TRANSACTION'S Of 'J,;

THE MISSOURI ACAD.EMVi' ;; ', ,'' ,,, ,, ,'' ' OF SCIENCE Transactions of The Missouri Academy of Science (Founded in 1934) Officers 1981-82 President ...... E. Allen McGinnes, Jr., University of Missouri-Columbia President-Elect ...... Edward M. Emery, Monsanto Industrial Chemical Company Vice-President . .... Albert R. Gordon, Southwest Missouri State University Past President ...... Dean A. Rosebery, Northeast Missouri State University Secretary ...... Nathan H. Cook, Lincoln University Treasurer ... ., ...... Richard McHugh, University of Missouri-Columbia Historian ...... Clayton H. Johnson, University of Missouri-Columbia Director, Collegiate Division ...... Roland A. Hultsch, University of Missouri-Columbia Director, Junior Division ...... Adell Thompson, University of Missouri-Kansas City AAAS Representative Dean A. Rosebery, Northeast Missouri State University, Kirksville, MO 63501 Editorial Staff Editor. John R. Jones, School of Forestry, Fisheries and Wildlife, University of Missoun­ Columbia, Columbia, MO 65211 Assistant to the Editor: Sandy Clark, School of Forestry, Fisheries and Wildlife, 112 Stephens Hall, University of Missouri, Columbia, MO 65211 Associate Editors: Biology: Jack R Wallin, University of Missouri, Columbia., MO 65211 Chemistry: John E. Bauman, University of Missouri, Columbia, MO 65211 Engineering: Gary Muller, University of Missouri, Rolla, MO 65401 Social Science: Rex Campbell, University of Missouri, Columbia, MO 65211

SEND ALL MANUSCRIPTS TO Dr. John R Jones, Editor, Transactions of the Missouri Academy of Science, School of Forestry, Fisheries and Wildlife, University of Missouri­ Columbia, 112 Stephens Hall, Columbia, MO 65211 Publications of the Missouri Academy of Science Transactions of the Missouri Academy of Science, Volumes 1-6, 9, 12, 13, 14, 15 & 16 6.00 Transactions of the Missouri Academy of Science, Double Volumes, 7 & 8, 10 & 11. .1000 Occasional Papers: 1. Symposium on Pests & Pesticides ...... 2.50 2. Enzymic Characterization of Brown Recluse Spider Venom ...... 2.50 3. Timber-Wildlife Management Symposium...... 3.50 4. Callaway Plant Units I and II, Preconstruction Environmental Monitoring ...... 3.50 5. Second Woody Ornamentals Disease Workshop...... 2.50

This joumal is abstracted by the Bioscience Information Service and Chemical Abstracts Services.

Cover art courtesy of David Hamilton Layout by Lynne/ Pappas !SSN 0544-540X Transactions of The Missouri Academy of Science

Table of Contents

Sponsors of the Missouri Academy of Science ...... 3

Biological Sciences Herpetofauna of the Ashland Wildlife Area, Boone County, Missouri M. E. Clawson and T. S. Baskett ...... 5 Food Habits of Microtus ochrogaster and Peromyscus maniculatus in Sympatry J. C. Cook, M. S. Topping and T. A. Stombaugh ...... 17 Study of the Secondary Xylem in Herbaceous Dicotyledons I. Boraginaceae, Cruciferae, and Euphorbiaceae C. U. Park ...... 25 Conversion of Forestland in the Missouri Ozarks S. T. Othic, W. B. Kurtz and D. C. Baumgartner ...... 37 Status of the River Otter (Lutra canadensis) in Missouri J. F. Choromanski and E. K. Fritzell ...... 43 Hemisinus Jacksonius (Gastropoda) as an Aid in Determining Potential of Marine Deposits for Yielding Remains of Terrestrial Mammals J. W. Westgate ...... 49 Effects of Barite Tailings on Benthos and Turbidity of Two Ozark Streams R. M. Duchrow ...... 55 Seasonally Flooded Missouri Wetlands: Early Winter Limnological Characteristics in Relation to Physiographic Regions M. V. Hoyer and F. A. Reid ...... 67 Distribution and Habitat Characteristics of the Painted Bunting in Missouri D. J. Norris and W. H. Elder ...... 77 Fire History of an Ozark Glade in Missouri R. Guyette and E. A. McGinnes, Jr...... 85 Problems in "Keeping-Up" With the Flora of Missouri D. B. Dunn ...... 95 Maximum Sizes of Amphibians and Reptiles From Missouri R. Powell, K. P. Bromeier, N. A. Laposha, J. S. Parmerlee and B. Miller ...... 99 Proestrus and Estrus of Missouri Coyotes D. H. Cary, J. D. Rhoades and D. Schmidt ...... 107 Physical Sciences Spin Glasses: A Two-Site Cluster Theory H. A. Brown ...... 111 Compactness ,of Missouri's Counties R. F. Austin and T. Dowell ...... 127

Social Sciences Determining Forest Management Dispositions of Missouri Eastern Ozark Nonindustrial Private Forest Landowners J. R. Fairweather and W. B. Kurtz ...... 131 Effects of a Regular Exercise Program on Geriatric Residents in a Nursing Home M. Masters and F. M. Haemmerlie ...... 137 Caregivers' Attitudes Toward the Aged and Levels of Authoritarianism C. B. Irvin, L. Isbell and M. H. Typpo ...... 143

Abstracts Annual Meeting of the Missouri Academy of Science, School of the Ozarks, Point Lookout, MO, April 23-24, 1982 Senior Division ...... 157 Collegiate Division ...... :...... 171 Index to Authors ...... 177 Sponsors of the Missouri Academy of Science

SUPPORT FOR THE MAS JUNIOR DIVISION has been received from Southwestern Bell Telephone Company and Union Electric Company. Awards for outstanding papers were made available in the fields of Communications and Energy.

Dr. John Paul Morris Award funds provided by Mrs. John Paul Morris have been used to award a travelling plaque each year to an outstanding young high school woman presenting a paper in the physical sciences.

SUSTAINING AND PATRON MEMBERS have contributed funds to support the activities of the Missouri Academy of Science. We wish to express our appreciation to members of the industrial community who have demonstrated their concern for the future of science by supporting the Academy.

PATRON MEMBERS SUSTAINING MEMBERS Monsanto Industrial Chemical Company ASARCO, Inc. St. Joe Lead Company MFA Oil Company Southwestern Bell Telephone Company Union Electric Company

INSTITUTIONAL MEMBERS have contributed funds to support the activities of the Academy on the basis of 1%¢ per FTE student, with a minimum of $15.00 for colleges with less than 1,000 students. We wish to express our appreciation to the following institutions who were Institutional Members during 1980-81:

Central Methodist College - Fayette Moberly Junior College - Moberly Central Missouri State University - Warrensburg Northeast Missouri State University - Kirksville Cleveland Chiropractic College - Kansas City Northwest Missouri State University - Maryville Columbia College - Columbia Park College - Parkville Conception Abbey - Conception RockhuFst College - Kansas City Crowder College - Neosho St. Louis College of Pharmacy - St. Louis Culver-Stockton College - Canton St. Mary's College- O'Fallon Drury College - Springfield Southeast Missouri State University- Cape Girardeau East Central Junior College - Union Southwest Baptist University - Bolivar Fontbonne College - St. Louis Southwest Missouri State University - Springfield Hannibal-La Grange College - Hannibal State Fair Community Co!Iege - Sedalia Kirksville College of Osteopathic Medicine -Kirksville Stephens College - Columbia Lincoln University - Jefferson City The School of the Ozarks - Point Lookout Lindenwood College - St. Charles University of Missouri - Columbia Maryville College - St. Louis University of Missouri - Kansas City Mineral Area College - Flat River University of Missouri - Rolla Missouri Baptist College - St. Louis University of Missouri - St. Louis Missouri Southern State College - Joplin William Jewell College - Liberty Missouri Western State College - St. Joseph

1982 Clawson & Baskett, Biological Sciences 5

Herpetofauna of the Ashland Wildlife Area, Boone County, Missouri1

Mary E. Clawson, 2 School of Forestry, Fisheries and Wildlife, University of Missouri Columbia, Missouri 65211 Thomas S. Baskett U.S. Fish and Wildlife Service Missouri Cooperative Wildlife Research Unit University of Missouri Columbia, Missouri 65211

Abstract: Trapping was undertaken on the Ashland Wildlife Research Area (AWRA) in Boone County, Missourt, to determine the species diversity, relative abundance and relative biomass of reptiles and amphibians on old field and upland forest sites. The purpose was to augment and update results of a baseline study of the AWRA completed in 1937. In total, 2545 individuals were captured (13 species of amphibians and 22 species of reptiles). The median numbers of species captured per site in old fields (M=24) differed significantly from those trapped in upland forest sites (M=20.5) (PO.lO) according to habitat type. Species diversity indices for herpetozoans (Shannon-Wiener function) were similar in old field (H' =2.92) and upland forest (H' =2.82) sites. The median values for estimated biomass of herpetozoans per site were significantly greater (P

The objective of this paper is to report the species diversity, relative abundance and relative biomass of reptiles and aTI)phibians on old field and upland forest sites of the AWRA, University of Missouri-Columbia. Its purpose is to augment and update a part of a baseline study of the AWRA conducted by Henning (1937), in which he descrtbed the reptiles and amphibians occurring on the area. The information presented is part of a larger study relating herpetofaunal abundance to habitat characteristics, seeking easily measured variables to incorporate into habitat models. Such models, in turn, can be used with Habitat Evaluation Procedures (HEP) (U. S. Fish and Wildlife Service 1980). This research was supported by Research Agreement# USDI14-16-0008-2014 be­ tween the U.S. Fish and Wildlife Service and the University of Missouri. We acknowledge the assistance of H. W. Campbell, H. Kochman, J. A. Ellis and M. J. Armbruster in planning and early implementation of the study design. Appreciation is also extended to many other

1Contrtbution from the Missourt Cooperative Wildlife Research Unit (U.S. Fish and Wildlife Service; Missouri Department of Conservation; Wildlife Management Institute; and School of Forestry, Fishertes and Wildlife, University of Missourt, cooperating), also from Missourt Agrtcultural Expertment Station Projects 182 and 184, Journal Sertes No. 9140. 2 Present address: Route 3, Box 52, Missourt Valley, lA 51555. 6 Transactions, Missourt Academy of Science Vol. 16 persons associated with the University of Missouri, Missouri Department of Conservation and the Habitat Evaluation Group of the U.S. Fish and Wildlife Service, who assisted in field work and gave statistical and editorial advice.

Study Area Twelve study sites were selected: 6 old field and 6 upland forest sites, ranging from about 1 to 2.5 ha. Each was selected to fit into a successional continuum, from recently disturbed (mowed) old field to mature sawtimber. Eleven of the sites were on the AWRA. One site, chosen for its size and successional stage, was immediately adjacent to the Ashland Area (Fig. 1). The AWRA is about 8 km east of Ashland, Missouri, in southeastern Boone County. Mean canopy closure in old field habitat varied from 0 to 44% among the sites. Predominant woody species were persimmon (Diospyros virginiana), Rubus spp., eastern red cedar (Juniperus virginiana) and plums (Prunus spp. ). Predominant herbaceous species were blue grasses (Poa spp. ), panic grasses (Panicum spp. ), poverty grass (Danthonia spicata), sedges (Carex spp. ), sericea lespedeza (Lespedeza cuneata) and goldenrods (Solidago spp.). Native prairie grasses, including broomsedge (Andropogon virginicus), big bluestem (Andropogon gerardi) and Indian grass (Sorghastrum nutans), were found on some sites; however, they were not the most frequently encountered species. White oak (Quercus alba) was the predominant overstory species on 4 of the upland forest sites. Overstory canopy closure ranged from 81 to 94%. Understory canopy closure ranged from 40 to 60% and was primarily sugar maple (Acer saccharum), white oak regeneration, and flowering dogwood (Comus florida). Three of the sites were sawtimber stands and 1 was a pole stand. The most frequently encountered herbaceous species were sedges, tick trefoils (Desmodium spp. ), Virginia creeper (Parthenocissus quinquefolia), fragrant sumac (Rhus aromatica) and panic grasses. Of the remaining 2 upland forest sites, 1 was intermediate between old field and upland forest and the other was an elm (Ulmus spp.) and black locust (Robinia pseudo-acacia) pole stand. These sites had mean overstory canopy closures of 54.5 and 80%, respectively. Ground layer species on the intermediate site included buck brush (Symphoricarpos orbiculatus), sedges, blue grasses, eastern red cedar and Queen Anne's lace (Daucus carota). The pole stand had elms, Rubus spp., multiflora rose (Rosa multiflora), sedges, thoroughworts (Eupatorium spp.) and jewelweeds (Impatiens spp.) in its ground layer.

