard the relationship as close, and the real con- necting links between the two sections have apparently become extinct. 1 Contribution from the Botanical Laboratory of the University of Pennsylvania. Previous articles in the series have appeared in Bartonia 11: 5. 1929; 12: 24. 1931; 13: 18. 1932; and 14: 14. 1932. 2 In Engler & Prantl's Pflanzenfamilien 43a: 47. 1891. (14) i^. 9^ d Bartonia, No. 15 Plate 2 > ^ [§ PHILADELPHIA BOTANICAL CLUB 15 % ► )i '^ mi ^ Fig. 1. Phlox amplifolia. One mile northeast of Willets, Jackson County, North Carolina. ""Ti Fig. 2. Phlox amplifolia. In cultivation; originally from Indiana. 14. Phlox amplifolia Britton. Broad-leaf Phlox. Plate 2. History. — Among the names regarded as synonyms of P. paniculata Linne by Gray^ in his revision of the Polemonia- ceae was included ^^P. glandulosa, Shuttleworth, coll. Rugel, pubescent form.'' So little description was thus given that the name lacks validity, but it may have represented a mem- ber of the Paniculatae which diifers from the typical species of this section in several important respects. Mohr^ collected the same Phlox in Alabama, considering it a ** well-marked variety,'' and identifying it with P. paniculata var. acumi- nata (Pursh) Chapman. His specimens, preserved in the U. S. National Herbarium, do not, however, agree with Pursh 's original description. Britton,^ on the other hand, recognized it to be an independent species, and gave it the appropriate name P. amplifolia. Not considering the publication of a description necessary to validate a name, Brand* adopted that ascribed by Gray to Shuttleworth, and gave the date of its original use as 1842, presumably on the basis of an annotated sheet bearing a Rugel specimen. Finally, Robinson and Fernald^ returned to Gray's view that the Phlox under discussion is identical with P. paniculata. Here the two are regarded as sufficiently differ- ent to deserve independent status, and Britton 's name as the only acceptable one. Geography. — Although herbaria do not contain many speci- mens of this Phlox, enough have been seen to show its range to center about the Interior Low Plateau province in Tennes- see, and to extend from east-central Alabama to eastern Mis- souri, southern Indiana, and western North Carolina. The Fall Line has formed a barrier to its migration southward, and it has been unable, during the period since the retreat of the last ice sheet, to invade the glaciated territory. These relations are brought out in the map (fig. 1) which appears on the following page. iProc. Amer. Acad. Arts Sci. 8: 249. 1870. 2 Plant Life of Ala. 684. 1901. 3 Manual Flora N. States & Can. 757. 1901. 4 In Engler's Pflanzenreich IV. 250: 61. 1907. 5 In Gray's Manual, ed. 7. 674. 1908. Barton I A, No. 15 Plate 2 Fig. 1. Phlox ampUfoUa. One mile iioitlieast of Willets, Jackson County, North Caiolinn, PHILADELPHIA BOTANICAL CLUB 15 III >• [> Fig. 2. Phlox am pU folia. In cultivation; originally from Indiana. 14. Phlox amplifolia Britton. Broad-leaf Phlox. Plate 2. History. — Among the names regarded as synonyms of P. paniciilata Linne by Gray^ in his revision of the Polemonia- ceae was included '^P. glandulosa, Shuttleworth, coll. Kugel, pubescent form.'' So little description was thus given that the name lacks validity, but it may have represented a mem- ber of the Paniculatae which differs from the typical species of this section in several important respects. Mohr^ collected the same Phlox in Alabama, considering it a ^ Svell-marked variety,'' and identifying it with P. panicitlata var. aciimi- nata (Pursli) Chapman. His specimens, preserved in the U. S. National Herbarium, do not, however, agree with Pursh's original description. Britton,^ on the other hand, recognized it to be an independent species, and gave it the appropriate name P. amplifolia. Not considering the publication of a description necessary to validate a name. Brand* adopted that ascribed b}' Gray to Shuttleworth, and gave the date of its original use as 1842, presumably on the basis of an annotated sheet bearing a Rugel specimen. Finally, Robinson and Fernald^ returned to Gray's view that the Phlox under discussion is identical with P. paniculata. Here the two are regarded as sufficiently differ- ent to deserve independent status, and Britton 's name as the only acceptable one. Geography. — Although herbaria do not contain many speci- mens of this Phlox, enough have been seen to show its range to center about the Interior Low Plateau province in Tennes- see, and to extend from east-central Alabama to eastern Mis- souri, southern Indiana, and western North Carolina. The Fall Line has formed a barrier to its migration southward, and it has been unable, during the period since the retreat of the last ice sheet, to invade the glaciated territory. These relations are brought out in the map (fig. 1) which appears on the following page. iProc. Amer. Acad. Arts Sci. 8: 249. 1870. 2 Plant Life of Ala. 684. 1901. 3 Manual Flora N. States & Can. 757. 1901. 4 In Engler's Pflanzenreich IV. 250: 61. 1907. 5 In Gray's Manual, ed. 7. 674. 1908. INTENTIONAL SECOND EXPOSURE 16 PROCEEDINGS OF THE Alabama : The report of this Phlox by Mohr has already been referred to under History; the county list is: Clay, Coosa, Lee, Madison, and St. Clair. Georgia : Known only in Floyd and Walker counties. [Illinois : Specimens from Wabash County labelled P. am- plifolia in several herbaria represent a pubescent form of P. panicnlata instead.] Indiana: Fairly common in 4 southern counties: Craw- ford^ Harrison^ Jefferson, and Perry. Kentucky: Short collected this species on the ''Barrens," perhaps in Barren or LfOgan County, and specimens have also been seen from Franklin, Green, Jefferson, and Woodford. Missouri : Rare in the eastern part, being known only from Reynolds and St. Louis counties. Fig. 1. Distribution of Fhlox ampUfolia. North Carolina: Occasional in the mountain counties: Buncombe, Haywood, Jackson^, and Madison. Tennessee: Five counties are represented in herbaria: Cocke, Davidson, Franklin, Hamilton^ and Knox^ Virginia : Known only in Lee, the southwesternmost county. [West Virginia: In the University of Pennsylvania her- barium there is a sheet labelled Hawks Nest, (Fayette County,) bearing a specimen of P. ampUfolia along with one of P. paniculata. In view of the uncertainty as to whether both belong with the label, the record requires confirmation.] PHILADELPHIA BOTANICAL CLUB 17 Ecology. — Unlike its more wide-spread relative. Phlox am- pUfolia is not in general an alluvial soil plant, but grows chiefly on rocky hillsides clothed with climax forests, where the soils are rich in humus and circumneutral in reaction. It begins to bloom in early June and continues throughout the Summer. The sweet-scented flowers are rather pale in color, and no doubt attract moths as well as butterflies. That the species lacks aggressiveness is clearly brought out by the scat- tering of the dots on its distribution map, as well as by its failure to enter the glaciated area during the many thousands of years since the ice retreated. It is probably a relic- endemic, which is in process of dying out. Variation. — In height this Phlox ranges from about 50 to 150 cm., and the stem varies from uniformly green to strongly red-maculate. Its leaves are larger than those of any other member of the genus, their average maximum size being 115 by 45 mm., with a range of 60 to 180 by 20 to 80 mm. They often have a broad petiole expanding rather abruptly into a rhombic-ovate blade, but the latter may be elliptic-lanceolate, ovate-oblong, etc., and rarely the petiole is absent. The upper surface of the leaves is normally beset with coarse bristles aris- ing from papillae, varying considerably, however, in size and abundance; the lower surface usually bears numerous fine hairs, sometimes mingled with coarse ones. Locally, espe- cially in Indiana, there occurs a well-marked form, in which the leaves are glabrous on both surfaces. The calyx does not vary significantly, but the corolla shows considerable range in size and in color, being usually pale to deep pink, and rarely tending toward phlox purple. While the anthers are sometimes nearly as intensely yellow as in the species of other sections, they are more often pale yellow to cream-colored, as in the second member of the present section. The style is moderately long, from 12 to 22 mm. Cultivation. — Being less vigorous and paler in corolla-color than the related P. paniculata^ the present species has appar- ently never been brought into cultivation. It is worthy of trial, however, in wild-flower gardens. 18 PROCEEDINGS OF THE 15. Phlox panlculata Linn6. Veint-leaip Phlox. Plate 3. History.—This species was first recorded by Plukenet' in 1700, with the characterization "Lychnidea Virginiana Blat- tariae accedens, umbellata, maxima, Lysimachiae luteae foliis amphoribus"; the source of his material is unknown. John Bartram sent it to Peter Collinson, and in a letter dated June rhu .^^"^ published by Darlington,=> referred to it thus- The other, which I brought from Virginia, grows with me about five feet high, bearing large spikes of different coloured flowers, for three or four months in the year, exceeding beau- titul. In the catalog of Collinson 's garden edited by Dill- wyn It IS entered as "Lychnidea folio Peraica, floribus in spicam depositis. 1744, a new lychnidea, sent by J. Bartram with a large spike of pale reddish purple flowers with peach- shaped leaves, flowered in July and August." Linne^ listed It as the first species of the genus, and noted that he had ob- tained It from Collinson. A figure published by Dillenius= under the name "Lych- nidea foho salicino" is often regarded as representing Phlox pamculata, but bears a closer resemblance to P. carolim A colored plate showing clearly the characteristic features of the former species appeared, however, in the second volume of the Sfrr- ifir"^ '"" '^' Gardeners' Dictionary issued bv Miller« m 1760, and it has been repeatedly figured during subsequent years. . ^ Far more names have been applied to variants of this species than to those of any other Phlox, but only the more important of these from the technical standpoint will be discussed here, ihe earliest, P. undulata, was apparently firet published by Alton though also used by various other writers of the same period. The supposed distinctive feature of undulate leaves ^ often shown by otherwise typical P. paniculata, a species which varies markedly in foliar characters. 1 Mantissa 121. 1700. 2 Memorials of Bartram and Marshall 164. 1849 3 Hortus Collinsonianus 39. 1843 ^ Species Plantarum (1) : 151 1753 f ?.^'"*"^,-^^*^^"'^^^^» 1- 205, pi. 166,* fig 203 17^2 7 l'^': ^^^^' ^^"^^ '« Gard. Diet. 2 / pf '205 fi/ 2 1 TfiO 7 Hortus Kewensis 1: 205 1789 ' ^* ^^• Bartonia, No. 15 Plate 3 Fig. 1. Phlox paniculata. One mile soutliwcst of Marticville, Lancaster County, Pennsylvania. Fig. 2. Phlox panicnlata. In cultivation; originally from Virginia. 18 PROCEEDINGS OP THE 15. Phlox paniculata Linne. Veiny-leap Phlox. Plate 3. iviS*'^"-'^'""^^"^ «P«"e« ^™s first recorded by Plukenet^ in I/UO, with the characterization "Lychnidea Virginiana Blat- tariae accedens, umbellata, maxima, Lysimachiae liiteae foliis amplioribus"; the source of his material is unknown. John Bartram sent it to Peter Collinson, and in a letter dated June 11 1743, later published by Darlington,^ referred to it thus • The other, which I brought from Virginia, grows with me about five feet high, bearing large spikes of different coloured flowers, for three or four month., in the year, exceeding beau- titul. In the catalog of Collinson 's garden edited by Dill- wyn^ It IS entered as "Lyehnidea folio Persica, floribus in spicam depositis. 1744, a new lyehnidea, sent by J. Bartram with a large spike of pale reddish purple flowers with peach' shaped leaves, flowered in July and August." Linne^ listed It as the first species of the genus, and noted that he had ob- tained It from Collinson. A figure published by Dillenius' under the name "Lyeh- nidea folio salieino" is often regarded as representing Phlox pamcnata, but bears a closer resemblance to P. Carolina A colored plate showing clearly the characteristic features of the former species appeared, however, in the second volume of the Miller m 1760, and it has been repeatedly figured durin<^ subsequent years. ° Far more names have been applied to variants of this species than to those of any other Phlox, but only the more important of these from the technical standpoint will be discussed here ine earliest, P. undidata, was apparently fii^t published bv Alton though also used by various other writers of the sam; period. The supposed distinctive feature of undulate leaves IS of en shown by otherwise typical P. panindata, a species which varies markedly in foliar characters. 1 Mantissa 121. 1700 2 Memorials of Bartram and Marshall 164. 1849 3 Hortus Collinsonianus 39. 1843 4 Species Plantarum (1) : 151. 1753 5 Hortus Elthamensis 1 : 205, T3l 166*fify 9(\'\ 17qo I Figs. Plants Miller 's Gard.^Di^t 2 • pf '205 fi^ 2 ' i -c^ 7 Hortus Kewensis 1: 205. 1789. ' ^* '^^• Barton I A, No. 15 Plate 3 j Fig. 1. Phlox paniculata. One mile southwest of Marticville, Lancaster County, Pennsylvania. ^f^ f Fig. 2. Phlox panicvlafa. In cultivation; originally from Virginia. INTENTIONAL SECOND EXPOSURE PHILADELPHIA BOTANICAL CLUB 19 ''i^ « T V » « I*- -« In a catalog of horticultural plants issued about 1812, Lyon included, according to Pursh,^ a ''Phlox decussata,'' but in describing this in his Flora the latter author renamed it P. acuminata. A close relationship to P. paniculata was shown in many of the features enumerated, especially in the pubes- cent corolla-tube, but it was considered to differ in having the leaves acuminate at both ends and pubescent beneath. The colored plate of it published by Sims^ in 1817 left no doubt as to the identity of the two. When he came to list the plants of South Carolina and Georgia, Elliott^ not only maintained Lamarck's and Pursh's species as distinct, but also (on page 244) proposed another name, P. cordata, for material with the leaves ' ' uniformly cor- date.'' No type specimen of this is preserved in his her- barium at the Charleston Museum, but there can be little question of its being a mere form of the species here under discussion. A list of seeds distributed by the Hamburg Botanical Gar- den in 1826 included a Phlox sickmanni, and two years later this was described and figured by Lehmann.