Methods To determine the relative herpetological abundance on the study sites, we used a removal method modified from the trapping array system described by H. W. Campbell and S. P. Christman (1977, Techniques for herpetofaunal community analysis, unpub­ lished paper presented at Joint Ann. Meet. Herpetol. League, Soc. Study Amphibs. and Reps., Lawrence, Kansas). Each array consisted of 4 drift fences: 7.6 m lengths of 50 em-high aluminum flashing, arranged in a cross-shaped pattern with a central separation of 15m (Fig. 2). A single-ended funnel trap was placed on the ground at each end of the drift fence. Two double-ended funnel traps were placed on either side at the center of each drift fence, flush to the ground and pressed against the fence. Dirt or ground litter was brushed into the mouth of each funnel to create a more natural entrance. Masonite boards were placed over the funnel traps in the old field sites to reduce heat stress on trapped animals during the summer months. 1982 Clawson & Baskett, Biological Sciences 7

1

TIMBER OlD fiElD-SHRUBS AND TREES OlD FIIEI.D-HIERBACEOUS CUlTIVATED ;_ C• ._ ...:, ROADS

Fi.gure 1. Ashland Wildlife Research Area, Boone County, Missouri, showing approximate locations and sizes of old field coiiecting sites nos. 1-6, and upland forest sites nos. 7-12. 8 Transactions, Missouri Academy of Science Vol. 16

-Single-Ended Funnel Trap 1 ~ -15m- ) l-7.6m-l

Figure 2. Schematic of a typical herpetological trapping array.

The trapping array provided definite advantages over other methods of trapping. Placement of the array was not dependent on natural obstructions, such as logs or rock outcroppings; the arrays provided data that were comparable among the study sites. Also, the array made trapping in old fields much easier because natural obstructions and debris were scarce on these sites. Thirteen arrays were laid out, 1 on each site, except for a site with especially complex vegetative structure that had 2 arrays. Funnel traps were put into place 17 and 18 May 1978. They were checked once per week, and all animals found in the funnel traps were removed, identified using Vertebrates of the United States (Blair et al. 1968) and recorded. The traps were removed on 9 November 1978 because the animals were inactive during cold weather. They were replaced 22 March 1979 and trapping continued until31 May 1979. Thus, a full season of trapping-summer, fall and spring-was completed. Initially, all captured animals were preserved for confirmation of field identification. Many of these animals were donated to the Natural History Museum at the University of Kansas, Lawrence. During the remainder of the trapping effort, animals were identified, measured and weighed, then released at least 6 km from the Ashland Area to prevent their return to the capture sites. Herpetofaunal catches on each site were tallied according to species (total number of species caught and number of each individual species caught for each site). The capture data also were divided according to season. A length-weight regression was determined for each species from the snout-vent length of each specimen and the weights of animals in good condition. These regressions were used to compute an estimate of biomass of the herpetofauna for each site. Diversity of herpetozoans trapped in the old field and upland forest sites was calculated by the Shannon-Wienerforrnula (Shannon and Weaver 1949): H' = - Ip1 loge PI, where p1 is estimated by dividing the number of individuals of species i in the sample by the total number of individuals of all species. 1982 Clawson & Baskett, Biological Sciences 9

Results and Discussion Traps on the old field and upland forest sites caught 2545 individuals, comprising 13 species of amphibians and 22 species of reptiles. Both numbers of individuals and estimates of total biomass in old field and upland forest sites were greatest in summer, intermediate in spring and lowest in fall (Table 1). The total number of species captured per site in old fields ranged from 20 to 25 (M = 24), whereas the number of species captured per site in upland forests ranged from 15 to 22 (M=20.5). These medians differed significantly by habitat type (P<0.05; Mann-Whitney U test). However, the total number of species captured in all old field sites was similar to that for all upland forest sites (Table 1). Species diversity, as reflected by the Shannon-Wiener index, was similar for amphibians in old field (H' =2.05) and upland forest (H' = 1.94) sites. Diversity of reptiles was also similar in both habitat types (H' = 2.44, old field; 2.40, upland forest).

Table 1. Summary of herpetofaunal trapping results: relative abundance and biomass estimate (g) of all animals caught, by habitat type and season.

Old Field Upland Forest

Species Captured 34 32 Individuals Captured Summer 667 585 Fall 322 208 Spring 474 289 TOTAL 1,463 1,082 Median No. per Site 160 173.5 Estimate of Total Biomass (g) Summer 12,068 5,166 Fall 4,867 1,492 Spring 8,629 3,133 --- TOTAL 25,564 9,791 Median Value per Site 3,304 1,121

Median numbers of individuals per site were not significantly different (P>0.10) between old field and upland forest sites, although total numbers caught in old fields were somewhat greater than those caught in the forest habitats (Table 1). Median values per site for estimated biomass of herpetozoans caught in old field sites were significantly greater (P<0.05) than median values for animals caught at forest sites, during all seasons. Estimates of biomass for combined old field sites were more than twice those for combined upland forest sites (Table 1). Individual old field site biomass estimates ranged from 2618 to 4686 g, whereas all but 1 upland forest site had total biomass estimates ranging from 828 to 1682 g. The remaining upland forest site had a total biomass 12 Transactions, Missouri Academy of Science Vol. 16

Table 2. (continued)

(number of individuals) Species Old Field Upland Forest

Scincel/a latera/is 109 86 (ground skink) Storeria d,ekayi wrightorum 14 8 (midland brown snake) Storeria o. occipitomaculata 21 51 (northern redbelly snake) Terrapene carolina triunguis 2 1 (three-toed box turtle) Thamnophis p. proximus 42 11 (western ribbon snake) Thamnophis sirtalis 13 4 (red-sided or eastern garter snake) Thamnophis spp. 2 0 (ribbon or garter snake) Unknown snake* 31 15 Virginia ualeriae elegans 105 86 (western earth snake) Unknown spp. * 0 1 TOTAL 1,463 1,082

*Specimens were partially decomposed and unrecognizable.

More species of reptiles showed a distinct pattern of utilizing either old field or upland forest habitat than did the amphibians. The larger snakes, such as the eastern yellowbelly racer, eastern hognose snake, prairie kingsnake, speckled kingsnake, western ribbon snake, red-sided or eastern garter snake and Osage copperhead were encountered more frequently in old field habitat. The prairie ringneck and the midland brown snakes, both small, cryptic snakes, also were found more frequently in old field habitat. The five-lined and the broadhead skinks were much more numerous in upland forest habitat. The northern redbelly snake was trapped more frequently in upland forest habitat than in the old fields. In the way that the herpetological arrays were used, we cannot estimate the area represented. If we had extended each array, lengthening the cross configuration, we might have been able to determine the "area of effect" ofthe trapping array. However, the scope of this study and time constraints did not allow extensions on the 13 arrays. The capture results therefore are strictly relative, permitting only comparisons among sites. Soon after the AWRA was established, Henning (1937) made a zoological reconnais­ sance of the vertebrates occurring on the area. His method of capture of herpetozoans was to traverse the area over a 11/z year period, searching for specimens. He also accepted specimens and accounts of species occurring on the area from residents and workers. 1982 Clawson & Baskett, Biological Sciences 13

Henning listed 30 herpetofaunal species (9 amphibians and 21 reptiles) occurring on the area (Table 3). One of these species, the leopard frog (Rana pipiens), has since been reclassified in this region into 2 species, the southern leopard frog (Rana sphenocephala) and the plains leopard frog (R. blairi; Mecham eta!. 1973). We encountered both the

Table 3. Herpetozoans recorded at Ashland Area by Henning (1937) and current synonyms (Collins et al. 1978).

Henning's Species List Current Synonyms

*Triturus v. viridescens Notophthalmus viridescens louisianensis (crimson-spotted newt) (central newt) P/ethodon g/utinosus Plethodon g. g/utinosus (slimy salamander) (slimy salamander) *Bujo americanus Bujo a. americanus (common toad) (eastern American toad) Bujojowleri Bujo woodhousei jowleri (Fowler's toad) (Fowler's toad) *Acris gryl/us Acris crepitans blanchardi (cricket frog) (Blanchard's cricket frog) *Pseudacris triseriata Pseudacris t. triseriata (swamp cricket frog) (western chorus frog) *Rana pipiens Rana blairi & Rana sphenocephala (leopard frog) (plains & southern leopard frogs) *Rana pa/ustris Rana palustris (pickerel frog) (pickerel frog) *Rana catesbeiana Rana catesbeiana (bullfrog) (bullfrog) *Sce/oporus undu/atus Sce/oporus undulatus hyacinthinus (fence swift) (northern fence lizard) *Eumeces jasciatus Eumeces jasciatus (blue-tailed skink) (five-lined skink) *Eumeces laticeps Eumeces /aticeps (red-headed lizard) (broadhead skink) *Leio/opisma laterale Scince//a latera/is (brown-backed skink) (ground skink) *Carphophis amoenus vermis Carphophis amoenus vermis (worm snake) (western worm snake) *Diadophis amyi Diadophis punctatus amyi (ring-necked snake) (prairie ringneck snake) *Heterodon contortrix Heterodon p/atyrhinos (hog-nosed snake) (eastern hognose snake) *Opheodrys aestivus Opheodrys aestivus (rough green snake) (rough green snake) 14 Transactions, Missouri Academy of Science Vol. 16

Table 3. (continued)

Henning's Species List Current Synonyms

*Coluber c. constrictor Coluber constrictor j/aviventris (blue racer) (eastern yellowbelly racer) *Elaphe o. obsoleta Elaphe o. obsoleta (pilot blacksnake) (black rat snake) *Lampropeltis getulus holbrooki Lampropeltis getulus holbrooki (Holbrook's king snake) (speckled kingsnake) *Lampropeltis cal/igaster Lampropeltis c. calligaster {Evans' king snake) (prairie kingsnake) *Natrix s. sipedon Nerodia s. sipedon (common water snake) (northern water snake) *Thamnophis sirtalis Thanmophis sirtalis (garter snake) (red-sided or eastern garter snake) *Thamnophis' proximus Thamnophis p. proximus (western ribbon snake) (western ribbon snake) Agkistrodon mokasen Agkistrodon contortrix phaeogaster (copperhead) (Osage copperhead) Crotalus horridus Crotalus horridus (common rattlesnake) (timber rattlesnake)

Chelydra serpentina Chelydra s. serpentina (common snapping turtle) (common snapping turtle) Terrapene c. carolina (northern box turtle) *Terrapene carolina triunguis Terrapene carolina triunguis (southern box turtle) (thee-toed box turtle) Terrapene ornata Terrapene o. ornata {painted box turtle) (ornate box turtle) Amyda spinifer TrionyX'spiniferus hartwegi (soft-shelled turtle) (western spiny softshell)

*Species also caught in present study.

southern and pla.ins leopard frogs, plus 23 of the remaining 29 species that Henning identified. We identified 10 additional species of herpetozoans that Henning did not list (Table 4). Differences in capture methodology probably accounted for the additional species captured in our study. The funnel trap array system presents far more opportunity to capture herpetozoans in a less biased manner than traversing the area on foot. Species captured in our study but not recorded in Henning's study may well have been present at the time of his reconnaissance, but simply not encountered. 1982 Clawson & Baskett, Biological Sciences 15

Table 4. Species of herpetozoans identified in the present study not recorded by Henning (1937).

Scientific Name Common Name

Ambystoma maculatum spotted salamander Hyla c. crucifer northern spring peeper Hyla versicolor!chrysoscelis gray treefrog Rana areolata circulosa northern crawfish frog Rana clamitans melanota green frog Lampropeltis triangulum syspi/a red milk snake Storeria dekayi wrightorum midland brown snake Storeria o. occipitomaculata northern redbelly snake Virginia valeriae elegans western earth snake Chrysemys picta belli western painted turtle

Henning listed 7 species of herpetozoans not captured in our study (Table 3). One of these, the timber rattlesnake, was seen on 1 of the upland forest sites after trapping had ceased. Four of the species were turtles, not vulnerable to trapping in our arrays. The funnel traps used had too small an opening for any but the smallest turtles. Also, being placed on land, they excluded primarily aquatic turtles, such as the western spiny softshell and the common snapping turtle, which Henning recorded. The AWRA has been used as a research facility by faculty and students of the University of Missouri and their associates since its establishment. In-depth studies of herpetozoans are included. Their resulting publications provide detailed information on some of the species included in the present paper. (See Korschgen and Moyle 1955, Willis eta!. 1956, Korschgen and Baskett 1963, Schroeder and Baskett 1968, Schroeder 1975 and Gerhardt 1978). This present study provides baseline information on the species and their relative abundance found on old field and upland forest habitat on the AWRA. It can be used to augment information from previous studies, or to provide a basis for future studies of selected species of reptiles and amphibians on the AWRA or central Missouri.

Literature Cited Blair, W. F., A. P. Blair, P. Brodkorb, F. R. Cagle, and G. A. Moore. 1968. Vertebrates of the United States, 2nd ed., McGraw-Hill Book Co., Inc., NY. 616pp. Collins, J. T., J. E. Huhley, J. L. Knight, and H. M. Smith. 1978. Standard common and current scientific names for North American amphibians and reptiles. Soc. Study Arnphibs. and Reps., Misc. Pub!., Herpetological Circular no. 7. 36pp. Gerhardt, H. C. 1978. Temperature coupling in the vocal communication system of the gray tree frog, Hyla versicolor. Science 199:992-994. Henning, W. L. 1937. Zoological reconnaissance of the Ashland Area with special reference to the vertebrates. MA Thesis, Univ. Missouri, Columbia. 153pp. 16 Transactions, Missouri Academy of Science Vol. 16

Korschgen, L. J. and T. S. Baskett. 1963. Foods of impoundment- and stream-dwelling bullfrogs in Missouri. Herpetologica 19:89-99. _____ and D. L. Moyle. 1955. Food habits ofthe bullfrog in central Missouri farm ponds. Am. Midi. Nat. 54:332-341. Mecham, J. S., M. J. Littlejohn, R. S. Oldham, L. E. Brown, and J. R. Brown. 1973. A new species of leopard frog (Rana pipiens complex) from the plains of the central United States. Occas. Pap. Mus. Texas Tech. Univ. 18:1-11. Schroeder, E. E. 1975. The reproductive cycle in the male bullfrog, Rana catesbeiana, in Missouri. Trans. Kansas Acad. Sci. 77:31-35. _____ and T. S. Baskett. 1968. Age estimation, growth rates, and population structure in Missouri bullfrogs. Copeia 1968:583-592. Shannon, C. E. and W. Weaver. 1949. The mathematical theory of communication. Univ. Illinois Press, Urbana. 117pp. Willis, Y. L., D. L. Moyle, and T. S. Baskett. 1956. Emergence, breeding, hibernation, movements and transformation of the bullfrog (Rana catesbeiana) in Missouri. Copeia 1956:30-41. U.S. Fish and Wildlife Service. 1980. Habitat evaluation procedures. Ecol. Services Publ., Washington, DC. 30pp. 1982 Cook, Topping & Stombaugh, Biological Sciences 17

food Habits of Microtus ochrogaster and Peromyscus maniculatus in Sympatry

John C. Cook\ Milton S. Topping and Tom A Stombaugh Biology Department Southwest Missouri State University Springfield, Missouri 65804

Abstract: The food habits of Microtus ochrogaster and Peromyscus manicu/atus co-occurring in old field succession were studied. M. ochrogaster was primarily herbivorous, while P. maniculatus ate both plant and animal material. The feeding niches of both species were relatively broad (B = 0. 76 forM. ochrogaster and B = 0. 70 for P. maniculatus), although overlap in food habits was always less than 0.6 and averaged 0.38. Key Words: Microtus ochrogaster, Peromyscus maniculatus, food habits, feeding niche, food overlap.

Introduction Co-occurrence of Microtus ochrogaster and Peromyscus maniculatus in areas of old field succession within their geographic ranges (Hall and Kelson, 1959) may imply differential adaptation to avoid competition. M. ochrogaster is known to be an herbivore (Fleharty and Olson, 1969; Fisher, 1945; Zimmerman, 1965), while P. maniculatus is considered to be a granivore/omnivore (Flake, 1973; Johnson, 1961; Vaughan, 1974; Whitaker, 1966; Williams, 1959). However, the food habits of these species have not been studied in situations of local sympatry. Feeding of these species was studied in populations occurring sympatrically in a relatively uniform stand of old field succession which was mowed inadvertently during the study.

Methods Specimens were live- and snap-trapped from an 8 ha portion ofthe Southwest Missouri State University Agricultural Station in southwest Springfield from April 1975 until April 1976. Live trapping was done with 25 Longworth live traps set 6.1 m apart in a square grid on 3 of the 8 ha to obtain fecal samples from live animals for evaluation of food habits. Fecal samples from 17 animals were collected, weighed ( ±0.01 mg), and preserved in 70% alcohol. During the same period, 60 M. ochrogaster and 63 P. maniculatus were collected using snap traps. Stomachs of each of these animals were removed and their contents weighed (± 0.01 mg) and preserved in 10% formalin in individually labeled vials. Presumed forage species of vascular plants were identified and their abundances recorded using Oosting's scale of abundance (Oosting, 1956). Stomach samples were analyzed using the methods of Dusi (1952) and Williams (1962). Samples were stained with eosin dye and spread on microscope slides from which 10 randomly selected fields were examined for type and amount of material at 100 X magnification. Fecal samples were analyzed qualitatively. Types of material encountered

1 Present address: 1400 E. McDaniel, Springfield, Mo. 65802 18 Transactions, Missouri Academy of Science Vol. 16 included vascular plant material, arthropod parts, hair, and miscellaneous material. Materials were identified by use of reference photographs of plant epidermal cells and arthropod exoskeletons collected from the study area. Bias may be introduced by these methods (Westoby et al., 1976), but we assume that the errors are relatively consistent. Differences in feeding between the species were evaluated using a three-way fixed effects model of AN OVA in which the main effects were species, month, and food item (Sokal and Rohlf, 1969). Food niche breadth was studied using Culver's (1972) measure of niche breadth (B = 0 when only one food type is eaten and B = 1 when all food types are eaten equally) and overlap in feeding habits using Colwell and Futuyma's (1971) measure of niche overlap (0 indicates no overlap and 1 indicates total overlap).

Results Vegetational analysis indicated that tall fescue (Festuca arundinacea), orchard grass (Dactylis g/omerata), timothy grass (Phleum pratense), purple top (Tridens flavus), and green foxtail grass (Setaria viridis) were abundant; the common dandelion (Taraxacum officina/e), henbit (Lamium amp/exicaule), downy chess brome grass (Bromus tectorum), and the nodding foxtail (Setaria faberi) were common; and shepherd's purse (Capsel/a bursa-pastoris) and running buffalo clover (Trifolium stoloniferum) were rare in the study area. As well, collection of marked M. ochrogaster and P. maniculatus at 21 of the 25 grid stations confirmed local sympatry. Analysis of gut contents indicated that M. ochrogaster tended to utilize more parts of each of the forage items than did P. maniculatus (Table 1). Seasonal variations in mean number of all food particles were similar for both species, although M. ochrogaster always contained more particles. A highly significant interaction (P

Table 1. food parts utilized.

Food Type M. ochrogaster P. maniculatus

Festuca arundinacea leaf, stem, seeds leaf, seeds Dacty/is g/omerata leaf, seeds Phleum pratense leaf leaf, seeds Tridens flavus leaf, stem leaf Setaria viridis leaf, seeds leaf, seeds Taraxacum officinale leaf leaf Lamium amp/exicaule leaf leaf Bromus tectorum leaf, stem, seeds leaf, seeds Setaria faberi leaf, stem leaf Capsella bursa-pastoris leaf leaf Trifolium stoloniferum leaf leaf Arthropod whole whole Other animal material whole whole Table 2. Mean number of food items (and SE) contained in stomach of M. ochrogaster (M) and P. maniculatus (P)...... 1.0 00 l\:) Species JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT AVG

Sample size M 5 5 6 4 4 11 15 13 5 5 p 7 3 6 6 2 8 11 10 5 3

Weight of stomach M 31.60 30.40 38.55 32.70 28.51 30.62 28.11 36.52 36.50 44.90 33.84 () p 0 contents (mg) 20:90 19.77 20.73 17.20 15.42 20.43 15.60 10.80 13.11 16.52 18.56 0 :F -l 0 FOOD ITEM "0 'g, ~ lO Festuca arundinacea M 30.8 27.0 25.0 16.5 20.0 25.2 25.1 24.6 10.8 10.3 21.5 R" (3.81) (2.51) (3.00) (4.55) (5.29) (1.10) (1.98) (1.31) (2.03) (2.90) (2.85) Ul 0 p 0.0 3.0 0.2 6.2 2.0 3 1.8 5.5 1.6 0.0 0.7 0.0 cr" (0.75) (4.27) (0.80) (0.00) (0.00) {1.22) {0.72) {0.00) (0.25) {1.25) {0.93) "'c ~ Dactylis g/omerata M 7.5 0.8 2.0 10.0 8.0 1.0 1.7 2.5 0.4 1.7 3.6 Q;l {1.63) Q. (2.72) {0.47) {1.00) (3.55) (4.61) {0.40) {1.05) {1.22) {0.40) {0.88) 0

ce.(") p 4.7 4.2 3.6 1.5 1.5 5.4 2.2 0.4 1.0 1.0 2.4 e>.. {1.52) {2.65) {0.64) {0.64) (1.50) {1.16) {0.58) {0.18) {0.70) {0.57) {0.95) Ul 0. ro ~ Phleum pratense M 9.5 1.2 4.0 10.2 9.0 2.0 2.6 1.5 1.8 2.3 4.4 (")ro (/) {2.72) {0.62) {0.80) {1.25) {2.51) {1.41) {1.11) (0.50) {0.58) {0.33) {1.11)

p 1.7 2.5 2.1 2.5 7.0 4.4 4.2 0.8 1.0 7.5 3.3 {0.83) {1.40) {1.04) {2.50) {2.00) {1.14) {1.07) (0.49) (0.57) {0.28) (1.14)

Tridens flauus M 3.2 18.8 19.0 14.0 18.3 10.5 13.3 9.4 2.4 1.3 11.0 {2.62) (4.69) {2.00) {0.70) {4.84) (2.06) (1.90) {1.77) (1.07) (0.88) (2.26)

p 3.1 0.0 5.3 2.5 0.0 2.2 1.0 1.5 1.0 1.2 1.7 ...... {2.03) {0.00) (1.26) (1.50) {0.00) {1.33) (1.00) (1.00) {1.00) {0.94) {0.80) 1.0 1\:) Table 2. (continued) 0

Species JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT AVG

Setaria viridis M 11.5 0.8 3.0 8.5 8.7 2.8 7.9 7.9 1.0 2.3 5.4 (4.05) (0.47) (1.00) (2.33) (2.33) (0.75) (1.86) (1.90) (0.77) (1.20) (2.26) p 0.0 0.0 0.2 2.0 0.0 0.4 0.7 0.2 0.0 0.5 0.4 (0.00) (0.00) (0.22) (2.00) (0.00) (0.36) (0.41) (0.17) (0.00) (0.50) (0.17)

::l~ Taraxacum officinale M 1.0 2.0 1.5 6.5 9.3 7.5 4.1 2.9 5.0 2.7 2.2 (J> (") (1.00) (0.70) (0.50) (1.19) (3.71) (2.87) (1.13) (0.76) (2.23) (2.66) (1.68) "'d: 0 ::l p 9.2 10.8 6.8 6.0 11.0 7.2 4.9 3.8 9.3 8.0 7.7 -"' (2.16) (4.25) (2.10) (5.00) (2.00) (1.75) (1.51) (1.89) (4.11) (0.57) (2.54) ~ (J> c0 Lamium amplexicaule M 4.8 2.0 3.5 1.8 6.3 1.8 3.7 3.1 4.6 5.7 3.7 ::l. (2.59) (1.68) '(3.50) (1.10) (3.84) (0.75) (0.99) (0.65) (0.65) (1.45) (1.91) i!;' p 8.9 8.2 4.0 9.0 25.0 5.6 1.4 1.6 1.2 0.0 6.5 "'g. 3 (2.11) (3.88) (0.97) (0.80) (3.00) (1.30) (0.59) (0.74) (1.25) (0.00) (1.39) 'C a. (fl Bromus tectorum M 3.8 5.8 3.5 2.2 2.7 4.0 5.9 4.1 0.2 4.7 3.7 o. (2.42) (4.00) (1.50) (1.43) (2.66) (2.27). (1.76) (1.47) (0.20) (2.90) (2.07) ::l "'(") p 2.1 2.5 2.9 0.0 0.0 1.9 0.0 0.0 2.8 5.5 1.8 "' (0.85) (1.32) (0.99) (0.00) (0.00) (0.67) (0.00) (0.00) (1.23) (1.93) (0.70)

Setaria faberi M 1.2 3.0 5.0 5.0 7.0 4.2 3.0 4.6 1.4 1.0 3.8 (0.75) (2.38) (3.00) (0.70) (1.00) (2.52) (0.61) (1.52) (0.60) (1.00) (1.41) p 3.6 5.8 0.7 2.0 2.0 1.3 0.0 0.0 0.0 0.6 1.6 (0.91) (4.41) (0.28) (0.00) (2.00) (0.84) (0.00) (0.00) (0.00) (0.60) (0.91) < ??- ..... 0\ Table 2. (continued) .... \0 00 -- 1\:) Species JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT AVG

Capse/la bursa-pastoris M 0.0 1.2 0.0 3.8 4.3 0.8 1.9 1.0 0.6 2.0 1.6 (0.00) (0.94) (0.00) (0.75) (2.18) (0.25) (0.70) (0.29) (0.40) (1.15) (0.67) p 5.6 4.5 6.1 4.0 2.5 4.9 1.7 3.1 7.2 3.5 4.6 (") 0 (1.25) (1.84) (1.16) (3.00) (1.50) (1.62) (0.75) (1.34) (2.28) (0.95) (1.57) 0 1' -l Trifolium stoloniferum M 0.0 2.8 4.5 1.0 1.7 1.0 0.8 0.7 1.0 0.0 1.3 0 "0 (4.50) (0.70) (1.20) (1.00) (0.25) (0.49) (0.00) (1.10) (0.00) (2.09) (0.77) ~- p tO 2.6 0.0 2.1 4.0 3.5 2.8 0.7 3.5 0.0 0.0 1.9 I

ce.(") Animal material M 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.6 0.4 0.0 0.1 e!.. (f) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.08) (0.28) (0.40) (0.00) (0.08) ~- :::> p 1.0 0.0 0.4 2.0 3.0 1.8 5.4 4.8 0.0 0.0 1.9 (") (J> (0.40) (0.00) (0.33) (0.87) (3.00) (0.32) (1.31) (1.83) (0.00) (0.00) (0.81) "'

Miscellaneous M 0.2 1.8 5.0 5.0 1.7 1.5 1.4 1.2 2.6 0.3 2.1 (0.25) (1.03) (1.00) (0.91) (1.66) (0.86) (0.49) (0.42) (1.16) (0.33) (0.81) p 3.9 1.0 2.3 0.0 0.0 1.8 0.8 1.6 0.0 0.0 1.1 (0.75) (1.00) (0.37) (0.00) (0.00) (0.51) (0.35) (0.73) (0.00) (0.00) (0.37) ....1\:) 22 Transactions, Missouri Academy of Science Vol. 16

top, and nodding foxtail were the most common food forM. ochrogasterfrom January until July or August, with their decrease presumably due to mowing. L. amplexicaule, T. officinale, and C. bursa-pastoris were the most commonly eaten forage items by P. maniculatus, with the first species most common from January until July and the latter two equally abundant throughout the study. As expected, insects comprised a relatively large share of their diet during summer. Forage items such as orchard grass, timothy, common dandelion, and running buffalo clover were rare in the diet of both species. Comparison of the breadths of the feeding niches indicated that both M. ochrogaster (B = 0. 76) and P. maniculatus (B = 0. 70) possess reasonably broad ranges of food habits. However, the types of food eaten were more variable seasonally for P. maniculatus, which showed the greatest selectivity during January (B = 0.27) and least during June (B = 0.91). Overlap in food habits ofthe species was greatest during March (0.50) and October (0.60) and <0.42 throughout the remainder of the year (Table 3).