* The name was used for a time by horticulturalists, but has long since become obsolete, as the plant was in every respect identical with P. paniculata. Between 1828 and 1831 Sweet^ published colored plates of three supposedly distinct species, P. scahra, P. cordata, and P. corymhosa. None of these, however, differs from P. paniculata in any respect which would now be considered of specific significance. Bentham,^ the first monographer of the Polemoniaceae, classed P. undulata, P. cordata, and P. sickmanni (misspelled *'siebmanni") as synonymous with P. paniculata, but main- tained P. acuminata, inclusive of P. corymhosa, as a distinct species. 1 Flora Americae Septentrionalis 2: 730. 1814. 2 Curtis 's Botanical Magazine 44: pi. 1880. 1817. 3 Sketch Botany S. C. & Ga. (1) : 242. [1817] * Act. Acad. nat. cur. 14: 814, pi. 46. 1828. 5 British Flower Garden 3: pi. 248. 1828; [2] 1: pi. 13. 1829; [2] 2: pi. 114. 1831. 6 In De Candolle's Prodromus 9: 303. 1845. 20 PROCEEDINGS OF THE PHILADELPHIA BOTANICAL CLUB 21 Chapman^ reduced P. acuminata to a variety of P. panicu- lata, adding to Pursh 's distinguishing characters ' ' calyx-lobes shorter. ' ' Gray,^ on the other hand, was unable to recognize any distinction, and made acuminata merely another syno- nym. Mohr^ followed Chapman's nomenclature, but misap- plied the name, as already mentioned under P. amplifolia. In his southeastern flora. Small* considered P. acuminata as worthy of being restored to the list of species, but his key- characters were quite different from those given by previous authors: *' Leaves conspicuously decurrent on the internodes: calyx-lobes about as long as the tube : corolla-tube pubescent. ' ' Actually, there is nothing in Pursh 's description to suggest decurrence of the leaves, and moreover the leaves of otherwise tjHpical P. paniculata are often more or less decurrent. The extent of union of sepals varies so greatly within single colo- nies as to be of no diagnostic value, even were the data of Chapman and Small in agreement. Finally, a pubescent corolla-tube is one of the most typical features of normal P. paniculata, so there seems no basis for specific separation be- tween the two. Brand^ followed Gray in classing most of the names enu- merated, including acuminata, as mere synonyms of panicu- lata; at the same time, he considered P. decussata to represent a hybrid of P. paniculata with P. m^aculata, but failed to fur- nish any supporting evidence. He also ventured to propose a variety laxifiora for a specimen from Missouri with narrow leaves and lax inflorescence, although this appears to repre- sent merely a shade-form such as can be found in many occur- rences of the species. White or near-white forms appear occasionally in colonies of P. paniculata, and the term variety alha was proposed for these by Don^ in 1838; numerous horticultural names have also been applied to them subsequently. 1 Flora Southern U. S. 338. 1860. 2Proc. Amer. Acad. Arts Sci. 8: 249. 1870. 3 Plant Life of Ala. 684. 1901. 4 Flora SE. U. S. 977. 1903. 5 In Engler's Pflanzenreich IV. 250: 59. 1907. 6 Gen. Hist. Dichl. Plants 4: 240. 1838. « I I *4 F-^v. > Geography. — Owing to the fact that Phlox paniculata es- capes from cultivation more frequently than any other species, the determination of its natural range is especially difficult, and some of the occurrences indicated by circles on the accom- panying map should probably have been marked by an x (for escape) instead. It seems undoubtedly native, however, from northern Georgia to Arkansas, northeastern Kansas, northern Indiana, and central New York. The map of its distribution shows it not only to be more wide-spread than its relative, pre- viously discussed, but also to have locally crossed both the Fall Line and the Wisconsin terminal moraine. Fig. 2. Distribution of Phlox paniculata. [Alabama: Kecorded from Montgomery County, but scarcely native.] Arkansas : Known in Clark, Madison, Newton, Sharp, and Washington counties.^ [Connecticut: A common escape throughout, the county list being: Fairfield^ Litchfield% Middlesex^ New Haven', New London^, and Windham*.] [Delaware: Escaped in New Castle County*.] 1 Prof. D. M. Moore kindly supplied this county list. f 'k ll ; 22 PROCEEDINGS OF THE District of Columbia^ : Native alon^ the Potomac and Rock Creek valleys. [Florida: Britton and Brown^ stated the range of this species to extend to Florida, but no evidence is at hand.] Georgia : Occasional in the mountains and inner Piedmont, having been seen from De Kalb and Eabun^ counties. Illinois: Common, especially southward, the county list being: Champaign**, Greene, Hancock, Jackson, Macoupin, Marion, Peoria, Pope, Pulaski, Richland, St. Clair, Union, Vermilion, and Wabash. Indiana: Present practically throughout, having been re- corded from 50 counties : Bartholomew, Brown, Carroll, Cass, Clark, Clay, Crawford, Daviess, Dearborn, Decatur, Fayette, Franklin, Gibson, Grant, Hamilton, Harrison^ Hendricks, Jay, Jefferson, Jennings, Knox, Kosciusko, Lake, Laporte, Marion, Marshall, Martin, Miami, Monroe, Montgomery, Owen, Parke, Perry, Porter, Posey, Putnam, Rush, St. Joseph, Shelby, Spencer, Sullivan, Tippecanoe, Union, Vermilion, Vigo, Warren, Warrick, Washington, Wayne, and Wells. [lowA: Escaped in Johnson'' County.] Kansas : The westernmost known native occurrence of the species is in Doniphan County. Kentucky : Collected in but 8 counties : Caldwell, Daviess, Edmondson, Fayette, Hardin^, Lee, Mercer, and Warren. [Louisiana: The statement under Florida applies here.] Maryland: Common in the upland counties: Allegany', Baltimore, Cecil, Frederick, Garrett, Montgomery^ Prince Georges, and Washington. [Massachusetts: Escaped in BristoP, Essex^ Middlesex^, Norfolk^ and Suffolk^ counties.] [Michigan: Reported to escape locally.] Mississippi: Present in Panola and Tippah counties; further south P. glaberrima has been mistaken for it. Missouri: Wide-spread, material being preserved from 19 counties: Barry, Carter, Christian, Cole, Greene, Holt, Jack- son, Jefferson, Laclede, McDonald, Madison, Osage, Pulaski, St. Francois, St. Louis, Shannon, Stoddard, Stone, and Taney. 1 Illustr. Flora N. U. S. 3 : 32. 1898. PHILADELPHIA BOTANICAL CLUB 23 [New Hampshire: Recorded as escaped in Belknap^ Cheshire'', and Sullivan'' counties.] [New Jersey : Specimens have been preserved from 6 coun- ties, but none are regarded as native : Atlantic'', Monmouth", Morris", Ocean", Somerset", and Warren".] New York: Occurs, in part as a garden escape, in Bronx Boro", Cattaraugus, Dutchess", Erie, Herkimer", Livingston", Onondaga, Otsego, Suffolk", Tompkins", Washington", and Yates counties. North Carolina: Frequent along stream-valleys in the mountains and higher Piedmont, the county list being : Bun- combe, Burke, CaldwelF, Forsyth, Haywood, Macon, Orange, Polk, Rockingham^ Rutherford^ Swain, Transylvania', and Watauga. Ohio: Wide-spread, having been collected in 22 counties: Adams, Athens, Auglaize, Cuyahoga, Defiance, Erie, Fairfield, Franklin^ Hamilton, Hancock, Hocking, Jackson, Lorain, Marion, Miami, Monroe, Ottawa, Perry, Richland, Ross, Sci- oto, and Wayne. Pennsylvania: Native except in the easternmost counties, where it often escapes : Allegheny, Armstrong, Beaver, Berl^", Butler, Columbia, Crawford, Dauphin, Delaware", Fayette', Franklin, Greene, Huntingdon, Lancaster', Lawrence, Le- high", Luzerne, Montgomery", Monroe", Northampton", Phila- delphia", Schuylkill, Somerset', Washington, Westmoreland and York. [South Carolina: Listed by Elliott;^ no specimens seen.] Tennessee : Rare, there being but 6 county records : Cheat- ham, Davidson, Franklin, Marion, Montgomery, and Sevier. [Vermont: Reported to escape occasionally.] Virginia: Common northward, in the counties: Arlington, Bedford, Culpeper', Fairfax', Frederick, Greene', Madison', Prince William', Rappahannock', Shenandoah', and Stafford'. West Virginia : Occasional throughout, records having been seen from : Fayette, Grant', Greenbrier, Harrison', Jefferson', Monongalia, Monroe, Pendleton, Pochahontas, Preston, Ran- dolph', and Tucker counties. 1 Sketch Botany S. C. & Ga. 1: 242. (1821) 24 PROCEEDINGS OF THE il [Wisconsin: Escaped in Kewaunee* County.] [Canada: Ontario: Has been seen from Kent* Co.] Ecology.— The typical native habitat of Phlox paniculata is a thinly wooded alluvial flat where the vegetation has reached an intermediate successional stage. The soil is corre- spondingly circumneutral in reaction and rich in plant-foods. When it escapes from cultivation, however, it may occupy all sorts of locations, including weedy roadsides and rubbish heaps. Its blooming period extends from about the first of July to the last of September, with occasional stragglers into early winter. The bright purple flowers attract numerous butterflies, which cross-pollinate it and lead to the production of abundant viable seed. Variution.— Under cultivation this species has given rise to an extraordinarily large series of forms, which are only par- tially foreshadowed by variations observed in nature. Its stem ranges in height from 75 to 200 cm., with 15 to 40 nodes, at a few of which the leaves are opposite, but at most dis- tinctly subopposite. The leaf -outline varies from narrowly elliptic-lanceolate to oblong-ovate, the blades being usually borne on a short margined petiole but sometimes sessile and more or less cordate. On their surfaces the leaves are often wholly glabrous, but on occasional individuals in many colo- nies they become downy-pubescent beneath, and less frequently bristly above. Considerable variation is shown in the size and compactness of the inflorescence, but this seems to be of ecological origin, and connected with the extent of crowding or shading. The inflorescence-herbage is usually somewhat pubescent, at least up to the ba^e of the calyx, the hairs being in most cases sharp- pointed, though exceptionally gland-tipped. The sepals range from 6 to 10 mm. long, and are united from ^ to f their length. Phlox-purple is the predominant corolla-color, with occa- sional trends toward amparo-purple (Kidgway 63 b) or mallow-purple (67 b), and toward white, but there seems no indication in the native colonies of the varied and brilliant hues which have been brought out by horticulturalists. Eye- PHILADELPHIA BOTANICAL CLUB 25 striae are also less conspicuous in nature than they have become in cultivated forms. The corolla-tube varies in length from 16 to 26 mm., and, while usually strongly pubescent, is glabrous in exceptional individuals. The broadly to narrowly obovate lobes have been observed in nature to range from 7 to 12 mm. in length and 5 to 11 in width, although in some of the garden forms they are over twice the maximum sizes stated. Stamens and style vary from wholly included to slightly exserted. No geographic segregation in any of the variable features above enumerated having been recognized, the species is not here separated into varieties; the more noteworthy of its forms may, however, be listed : Forms of Phlox paniculata Cordate-leaved. P. cordata Elliott. Occasional, grading into truncate- leaved. Rough-pubescent on upper leaf -surf aces. Phlox scdbra Sweet. Eare. Soft-pubescent on lower leaf -surf aces. Fhlox acuminata Pursh. P. de- cussata Hort. ex Pursh. P. paniculata acunninata Chapman. Frequent. Lax-flowered. P. paniculata laxiflora Brand; the plant to which this name was applied also had unusually narrow leaves, and apparently represented an ecological form. Glandular as to inflorescence-foliage. Typified by specimen from road- side thicket 3 miles west of Amissville, Rappahannock Co., Va., col- lected by E. T. W. August 23, 1927. Infrequent. Glabrous-tubed. Typified by specimen from woods along Rock Creek one mile south of Kensington, Montgomery Co., Md., collected by E. T. W. August 13, 1927. Very rare. Light-colored (pink, white, etc.). P. paniculata alha Don; also bearing numerous horticultural names. Frequent. Cultivation.^ — First introduced into horticulture, according to Aiton,2 by James Sherard in 1732, this Phlox has received far greater development than any other species. In Stand- ardized Plant Names Olmsted, Coville, and Kelsey^ listed 270 named ** horticultural varieties of Phlox paniculata or hybrids between that species and P. maculata'' known to be in the trade in 1923, and others have been introduced subsequently. These have apparently largely been produced by cross-breed- ing of mutants which have chanced to appear in cultivation ; 1 Cf. Pridham, Proc. Amer. Soc. Hort. Sci. 1931 ; 419-421. 2Hortus Kewensis 1: 205. 1789. 3 Standardized Plant Names 365. 1923. 