Table 3. Seasonal variation in food niche breadth and niche overlap.

Food Niche Breadth Month M. ochrogaster P. manicu/atus Niche Overlap

January 0.669 0.267 0.356 February 0.673 0.784 0.354 March 0.749 0.897 0.502 April 0.860 0.822 0.380 May 0.859 0.666 0.397 June 0.729 0.915 0.419 July 0.772 0.664 0.251 August 0.766 0.621 0.204 September 0.766 0.668 0.330 October 0.788 0.709 0.603

AVERAGE 0.763 0.701 0.380

Discussion Vegetation present in the study area from April until August 1975 was characteristic of old field succession. However, during August, the study area was inadvertently mowed for hay. Although the species remained the same, vegetation was rendered structurally homogenous and physical habitat had little opportunity to account for co-occurrence of the species. In general, food habits of M. ochrogaster were similar to those reported by Fisher (1945), Zimmerman (1965), and Fleharty and Olson (1969), i.e., the species is an herbivore feeding on a reasonably broad spectrum of Gramineae species. Of the five species which were abundant in the study area, only tall fescue, purple top, and nodding foxtail were common foods, thereby suggesting feeding preferences. Decline in their 1982 Cook, Topping & Stombaugh, Biological Sciences 23

importance during late summer and fall was likely due to the field being mowed. Breadth of the feeding niche occupied by M. ochrogaster remained relatively constant throughout the study, as did consumption of the amounts of specific food types (Tables 2 and 3). Analysis of Zimmerman's (1965) data indicated a similar breadth of feeding niche (B = 0.87). The food habits of P. maniculatus also were similar to those reported in the literature (Williams, 1959; Johnson, 1961; Whitaker, 1966; Flake, 1973; Vaughan, 1974) in that they included both plant and arthropod material. However, arthropods and seeds per se comprised a smaller portion of their diet than reported by Vaughan (1974) and Whitaker (1966). Dominance of henbit, common dandelion, and shepherd's purse as forage items also suggest feeding preferences in that the latter two species were common and rare, respectively, in the study area. Hen bit was eaten commonly from January until August with its decline likely being due to mowing, while the common dandelion and shepherd's purse were eaten approximately equally commonly throughout the study. Although the study area was rendered structurally homogenous after mowing during late summer, niche overlap and breadths of feeding niches of the two species showed little difference. Whether qualitative measures of relative abundance of forage items are sufficient to permit contrasts with their relative importance as food to be interpreted as species preferences is a moot point. However, relative differences in the foods eaten by the two species, together with similarity of breadth of feeding niches and feeding niche overlap both before and after mowing the study area, suggests the species are partitioning their food resources.

Acknowledgments Dr. H. Steven Logsdon assisted in identification of forage items.

Literature Cited Colwell, R. K., and D. J. Futuyma. 1971. On the measurement of niche and breadth and overlap. Ecology, 52:567-576. Culver, D. C. 1972. A niche analysis of Colorado ants. Ecology, 53:126-131. Dusi, J. L. 1952. Food habits of several populations of cottontails in Ohio. J. Wild!. Mgt., 13:295-298. Fisher, H. J. 1945. Notes on voles in central Missouri. J. Mamm., 26:435-436. Flake, L. D. 1973. Food habits of four species of rodents on a short grass prairie in Colorado. J. Mamm., 54:636-647. Fleharty, E. D., and L. E. Olson. 1969. Summer food habits of Microtus ochrogaster and Sigmodon hispidus. J. Mamm., 50:475-486. Hall, E. R., and K. R. Kelson. 1959. The mammals of North America. Ronald Press Co., New York, 2: viii + 547-1083 + 79. Johnson, D. R. 1961. The food habits of rodents in range lands of southern Idaho. Ecology, 42:407-410. Oosting, J. H. 1956. The study of plant communities. W. H. Freeman and Co., San Francisco, 440 pp. Sokai, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Francisco, 776 pp. Vaughan, T. A 1974. Resource allocation in some sympatric subalpine rodents. J. Mamm., 55:764-795. Westoby, M., G. R. Rost, and J. A Weis. 1976. Problems with estimating herbivore diets by microscopically identifying plant fragments from stomachs. J. Mamm., 57:167-172. Whitaker, J. 0., Jr., 1966. Food of Mus musculus, Peromyscus maniculatus, Peromyscus /eucopus in Vigo County, Indiana. J. Mamm., 47:473-486. Williams, 0. 1959. Food habits of the deer mouse. J. Mamm., 40:415-419. -----· 1962. A technique for studying microtine food habits. J. Mamm., 43:365-368. Zimmerman, E. G. 1965. A comparison of habitat and food of two species of Microtus. J. Mamm., 46:605-612.

1982 Cook, Topping & Stombaugh, Biological Sciences 23 importance during late summer and fall was likely due to the field being mowed. Breadth of the feeding niche occupied by M. ochrogaster remained relatively constant throughout the study, as did consumption of the amounts of specific food types (Tables 2 and 3). Analysis of Zimmerman's (1965) data indicated a similar breadth of feeding niche (B = 0.87). The food habits of P. maniculatus also were similar to those reported in the literature (Williams, 1959; Johnson, 1961; Whitaker, 1966; Flake, 1973; Vaughan, 1974) in that they included both plant and arthropod material. However, arthropods and seeds per se comprised a smaller portion of their diet than reported by Vaughan (1974) and Whitaker (1966). Dominance of henbit, common dandelion, and shepherd's purse as forage items also suggest feeding preferences in that the latter two species were common and rare, respectively, in the study area. Henbit was eaten commonly from January until August with its decline likely being due to mowing, while the common dandelion and shepherd; s purse were eaten approximately equally commonly throughout the study. Although the study area was rendered structurally homogenous after mowing during late summer, niche overlap and breadths of feeding niches of the two species showed little difference. Whether qualitative measures of relative abundance of forage items are sufficient to permit contrasts with their relative importance as food to be interpreted as species preferences is a moot point. However, relative differences in the foods eaten by the two species, together with similarity of breadth of feeding niches and feeding niche overlap both before and after mowing the study area, suggests the species are partitioning their food resources.

Acknowledgments Dr. H. Steven Logsdon assisted in identification of forage items.

Literature Cited Colwell, R. K., and D. J. Futuyma. 1971. On the measurement of niche and breadth and overlap. Ecology, 52:567-576. Culver, D. C. 1972. A niche analysis of Colorado ants. Ecology, 53:126-131. Dusi, J. L. 1952. Food habits of several populations of cottontails in Ohio. J. Wild!. Mgt., 13:295-298. Fisher, H. J. 1945. Notes on voles in central Missouri. J. Mamm., 26:435-436. Flake, L. D. 1973. Food habits of four species of rodents on a short grass prairie in Colorado. J. Mamm., 54:636-647. Fleharty, E. D., and L. E. Olson. 1969. Summer food habits of Microtus ochrogaster and Sigmodon hispidus. J. Mamm., 50:475-486. Hall, E. R., and K R. Kelson. 1959. The mammals of North America. Ronald Press Co., New York, 2: viii + 547-1083 + 79. Johnson, D. R. 1961. The food habits of rodents in range lands of southern Idaho. Ecology, 42:407-410. Oosting, J. H. 1956. The study of plant communities. W. H. Freeman and Co., San Francisco, 440 pp. Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Francisco, 776 pp. Vaughan, T. A 1974. Resource allocation in some sympatric subalpine rodents. J. Mamm., 55:764-795. Westoby, M., G. R. Rost, and J. A Weis. 1976. Problems with estimating herbivore diets by microscopically identifying plant fragments from stomachs. J. Mamm., 57:167-172. Whitaker, J. 0., Jr., 1966. Food of Mus musculus, Peromyscus maniculatus, Peromyscus /eucopus in Vigo County, Indiana. J. Mamm., 47:473-486. Williams, 0. 1959. Food habits of the deer mouse. J. Mamm., 40:415-419. -----· 1962. A technique for studying microtine food habits. J. Mamm., 43:365-368. Zimmerman, E. G. 1965. A comparison of habitat and food of two species of Microtus. J. Mamm., 46:605-612.

1982 Park, Biological Sciences 25

Study of the Secondary Xylem in Herbaceous Dicotyledons I. Boraginaceae, Cruciferae, and Euphorbiaceae

Chung Uk Park Department of Natural Sciences and Mathematics Lincoln University Jefferson City, Missouri 65101

Abstract: The secondary xylem of three herbaceous families (Boraginaceae, Cruciferae, and Euphorbiaceae) was studied. Boraginaceae and Cruciferae have short vessel elements with small vessel diameters. The majority of vessel elements have oblique end walls with simple perforations. Euphorbiaceae has medium size vessel elements. All three families have short libriform fibers with thin walls and numerous pits. Axial parenchyma is scanty paratracheal or it is absent. Rays are absent in some species of Boraginaceae. Species with rays exhibit both uniseriate and multiseriate rays. Homocellular and heterocellular rays are present in Cruciferae. Euphorbiaceae has only homocellular rays. Key Words: Secondary xylem, herbaceous dicotyledons Introduction Investigations of evolutionary trends in the secondary xylem of angiosperms were conducted by numerous workers as the result of the profound impetus generated by the work of Bailey and Tupper (1918). The extensive and comparative data obtained from the survey of Bailey and Tupper revealed consistent changes in size, form, and pitting of tracheary cells at successive levels in the evolution and specialization of vessels in the dicotyledons. Several features of xylem which are considered as major trends of evolution in angiosperms are based upon the vessel element and its evolution (Carlquist, 1961). Major trends of xylem evolution other than vessel element were conducted by various investigators. Baily and Tupper (1918) and Bailey (1936, 1944, 1953) investigated evolutionary trends of imperforate tracheary elements. Kribs (1937), Bailey and Howard (1941), and Hess (1950) expounded the structural specialization ofthe wood parenchyma in dicotyledons. Major trends of xylem rays were investigated by Kribs (1935) and Barghoom (1940, 1941a, 194lb). Existing knowledge of major trends of xylem evolution has been obtained primarily from woody dicotyledons. The relative scarcity of information concerning structural specialization in the xylem of herbaceous dicotyledons led me to propose the study of structural variability in the secondary xylem of herbaceous dicotyledons in many families. The information obtained will: (1) establish a more complete inventory of xylem features for use in phylogenetic studies, (2) provide a critical evaluation of the structural variabilities between related herbaceous and woody species. This study of the Boraginaceae, Cruciferae, and Euphorbiaceae is the first in a series of comparative studies that have been planned to compile more complete information on features of the secondary xylem in herbaceous dicotyledons. 26 Transactions, Missouri Academy of Science Vol. 16

Materials and Methods The majority of fresh material utilized in this study was collected by Dr. B. G. Cumbie and was killed and fixed in formalin-acetic acid-alcohol (FAA). Additional plants were collected in the field and some were grown from seeds in the research greenhouse at the University of Missouri-Columbia. Dried specimens were obtained from the herbarium at the University of Missouri. Annotations were made on herbarium sheets from which materials were removed. Each species utilized in this study is listed in Table 1, together with the source of herbarium materials and locations of voucher specimens indicated according to Lanjouw and Stafleu (1964). In order to obtain as much secondary xylem as possible, segments were taken from near the base of the stems. Dried herbarium specimens were boiled in water to remove air from tissues. Stem segments were then treated with hydrofluoric acid for one week to soften tissues. After washing in water, segments were processed for embedding in paraffin by standard procedures (Sass, 1958). Transverse, radial, and tangential sections were cut in 12 J.Lm, stained with Heidenhain' s iron alum hematoxylin and safranin, and mounted in Harleco synthetic resin. Macerations were made by following Jeffrey's method outlined by Johansen (1940). Diagnostic characters to be included in the descriptions of each species were selected from the list suggested by Tippo (1941). Measurements of the length of fibers and vessel elements were made from macerated materials. In measuring length of vessel elements and fibers as well as measurements of vessel diameter, a total of 50 random measurements of the respective dimensions were made for each species. From these measurements, a range, a most frequent range (50% or more of the measurements and which will be referred to as MFR from this point on), and a mean (average) were calculated. T]le terminology used in this investigation is generally in accord with that suggested by the Committee on Nomenclature, International Association of Wood Anatomists (1957).