26 PROCEEDINGS OF THE whether hybridization with other species has entered in de- serves further inquiry. The view that at least part of the horticultural forms of P. paniculata or ''P. decussata" represent hybrids of this species with P. maculata has frequently been expressed. If that cross had actually occurred, however, some traces of the features characterizing P. maculata, such as obscure veining in the leaves, short calyx-lobes, glabrous corolla-tube, and deep yellow anthers, would be expected to appear in the hybrid plants. All the horticultural material which has been avail- able for examination, however, has proved to exhibit the prominent leaf -veins, long calyx-lobes, pubescent corolla-tube, and cream-colored anthers typical of P. paniculata itself. Moreover, most of the cultivated plants produce abundant viable seed, and when this germinates the seedlings all show the features of the latter species (including a dingy corolla- color, as gardeners know to their sorrow,) whereas were a bispecific hybrid represented, a certain proportion of any seedlings would correspond to the other parent. All the evi- dence thus indicates that the presumed cross has not entered into the development of the horticultural material in question. By way of contrast, the so-called *' Phlox arendsii'' com- bines the features of P. paniculata with those of P. divaricata in such a striking way as to leave no doubt that it represents the result of crossing these two species. So far as known, no viable seeds are produced, which also favors the hybrid inter- pretation. The Phloxes discussed in this paper complete the list of those known to occur native in the United States east of the Mississippi River. It is proposed to publish in the next num- ber of *'Bartonia,'' however, a supplementary article in which data which have come to hand since the series was started will be placed on record. The writer will greatly appreciate the receipt of information as to errors or omissions which anyone may have noted, and in particular names of additional coun- ties in which individual species have been found, for inclusion in this supplement. Reprinted from The Scientific Monthly, January, 1934, Vol. XXXVI 11, pages 80-85 ill EXPLORING FOR PLANTS IN THE SOUTHEASTERN STATES By Dr. EDGAR T. WHERRY ASSOCIATE PROFESSOR OF BOTANY, UNIVERSITY OF PENNSYLVANIA One of the most pressing tasks of the plant geographer is to ascertain as fully as practicable the present distribution of the various kinds of plants over the surface of the earth, before civilized man succeeds in destroying all natural habi- tats and exterminating their occupants. While many manuals and floras give in a general way the ranges of such species as occur within the areas covered, accu- rate distributional data are at hand for very few. The lack of such informa- tion is especially serious in the case of plants endemic to the southeastern United States, where there has been so little collecting that the range of even conspicuous objects like the pitcher- plants is but imperfectly reflected by specimens in herbaria. I was accord- ingly especially glad to be invited by Mr. Louis Burk, the well-known Phila- delphia horticulturist, to obtain for him a complete collection of the species and varieties of Sarracenia, in the summer of 1932. Not only would such a trip make it possible to fill in many gaps in the recorded ranges of these plants, but also there would be a chance to study in the field undescribed ones as to which more or less unsatisfactory data were at hand. Late in June I drove to Washington, D. C, and was fortunate in having Mr. James E. Benedict, Jr., join me for the trip. Continuing southward on U. S. Route 1, our first stopping point was Raleigh, North Carolina, where we called on Professor Bertram W. Wells. He not only furnished us information as to pitcher-plant localities in the southern part of that state, but also showed us a tiny meadow not far from the city where by good fortune a few pitcher-plants still survived the encroachments of agri- culture. Two species were represented, the wide-spread yellow pitcher-plant (S. flava) and a relative of the side-saddle pitcher-plant {S. purpurea), which we especially wished to study. In his *'Au- tikon Botanikon, ' ' published in 1840, Rafinesque had pointed out what he con- sidered specific differences between the northern and southern representatives of this species, and had named the south- ern one S. venosa; but his work has been ignored by all subsequent students of these plants. At this locality its aspect was certainly quite unlike that of the familiar pitcher-plant of New England and the Great Lakes region, and we felt disposed to accept Rafinesque 's interpre- tation of it; but other occurrences seen in the course of the trip indicated the two to intergrade too much to be main- tained as independent species. Addi- tional data as to pitcher-plant localities were obtained from Professor William C. Coker at Chapel Hill, and we set out for central Georgia. At Macon we were joined by Dr. Charles C. Harrold for a two-day trip on the coastal plain of the state. As we traveled southeast, pitcher- plants began to appear in the swamps in the vicinity of Swainsboro ; these com- prised not only the tall and conspicuous Sarracenia flava, but also the diminutive hooded (S. minor) and parrot pitcher- plant {S. psittacina). Michaux had re- ported the latter from ''Augusta, Geor- gia, to Florida,'' and, desiring to obtain roots from as far north as possible, we kept searching for it in one county after another, but the most northern colony to be found lay 10 miles south of Millen and thus fully 50 miles south of Au- gusta. Several rooted clumps were col- lected, packed in wet moss, carried with us until we could find a state inspector and get them certified as pest-free, and then shipped home. Some of these were planted outdoors in a wild-life preserve controlled by Mr. Burk in southern New Jersey, where they have survived the first winter, at least. The remainder w^ere held in a cool greenhouse, and bloomed freely during early spring. Pitcher-plants were, however, not the only thing to claim our attention in this part of the country. AVe planned to make an effort to rescue a native tree which is on the verge of extinction. This plant, discovered by Stephen El- liott in the early 1800 's and named in his honor Elliottia by Muhlenberg, is a primitive member of the heath family. The genus is monotypic, being repre- sented by the single species E. racemosa, and its nearest relative is the genus Tri- petalem of Japan. These are evidently relics of the late Cretaceous and early Tertiary floras which spread widely over northern lands, but have been restricted by subsequent geological events, espe- cially the Pleistocene glaciation, to re- mote isolated areas. Elliottia is a small tree, attaining a height of about 15 feet and a trunk di- ameter of 2 or 3 inches. It spreads by rootstocks into colonies of a score or two of individuals, and about the end of June produces attractive large panicles of small white delicately scented flowers. These attract various sorts of bees, which carry pollen from flower tx) flower ; as a rule, however, no fertilization occurs, and the ovaries soon drop from the pedi- cels. Evidently individual plants are sterile to their own pollen, and as each of the 5 or 6 known colonies is appar- ently the result of vegetative propaga- tion from a single seedling, this sterility extends throughout. Before the coming of the white man colonies must have grown close enough together for pollen to be borne by insects from one to an- other, and seed was sometimes produced. Clearing the land for agriculture and burning over the woods destroyed so many, however, that this no longer oc- curs, and the seed of the species is actu- ally unknown to science. >■-* ^ FIG 1 OUR SHOWIEST SPECIES OF PITCHEK^PLANT IS S. DEUMMONDII WITH THE UPPER PART OF THE LEAVES WHITE, VEINED WITH GREEN AND RED. THIS VIEW WAS TAKEN AT ITS NORTHERNMOST KNOWN STATION, NEAR AmERICUS, GEORGIA. The current practise in that region of burning the low-growing vegetation every year or two causes the Elliottia^ like many other plants which, when un- disturbed, have an arborescent habit, to send up numerous small shoots from their woody underground parts, and thus produce shrubby thickets. These often become so dense that reproduction of the longleaf pine and other valuable trees is prevented, until another fire de- stroys the brush to which the preceding one gave rise, which has led to the curi- ous idea held in many circles that fre- quent burning is natural and desirable. No doubt the great pine forests of the coastal plain got started in the first place when particularly severe fires de- stroyed whatever deciduous climax for- est formerly occupied the areas ; but the infrequency of charred rings in stumps and of charcoal layers in peat deposits shows that before the white man came fires occurred only at intervals of many years. Unless and until the present fre- quency of fires shall be reduced to that of primeval times by protective measures and by education, all but the most vigor- ous and aggressive of the native plants of that region are doomed to extinction in the near future, and it seems idle to talk about *' reforesting the south.'' Because, then, of the impending disap- pearance from native habitats of this relic of past geologic times, Elliottia, all possible efforts to get it into cultivation are worth while. With this in view, Mr. Harry W. Trudell and I had twice be- fore visited this region and had located certain of the remaining colonies of the tree, in part through directions kindly furnished by Dr. Roland M. Harper (who, I should state at this point, dis- agrees with me completely as to the fire situation ) . Both times w^e had found but a single colony in bloom, the others hav- ing been seriously damaged by the fires of those years. On the present trip, how- ever, conditions were more favorable; not only were two previously known colo- nies blooming, but Dr. Wallace Ken- nedy, of Metter, had discovered near there a new one, which had escaped burning for a number of years and ap- proached the normal arborescent habit of the species. Pollen was accordingly carried, by w^hat I can not refrain from terming an *'automobee,'' from one locality, which may be designated A, some 5 miles tc locality B, and from the latter 75 miles to locality C. The pollinated plants were carefully located by landmarks, and Dr. Harrold planned to return in the fall to see if any seed had matured. He was unfortunately prevented from doing so by serious illness, so what oc- curred at locality C is indeterminate. During the winter Dr. Kennedy w^ent out to locality B, and found that cap- sules had actually formed on the pol- linated plants, but by that time de- hiscence had occurred, and the contents had fallen out, so the seed of the species is still unknown to science. Horticul- turists have now become interested, how- ever, and clumps from different colonies have been planted side by side on the grounds of Dr. Lee, in Macon, and of Professor De Loach, in Statesboro, where fire can be kept out and the plants watched closely, so by another year we should know what the seed is like, and have some from which seedlings can be grown for cultivation elsewhere. Another group of plants on which data as to geographic range were being sought on this trip was the phloxes. Many of the counties of Georgia tra- versed yielded one which, though ex- ceedingly variable in habit and leaf- shape, could only be classed as P. glahernma L. In an alder thicket near La Grange, Troup County, we found a colony of the tallest plants of this species on record, attaining a height of 175 cm. Driving south through Webster County and watching the roadsides for plants of interest, we suddenly caught a flash of purple on a plant which looked different be found lay 10 miles south of Milieu and thus fully 50 miles south of Au- gusta. Several rooted clumjjs were col- lected, packed in wet moss, carried with us until we could find a state inspector and get them certified as pest-free, and then shipped home. Some of these were planted outdoors in a wild-life preserve controlled by Mr. Burk in southern New Jersey, where they have survived the first winter, at least. The remainder were held in a cool greenhouse, and bloomed freely during early spring. Pitcher-plants were, however, not the only thing to claim our attention in this part of the country. We planned to make an effort to rescue a native tree which is on the verge of extinction. This plant, discovered by Stephen El- liott in the early 1800 's and named in liis honor ElUottm by Muhlenberg, is a primitive member of the heath family. The genus is monotypic, being repre- sented by the single species E. racemosa, and its nearest relative is the genus Tri- petaleia of Japan. These are evidently relics of the late Cretaceous and early Tertiary floras which spread widely over northern lands, but have been restricted by subsequent geological events, espe- cially the Pleistocene glaciation, to re- mote isolated areas. ElUottm is a small tree, attaining a height of about 15 feet and a trunk di- ameter of 2 or 3 inches. It spreads by rootstocks into colonies of a score or two of individuals, and about the end of June produces attractive large panicles of small white delicately scented flowers. These attract various sorts of bees, which carry pollen from flower to flower ; as a rule, however, no fertilization occurs, and the ovaries soon drop from the pedi- cels. Evidently individual plants are sterile to their own pollen, and as each of the 5 or 6 known colonies is appar- ently the result of vegetative propaga- tion from a single seedling, this sterility extends throughout. Before the coming of the white man colonies must have growai close enough together for pollen to be borne by insects from one to an- other, and seed was sometimes produced. Clearing the land for agriculture and burning over the woods destroyed so many, however, that this no longer oc- curs, and the seed of the species is actu- ally unknown to science. i 4 i A^ «> ^1 V I -^ ■- /■ — » r1> t ►-* /-■^ FIG 1 OUR SHOWIEST SPECIES OF PITCHER-PLANT IS S. VRUMMONDII WITH THE I PPER PART OF THE LEAVES WHITE, VEINED WITH GREEN AND RED. THIS VIEW WAS TAKEN AT ITS NORTHERNMOST KNOWN STATION, NEAR AmERICUS, GEORGIA. The current practise in that region of burning the low-growdng vegetation every year or two causes the Elliot tia, like many other plants which, when un- disturbed, have an arborescent habit, to send up numerous small shoots from their w^oody underground parts, and thus produce shrubby thickets. These often become so dense that reproduction of the longleaf pine and other valuable trees is prevented, until another fire de- stroys the brush to which the preceding one gave rise, which has led to the curi- ous idea held in many circles that fre- quent burning is natural and desirable. No doubt the great pine forests of the coastal plain got started in the first place when particularly severe fires de- stroyed whatever deciduous climax for- est formerly occupied the areas ; but the infrequency of charred rings in stumps and of charcoal layers in peat deposits shows that before the white man came fires occurred only at intervals of many years. Unless and until the present fre- quency of fires shall be reduced to that of primeval times by protective measures and by education, all but the most vigor- ous and aggressive of the native plants of that region are doomed to extinction in the near future, and it seems idle to talk about *' reforesting the south.'' Because, then, of the impending disap- pearance from native habitats of this relic of past geologic times, Elliottia, all possible efforts to get it into cultivation are worth while. With this in view, Mr. Harrv W. Trudell