Results Boraginaceae-Cryptantha jamesii (Torr.) Payson-The secondary xylem is not differentiated into interfascicular and fascicular regions. The primary xylem is in the form of a continuous cylinder. Vessels, formed throughout the secondary xylem are solitary, in multiples, and clusters. Average length of vessel elements is 125 J.Lmand range 72-171 J.Lm with a MFRof 100-149 J.Lm. Average diameter of vessel elements is 29 J.Lffi. End walls vary from oblique to transverse-oblique to transverse. The perforations are simple and circular. Pits are bordered and alternate. Mean length of fibers is 99 J.Lm and range 162-343 J.Lm with a MFR of 200-243 J.Lm. The wall is thin and pits are simple, circular, and numerous. Protoplasts are present and the nucleus is visible in many fibers. Axial parenchyma is scanty para tracheal and composed of two to five strands. Rays are absent. Echium p/antagineum L.-The secondary xylem is not divided into fascicular and interfascicular regions. The primary xylem is in the form of separate bundles. Average length of vessel elements is 295 J.Lm and range 117-495 J.Lm with a MFR of 180-297 f.1ill. Average diameter of vessel elements is 43 J.Lm. End walls are mostly oblique and rarely transverse. Perforations are simple and circular. Average length of fibers is 487 J.Lm and range 270-810 J.Lm with a MFR of 359-395 J.Lm. The wall is thin and pits are simple and circular. The number of pits range from few to numerous. 1982 Park, Biological Sciences 27

Table 1. list of species invesil:igated.

Species Herbarium Collector

Boraginaceae Cryptantha jamesii (Torr.) Payson TIC Cumbie 213 Echium p/antagineum L. UMO Cumbie 486 Heliotropium convolvulaceum (Nutt.) Gray TIC Rowell60 H. corymbosum Ruiz. and Pav. UMO Cumbie 444 Lappula echinata Gilib. UMO Cumbie 507 Onosmodium mol/e Michx. UMO Stephens 8984

Crudferae Alyssum a/yssoides L. UMO Ahles 67921 Brassica nigra (L.) Koch UMO Cumbie 495 Came/ina microcarpa Andrz. UMO Palmer Caulanthus crassicaulis (Torr.) Wats. UMO Dunn Dithyrea wis/izenii Euglem TIC Cumbie 189 Stanleya elata Jones UMO Dunn 6712

Euphorbiaceae Acalypha virginica L. UMO Bojansky NA-69-41 Cnidosco/us stimulosis (Michx.) Englem and Gray TIC Cumbie 58 Croton texensis (K) Muell. -Arg. TIC Cumbie 38 Euphorbia dentata Michx. TIC Cumbie 111 E. maculata L. UMO Cumbie 459 Tragia nepetaefolia Cav. UMO Palmer 16278

Axial parenchyma is scanty paratracheal. Rays are absent. Heliotropium conuoluulaceum (Nutt.) Gray-The arrangement of the primary and secondary xylem as well as the distribution of vessel elements in the secondary xylem are identical with C. jamesii. Average length of vessel elements is 289 ,_._.m and range 162-450 fLm with a MFR of 225-297 fLm. Average diameter of vessel elements is 75 fLm. End walls are mostly oblique but a few are transverse-oblique among short vessel elements. Perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 452 fLm and range 279-720 f.lm with a MFR of 405-495 f.lm. The wall is thin. Pits are simple, circular, and range few to numerous. Axial parenchyma is scanty paratracheal. Rays are composed of upright and square cells. Multiseriate and uniseriate rays are present in the outer and inner xylem. In the inner xylem, uniseriate rays are predominant 28 Transactions, Missouri Academy of Science Vol. 16 and range one to 20 cells in height Multiseriate rays are chiefly two, rarely three, cells wide and range from four to 20 cells in height. H. corymbosum Ruiz and Pav.-Most of the xylem features are the same or very similar to that of H. conuoluulaceum except the dimensions of the vessel elements and fibers. Mean length of vessel elements is 238 ILm and range 180-324 ILm with a MFR of 180-288 jLm. Average diameter of vessels is 48 !Lffi. Mean length of fibers is,495 ILm and range 180-549 jLm. Lappula echinata Gilib.-Most of the xylem features are the same as C. jamesii. Average length of vessel elements is 290 ILm and range 162-414 ILm with a MFR of 200-299 jLm. Average diameter of vessel elements is 22 jLm. End walls are oblique and the perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 372 ILm and range from 288-488 ILm with a MFR of 300-405 jLm. Wall is thin. Pits are simple, circular, and numerous. Rays are composed of upright cells. Uniseriate rays range from one to 18 cells in height. Multiseriate rays are two cells wide and range three to 21 cells in height Osmodium molle Michx.-Arrangement of the primary and the secondary xylem are identical with that of C. jamesii. Average length of vessel elements is 309 ILm and range 198-441 ILm with a MFR of 252-349 jLm. Average width of vessel elements is 44 jLm. End walls are oblique. Perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 431~-Lm and range 297-630 ILm with a MFR of 405-486 jLm. Wall is thin. Pits are simple, circular, and numerous. Axial parenchyma is scanty paratracheal and rays are absent. Cruciferae-Alyssum alyssoides L.-The primary xylem is arranged in separate bundles. The secondary xylem is not differentiated into fascicular and interfascicular regions. Vessels, formed throughout the secondary xylem, are solitary, in multiples, and in clusters. Mean length of vessel elements is 179 ILm and range 108-270 ILm with a MFR of 153-243 !Lffi. Average diameter of vessels is 18 jLm. The outlines of vessels range from angular to circular. End walls are mostly oblique with a rare transverse-oblique. Perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 290 ILm and range 216-387 ILm with a MFR of 252-297 !Lffi. Wall is thin and pits are simple, circular, and numerous. Axial parenchyma and rays are absent. Brassica nigra (L.) Koch-Arrangement of the primary and secondary xylem are identical as the A a/yssoides. Vessels, formed throughout the secondary xylem, are solitary, in multiples, and in clusters (Fig. 1). Average length of vessel elements is 167 ILm and range 99-288 ILm with a MFR of 153-198 jLm. Average diameter of vessels is 39 !Lffi. End walls are mostly oblique, transverse-oblique, with a few transverse. Perforations are simple and circular. Pits are bordered, aJternate, and circular. Length of fibers average 366 ILm and range 261-495 ILm with a MFR of 315-387 jLm. Wall is thin. Pits are simple, circular, and numerous. Axial parenchyma is absent. Rays consist of erect, square, and procumbent cells. Multiseriate rays are predominant in the inner as well as the outer xylem. Uniseriate rays are one to five cells high. Multiseriate rays are two to six cells wide and 5-30 cells high (Fig. 2). Came/ina microcarpa Andrz.-Most of the xylem features are identical as A alyssoides (Fig. 3). 1982 Park, Biological Sciences 29

Average length of vessel elements is 197 1-1m and range 99-333 1-1m with a MFR of 153-243 f.Lm. Average diameter of vessels is 24 f.Lm. The outlines of vessels range from angular to circular. End walls are oblique and transverse-oblique. Perforations are simple and circular. Pits are bordered and alternate. Fibers average 346 J.Lm in length and range 198-468 1-1m with a MFR of 306-396 J.Lm. Wall is thin and pits are simple, circular, and numerous. Axial parenchyma and rays are absent. Caulanthus crassicaulis (Torr.) Wats.-Arrangement of the primary and the secondary xylem as well as the distribution of vessels, angle of end walls, pitting, perforations on the vessel elements are similar to A alyssoides (Fig. 4). Average length of vessel elements is 156 J.Lm and range 72-243 1-1m with a MFR of 153-189 fLm. Average diameter of vessels is 44 f.Lm. Fibers average 259 J.Lm in length and range 171-360 1-1m with a MFR of 252-297 J..Lm. Axial parenchyma is absent. Rays consist of upright and square cells. Uniseriate rays are one to eight cells high. Multiseriate rays are two to three cells wide and four to more than 20 cells high and predominant (Fig. 5). Dithyrea wislenzeni Engelm-Many of the anatomical features of the xylem are the same as the A alyssoides (Fig. 6). Average length of vessel elements is 122 f.Lm and range 90-171 J.Lm with a MFR of 108-144 J..Lm. Average diameter of vessels is 52 J.Lm. Fibers average 261-297 J..Lm in length of range 216-297 J.Lm with a MFR of 216-297 1-1-m. A protoplast is present. Axial parenchyma and rays are absent. Stanleya elata Jones--Average length of vessel elements is 170 J.Lm and range 108-252 J.Lm with a MFR of 153-198 J.Lm. Such xylem features as vessel distribution, pitting, end walls, arrangement of the primary and the secondary xylem, and inclination of end walls are identical with that of A alyssoides (Fig. 7). Average diameter of vessel elements is 43 J.Lffi. Average length offibersis331 J..Lmand range 252-450 1-1m with aMFR of315-369 J.Lm. Axial parenchyma is scanty paratracheal. Rays consist of upright and square cells. Uniseriate rays are predominant and range one to ten cells in height. Multiseriate rays are two to six cells wide and four to 26 cells high. Euphorbiaceae-Aca/ypha virginica L.-The secondary xylem is not differentiated into fascicular and interfascicular regions and the primary xylem is in the form of separate bundles. Vessels, formed throughout the secondary xylem are solitary, in multiples, and in clusters. Average length of vessel elements in 461 J.Lm and range 297-603 J.Lm with a MFR of 423-549 J.Lm. Average diameter of vessels is 36 J.Lm. End walls are oblique. Perforations are simple, circular, and pits are bordered, alternate, and circular. Average length of fibers is 513 J.Lm and range 405-630 J.Lm with a MFR of 450-567 1-1-m. Pits are simple, circular, and rare. Axial parenchyma is scanty paratracheal. Rays consist of erect and square cells. Uniseriate rays are one to 21 cells high. Multiseriate rays are two to six cells wide and eight to 20 cells high. Cnidoscolus stimulosus (Michx.) Engelm. and Gray-Arrangement of the primary and the secondary xylem are identical as A virginica. Vessels, formed throughout the secondary xylem, are solitary, in radial multiples, and in clusters (Fig. 8). Average length of vessel elements is 292 f.Lm and range 171-405 f.Lm with a MFR of 243-324 1-1m. Average diameter of vessels is 104 1-1m. End walls are oblique, and a 30 Transactions, Missouri Academy of Science Vol. 16 few transverse. Perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 618 J.Lm and range 315-765 J.Lm with a MFR of 450-549 J.Lm. Wall is thin. Pits are simple, circular, and numerous. Axial parenchyma is scanty paratracheal. Rays are composed of erect and square cells. Uniseriate rays are one to 25 cells high. Multiseriate rays are two to three cells wide and most of them reach the entire height of the section. Croton texensis (K.) Muell-Arg.-Arrangement of the primary and the secondary xylem are identical as A. uirginica. Vessels, formed throughout the secondary xylem, are mainly in multiples and are solitary, with a few clusters (Fig. 9). Average length of vessel elements is 4 72 J.Lm and range 288-711 J.Lm with a MFR of 405-540 J.Lm. Average diameter of vessels is 42 J.Lm. End walls are oblique. Perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 557 J.Lm and range 360-738 J.Lm with a MFR of 450-594 J.Lm. Wall is thin and pits are simple, circular, and numerous. Axial parenchyma is scanty paratracheal. Rays consist of erect and square cells. Uniseriate rays are predominant and range two to 20 cells high. Multiseriate rays are two to three cells wide and eight to ten cells high (Fig. 10). Euphorbia dentata Michx.-Like other species of the family described already in this paper, primary xylem is in separate bundles and the secondary xylem forms a continuous cylinder without differentiating into fascicular and interfascicular regions. Vessels, formed throughout the secondary xylem, are solitary, in multiples, and in clusters (Fig. 11). Average length of vessel elements is 6 70 J.Lm and range 342-945 J.Lm with a MFR of 513-675 J.Lm. Average diameter of vessels is 50 J.Lffi. End walls are oblique and perforations are simple and circular. Pits are bordered, alternate, and circular. Average length of fibers is 695 J.Lm and range 315-999 J.Lm with a MFR of 495-738 J.Lm. Wall is thin and pits are simple, circular, and numerous. E. maculata L. -Most of the anatomical features of this species are similar to that of E. dentata except the length of vessel elements and fibers (Fig. 12). Average length of vessel elements is 299 J.Lm and range 198-450 J.Lm with a MFR of 252-297 J.Lm. Average diameter of vessel elements is 44 J.Lm. Mean length of fibers is 380 J.Lm and range 243-513 J.Lm with a MFR of 360-432 J.Lm. A protoplast is present. Tragia nepetaefo/ia Cav.-Arrangement of the primary and the secondary xylem are identical as that of A. uirginica. Vessels are formed throughout the secondary xylem mainly in solitary and clusters, with a few multiples. Average length of vessel elements is 220 J.Lm and range 126-396 J.Lm with a MFR of 207-288 J.Lm. Average diameter of vessels is 64 J.Lffi. End walls are mostly oblique with a rare transverse-oblique or transverse. Perforations are simple and circular. Mean length of fibers is 408 J.Lm and range 315-621 J.Lm with a MFR of 405-477 J.Lm. Wall is thin and pits are simple, circular, and numerous. Axial parenchyma is scanty paratracheal. Rays consist of erect and square cells. Rays are exclusively uniseriate and range from one to ten cells in height. 1982 Park, Biological Sciences 31