and I had twice be- t. fore visited this region and had located certain of the remaining colonies of the tree, in part through directions kindly furnished by Dr. Roland M. Harper (who, I should state at this point, dis- agrees with me completely as to the fire situation ) . Both times we had found but a single colony in bloom, the others hav- ing been seriously damaged by the fires of those years. On the present trip, how- ever, conditions were more favorable; not only were two previously known colo- nies blooming, but Dr. Wallace Ken- nedv, of Metter, had discovered near there a new one, which had escaped burning for a number of years and ap- proached the normal arborescent habit of the species. Pollen was accordingly carried, by what I can not refrain from terming an ''automobee,'' from one locality, which may be designated A, some 5 miles tc locality B, and from the latter 75 miles to locality C. The pollinated plants were carefully located by landmarks, and Dr. Harrold planned to return in the fall to see if any seed had matured. He was unfortunately prevented from doing so by serious illness, so what oc- curred at locality C is indeterminate. During the winter Dr. Kennedy went out to locality B, and found that cap- sules had actually formed on the pol- linated plants, but by that time de- hiscence had occurred, and the contents had fallen out, so the seed of the species is still unknown to science. Horticul- turists have now become interested, how- ever, and clumps from different colonies have been planted side by side on the o-rounds of Dr. Lee, in Macon, and of Professor De Loach, in Statesboro, where fire can be kept out and the plants watched closely, so by another year we should know what the seed Ls like, and have some from which seedlings can be grown for cultivation elsewhere. Another group of plants on which data as to geographic range were being sought on this trip was the phloxes. Many of the counties of Georgia tra- versed yielded one which, though ex- ceedingly variable in habit and leaf- shape, could only be classed as P. glaherrima L. In an alder thicket near La Grange, Troup County, we found a colony of the tallest plants of this species on record, attaining a height of 175 cm. Driving south through Webster County and watching the roadsides for plants of interest, we suddenly caught a flash of purple on a plant which looked different INTENTIONAL SECOND EXPOSURE FIG 2 ONE OF THE FEW SARRACENIA MEADOWS STILL PRESERVED NEAR THEODORE, Al^BAMA. THE SPECIES ARE S. ^'"^^Z^'oOuZL^sZ WH INNUMERABLE HYBRIDS BETWEEN THEM, SHOWING ALL SORTS OP COMBINATIONS OF CHARACTERS. from any previously seen. On investi- gation this proved to be Phlox floridana, which had not been previously reported north of Thomas County, so that our find extended its known range by more than 100 miles. Other new stations for this species in Georgia and Alabama were also found later. Showiest of all the species of Sarra- cenia is the white-top pitcher-plant, usu- ally known technically as S. drummondn, although Rafinesque's name S. leitco- pJiylla has many years' priority. Ama- teur botanists have reported it to grow as far up as North Carolina, but the northernmost locality represented by specimens in herbaria is Americus, Geor- gia. After an hour's search in that vicinity we found in a swamp a small colony which, by a fortunate chance, had not been destroyed by cultivation. Here the stock of the species for Mr. Burkes collection was obtained, and although it is still too early to tell whether the clump planted outdoors in southern New Jersey will survive, those wintered over in the cool greenhouse have grown and bloomed well. (See Fig. 1.) Another member of the genus does not grow east of Mobile, Alabama, so we traveled slowly toward that place, col- lecting various plants of interest along the way. The technical name of the species in question is S. sledgei Macfar- lane, and, as its flowers are lighter in color than those of any other species, it seems most aptly termed the pale pitcher-plant. This proved to occur in a number of swamps, and we soon had some plants ready to send off. Here we were so fortunate as to meet Mr. T. S. Van Aller, who not only inspected our plants and certified them as safe for shipment, but also guided us to several pitcher-plant meadows which we would not have found otherwise. In most of > » •t- the localities draining, burning and other destructive activities of civilized man have greatly reduced the numbers of these plants, but one locality near Theodore proved to be still undisturbed. (See Fig. 2.) Here countless thousands of S. sledgei and S. drummondii grew together, along with a host of hybrids showing every conceivable gradation be- tween and combination of the characters of the two parents. It seems a pity that there is no one in the region sufficiently interested in conservation to buy up this bit of meadow and save it for investiga- tion by geneticists and enjoyment by nature lovers of the future. In April, 1910, while carrying on his fascinating studies of the relations be- tween insects and pitcher-plants, Dr. Frank Morton Jones had spent some time at Theodore, and had observed in one near-by meadow a pink-flowered form of S. venosa. He had furnished us approximate directions as to its location, and we soon found what appeared to be the right spot. In July, of course, pitcher-plant petals are withered, but we dug a few plants and shipped them to Philadelphia in the hope that they mio^ht bloom in the greenhouse the fol- lowing spring. This hope has now been realized ; and it turned out that we had struck the right spot. The parts of the flower which in most pitcher-plants are green or bronzy— the bracts, sepals and style-umbrella— are in this one nearly white, while the petals have a lovely rose color, unlike that of any other Sarra- cefi^d. Few herbarium records existing for Mississippi plants, we visited bogs in two of the eastern counties of that state, ob- taining specimens of several Sarracenias and phloxes. Next we went to Havana, Alabama, and found the famous colony of the hybrid spleenwort {Asplemum ehenoides)—the only one in which fer- tility has been attained— to be m good condition. We then called on Dr. Ro- land M. Harper at Tuscaloosa, and ob- tained from him directions as to certain Sarracenia localities in the northern part of this state. Along the Sipsey River we found a colony of the Allegheny filmy- fern {Trichomanes hoscianum), but search for its diminutive relative, T. peiersii, was unsuccessful. In Chilton County we located a colony of a red-flowered pitcher-plant, but it was not in good enough condition to es- tablish its identity. We then set out for the valley of the Little River east of Fort Payne, where a yellow-flowered one was reported. In spite of many hours' search in every conceivable type of habi- tat, we were unsuccessful in finding it there, but Dr. Harper had fortunately observed it, also, near Center. On reach- in"- that place we found that, although recent clearing of the land for agricul- ture and burning over of the swamps, even where no such use was practicable, had nearly exterminated it, a few small clumps had somehow managed to escape destruction. Both in the field and m the o-reenhouse, where it bloomed the f ollow- hig spring, this plant showed a number of differences from its nearest relative, S. ilava, and is to be classed as an inde- pendent species. The mountains of North Carolma were our next objective, for there grows the red-flowered pitcher-plant known as Sar- racenia jonesii, the distinctness of which had only been recognized in 1929. its colonies proved to have been nearly de- stroyed by drainage of the swamps and by the raids of vandals from the towns, but enough remained to enable this spe- cies to bemadded to the collection. With it n-rew some beautifully veined Sarra- cenia venosa. Ordinarily, when two closely related species or varieties exist, the more southern one tends to grow m the coastal plain, the more northern m the mountains ; in this case, however, the southern representative grows both at low and high elevations. We also found hybrids between S. venosa and S. jonesn as yet undescribed. K 4 i FIG '' ONE OF THE FEW SAEEACENIA MEADOWS «^„ , PRFSERVED neab'theodore, ALABAMA. TiiE SPECIES ABE S. drummondii AND S. sUdgci, :;::™ ZTZLZL.... bUeek them, ..o..... ... sorts o. combxnatxoks or CHARACTERS. from any previousl}^ seen. On investi- gation this proved to be Fhlox jloridana, which had not been previously reported north of Thomas County, so that our find extended its known range by more than 100 miles. Other new stations for this species in Georgia and Alabama were al^o found later. Showiest of all the species of Sarra- cenia is the white-top pitcher-plant, usu- ally known technically as S. drummondii, although Rafinesque's name S. leuco- phiflla has many years' priority. Ama- teur botanists have reported it to grow as far up as North Carolina, but the northernmost locality represented by specimens in herbaria is Americus, Geor- gia. After an hour's search in that vicinity we found in a SAvamp a small colony which, by a fortunate chance, had not been destroyed by cultivation. Here the stock of the species for Mr. Burk\s collection was obtained, and although it Is still too early to tell whether the clump planted outdoors in southern New Jei-sey will survive, those wintered over in the cool greenhouse have grown and bloomed well. (See Fig. 1.) Another member of the genus does not grow east of Mobile, Alabama, so we traveled slowly toward that place, col- lecting various plants of interest along the way. The technical name of the species in question is 8. sledgei Macfar- lane, and, as its flowers are lighter in color than those of any other species, it seems most aptly termed the pale pitcher-plant. This proved to occur in a number of swamps, and we soon had some plants ready to send off. Here we were so fortunate as to meet Mr. T. S. Van Aller, who not only inspected our plants and certified them as safe for shipment, but also guided us to several pitcher-plant meadows which we would not have found otherwise. In most of '*- V> i ' > Vv I 1^1 - / % *• 1 -* v, -4- >•♦ 4 \ • the localities draining, burning and other destructive activities of civilized man have greatly reduced the numbers of these plants, but one locality near Theodore proved to be still undisturbed. (See Fig. 2.) Here countless thousands of S. sledgei and S. dntmmondii grew together, along with a host of hybrids showing every conceivable gradation be- tween and combination of the characters of the two parents. It seems a pity that there is no one in the region sufficiently interested in conservation to buy up this bit of meadow and save it for investiga- tion by geneticists and enjoyment by nature lovers of the future. In April, 1910, while carrying on his fascinating studies of the relations be- tween insects and pitcher-plants. Dr. Frank Morton Jones had spent some time at Theodore, and had observed in one near-by meadow a pink-flowered form of S. venosa. He had furnished us approximate directions as to its location, and we soon found what appeared to be the right spot. In July, of course, pitcher-plant petals are withered, but we dug a few plants and shipped them to Philadelphia in the hope that they mio^ht bloom in the greenhouse the fol- lowing spring. This hope has now been realized ; and it turned out that we had struck the right spot. The parts of the flower which in most pitcher-plants are green or bronzy— the bracts, sepals and stvle-umbrella— are in this one nearly white, while the petals have a lovely rose color, unlike that of any other Sarra- cejttO/, Few herbarium records existing for ]\Iississippi plants, we visited bogs in two of the eastern counties of that state, ob- taining specimens of several Sarracenias and phloxes. Next we went to Havana, Alabama, and found the famous colony of the hybrid spleenwort {Asplenmm ehenoides)—i\ie only one in which fer- tility has been attained— to be m good condition. We then called on Dr. Ro- land M. Harper at Tuscaloosa, and ob- tained from him directions as to certain Sarracenia localities in the northern part of this state. Along the Sipsey River we found a colony of the Allegheny filmy- fern {Trichommies hoscianum), but search for its diminutive relative, T. peiersii, was unsuccessful. In Chilton County we located a colony of a red-flowered pitcher-plant, but it was not in good enough condition to es- tablish its identity. We then set out for the valley of the Little River east of Fort Payne, where a yellow-flowered one was reported. In spite of many hours' search in every conceivable type of habi- tat, we w^ere unsuccessful in finding it there, but Dr. Harper had fortunately observed it, also, near Center. On reach- ino- that place we found that, although recent clearing of the land for agricul- ture and burning over of the swamps, even where no such use was practicable, had nearly exterminated it, a few small clumps had somehow managed to escape destruction. Both in the field and in the crreenhouse, where it bloomed the follow- Tncr spring, this plant showed a number of'differences from its nearest relative, S. fava. and is to be classed as an inde- pendent species. The mountains of North Carolina were our next objective, for there grows the red-flowered pitcher-plant known as bar- racenia jonesii, the distinctness of which had onlv been recognized m 1929. its colonies^ proved to have been nearly de- stroved bv drainage of the swamps and bv the raids of vandals from the towns, but enough remained to enable this spe- cies to be" added to the collection. With it o-rew some beautifully veined Surra- cenia venosa. Ordinarily, when two closely related species or varieties exist, the more southern one tends to grow m the coastal plain, the more northern m the mountains; in this case, however, the southern representative grows both at low and high elevations. We also found hybrids between S. venosa and S. jonesii as yet undescribed. INTENTIONAL SECOND EXPOSURE Never having had an opportunity to make habitat photographs of and color notes on the rather rare Phlox ampli- folia Britton, we then made an effort to find this plant in several counties where it had been reported. The roadsides were gay with another member of the genus — P. Carolina L., long mistaken for various other species — but for some time we were unable to find a single colony of the one especially sought. Finally, however, it turned up in thickets in the vicinity of Willets, Jackson County, and the desired data upon it were obtained. Leaving the mountains, we next drove to Charleston, South Carolina, where some of the specimens in the Elliott Herbarium, preserved at the Charles- ton Museum, were studied, and then made for Summerville. How abundant pitcher-plants formerly were here is well shown by the splendid photograph in Macfarlane's Monograph on the fam- ily in Engler's * * Pflanzenreich ' ' ; but when w^e reached the spot where this had been taken, a very different sight met our eyes. Drainage of the swamps and burning of the woods had destroyed practically everything, and it was only after considerable search that we found even a single pitcher-plant in the midst of the rank, weedy grass and brush that had come in. Three species of pitcher-plants re- mained to be collected at the northern- most margin of their range, so that they would be as hardy as possible. Eastern South Carolina proved, however, to be poor collecting ground, for droughts ex- tending over a period of years had so lowered the water-table that many for- mer swamps were now dry land. More- over, the local farmers had taken to planting crops in the lower areas, and many a time when we pushed through pine woods toward what should have been a Sarracenia bog we found only a Zea Mays or Gossypium bog instead. Sarracenia minor was finally obtained in the neighborhood of Florence, South Carolina, and we then made for Lake Waccamaw, North Carolina. A few months before, Mr. Benedict had dis- covered here a northern outpost of the Florida swamp-fern, Dryopteris flori- dana, and of this we were able to obtain a good series of pressed specimens. The sweet pitcher-plant, S. rubra, eluded us for some time, but we finally located it in wet woods on the outskirts of Fayetteville, and roots were duly col- lected. Before leaving this part of the country, an attempt was made to obtain some of the remarkable little insectivo- rous plant, Dionaea muscipula, from a northern marginal occurrence, but the drought proved to have destroyed prac- tically all of it, and only a very small clump could be obtained for planting out in the New Jersey preserve. Here it has survived the first winter, however, so there is some hope that it may become established there. The last pitcher- plant, S. flava, was obtained near New Bohemia, Prince Charles County, Vir- ginia, and the series was complete. ^ A V Mm : I : »^ • f 1 t K \ y t < > -■ ^ ^ ^ r K Reprinted from American Fern Journal, Vol. 23, No. 4, October-December, 1933. Fern Field Notes, 1933 Edgar T. Wherry^ AsPLENiUM MONTANUM. — As its species name implies, this Spleenwort is best developed in the mountains (of the eastern United States), but it occasionally descends to fairly low altitudes. Thus, it has long been known to occur on a cliff rising from the Potomac River 21 miles east of Dean wood, Fairfax County, Virginia (recorded in the Flora of the District of Columbia and Vicinity as *^ above Great Falls'^). The contour lines on the Seneca quadrangle of the U. S. Geological Survey indicate this to lie 175 feet above sea-level. In August, 1932, it was discovered by a young amateur botanist, Arlton Murray, still nearer to the District of Columbia, namely on a cliff along Northwest Branch, 1^ miles north of Burnt Mills, in Montgomery County, Maryland. This lies so much nearer the Fall Line that it might have been expected to be even lower in altitude ; the stream has, however, a rather steep gradient, and a visit to the locality in com- pany with Dr. William R. Maxon and Mr. J. E. Benedict, Jr., on February 22, 1933, enabled it to be located on the topographic map, which showed it to be 250 feet above sea-level. The colony is a very small one, only 7 or 8 plants being in evidence on the rock. The occur- rence is worth placing on record, however, both as rep- resenting a range-extension of the species into a new county, and as illustrating the remarkable fact that even in a region supposedly well explored there are still dis- coveries in plant geography to be made. On July 31, 1933, this fern was collected on slate cliffs along the James River at both ends of the bridge running from Bremo, Fluvanna County, to New Canton, Buck- 1 Contribution from the Botanical Laboratory and Morris Arbo- retum of the University of Pennsylvania. • ^ -• « 110 American Fern Journal Fern Field Notes, 1933 111 ingham County, Virginia. The Palmyra topographic sheet shows these colonies also to have an elevation of 250 feet. The record for the growth of the species at low altitudes is thus still held by the stations along the Susquehanna River in the southern parts of York and Lancaster Counties, Pennsylvania, several of which lie little more than 100 feet above mean sea-level. Asplenium bradleyi. — This rare species has appar- ently been recorded but twice in West Virginia, on Muddy Creek Mountain in Greenbrier County, near the southern boundary of the state (Fred W. Gray), and near Fayetteville, Fayette County (Maurice Brooks). It has now turned up considerably further north, on the southwest-facing cliffs of New Creek Mountain, 2 miles southwest of Corners P. 0., in Grant County. Asplenium resiliens. — For many years the northern- most known occurrence of the Black-stem Spleenwort has been in the vicinity of Front Royal, Warren County, Virginia, latitude 38° 55\ In 1930, however, its range was extended to latitude 39° 10' by its discovery along the North Fork of Patterson Creek, in Grant County, West Virginia (Natalie B. Kimber and Mary F. Wright, specimen in herbarium Academy Natural Sciences Phila- delphia.) On July 23, 1933, J. E. Benedict, Jr., and I decided to make an effort to find it still further north, and selected Jefferson County, West Virginia, as a prom- ising region, in view of the extensive outcrops of lime- stone rock occurring there. At Bloomery, 3 miles southeast of Charles Town, a new bridge has been con- structed across the Shenandoah River, and from this bridge limestone cliffs were seen to extend southward along the east bank of the river. At first only the com- moner spleenworts could be found there, but we finally came upon one especially moist, sheltered crevice, sup- porting 6 or 8 plants of the desired species, thus extend- ing its known range north to latitude 39° 15'. J This record was able to stand, however, for but a few weeks. On August 8, 1933, Mr. Walter S. Lapp and I visited a cliff on the north bank of the South Branch of the Potomac River at Grace, two miles southwest of Springfield, Hampshire County, West Virginia. This cliff is interesting in that the upper part is of a fossilifer- ous sandstone, the soils in crevices tending to become acid there, while the lower part is limestone, and the soils correspondingly circumneutral throughout. Polypodium virginiamim and Selaginella ritpestris proved to be lim- ited to the more acid situations, and Cheilanthes layiosa to be best developed there. Pellaea atropitrpurea and Asplenium cryptolepis {ruta-muraria of old) grew only, and Woodsia ohtusa and Asplenium platyneuron most frequently, in the more limy substrata. In sheltered places at the lower levels a few small plants of Asplenium trichomanes were seen. Fortunately, just before leav- ing, inspection was made of a crevice near the base, where the limestone had weathered back to a depth of about 25 cm. for several meters along the stratum, and this proved to hold in its inner recesses several plants of unmistakable Asplenium resiliens. Latitude 39° 2;')' thus becomes its present known northern Jimit. Thelypteris simulata. — While the books give Mas- sachusetts Fern as the common name of this species, that name is apparently without significance in connection with it, and the suggestion is here made that it be called Bog Fern, because it is about as characteristic of sphagnum bogs as its relative, the Marsh Fern, is of marshes. The Bog Fern is very rare southward, and so far as I know the only published records for Maryland have referred to Coastal Plain occurrences. It was therefore interesting to find it, on August 4, in a boggy woods south of the National Highway 3 miles east of Grantsville, Garrett County, at an altitude of 2,575 feet. 112 American Fern Journal Ophioglossum engelmanni. — The record of the find- ing of the Limestone Adders-tongue in Frederick County, Virginia, by Mr. Hunnewell in a recent number of this Journal brings the number of counties of this state in which it is thus far known to five. A sixth may now be added: it was collected on July 26 by J. E. Benedict, Jr., and the writer one mile west of Harrison- burg, Rockingham County. Though not representing a new county, the following report may also be of interest. About 2 miles northeast of Staunton, U. S. Highway No. 11 forks, the left-hand branch avoiding traffic by entering only the outskirts of the town. Southwest from the filling station in the fork there extends a long series of limestone ledges with clayey depressions between, and here this fern is probably as abundant and accessible as anywhere in the northeastern part of its range. It grows in dense clay, mingled with the usual weeds of pastured land, and can best be found after a rainy spell. WooDsiA IN North Carolina.— The manuals give the range of Woodsia ilvensis as extending into this state, but no specimens appear to be preserved in any her- barium, so that this report must be considered as lacking adequate foundation. On the other hand, Woodsia sco- pulina (or its southeastern representative) is unquestion- ably represented by a specimen in the New York Bo- tanical Garden. This had been collected by the late J. A. Ferriss, supposedly on a cliff in the Craggy Moun- tains, in Buncombe County. Mr. William A. Knight, Mr. J. E. Benedict, Jr., and I recently spent a day in that region, exploring a number of cliffs, but were un- able to find any trace of it. As its reported occurrence on White Oak Mountain in Polk County has also never been rediscovered, the presence of any species other than W, ohtusa in North Carolina remains doubtful. Philadelphia, Pa. From Proceedings of the Pennsylvania Academy of Science, Vol. VII, 1933 FOUR SHALE-BARREN PLANTS IN PENNSYLVANIA^ Edgar T. Wherry Associate Professor of Botany, University of Pennsylvania In the writer's summary- of data as to the endemic plants of the Ap- palachian shale-barrens, this type of habitat wa« mapped as developed only between latitudes 37° 15' and 39° 45' north. At the same time, how- ever, it was pointed out that at least one of the endemics in question — the larp^e flowered Evening-primrose, Oenothera argillicola Mackenzie — occurs on shale of the same geological age as that underlying the bar- rens, in Perry County, Pennsylvania. Other similar range-extensions can now be reported. Fig. 17. Map of the ranges of Trifolium, virginicum and T. reflexum as known up to April, 1933, bringing out their interpretation as the only surviving descendants of an ancestor which lived in the upper Great Lakes region during pre-Glacial time. Trifolium virginicum Small. Discovered in 1893, this plant was described by SmalP the follow^ng year, the type locality being Kates Mountain, Greenbrier County, West Virginia, which for many years remained the only known station for it. In 1008 Miss McDermott-* stated it to be ''abundant throughout the Ap- palachian Mountains,'' and although this is somewhat exaggerated, it ^ Contribution from the Botanical Laboratory and Morris Arboretum of the Uni- versity of Pennsylvania. * Journal Washington Academy Sciences 20 : 43. 1930. ^Mem. Torrey Botan. Club 4: 112. 1894. * North American Species Trifolium : 273. 1908. has subsequently been found in at least 10 new localities. Since several of these lie not far south of the Pennsylvania line, there seemed a possi- bility that it might extend locally into the latter State, and in 1932 Pro- fessor S. C. Palmer, of Swarthmor^ College, joined me in a search for it there. Several hours were spent in scanning shaly slopes of various de- grees of sterility without results, but finally, on June 17th, a colony of it was found one mile south of the village of Artemas, in Bedford County, a like distance north of the Mason and Dixon line. Its closest relative, T. reflexum L., ranges far and wide over the interior provinces. The Shale-barren Clover differs in being perennial with, more elongated, fleshier roots, dwarfer stature, and narrower leaflets, though does not ex- hibit any recognizable floral differences. Miss McDermott, in the paper cited, considered these features to justify only varietal separation ; since, however, no intermediates between the plants occur, they may as well be maintained as distinct species. Oenothera argillicola Mackenzie. Like the preceding, this plant was discovered at Kates Mountain, and subsequently found to be wide-spread in the shale-barren country. In PENNSYLVANIA ACADEMY OF SCIENCE 163 '4'y \ ^\~>^^s^ Fig. 18. Map showing by dots the 12 localities of Oenothera argillicola known up to April, 1933. 1920 I observed it opposite Losh Run Station in Perry County, Pennsyl- vania,"' but the construction of a highway (U. S. No. 22) along the north- east bank of the Juniata subsequently exterminated it there. It was col- lected 2 miles southeast of Huntingdon by State Botanist E. M. Gress a few years later, and in the State College herbarium there is a specimen of it from Hawn Bridge, also in Huntingdon County. A visit to out- crops of Devonian shale during August, 1932, resulted in finding it at two new localities — on a stream bank 3 miles southwest of Orbisonia, Huntingdon County ; and on steep slopes between highway and river two miles west of Newton-Hamilton, Mifflin County. Convolvulus pursiiianus Wherry. In announcing the discovery of three shale-slope plants in Maryland,^ ^ This occurrence has recently been discussed, and the plant illustrated in color in Addisonia 17: 55, pi. 572. 1932. «Torreya 29: 105. 