Figures 1-6. 1. Transverse section of Bmssica nigra showing solitary and multiple pores. X 125. 2. Tangential section of B. nigra showing uniseriate and multiseriate rays. X 125. 3. Transverse section of Camelina micmcarpa showing pore multiples, pore dusters, and solitary vessels. X 125. 4. Transverse section of Caulani:hus crasskaulis. X 125. 5. Tangential section of C. cmssicatdis. X 125. 6. Transverse section of Dithyrea wislenzeni showing pore distribution. X 125. 32 Transactions, Missouri Academy of Science Vol. 16

figures 7-12. 7. Transverse section of Stanleya elata showing vessel distribution. X 125. 8. Transverse section of Cnidoscolus stimtdosus showing pore multiples and pore dusters. X 125. 9. In.msverse section of Croton texensis. X 125. 10. Tangential section of C. texensis. X 125. 11. Transverse section of Euphorbia dentata. X 125. 12. Cmss section of E. maculata. X 125. 1982 Park, Biological Sciences 33

Discussion Boraginaceae and Cruciferae have short vessel elements (258 fLm and 165 fLm respectively). Euphorbiaceae has mean vessel element length of 402 fLm, which is medium (Metcalfe and Chalk, 1950). Length of tracheary elements decreases as the tracheophytes advance in the course of evolution and as they obtain structural specializations in the xylem (Bailey and Tupper, 1918; Cheadle, 1943a; Frost, 1930a). Reductions in length of tracheary elements originated in fusiform cambial initials (Bailey, 1920). All the families studied for this investigation have small vessel diameters (44 fLm, 37 fLm, and 57 fLm respectively). Bailey and Tupper (1918), Bailey (1920), and Frost (1930a) postulated that vessel elements which are long, small in diameter, and angular in cross section preceded those which are short, broad, and circular in cross sectional outline. In his study of wood anatomy of Asterae, Carlquist (1966) found the decrease in both vessel diameter and vessel element length with increasing aridity. Herbaceous dicots generally have short vessel elements with small diameters. The most common type of end walls in all the species studied is oblique and the less common type is transverse-oblique; the least common type is transverse on both ends. All the species have vessel elements with a simple perforation plate. The vessel elements with scalariform plates appeared in plants before elements with a simple perforation plate (Jeffrey, 1917; Bailey and Tupper, 1918; Frost, 1930a, 1930b; Cheadle, 1942, 1943a, 1943b, 1944). All the families studied for this investigation have vessel elements with small, bordered, and alternate pits. Boraginaceae, Cruciferae, and Euphorbiaceae have short libriform fibers (406 fLm, 306 fLm, and 529 fLm respectively). The greater length of fibers compared with the length of vessel elements indicates substantial elongation of fusiform cambial initials during maturation. In woody species there is a tendency toward greater wall thickness as imperforate cells evolve to become libriform fibers. Unlike woody species, the most herbaceous dicotyledons have very thin walls with simple pits which are more or less circular in outline and numerous in number. Cumbie (1960) noted the presence of nucleated fibers in the secondary xylem. In order to determine presence or absence of nucleated fibers, one must utilize living plants as a source of materials. Cumbie emphasized that all the specimens secured from living plants did not retain nuclei at maturity. Arnold (1951) considers nucleated fibers to be a specialized feature among libriform fibers. Axial parenchyma in Boraginaceae is scanty paratracheal and composed of two to five strands. All the species of Cruciferae except Stanleya elata Jack axial parenchyma. Axial parenchyma inS. e/ata is scanty paratracheal. Euphorbiaceae has scanty paratracheal axial parenchyma. Cumbie (1960) found that in Leguminosae, herbaceous species generally exhibit only a sparse amount of axial parenchyma. The trends of axial parenchyma specialization were statistically correlated with already established vessel specialization by Kribs (1937). Bailey and Howard (1941) and Hess (1950) suggested that additional trends to be included in addition to those established by Kribs. Rays are absent in some species of Boraginaceae. Species which possess rays exhibit only homocellular types. Both uniseriate and multiseriate rays are found in species which possess rays. Cruciferae, like Boraginaceae, have species with rays and others without rays. When present, both uniseriate and multiseriate rays are found. Homocellular as well as heterocellular types are found in the family. Rays in the Euphorbiaceae studied in this investigation are all homocellular. Both uniseriate and multiseriate rays are present in most species except in Tragia nepetaefo/ia which possess only uniseriate rays. Statistical analysis of various dicot ray types compared with levels of vessel specializa- 34 Transactions, Missouri Academy of Science Vol. 16 tion has revealed significant correlations. Kribs (1935) concluded that ray evolution belongs to the major trends of xylem evolution. Barghoorn (1940, 1941a, 194lb) demonstrated a loss of heterogeneity in rays composed of erect cells only. These changes result from increase in length of ray initials. The ultimate result of continued elongation of ray initials is conversion of ray initials into fusiform initials (Barghoorn, 1941b). Thus, raylessness is the end result of this process. There is a strong need for more extensive investigations of individual genera and species of herbaceous families to establish the structural specialization of the secondary xylem with such habits.

Acknowledgement I would like to give my sincere thanks to Dr. David B. Dunn and Dr. B. G. Cumbie for their guidance and assistance. They have been very generous in allowing me to use their laboratories and herbarium even after I left the University of Missouri-Columbia. The major proportion of the work was completed at the University of Missouri-Columbia.

Literature Cited Arnold, C. A. 1951. Some observations on the anatomy ofthe common geranium. Papers Mich. Acad. Sci. 37: 3-11. Bailey, I. W. 1920. The cambium and its derivative tissues. Il. Size variations of cambial initials in gymnosperms and angiosperms. Amer. Jour. Bot. 7: 355-367. -----· 1936. The problem of differentiating and classifying tracheids, fiber-tracheids, and libriform fibers. Trop. Woods. 45: 18-23. -----· 1944. The development of vessels in angiosperms and its significance in morphologi­ cal research. Amer. Jour. Bot. 31: 421-428. -----· 1953. Evolution of the tracheary tissue of land plants. Amer. Jour. Bot. 40: 4-8. _____ and Howard, R. A. 1941. The comparative morphology of the lcacinaceae. Ill. Imperforate tracheary elements and xylem parenchyma. Jour. Arnold Arb. 22: 432-442. _____ and Tupper, W. W. 1918. Size variations in tracheary cells. I. A comparison between the secondary xylems of vascular cryptogams, gymnosperms, and angiosperms. Amer. Acad. Arts Sci. 54: 149-204. Barghoorn, E. S. 1940. The ontogenetic development and phylogenetic specialization of rays in the xylem of dicotyledons. I. The primitive ray structure. Amer. Jour. Bot. 27: 918-928. -----· 1941a. The ontogenetic development and phylogenetic specialization of rays in the xylem of dicotyledons. II. Modification of the multiseriate and uniseriate rays. Amer. Jour. Bot. 28: 273-282. -----· 1941b. The ontogenetic development and phylogenetic specialization of rays in the xylem of dicotyledons. III. The elimination of rays. Bull. Torrey Bot. Club. 68: 317-325. Carlquist, S. 1961. Comparative plant anatomy, Holt, Rinehart and Winston. New York. -----· 1966. Wood Anatomy of Composite: A summary with comments orfactors controlling wood evolution. Aliso 6(2): 25-44. Cheadle, V. I. 1942. The occurrence and types of vessels in the various organs of the plant in the Monocotyledoneae. Amer. Jour. Bot. 29: 441-450. -----· 1943a. The origin and certain trends of specialization of the vessel in the Monocotyledoneae. Amer. Jour. Bot. 30: 11-17. -----· 1943b. Vessel specialization in the late metaxylem of the various organs in the Monocotyledoneae. Amer. Jour. Bot. 30: 484-490. -----· 1944. Specialization of vessels within the xylem of each organ in the Monocotyle­ doneae. Amer. Jour. Bot. 31: 81-92. Cumbie, B. G. 1960. Anatomical studies in the Leguminosae. Trop. Woods 113: 1-47. 1982 Park, Biological Sciences 35

Frost, F. H. 1930a. Specialization in secondary xylem of dicotyledons. I. Origin of vessels. Bot. Gaz. 89: 67-94. -----· 1930b. Specialization in secondary xylem of dicotyledons. II. Evolution of end wall of vessel segment. Bot. Gaz. 90: 198-212. -----· 1931. Specialization in secondary xylem of dicotyledons. III. Specialization of lateral wall of vessel segment. Bot. Gaz. 91: 88-96. Hess, R. W. 1950. Classification of wood parenchyma in dicotyledons. Trop. Woods 96: 1-20. International Association of Wood Anatomists. Committee on Nomenclature. 1957. International glossary of terms used in wood anatomy. Trop. Woods 107: 1-36. Jeffrey, E. C. 1917. The anatomy of woody plants. Univ. Chicago Press, Chicago. Johansen, D. A 1940. Plant microtechnique. McGraw-Hill Book Co., New York. Kribs, D. A 1935. Salient lines of structural specialization in the wood rays of dicotyledons. Bot. Gaz. 96: 547-557. -----· 1937. Salient lines of structural specialization in the wood parenchyma of dicotyle­ dons. Bull. Torrey. Bot. Club 64: 177-187. Lanjouw, J. and Stafleu, F. A 1964. Index herbariorum. Part I. The herbaria of the world. 5th ed. Regnum Vegetabile, Vol. 31. -----· Metcalfe, C. R. and Chalk, L. 1950. Anatomy of the dicotyledons. Clarendon Press, Oxford. Sass, J. E. 1958. Botanical microtechnique. 3rd ed. Iowa State Univ. Press, Ames, Iowa. Tippo, 0. 1941. A list of diagnostic characteristics for descriptions of dicotyledonous woods. Trans. Illinois Acad. Sci. 34: 105-106.

1982 Othic, Kurtz & Baumgartner, Biological Sciences 37

Conversion of Forestland in the Missouri Ozarks1

S. T. Othic Missouri Department of Natural Resources Jefferson City, Missouri 65202 W. B. Kurtz School of Forestry, Fisheries and Wildlife University of Missouri-Columbia Columbia, Missouri 65211 D. C. Baumgartner USDA Forest Service East Lansing, Michigan 48823

Abstract: Conversion of forested areas to agricultural uses in the Missouri Ozark Region has generated concern over future timber supplies from this region. Thirty tracts within the USDA Forest Service Southwestern and Eastern Ozark Survey Units were examined to determine differences in quality of forest stands prior to conversion as well as the degree of conversion success. With the exception of bottomland sites in the Eastern Unit, original stand quality was found to be relatively poor with no apparent differences between regions in upland site quality. Due to poor tree form, low merchantable volumes and low stocking rates the pre-conversion forest stands of both regions were of questionable commercial value. The relative success of conversion appears to be greatly influenced by post-conversion treatments. No strong relationships were apparent among site characteristics, method of conversion and success of conversion. Key Words: Forest conversion, Ozarks, Forestland, Oak-hickory, Site quality, Pasture.

Missouri's total commercial forest growing stock increased nearly five percent between the 1959 and 1972 Forest Inventories, despite the conversion to other uses of approximately 1. 9 million acres of commercial forestland during the same time period (Spencer and Essex 1976). This paradox raises questions as to the actual quality of converted forestland and whether or not these losses will be of future consequence to Missouri's timber industry. Further, given that the majority of losses occurred as a result of conversion of forestland to pasture (Essex 1972), it is important to know success of the conversion efforts since landowners typically make land use decisions based on relative success or failure of others (Ralston 1965). The purpose of this study was to determine the quality and productivity of forestland converted to other land uses. Specific objectives were: 1. Determine the pre-conversion timber stand characteristics for sample plots within the Southwestern and Eastern Ozarks Survey Units. 2. Document the conversion techniques utilized on sample plots, current land use and apparent success of the conversion efforts.