1929. the name Convolvulus starts Michaux was revived for the derivative of C spithamaeus L. occurring on the shale-barrens, since his description seemed to apply to the Appalachian plant, although the locality cited was ''Canada near Lake Champlain.'' In September, 1932, under the guidance of Brother Marie-Victorin, I visited the sand-plains east of Montreal and collected what may be regarded as topotype material of Michaux 's species. It proved to be merely a hairy extreme of C. spitha- maeus, not identical with the shale-barren plant, so that the name C. stems is not correctly applied to the latter, after all. Pursh's name,^ Calystegia tomentosa, can not be recombined because there is already a Convolvulus tomentosus in another part of the world. Accordingly, the Velvet Convolvulus of the shale-barren country is here renamed in honor of its discoverer. Fig. 19. Map of the ranges of Convolvulus spithamaeus ^ C. purshianus, and C. stans as known up to April, 191^3, bringing out the interpretation of the last two as regional segregates from the first. Convolvulus purshianus, nomen novum. Calystegia tomentosa Pursh, 1814, not Convolvulus tomentosus L. 1753. Convolvulus stans Wherry, 1929, not Michaux. Plant spreading into large colonies on shale-slopes by rootstocks; aerial branches 10 to 40 cm. tall, the internodes little exceeding the peti- oles ; herbage densely white velvety-pubescent ; leaf-blades mostly oblong or elliptic-sagittate with conspicuous auricles 5 to 10 mm. long ; petioles 10 to 20 mm. long, about Ys the length of the blades ; bracts ovate, often cordate, and rather strongly keeled ; corolla white. ' Flora America Septentrionalis 1 : 143. 1814. Type specimen collected in dry woods on *'Top of the rid^e behind Rattlesnake Den, Sweet Sprinp^s/' Monroe County, West Virginia, by . Frederick Pursh in 1806, in herbarium Academy Natural Sciences, Philadelphia. The ran^e of this Convolvulus hitherto reported is from Alleghany County, Virginia, to Allegany County, Maryland. In June, 1920, it was collected at Charter Oak, Huntingdon Co., Pa., by W. C. Muenscher, but distributed to herbaria as C. spithamaeus. Search for it in 1932 resulted in finding it at several places in Bedford, Fidton, and Huntingdon Counties, Pennsylvania, on slopes of Devonian shale. Senecio antennariifolius Britton The occurrence of this plant in Pennsylvania has already been re- corded by State Botanist Gress,^ and no new stations for it have been dis- covered. This plant shows a different geographic relation than the other species here included : instead of being related, like these, to plants which approach or enter the Appalachians, its nearest relative occurs in the Rocky Mountains, two thousand miles away. The only reasonable ex- FiG. 20. Map of the ranges of Senecio canus and S. antennariifolius as known up to April, 1933, bringing out their interpretation as the only surviving descendants of an ancestor which lived in the Hudson Bay region during pre-Glacial time. planation of such a distribution is that during Tertiary times the an- cestor of both species grew somewhere in what is now central Canada, and descendants chanced to migrate out far enough to escape destruction by the Quaternary ice sheets in two regions. Since the ice retreated the western species has been able to regain some of the lost territory, but the eastern one, having seemingly become more conservative, has not yet returned even as far as the Wisconsin terminal moraine. *Proc. Penna. Acad. Sciences 4: 29. 1930. Bartonia, No. 15 Plate 1 Reprinted from Bartonia, No. 15, 1933 Fig. 1. Sarracenia p^irpurea. A, gihhosa; B, venosa. Fig. 2. Sarracenia oreophila. Type specimen. |! Fig. 3. Sarracenia oreophila. Flat Rock, Jackson County, Alabama, June 11, 1933. The Geographic Relations of Sarracenia purpurea^ Edgar T. Wherry Early in 1931 there appeared in a German seriaP a set of maps purporting to show the distribution of the species of Sarracenia. These maps indicated the ranges as extending in some directions many miles beyond any localities cited in the trustworthy literature on these plants, and in other direc- tions not including nearly all the territory covered by definite records. Sarracenia purpurea is shown as occurring in Ken- tucky, West Virginia, and upland Virginia, and as abundant in Tennessee, although as had been pointed out by Harper^ * * no botanist now living seems to have seen it in those states, ' ' and no specimens from them are included in the principal herbaria of this country. At the same time, southern New Jersey is excluded from the range, in spite of the fact that Macfarlane* cited twelve and Stone^ twenty-four collections. The map presented herewith is based on the records ob- tained from seven comprehensive herbaria: Academy of Natural Sciences of Philadelphia, Canadian National Her- barium, Cornell University, Gray Herbarium, New York Botanical Garden, U. S. National Herbarium, and University of Pennsylvania, the last including the material collected by Macfarlane in the preparation of his monograph on the Sarraceniaceae. The literature has also been reviewed. 1 Contribution from the Botanical Laboratory and Morris Arboretum of the University of Pennsylvania. 2 Die Pflanzenareale, Ser. 3, No. 1, maps 1 to 3. 1931. 3 Journ. Elisha Mitchell Sci. Soc. 34: 115. 1918. 4 Sarraceniaceae, in Engler 's Pflanzenreich IV. 110 : 33. 1908. 5 Plants of Southern New Jersey : 467. 1911. (1) f\ Baktonia, No. 15 Plate 1 Reprinted from Bartonia, No. 15, 1933 V Fig. 1. Sarracenia purpurea. A, (fibho.sa: B, renosa. Fig. 2. Sarracenia oreophila. Type speeiinen. Fig. 3. Sarracenia oreophila. Flat Rock, Jackson County, Alabama, June 11, 1933. The Geographic Relations of Sarracenia purpurea^ Edgar T. Wherry Early in 1931 there appeared in a German seriaP a set of maps purporting to show the distribution of the species of Sarracenia. These maps indicated the ranges as extending in some directions many miles beyond any localities cited in the trustworthy literature on these plants, and in other direc- tions not including nearly all the territory covered by definite records. Sarracenia purpurea is shown as occurring in Ken- tucky, West Virginia, and upland Virginia, and as abundant in Tennessee, although as had been pointed out by Harper^ * * no botanist now living seems to have seen it in those states, ' ' and no specimens from them are included in the principal herbaria of this country. At the same time, southern New Jersey is excluded from the range, in spite of the fact that Macfarlane* cited twelve and Stone^ twenty-four collections. The map presented herewith is based on the records ob- tained from seven comprehensive herbaria: Academy of Natural Sciences of Philadelphia, Canadian National Her- barium, Cornell University, Gray Herbarium, New York Botanical Garden, U. S. National Herbarium, and University of Pennsylvania, the last including the material collected by Macfarlane in the preparation of his monograph on the Sarraceniaceae. The literature has also been reviewed. 1 Contribution from the Botanical Laboratory and Morris Arboretum of the University of Pennsylvania. 2 Die Pflanzenareale, Ser. 3, No. 1, maps 1 to 3. 1931. 3 Journ. Elisha Mitchell Sci. Soc. 34: 115. 1918. 4 Sarraceniaceae, in Engler's Pflanzenreich IV. 110: 33. 1908. 5 Plants of Southern New Jersey: 467. 1911. (1) INTENTIONAL SECOND EXPOSURE ._ JMH i^ ii til WJ-^iWrjiffia aifilHWMiTMCiWlifrrrin 2 PROCEEDINGS OF THE Sarracenia purpurea Linne is not altogether constant in its characters throughout its range, and three attempts have been made to subdivide it. In 1822 Eaton^ described under the name 8. heterophylla a plant from Massachusetts, character- ized as slender and yellowish throughout, and as having the outer leaves longer than the inner ones. He considered this ''a remarkably distinct species,'' but field study indicates that slender habit and elongated outer leaves represent merely the result of growing in the shade, leaving only the color as a distinctive feature. Eaton's species has accordingly been reduced in status by later workers, Torrey^ making it a variety, and Fernald^ a form. In 1929 the writer* termed it ''mutation heterophylla," but has subsequently come to the view that descriptive vernacular phrases are preferable to technical designations for such plants. Accordingly, when leaf-outline is under consideration, it may be called ''a slender- leaved ecad, ' ' while when attention is being paid to the colora- tion, it should be classed as ''an anthocyan-free mutation." The other two articles on this subject have been overlooked by most subsequent writers. The first was the characteriza- tion of a Newfoundland plant by La Pylaie^ as follows : *'p. Terrae-Novae, N. PhyUodiis brevioribus, fauce venis ramosis anastomosantibus vivid^ sanguineis; petalis minus spathulato-dilatatis, stigmatis angulis parum productis, non bifidis. ' ' Such features as short leaves, prominent red veining, and narrowed petals are apparently of ecological origin, develop- ing throughout the range in plants which grow high up on sphagnum hummocks or in marly depressions where nutrient elements are relatively unavailable. As to the stigma-char- acters, Newfoundland and east-Canadian material preserved in herbaria shows gradation from a condition where the sinuses between the projections are nearly filled with tissue, as in La Pylaie's specimen, to one where these sinuses reach nearly to the stalk, yielding a five-pointed star. Should one 1 Manual of Botany, ed. 3 : 447. 1822. 2 Rept. Bot. Dept. Survey N. Y. Assembly No. 50 : 120. 1839. sRhodora 24: 174. 1922. 4 Journ. Wash. Acad. Sci. 19 : 382. 1929. 5 Mem. soc. linn. Paris 6: 389, pi. 13. 1827. PHILADELPHIA BOTANICAL CLUB 3 extreme in this series receive a varietal name, the opposite extreme as well as various intermediates would also require naming. Until some one has the opportunity to work out the significance of these variations, it seems undesirable to burden the literature with such a series of names. Being in spite of his eccentricities a keen observer, Rafinesque^ divided the Linnean species geographically: * ^ 261, Sarazina gihhosa Eaf . (vel grandiflora) purpurea L. non omnis — several sp. or var. are blended in this remarkable plant, difficult to characterize and none are really purple — fol. conformis subsessilib. obo- vato gibbosis, lutescens, ala ampla gibbosa, appendice renif. setis retrorsis, scapis flexuosis, cal. obt. vel retusis, petalis spatulatis — Canada to Virginia, swamps. . . . **263, Saraz. venosa Eaf. differs from gibbosa, by leaves short with small wings, venose reticulate of red chiefly in the lid, scape streight flowers smaller — Virg. ad Florida." Field and herbarium study have shown that there really are foliage differences between average northern and southern material of 8. purpurea, although ecads and other variants of either one may simulate the features of the other, so that sepa- ration into independent species is scarcely justified. The areas occupied by the two being, however, distinct, they are here classed as subspecies. Sarracenia purpurea venosa (Raf.) Wherry, status novus. Plate 1, fig. IB. Of the differentiating characters given by Rafinesque for his ''8arazina venosa'^ only the first, the leaf -length is really valid, as in both northern and southern plants the veining, length of scape, and size of flower vary in relation to state of nutrition. The leaves of the southern one are character- istically covered with bristly hairs on the exterior surface, and are short and broad in outline, the hollow part averaging less than three times as long as wide. The hood above their orifice is relatively large, and its wings (when the leaf is laterally flattened on a herbarium sheet) may extend well beyond the lip of the hollow part. This type of leaf was figured by two pre-Linnean writers, Plukenet,^ who termed the plant 1 Autikon Botanikon: 33. 1840. 2 Amalth. Bot.: 46, pi. 376, fig. 6. 1705. 4 PROCEEDINGS OF THE ' ' Bucanephyllon americanum'' and Catesby/ who described it a^ ''Sarracena foliis hreviorihus.'' When growing in espe- cially wet places, the leaves tend to elongate and the hood to decrease in size, thereby approaching the northern segregate in aspect, and decision as to which is represented by isolated herbarium specimens is sometimes difficult. Sarracenia purpurea venosa is fairly common along the Gulf Coastal Plain from eastern Louisiana to western Florida, and rare in Georgia and South Carolina. It reaches ite maxi- mum development in North Carolina, growing not only in the lower country but also up to an elevation of at least 3,500 feet in the Blue Ridge. In Virginia, on the other hand, it is ap- parently absent from the uplands, and there are authentic records from but three lowland counties, although it is abun- dant enough around Richmond to be offered for sale along the streets. Skipping northeastern Virginia, Maryland, and Dela- ware, it appears again in southern New Jersey, intergradation with the northern segregate being there especially frequent. Migration to this isolated area presumably took place during late Tertiary or early Glacial times, over a strip of Coastal Plain which has subsequently sunk beneath the sea. No anthocyan-free mutation has been observed in the south- ern segregate, but there is a beautiful color-form connected with partial loss of plastids from the flower, resulting in the sepals becoming pale green, the style-umbrella white, and the petals rose-pink instead of the usual deep red. This was dis- covered by Dr. Frank Morton Jones near Theodore, Alabama, in 1910, but has apparently never been recorded in print. In the course of an expedition for the collecting of pitcher- plants undertaken by the writer during July, 1932, at the instance of Mr. Louis Burk, of Philadelphia, three small clumps of this plant were obtained, and the following Spring these flowered in the greenhouses at Latham Park. No tech- nical form-name will be given to this, a^ when being discussed from the scientific standpoint it can be referred to as a pallid mutation; in cultivation, however, it may well be known as Sarracenia purpurea venosa, horticultural variety Louis Burk. iNat. Hist. Carolina, etc. 2: 70. 