1Funding for this study was provided by the USDA Forest Service, North Central Forest Experiment Station, Carbondale, Illinois, under Cooperative Agreement Number 13-570, Forest Land Conver­ sion in Southwest Missouri, and Cooperative Agreement Number 13-658, Factors Influencing the Conversion of Forest Land in Southern Missouri. 38 Transactions, Missouri Academy of Science Vol. 16

The study was limited to 30 plots evenly distributed between the Southwestern Ozarks and Eastern Ozarks Forest Survey Units in southern Missouri. These two Survey Units were chosen because the Eastern Ozarks Unit has the greatest acreage of commercial forestland in the state, while the adjacent Southwestern Ozarks Unit experienced an 18 percent decrease in acreage of commercial forestland between the 1959 and 1972 Forest Surveys (Spencer and Essex 1976). This contrast provided a basis for comparison of conversion results between the two areas. (Fig. 1)

I Southwestern Ozarks Eastern Ozarks

figure 1. Southwestern Ozarks and Eastern Ozarks Forest Survey Units.

Study plots were randomly selected from permanent sampling plots previously established by the USDA Forest Service for their continuing nation-wide forest inventory. Inventory data for the sample plots were obtained from the North Central Forest Experiment Station, St. Paul, Minnesota. The investigation to document conversion characteristics was conducted during the summer of 1978. 1982 Othic, Kurtz & Baumgartner, Biological Sciences 39

Preconversion Timber Stand Characteristics Based on 1959 Forest Survey records no appreciable differences in forest types were found between Survey Units, with the exception of four plots located in the Eastern Ozarks Survey Unit (Table 1) belonging to the oak-gum-cypress association. Twenty-four plots, or

Table 1. forest types, size and stocking classification, and site quality of sample plots, 1959 forest Survey.

Southwestern Ozarks Eastern Ozarks Forest Survey Unit Forest Survey Unit Number Number of plots Percent of plots Percent Percent

Species association Black-scarlet oak 8 53.3 5 33.3 43.3 White oak 1 6.7 1 6.7 6.7 Post-blackjack oak 5 33.3 4 26.7 30.0 Oak-gum-cypress 0 0 4 26.7 13.3 Elm-ash-cottonwood 6.7 6.7 6.7

Size and stocking classification Nonstocked 9 60.0 2 13.3 36.7

'01 Ole; &;g Poor 2 13.3 1 6.7 10.0 O.OJ co OJ (f) U) Medium 1 6.7 2 13.3 10.0 ... OJ ..D Poor 1 6.7 3 20.0 13.3 E OJ Medium 1 6.7 2 13.3 10.0 0"" 0.. ... OJ Poor 0 0 2 13.3 6.7 ..D E Medium 0 0 3 20.0 10.0 ""co:s (f) Well 1 6.7 0 0 3.3

Site quaiity1 / Poor 2 13.3 6 40.0 26.7 Medium 13 86.7 6 40.0 63.3 Good 0 0 2 13.3 6.7 Excellent 0 0 1 6.7 3.3

1Based on number of 16-foot Jogs produced by mature, dominant trees of commercial species on each one-fifth acre sample plot. If mature trees were not present, potential was estimated from existing dominant trees or by referral to adjacent stands. 40 Transactions, Missouri Academy of Science Vol. 16

80 percent of the total, were classified as oak type associations with the black-scarlet oak type being the most widely distributed and consequently the most affected species association. Approximately two-thirds of all sample plots within both Survey Units were either non-stocked or poorly stocked with commercial species (Table 1). The majority of non-stocked plots were located in the Southwestern Ozarks Survey Unit, where 60 percent of the plots had less than 9 percent of potential growing space on the sample areas occupied. In contrast, in the Eastern Ozarks Survey Unit only 13 percent of sample plots were non-stocked, though 40 percent were poorly stocked with commercial species. The average growing stock volume prior to conversion for the 30 sample plots was approximately 375 cubic feet per acre. This is somewhat greater than the estimated average volume of 327 cubic feet per acre for the land converted between the 1959 and 1972 Forest Inventories (Jakes eta!. 1979). Sample plots in the Eastern Ozarks Survey Unit had a much higher total average growing stock volume (431.2 cu ft/ac) than those in the Southwestern Ozarks Survey Unit (319.3 cu ft/ac). Little variation in site quality was recorded for the 15 sample plots in the Southwestern Ozarks Survey Unit. All plots were in the poor to medium categories (Table 1). Plots were more evenly distributed over the site quality categories within the Eastern Ozarks Survey Unit; however, the majority (80 percent) of plots were in the two lowest categories. In all, 90 percent of plots surveyed were of poor to medium site quality, based upon sampling criteria utilized in the 1959 Forest Survey.

Method and Success of Conversion Overstory elimination was nearly complete in the Eastern Ozarks Survey Unit with dozing operations resulting in total kill on 14 of 15 sample plots (Table 2). Considerably less success was obtained in the Southwestern Ozarks Survey Unit where differing techniques,

Table 2. Conversion method and percent overstory elimination on sample plots, 1978 investigation.

Southwestern Ozarks Eastern Ozarks Forest Survey Unit Forest Survey Unit All plots Method of Number Overstory Number Overstory Number Overstory conversion of plots elimination of plots elimination of plots elimination

(percent) (percent) (percent) Bulldozer 7 100.0 14 100.0 21 100.0 Spray 4 67.5 4 67.5 Cut-girdle 4 50.0 1 30.0 5 46.0 i.e., spraying and cut-girdle operations, resulted in lower levels of overstory elimination. Complete overstory kill was achieved on 70 percent of plots. Only in three cases were conversion efforts failures due to an incomplete kill of the overstory. The remaining failures were due to post-conversion treatment. Considerable variation in success of vegetation establishment was found within Survey Units and this is believed to be due to intensive grazing pressures (particularly within the 1982 Othic, Kurtz & Baumgartner, Biological Sciences 41

Eastern Ozarks Unit) and hardwood overstory competition. Tall fescue (Festuca arundinacea) was the most widely seeded grass species in both Units. No relationship between vegetative establishment and site characteristics could be determined in this study.

Land Use Changes A shift in management priorities occurred in the interim between the 1972 Forestry Survey and our 1978 investigation (Table 3). Wooded pasture had been the most frequent use recorded for the sample plots in 1972, but by 1978 the predominant use had switched to improved pasture. This change was most likely linked to rising cattle prices during the period. The majority of the land-use change occurred in the Southwestern Unit.

Table 3. Land use of sample plots, 1972 forest Survey and 1978 investigation.

Southwestern Ozarks Forest Survey Unit 1978 Improved Wooded pasture pasture Reverting Total (number of plots) Improved pasture 2 2 4 C'-l [""- Wooded pasture 6 5 11 ,.....0'\ Total 8 5 2 15

Eastern Ozarks Forest Survey Unit 1978 Improved Wooded Cropland pasture pasture Total (number of plots) Cropland 3 3 Improved pasture 6 6

C\l ['-.. Wooded pasture 3 1 4 0'\ ,...; Other farmland 2 2 Total 3 11 1 15

Conclusions and Recommendations The most important finding of this investigation is the major shift in management priorities, toward improved pasture development, between the 1972 Forest Survey and the 1978 fall study. This was evidenced primarily within the Southwestern Ozarks Forest Survey Unit. The significance of such a shift in emphasis to more intensive management is that it essentially precludes any later reestablishment of commercial forestland should grazing be discontinued. The long-term effect of this should have little consequence on the forest industry because of low quality, etc.; however, loss of wildlife habitat could have 42 Transactions, Missouri Academy of Science Vol. 16 far-reaching negative consequences within the region. The greatest negative impact will most likely be levied against wildlife populations in the region due to loss of wildlife habitat. However, conversion of the predominantly tall fescue pastures to a mixture of warm season grasses and forbs might mitigate the situation to some extent because of their diversity in height, density and food source. In general, timber stands prior to conversion were of marginal commercial value. Greatest production losses occurred in the Eastern Ozarks Unit. Overall conversion efforts were successful, though somewhat variable within the respective Survey Units. However, no substantial differences in completeness of conversion were found between the Survey Units. Although not specifically addressed in this paper, no significant relationships between physical site characteristics and conversion success could be shown. Production losses incurred through agricultural encroachment upon Missouri's timber resources, though substantial, are variable and dependent upon individual site and stand characteristics. Efforts should be undertaken to accurately identify those areas of forestland which are of higher quality, and therefore capable of responding to more intensive timber management. Once these areas are identified and properly managed, many of the past losses may be regained and possible future losses offset by increased production of quality timber.

Literature Cited Essex, B. 1972. Forest area in Missouri counties. USDA Forest Service Resource Note NC-182, 4 p. North Central Forest Experiment Station, St Paul, Minnesota. Jakes, P., J. Spencer, Jr. and B. Essex. 1979. Land use change in Missouri, 1959-1972. USDA Forest Service Resource Bulletin NC-40, 19 p. North Central Forest and Range Experiment Station, St. Paul, Minnesota. Ralston, R. 1965. Economic evaluation of land uses in Dent County, Missouri. Unpublished Ph.D. dissertation, University of Missouri, Columbia. Spencer, J., Jr. and B. Essex. 1976. Timber in Missouri, 1972. USDA Forest Service Resource Bulletin NC-30, 108 p. North Central Forest Experiment Station, St. Paul, Minnesota. 1982 Choromanski & Fritzell, Biological Sciences 43

Status of the River Otter (Lutra canadensis) in Missouri1

Jane F. Choromanski and Erik K Fritze]] School of Forestry, Fisheries and Wildlife University of Missouri-Columbia Columbia, Missouri 65211

Abstract: Records of reliable sightings, sign, and accidental captures of river otters during the past 20 years were compiled to describe the status and distribution of the species in Missouri. These records suggested that the number of otters in the state has increased since about 1975. A self-sustaining population exists in the Mississippi lowlands. An estimated 20-30 individuals inhabit Mingo National Wildlife Refuge and Duck Creek Wildlife Management Area. A second, smaller population was identified in Henry County in west-central Missouri. Otters reported outside these two areas probably were transient and not members of a local population. Most reports were along major river systems. Recent increases in otter records in Missouri probably resulted from an increase in well-established populations in states to the south, improved quality of aquatic habitats, and legal protection. Key Words: River otter, Lutra canadensis, Distribution, Status, Missouri

River otters were once common throughout North America with the exception of the arid southwestern United States (Ha111981, Seton 1929). Populations in many regions of the country, however, were extirpated during the last century. Although populations are sufficient to warrant annual harvests in some states (Deems and Pursley 1978), otters are rare or absent in the midwest south of the Great Lakes states. Although currently classed as endangered in Missouri (Nordstrom et al. 1977), otters probably occurred formerly in all major watersheds of the state. Bennitt and Nagel (1937: 136) reviewed county histories for 1830-1880 as follows: "abundant" (Nodaway, St. Charles, Montgomery, Monroe, Johnson); "plentiful" (Saline, Audrain); "in large numbers" (Shelby, Dunklin); "numer­ ous" .(Atchison); "profusely occupied this county" (Cass); "Store bills paid with coon, deer, otter, and other skins" (Ray); "Dr. Talbott had caught otter in sufficient numbers to make himself a large overcoat" (Nodaway); "present" (Miller, Platte, Lafayette, Linn). By the mid-1930's otter populations had dwindled such that Bennitt and Nagel (1937:135) reported "the otter is virtually extinct in Missouri." Based on reports of surviving remnants, they estimated a state population of about 70 individuals. The largest concentration occurred in southeastern Missouri; and a group of approximately 10 otters also remained in Atchison County in northwestern Missouri. The decline of otter populations was attributed to excessive exploitation and the effects of abusive land use practices on water quality (Bennitt and Nagel 1937:136). This study is part of an effort by the Missouri Department of Conservation to evaluate methods to restore the otter to the state. The objective of this report is to describe current status and distribution of the river otter in Missouri.

Contribution from Federal Aid in Wildlife Restoration Project W-13-R-36, Missouri Department of Conservation; Missouri Agricultural Experiment Station Journal Series No. 9093; and the Missouri Cooperative Wildlife Research Unit (U.S. Fish and Wildlife Service; Missouri Department of Conservation; School of Forestry, Fisheries and Wildlife; and Wildlife Management Institute, cooperating). 44 Transactions, Missouri Academy of Science Vol. 16

Methods Information regarding the status of river otters in Missouri was collected from a number of sources. Pertinent literature and records on file with state and federal wildlife agencies were reviewed. Colleges and major museums in Missouri and elsewhere were contacted for information on specimens in collections or other records of otters in the state. State and federal conservation officials, university biologists, and knowledgeable private individuals (i.e., trappers, fur buyers) were questioned about recent otter occurrence. Questionnaires were sent to state conservation agents requesting information regarding river otters in their respective counties. Names of individuals thought to have knowledge of otters were also requested in the questionnaires. Individuals reporting sightings of otters or otter sign in the past 20 years were contacted for verification and possible ''updates'' on the information reported. Field checks were conducted at areas where river otter presence was suspected. Field work was concentrated in southeast Missouri, especially on Mingo National Wildlife Refuge (NWR) and adjacent Duck Creek Wildlife Management Area (WMA), a combined area of over 11,000 ha of lowland hardwood forest.