1731. PHILADELPHIA BOTANICAL CLUB 5 Sarracenia purpurea gibbosa (Raf.) Wherry, status novus. Plate 1, fig. lA. In contrast to the leaves of the southern plant, those of the northern one are only exceptionally bristly on the outside, and are normally long and narrow, the hollow part averaging over three times as long as wide. The hood is smaller, and its wings, when laterally flattened, extend little if at all beyond the pitcher-lip. The earliest known figure of any Sarracenia, that published by Clusius^ in 1601, showed a leaf of this type, so must have represented a northern occurrence. Here also ecological conditions may lead to variation, and in unfavorable situations the leaves of the northern segregate tend to become shorter and to resemble those of the southern one. Unlike most northern plants, this Sarracenia did not sur- vive the Glacial epoch in the southern Alleghenies, but on the Coastal Plain and Piedmont of the middle Atlantic states, for it occurs well south of the ice limit only in the District of Columbia, northeastern Maryland, and Delaware. After the ice retreated it spread rapidly into the boggy areas which developed in the north, reaching Newfoundland, Lake Mel- ville and East Main River, Labrador,^ Athabasca River val- ley,^ and Fort Chipewyan, according to a citation by Mac- f arlane.* Its northwestern limits are still uncertain ; Hooker^ reported it to occur on Bear Lake, and several compilers of manuals and floras have attributed it to the Canadian Rocky Mountains, but no specimens from these regions are extant. The two subspecies may be keyed out as follows: Leaves externally densely hirsute or rarely glabrous, tending to be short and broad, the hollow part averaging less than three times as long as wide; hood relatively large, its wings, when laterally flattened, often extending weU beyond the pitcher-lip ; petals rather light red, or pink in a mutant - S- P- '^^^^« Leaves externally glabrous or rarely sparingly hirsute, tending to be long and narrow, the hollow part averaging over three times as long as wide; hood relatively small, its wings, when laterally flattened, extending little if at all beyond the pitcher-lip ; petals intense red, or yellow in a mutant S. p. gihhosa iHist. Plant. Ear.: Ixxxij. 1601. 2 Geol. Surv. Canada Rept. 1895, App. vi. : 354 L. 3 Geol. Surv. Canada Rept. 1875-76: 188. 4 In Engler's Pflanzenreich IV. 110: 34. 1908. 5 Flora Boreali- Americana 1 : 33. 1833. PROCEEDINGS OF THE III 1 ! •1 II ^1 I They have been seen from the following counties : Sarracenia purpurea glbbosa (southern margin only). Delaware: Kent, Newcastle, Sussex; Illinois : McHenry ; Indiana: DeTawaX Varren Maryland: Anne ^-nde^ Prance Georges ; Minne- sota- Pope, Winona; New Jersey: Camden, Cumberland, Salem, Otao. Loganl Pe^sylvania: Center, Lancaster, Montgomery Somerset; Wis- consin: Dane; Canada, Alberta: Edmonton; Saakatchewan: Prince Albert. Fig. 1. Distribution of the subspecies of Sarracenm purpurea. The oppositely inclined rulings mark the respective areas occupied. The ^^w of circles indicates the approximate southern margin of the Wisconsin ice sheet. The X 's represent unconfirmed reports. Sarracenia purpurea venosa (complete list). Alabama: Baldwin, Clarke, Coffee, Escambia, Geneva, Henry Mobile, Washington; Florida: Calhoun, Gadsden, Holmes Leon, Liberty Oka- loosa, Walton; Georgia: Lee, Randolph, Tattnall ; Louisiana : St. Helena ; New Jersey: Atlantic, Burlington, Camden, Cape May, Cumberland, Gloucester, Monmouth, Ocean; North Carolina: Bertie, Buncombe Car- teret, Catawba, Columbus, Craven, Cumberlajid, Duplm, Forsyth Hender- son, Hoke, Jackson, Macon, Martin, Moore New Hanover, Richmond, Robeson, Scotland, Wake; South Carolina: Beaufort, Chesterfield, Dar- lington, Horry, Kershaw; Virginia: Brunswick, Henrico, James City. The Appalachian Relative of Sarracenia flava^ Edgar T. "Wherry In an article on undescribed and little known plants of Ala- bama, which appeared in 1897, Mohr^ mentioned a pitcher- plant growing along Little River, in De Kalb County. He decided it to be identical with Sarracenia cateshaei Elliott, but as it was obviously related to S. flava, he reduced its status and named it S. flava cateshaei. Three years later Kearney^ applied to this mountain plant the name S. flava var. oreo- phila, but failed to validate it by furnishing any descriptive data. When he came to prepare his volume on the flora of the state, Mohr^ noted this plant to be ''readily distinguished from the very closely allied Sarracenia flava by the strictly erect leaves with ventral wing narrower and the sides of the broad dark purple veined lamina scarcely if at all reflexed,'' and used Elliott's name for it. The same usage was followed by Harbison^ in reporting a new locality. Elliott's S, cateshaei was subsequently found by Macfar- lane^ to represent a hybrid between S. flava and S. purpurea, showing Mohr's application of the name to be erroneous; he did not regard Mohr's plant as distinct from S. flava. Some years later, however. Harper^ called attention to another point of difference, namely the presence of short recurved ensif orm leaves. In 1932 living plants were collected near Center, Cherokee County, Alabama, and the following March these bloomed in Mr. Burk's greenhouse at Latham Park. Many differences from S. flava were shown, and as no intergradation appears to occur, and their ranges are distinct, they are here classed as independent species, the new one becoming : 1 Contribution from the Botanical Laboratory and Morris Arboretum of the University of Pennsylvania. 2 Bull. Torrey Botan. Club 24: 23. 1897. 3 Science 12 : 833. 1900. 4 Plant Life Ala. : 531. 1901. 5 Biltmore Botan. Stud. 1 : 155. 1902. 6 Rept. Conf . Genetics, 1907 ; Journ. Botany 45 : 4. 1907. 7 J. Elisha Mitchell Sci. Soc. 34: 120. 1918. (7) 8 PROCEEDINGS OF THE Sarracenia oreophila (Kearney) Wherry, species nova Plate 1, figs. 2 and 3. S. flava oreophila Kearney, nomen nudum; Harper, nomen subnudum. S. cateshaei Mohr et al., not Elliott. Flat ensiform leaves short and numerous; hollow leaves (pitchers) up to 75 cm. tall, moderately constricted at junc- tion with hood, well developed at flowering time; scent of flowers faint, musty but not unpleasant ; style-umbrella spar- ingly pubescent to glabrate; petals little longer than the sepals, the blade narrow, greenish yellow, firm in texture and remaining outspread during anthesis. (Folia ensiformia numerosa; amphorae ad 75 cm. longae, opercuU marginibus inf erioribus parum constrictae ; odor floris non ingratus ; umbraculum styli sparse pubescens vel glabra- tum ; petala parum sepalis excedentia, laminis angustioribus viridi-flavescentibus rigidis per anthesem expansa.) Sarracenia flava L. differs in having longer leaves, flowers with disagreeable feline scent, pubescent style-umbrella, and larger pale yellow petals of more delicate texture. Alabama: Cherokee Co.: 2 miles northeast of Center, De Kalb Co. : ''Little River, Lookout Mt., Val- ley Head'' (Biltmore No. 371, May 22, 1899, TYPE in U. S. National Herbarium) ; actual lo- cality probably De Soto Falls, southeast of Mentone; also Little River southeast of Fort Payne. Jackson Co.: Flat Rock, Higdon, Pisgah, and probably elsewhere on Raccoon or Sand Mountain. Mar- shall Co.: Albertville and Carpen- ter, Sand Mountain. Georgia: Reports from Cloudlajid, Chattooga Co., and La Grange, Troup Co., not authenticated. . Taylor Co. : Butler (Neisler, iden- tified as S. cateshaei and annotated *' flowers not odorous,'' in her- barium N. Y. Botanical Garden). M^. Fig. 1. Distribution of Sarracenm oreophila. ■V'"t"- .■.'■'.''- '■^rrrr^- ; I Reprinted from Rhodora, Vol. 35, April, 1933. HEUCHERA HISPIDA PURSH REDISCOVERED^ Edgar T. Wherry During the year 1805 Frederick Pursh collected plants in the Appalachian mountain region of Virginia and West Virginia, a number of which were described as new species in his Flora Americae Septen- trionalis, which appeared 9 years later. Among these was a Ileuchcra hispida, stated to have the leaves hispid above but glabrous beneath, the peduncles glabrous, and the flowers medium-sized with purple petals and exserted stamens.^ Through misunderstanding, subse- quent authors came to apply this name to a western plant havmg the peduncles and lower leaf-surfaces more hispid than the upper surfaces. This situation was recognized in the course of a revision of the genus undertaken at the University of Minnesota by Miss Olga Lakela and Professors Rosendahl and Butters,^ but on borrowmg material from various herbaria they were unable to find a specimen corresponding to Pursh's description in any subsequent collection, except a few of material grown by Gray from roots collected m Giles County, Virginia, in 1843. A Pursh specimen of //. hisptda is for- tunately preserved, however, in the herbarium of the Academy of Natural Sciences of Philadelphia, and the label gives its place of collection as " high mountains between Fincastle & the Sweet Springs. On being advised of these facts in the Spring of 1932, the writer decided to endeavor to rediscover the plant, and as soon as the term's class work was over started on a trip, in company with Professor S. C. Palmer of Swarthmore College. Leaving Swarthmore, Pennsylvania, on June 9th, we made several stops to collect plants en route, and reached Fincastle, Botetourt County, Virginia, in the afternoon of June 12th. Continuing north- » Contribution from the Botanical Laboratory of the University of Pennsylvania. This account of the incidents of the trip supersedes any which has appeared m news- papers and popular magazines. « Flora Americae Septentrionalis 1: 188. 1814. • Cf. the preceding article. 1933] Wherry, — Heuchera hispida Pursh rediscovered 119 westward from this town, we took an unpaved but fairly good road which led over higher mountain passes than the modern highway, but found only Heuchera pubescem Pursh, a widespread species, between there and Newcastle, Craig County. The main highway running north from the latter place (State No. 22) proved to have been recently reconstructed, and though not yet surfaced was wide and well-graded, so even though night was approaching and clouds could be seen to be gathering along the mountain ridges, we ventured to continue on 11 miles to the summit of Potts Mountain, which was reached about 8 P.M. At the point where the highway crosses the divide, elevation about 2400 feet, we found an openly wooded rocky flat, and had soon selected a parking place for the night. Then, before making any preparations to retire, we got out our flashlights and started to look around to see if any Heuchera might be growing there. The fog was almost impenetrable, but in a few moments Palmer^s flashlight beam struck a clump of one of them, and a leaf was soon brought closer to the light. It proved to be hispid on top but not beneath, just as Pursh had said, and the flowers, though just beginning to open, agreed wholly with his description. We had rediscovered, at or near the type locality, the real Heuchera hispida, not seen growing there for 127 years, and last seen in the wild, in the county next adjoining on the west, by Gray 89 years before. The following morning we found two or three additional plants in bloom, and several in bud, along with another member of the genus, the well-known //. villosa Michaux, not yet showing its inflorescence. On descending the north side of the mountain, the H. hispida proved to be present down to about 400 feet below the summit. Continuing on toward Sweet Springs, we saw a few additional plants of this species high up on Peters Mountain, in both Craig County, Virginia, and Monroe County, West Virginia. None could be found, however, in any other part of the Appalachians visited, from western Virginia to southern Pennsylvania, so it is evidently endemic in a decidedly restricted area. University of Pennsylvania. ■f" MM -iia-ftTHiry ■ een most interest- ingly described by Morris and Eames ;^ and at least as many others occur in the southeastern states. To discuss these all at once would occupy too much space, so it is proposed to take them up in small groups at a time in successive numbers of this Bulletin. The rela- tively primitive Slipper-orchids {Cyp- ripedium and allied genera) may be treated first. The Rams-head Orchid, Cypripe- dium (Criosanthes) arietinum Br. Before the Glacial Epoch this odd little orchid evidently lived in the Arc- tic regions, but was able to migrate far enough southward in both eastern Asia and eastern North America to escape destruction by the advancing ice-sheets. Surviving glaciation somewhere in the Alleghenies, it followed the retreating ice to Minnesota, Manitoba, central Ontario, Massachusetts, and Quebec, but died out in its survival-area, the exact location of which is accordingly indeterminate. Reprinted from The American Orchid Society Bulletin, March, 1933 March, 1933 THE AMERICAN ORCHID SOCIETY BULLETIN 101 Moisture conditions, usually re- garded as an extremely important fac- tor in plant growth, seem to make no difference to this species, for it thrives alike in wet mossy swamps and on dry wooded rocky slopes. Its soil is, however, always well provided with humus and thoroughly aerated. So far as my tests have shown, the reaction is usually subacid or minim- acid, and as in the case of most orchids, rather sterile or poor in avail- able plant-foods. These factors should not be so difficult to match as to prevent its cultivation ; but there re- mains one which is evidently far more critical, namely the summer tem- perature. Throughout the region where it is native the air rarely gets hotter than 80 degrees F, and the soil at root level probably never ex- ceeds 70 degrees. Its inability to erow in nature below central New York and Massachusetts clearly in- dicates that any attempt to cultivate it in gardens where the summer sun heats the soil to 75 or 80 degrees for long periods is foredoomed to failure. The only way I can suggest to grow it south of its natural range for any length of time is to provide a means for keeping the bed cool. In the type of rock-garden more or less inaccurately termed the moraine, this is accomplished by causing a stream of cold water to trickle among loosely set rocks beneath the surface soil. The same result could be more effectively attained by running beneath the bed a pipe carrying brine from an electric refrigerating machine. The best plan of all would be to turn an area where it grows naturally into a wild-flower preserve, and to make an annual pilgrimage there during its blooming season. I Edgar T. Wherry Cypripedium arietinum Saiible Beach, Hepworth, Ontario Photographed, June, 1926 References ^Contribution from the Botanical Laboratorv and Morris Arboretum of the University of Pennsylvania. -Journal Wash. Acad. Sci. 8: 589. 1918; Smithsonian Annual Rept. 1920: 263; Rhodora 22: 47. 1920 and 23: 127. 1921 ; J. Wash. Acad. Sci. 17: 35. 1927 and 18: 212. 1928. ^Our Wild Orchids. Scribners, New York, 1929. r 100 THE AMERICAN ORCHID SOCIETY BULLETIN March, 1933 Native Orchids Our Eastern Orchids and their Cultivation Cypripedium arietiuuni Edgar T. Wiikrry Uniz'crsity of Pcjiiisvh'aiiia^ SoMK of our native ])lants are easy to transplant to the garden, and thrive tliere without si)ecial attention. Others, inckiding most of the orchids, usually last hut a year or two in cultivation, and have to he continually replaced. The removal of such plants from their na- tive haunts to situations where they die without reproducing themselves is leading to their disappearance from many localities. Something must he done to save them from extinction. For many years the writer has heen studying the more or less untamahle wild flowers of the eastern United States in the hojx^ of ascertaining the reasons for their peculiar hehaviors. While fmal solutions of the difficulties have not heen attained in all cases, many things have heen learned which seem worth i)utting on record for the henefit of other wild-flower cultivators. Data as to the soil-reaction preferences of native orchids have already heen ])ul)lishe(l,- hut other factors which must he taken into account if they are to l)e successfully cultivated remain to he discussed. There are over 70 species found in the northeastern United States and adjacent Canada, the hah- itats of which have heen most interest- ingly descrihed hy Morris and Eames r"^ and at least as many others occur in the southeastern states. To discuss these all at once would occupy too much space, so it is proixjsed to take them up in small groups at a time in successive numhers of this P>ulletin. The rela- tively primitive Slipper-orchids (Cv/^- ripcdium and allied genera) may he treated first. The Rams-hkai) Orchid, Cypripe- dium (Cri(fsanthcs) arietiuuni V)V. Before the Glacial Epoch this odd little orchid evidently lived in the Arc- tic regions, hut was ahle to migrate far enough southward in hoth eastern Asia and eastern North America to esca]x? destruction hy the advancing ice-sheets. Surviving glaciation somewhere in the Alleghenies, it followed the retreating ice to Minnesota, Manitoha, central Ontario, Massachusetts, and Ouehec, hut died out in its survival-area, the exact location of which is accordingly indeterminate. t March, 1933 THE AMERICAN ORCHID SOCIETY P.CLLETIN 101 Moisture conditions, usually re- garded as an extremely important fac- tor in plant growth, seem to make no difference to this species, for it thrives alike in wet mossy swamj)s and on dry wooded rocky slopes. Its soil is, however, alwavs well ]>rovi(led with humus and thoroughly aerated. So far as my tests have shown, the reaction is usually suhacid or minim- acid, and as in the case of most orchids, rather sterile or poor in avail- ahle ])lant-foods. lliese factors should not he so difficult to match as to prevent its cultivation ; hut there re- mains one which is evidently far more critical, namely the summer tem- perature. Throughout the region where it is native the air rarely gets hotter than 80 degrees F, and the soil at root level prohahly never ex- ceeds 70 degrees. Its inahility to grow in nature l)elow central New York and Massachusetts clearly in- dicates that any attem])t to cultivate it in gardens where the summer sun heats the soil to 75 or 80 degrees for long i:)eriods is foredoomed to failure. The onlv wav T can suggest to grow it south of its natural range for any length of time is to provide a means for keeping the hed cool. In the ty]>e of rock-garden more or less inaccurately termed the moraine, this is accomplished hy causing a stream of cold water to trickle among loosely set rocks IxMieath the surface soil. The same result could he more effectively attained hy running heneath the hed a pi])e carrying hrine from an electric refrigerating machine. The hest i)lan of all would he to turn an area where it grows naturall\- into a wild-flower ])reserve. and to make an annual ]>i1grimage there during its hloommg season. Edijar T. Wherry Cypripedium aricfinum Sauble Reach, Hcpwortli, Ontario Pli()tni»Taphe(l, Jiuie, 1*^26 Rkfkrkxces ^Contrihution from the Botanical Lahoratory and Morris Arhorelum ot the University of Penns\lvania. -lournal \\\ash. Acad. Sci. 8: S^^. 1918; Smithsonian Annual Kept. H>20 : 263; Rhodora 22: 47. 1920 and 23; 127. 1921 ; I. Wash. Acad. Sci. 17; 35. 1927 and 18; 212. 1928. •"^Our ^^1ld Orchids. Scrihners, New York, 1929. Reprinted jrom Thk A.mlrica.n Orchid Society Bulletin. March, l^'VS INTENTIONAL SECOND EXPOSURE f ! ! I Reprinted from June, 1933, American Orchid Society Bulletin !li Native Orchids Our Eastern Orchids and Their Cultivation 2. The Yellow and the White Cypripediums Edgar T. Wherry University of Pennsylvania^ Gor.DKN-sr.TPPKR Orchid, Cypripe- dium parvifloruui Salisb. Most current Ixytanical works class all the eastern yellow Ladyslipper Or- chids as l)elonging to this single small- flowered s]>ecies, but separate es- ])ecially large-flowered phases as a dis- tinct variety. Since the small and large-flowered ones differ in habitat and accordingly in cultural require- ments, they will here be treated as inde|)endenit species. The name Cypripediuni parviflornm is here applied to the one with rather deep yellow and often purple-blotched slipi>ers an inch or less in length. It is normally a swam]) dweller, and ranges from the mountains of North Carolina to Minnesota, central On- tario, and Nova Scotia. So far as my tests have shown it is a circumneutral soil ]>lant, and even though the sur- face may be covered with a layer of acid litter, its roots usually extend dee]) enough to reach non-acid muck benCcDth. To cultivate it successfully, these natural conditions should be matched ' C\>ntril)Uti()n froni the Botanical T^aboratory and Morris Arboretum of tlie University of Pennsvlvania. as closely as practicable. Its roots should be s]>read out on to]) of a bed of muck soil, provided with a never- failing supply of moisture, and then covered with an inch or two of some sort of ])orous litter, which can be added to from time to time. The temi>erature a]>j)arently does not have 'to l)e kept as low as with the Rams- head Orchid, yet markedly warm situations should be avoided. The delightful fragrance of the flowers of this species is unfortunately very attractive to slugs, and while in native habitaits these are kept in check by enemies, in the garden they may become so abundant as to injure or kill the plants. In some cases if a ])iece of an orange or lemon is laid in a shady ]>lace nearby, the ]:>ests will congregate on that during the night, and can be caught early in the morn- ing. Another ])lan is to surround the bed, during blooming season, with a stri]) of mulch pa]>er on which is laid a ridge of sand mixed with iron sul- ])hate or copperas, which they will refuse to cross. If these ])lans fail, they can l)e ]>icked from the ])lants at night, with the aid of a flashlight. I'i '> Yellow Ladyslipper Orchid, Cv- pripediimi puhcscens Wild. So far as known this is the easiest of our eastern orchids to cukivate. Iliere are records of i)lants lasting for a score of years, blooming freely, and setting abundant seed, in more than one flower garden. Clumps of it are offered by many dealers, and it is the s})ecies which should be tried first by the beginner in native orchid growing. While the slii)i)ers are usually around two inches long, individual ])lants with decidedly smaller flowers ai)i)ear in many of its colonies, and since when i)ressed these are indistin- guishable from those of the next- l)receding si>ecies, herbarium workers often assume the two to be identical. In the field, however, they ap])ear dis- tinct, in tliat, whatever their variabil- ity in size, the flowers of this one are ])aler in color and less fragrant than the other sj^ecies. It differs, too, in habitat, normally inhabiting uj)land woods or the drier j)arts of swamps. Not only is there confusion as to its technical nomenclature, but its common name is often stated in a misleading way, as "Ijarge Yellow Lady's Sli])}>er." As most yellow la- dies, whether large or small, wear sandals instead of slij>])ers, this ty])e of name arrangement seems to me to be undesirable ; to make such a name unambiguous, the words *'lady" and *'sli]>])er" should l)e united, either with a hyi)hen or into a single word, as has l>een done herewith, following Stand- ardized Plant Names. As this Ladyslij)})er ranges from Mississippi and Alabama on the Gulf Coast north to the lower borders of Canada, temperature is evidently of no imiK>rtance to it. Its soils' are often rich in i>lant foods, and vary considerably in reaction, which ac- counts for its ada])tability to average garden conditions. About the only sort of treatment to avoid is i)lanting It in the oi)en sun in clay so dense as to impede the circulation of the air, which, like all orchids, it needs around its roots. White Ladyslipper, Cypripe dium candidum Muhl. Chancing to survive glaciation in Kentucky, this orchid followed the retreating ice north to Minnesota and the other states bordering on the Great I^kes, also ])ushing eastward as far as the highlands of New Jersey. In nature its roots usually s])read over alkaline muck, i)lenti fully sui)i)lied with cool, lime-rich water, under a fairly thick layer of litter. If these conditions are matched, there seems no reason why it can not be grown in cultivalcion, although it is more delicate and easily injured by unfavor- able environment and by ])arasitic fungi than the related Golden-slii)per Orchid. An excellent cut of Cypripedium pannfloruni var. piibescens appears in The American Orchid Society Bulletin, Vol. 1, No. 1, June, 1932. -Ed. % I Reprinted from September, 1933, American Orchid Society Bulletin Native Orchids Our Native Orchids and Their Cultivation 3. The Pink Slipper Orchids* Edgar T. Wherry ^4 I Queen Slipper-orchid (Cypripedium reginae Walt.) Though the largest and seemingly most vigorous of all our native or- chids, this species is not easy to cul- tivate. The difficulty is apparently not a matter of temperature, for un- til destroyed by man it grew at fairly low elevations in Virginia and the Carolinas — ^liaving survived the Gla- cial Period there, — where the soil be- comes decidedly warm in the sum- mer. After the last ice sheet re- treated, it migrated far north into Canada, so it can evidently with- stand severe cold in winter. Soil acidity is also not concerned, for contrary to popular belief this is not at all an acid-loving plant. It may grow, to be sure, in swamps where sphagnum and other acid mosses cover the surface, but its roots then lie deep down, where the reaction is neutral or essentially so. It needs a more constant supply of moisture than the average garden affords, but even in especially moist situations it will often thrive for a year or two and then suddenly van- ish. Susceptibility to attacks of para- sitic fungi is the only explanation which seems to account for this be- havior, and no way to prevent such attacks is known. If it must be cul- tivated, then the only thing to do is to renew one's stock at frequent in- tervals. Clumps can be purchased from many dealers in native plants, although as they are being collected by thousands from the wild, its col- *rontribution from the Botanical Laboratory an Native Orchids Our Eastern Orchids and Their Cultivation 4. The Spurred Orchids (Orchis and Habenaria)* Edgar T. Wherry 4V 41 »^ I 4- The orchids of this group are ([uite as difficult to transplant as the most intractahle of the Cypripediums, already discussed in preceding- ar- ticles.' Dealers sell them, to be sure, in large numbers, but in most cases the result is simply wild tlow- er destruction ; usually they fail to come up at all the following- year, or if they do manage to pull through, they die inunediately after blooming^. The root-systems of these plants consist of radiating- clusters of elong^ated tuberoids, one of which is a rootstock, bearing: at some point along^ its upper surface the bud which is to g^row into the next year's ])lant. The sim])le tul)er- oids will wither away at the close of each season in any case, so in- jury to them does no particular harm. Injury to the bud-bearing- one, however, may ])ermit the in- vasion of parasites, and result in the death of that individual. The aver- ag^e commercial collector yanks the leafy stalks out of the toug^h mass of peaty material through which the orchid's root-system spreads, as likelv as not breaking off the bud in the process. He then hacks off the ends of both kinds of tuberoids indiscriminately, so it is small won- der that the mortality among^ pur- chased stock is hig^h. If these orchids are to be trans- ])lanted with any deg^ree of success, thev should be collected by the ex- ])erimenter personally. The approxi- mate root-spread of the particular species should be ascertained in ad- vance, and the dig^g^ingf implement Tdtitribulion from the Botanical Laboratory and Morris Arboretum of the l:niversity ot Pennsylvania. //'//(/ I'lo'Zi'cr Prcscr7'atio)i SiH-irty O nil is sp ccta h His Pierce Mill WckxIs, D. C. inserted far enoutrh out from the crown to avoid cutting even the longest tuberoids. The ball of earth thus obtained may be transported as it stands, but, as mentioned in the next preceding- article, there is considerable question whether this is a desirable procedure. Mere dis- turbance of a soil mass often results in a serious upset of the 1iaj^ninn moss, wliich is likely to be free from parasitic ore worth while. Soil temperature is not an im- portant factor in the case of this species, for I have seen it both in Georgia and in Arkansas, where even in the woods the soils heat up considerably in midsummer. After the retreat of the last ice sheet it was able to migrate into the glaci- ated territory as far as Minnesota. Ontario, and southern Quebec, so it is evidently also resistant to win- ter's cold. It always grows where there is shade, a moderate though not excessive supply of moisture, and a chance for air to circulate around its roots. The