Results and Discussion The current status of the river otter in Missouri appears to be somewhat improved from the estimated 70 animals present in 1934 (Bennitt and Nagel1937). The greatest number of river otters still is found in southeastern Missouri. The number of otter reports and accidental trappings have increased in recent years throughout the Mississippi lowlands (otters accidentally trapped are relinquished to the Missouri Department of Conservation). Verified reports typically occurred on the major rivers and state wildlife areas, including portions of the Mississippi, Castor, Whitewater, Current, Black and St. Francis rivers and Otter Slough WMA (Stoddard County), Ben Cash WMA (Dunklin County), Wolf Bayou WMA (New Madrid County) and Coon Island WMA (Butler County). These reports, together with the distribution of otters harvested in Arkansas during the 1980-81 trapping season (Lew Johnston, pers. comm.), suggest that a contiguous population exists throughout the upper portion of the Mississippi River Delta but also that few otters have penetrated far into the Ozark plateau (Fig. 1). The largest concentration of otters in southeastern Missouri was found in the Mingo/Duck Creek area. In 1967, a refuge biologist sighted an otter on Mingo NWR. This was the first sighting since establishment of the refuge in 1945. The following year, 1968, an otter was also sighted, but no more sightings occurred untill975. Since then, however, the number of otter reports has continued to increase. Originally these sightings were concentrated on a few areas within the refuge, possibly as a result of fish concentrations. Presently, however, otters and their sign can be found in most parts of Mingo NWR and Duck Creek WMA. Attempting to estimate the size of an otter population is difficult because there is no simple technique yet developed for censusing otters. The number of sightings and the amount of sign in the Mingo/Duck Creek area might lead one to conclude that there is a large number of otters present. Because otters are known to travel over large areas, however, estimating population size from sightings and sign may result in an overestimate of the actual population size. Considering these factors, we estimated that 20-30 otters reside on the Mingo/Duck Creek area. Possibly related to the increase in otters is the simultaneous increase in the beaver (Castor canadensis) population in this area and its effect on aquatic habitats. Narrative reports from Mingo NWR suggested that beavers were very rare during the 1950's; they apparently invaded the refuge by 1960, and in 1961 a population of40was estimated. By 1982 Choromanski & Fritzell, Biological Sciences 45

* 1960- Hll69 II 1970 - Hl79

• HISO- 1981

Figure 1. Locations of otter reports in Missouri, 1960-1981. (Shaded portion represents extent of southeastem population.)

1968 the population had risen to 120; in 1976 the refuge held an estimated 600 beavers and a trapping program was begun in 1977. From 1979 to 1982 a permittee trapper caught an average of 500 beavers annually on the refuge. Although this study did not reveal any factors responsible for the parallel increase exhibited by these 2 species, river otters may have benefitted from beaver activity for 2 reasons: 1) beaver activity created additional wetlands important as potential otter habitat and 2) otters commonly used abandoned beaver lodges and bank dens for shelter and whelping sites. Further investigation of these phenomena may be warranted. A second, but considerably smaller, population of river otters was found in west -central Missouri. Sightings and sign in this area have been concentrated in Henry County (Fig. 1, Table 1). When an otter was killed by an automobile in 1971, subsequent investigation revealed that otter activity had been observed by area residents for the previous 5 years. River otters have also been accidentally trapped in this area; 1 trapping was verified in 1980 and 5 others were reported to have occurred between 1974 and 1980. Not far from this area, river otter activity was documented at the Schell-Osage WMA in Vernon County. Otter sign was recorded in 1967 and 1978, and in 1981 an otter was sighted on the area. Although the majority of otter reports occurred in southeast and west -central Missouri, recurring reports of otter activity came from 3 other areas during the past 20 years (Fig. 1, 46 Transactions, Missouri Academy of Science Vol. 16

Table 1. Reports of l!'iver otter iin Missouri, exclusive of the Mississippi lowlands, 1960-1981.

Year Location Evidence

1960's Schell-Osage WMA, Vernon and St. Clair Cos. Tracks of at least 2 animals 1965 Big Creek, SW Johnson Co. and eastern Cass Co. Tracks and slide 1966 Yellow Creek, Swan Lake NWR, Chariton Co. Otter trapped 1 1969 Big Creek, Henry Co. Sec. 9, Twnshp. 42, Range 27 Adult and 2 pups sighted 1971 Big Creek and Highway #7, Henry Co., Sec. 21, Twnshp. 21, Range 27 Otter hit by ca~ 1971 Big Creek, Henry Co. Sec. 2, Twnshp. 42, Range27 Tracks and scats 1971-1975 Big Creek, Henry Co. Sec. 29, Twnshp. 43, Range 27 Tracks and scats Mid-1970's Big Lake, Holt Co. Otter sighted Mid-1970's 1 mi. N of Cleveland, Cass Co. Otter sighted Mid-1970's 8 mi. W of Maysville, DeKalb Co. Otter sighted and slides 1974 or 1975 Platte River, near Platte/Buchanan Co. line Otter sighted 1975 Swan Lake NWR, Chariton Co. Otter sighted 1977 or 1978 Nishnabotna River, Atchison Co. Sign 1977 Renner WMA, Hartsburg, Boone Co. 1 adult and 3 young sighted 1974-1980 Near Hartwell, Henry Co., on Big Creek tributaries 5 trapped 1980 Chariton River near Bynumville, Chariton Co. Sign 1980 Big Creek, 2-3 mi. N of Urich WMA, Henry Co. 1 trapped3 1981 Watson, Atchison Co. 1 sighted and sign 1981 Junction of N and S Fabius River, Marion Co. Slide and tracks 1981 Schell-Osage WMA, Vernon and St. Clair Cos. 1 sighted 1981 Missouri River, 3 mi. S of Rocheport, Boone Co. 1 sighted 1981 Merarnec River near Steeleville, Crawford Co. Family sighted

10n display at the Missouri Department of Conservation Research Center, Columbia, Missouri. 2 0n display at the Benton County Historical Museum, Warsaw, Missouri. 3T ernporarily being held by the Missouri Department of Conservation, Information and Exhibits Section.

Table 1). The first of these areas is in northwestern Missouri, from Platte County north to the Missouri border. This area includes Atchison County where 10 otters were estimated to remain in 1934 (Bennitt and Nagel 1937:136). The second area from which recurring reports of otter activity have come is in Chariton County in north-central Missouri, where an otter was accidentally trapped on the Swan Lake NWR in 1966. The third area is in central Missouri where otter activity has been reported along the Missouri River from Osage 1982 Choromanski & Fritzell, Biological Sciences 47

County west to Boone County. These reports probably represent transient animals or temporary residents and not self-sustaining populations. The notable absence of recent otter records from the Ozark plateau and the concentration of records along the major rivers of the state (Fig. 1) suggest that transient otters have traveled northward via the Mississippi and Missouri rivers and their major tributaries. Otters have not penetrated deeply into the Ozark plateau in either Missouri or Arkansas (Sealander 1979, L. Johnston, pers. comm.). The relatively high gradient streams may be less suitable habitat for otters than low gradient habitats of the Mississippi lowlands and larger rivers in the remainder of the state. It is doubtful that river otters were ever extirpated from Missouri (Schwartz and Schwartz 1959, 1981 ). We hypothesize that populations continued to decline or remained at 1934levels following 1937 when full protection of the species was implemented. Otter numbers probably remained at their lowest levels during 1940-1975 and most sightings were of transient animals probably entering the state from the south. By the mid-1970's numbers increased and self-sustaining populations became established, especially in the Mississippi lowlands. The period of increase coincided with an approximately 6-7 fold increase in the otter harvest in Arkansas (L. Johnston, pers. comm.) which probably reflected an increase in that well-established population. The harvest in Louisiana also reached an all-time high of 11,900 otters during the 1976-77 season (Ensminger and Linscombe 1980). Assuming that excessive exploitation and habitat loss and degradation were largely responsible for the decline in Missouri's river otter population, it is important to recognize that protection from harvest and improvement in habitat may have contributed to its subsequent increase. However, legal protection and habitat improvement, although crucial to the reestablishment of otter populations, probably did not by themselves result in population increases. Increases in established otter populations to the south, as mentioned previously, probably played a significant role in the increase of Missouri's river otter population.

Acknowledgments We are grateful to the many individuals contributing information to this study. The staff of the Missouri Department of Conservation, the Gaylord Memorial Laboratory, and the Mingo National Wildlife Refuge were especially helpful. David Erickson, research biologist of the Missouri Department of Conservation, provided assistance in many portions of the study from its initiation to final editing. The help and knowledge of James Ware during field checks in southeastern swamps were essential.

Literature Cited Bennitt, R. and W. 0. Nagel. 1937. A survey of the resident game and furbearers of Missouri. Univ. Missouri Studies 12(2):215pp. Deems, E. F. and D. Pursley (eds. ). 1978. North American furbearers. Internatl. Assoc. Fish and Wild!. Agencies. 157pp. Ensminger, A. and G. Linscombe. 1980. The fur animals, the alligator, and the fur industry in Louisiana. Louisiana Dept. Wild!. and Fisheries. 69pp. Hall, E. R. 1981. The mammals of North America. John Wiley and Sons, New York, 2:601-1181 + 90. Nordstrom, G. R., W. L. Pflieger, K C. Sadler and W. H. Lewis. 1977. Rare and endangered species of Missouri. Missouri Dept. Conserv. and USDA, SCS. 129pp. 48 Transactions, Missouri Academy of Science Vol. 16

Schwartz, C. W. and E. R. Schwartz. 1959. The wild mammals of Missouri. Univ. Missouri Press and Missouri Dept. Conserv., Columbia. 341pp. Schwartz, C. W. and E. R. Schwartz. 1981. The wild mammals of Missouri. Univ. Missouri Press and Missouri Dept. Conserv., Columbia. 356pp. Sealander, J. A 1979. A guide to Arkansas mammals. River Road Press, Conway, Arkansas. 313pp. Seton, E. T. 1929. Lives of game animals. Doubleday, Doran and Co., Inc., Garden City, New York, Vol. II, pp. 369-746. 1982 Westgate, Biological Sciences 49

Hemisinus Jacksordus as an Aid in Determining Potential of Marine Deposits for Yielding Remains of Terrestrial Mammals

James W. Westgate Department of Geography and Geology Southwest Missouri State University Springfield, Missouri 65804

Abstract: Recent quarrying operations in marl deposits of late Eocene age (upper Jackson Group) exposed near Madison, Arkansas, have produced unexpected finds of mammalian remains. No prior record of terrestrial mammals in Paleogene deposits of the Mississippi Embayment exists, and few terrestrial mammals have been discovered in equivalent strata throughout the Atlantic and Gulf Coastal Plain. Remains of oysters and mussels in life position indicate deposition occurred under estuarine conditions. Large numbers of unworn shells of Hemisinus jacksonius are also present. Modern relatives of this snail live in fresh to slightly brackish waters. Deposition of the marls and terrestrial mammals apparently occurred on the landward margin of the estuary. Conspicuous size and preference for low salinity habitats make H. jackson ius an excellent field indicator of deposits having a high potential for yielding terrestrial mammalian remains. Key Words: Terrestrial mammals, late Eocene, Mississippi Embayment, Hemisinus jacksonius, Protoreodon pumilus

Introduction Occurrences of late Eocene terrestrial mammals are rare in eastern North America, although complex faunas are well known from the Rocky Mountain region. The recent discovery of a primitive artiodactyl (Protoreodon pumilus) at the Crow Creek Quarry in eastern Arkansas (Westgate 1981) represents the first Paleogene terrestrial mammal from deposits of the Mississippi Embayment. The following analysis of fossil remains associated with the P. pumilus specimen, especially the invertebrate species Crassostrea alabamensis and Hemisinus jacksonius, was undertaken to account for the anomalous occurrence of this species and to determine whether continued excavation might result in discovery of additional terrestrial mammal specimens. Crow Creek Quarry is located on the east bank of Crow Creek, 1 km west of Madison, St. Francis County, Arkansas (Fig. 1). The "quarry" is a group of excavations dug in exposures of the Jackson Group (undifferentiated) to sample the fragmentary vertebrate remains which were concentrated at the site. Although the Arkansas Geological Commission has not defined these strata at the formation level (J. D. McFarland, pers. commun.), Palmer and Brann (1966:707) referred the deposits to the "d. upper Jackson" Group; thus they are considered latest Eocene in age. Sediments at the quarry horizon are marly sands with up to 50% of the volume consisting of valves of the oyster Crassostrea alabamensis. Detailed studies of the molluscan fauna preserved at Crow Creek were conducted by Harris (1946) and Palmer (1947). Crow Creek Fauna. The invertebrate fauna at Crow Creek Quarry is dominated by the oyster Crassostrea a/abamensis and by the mussel Mytilus hamatoides (Table 1). Many individuals display articulated valves indicating that the massive concentration of 50 Transactions, Missouri Academy of Science Vol. 16

.s lKM

Figme 1. Location of Crow Creek Quarry, St. Francis County, Arkansas.

shells occurred due to an in situ biohermal accumulation and not as a result of post-mortem transport and deposition. Palmer (194 7:278, 364) suggested the reefal deposit exposed on Crow Creek and nearby Little Crow Creek formed under intertidal conditions. Analysis of the habitat ofthe modem reef -forming oyster Crassostrea virginica in Texas coastal lagoons has shown a preference for shallow intertidal waters with salinities ranging between 9 and 20 ppt (Ladd, et al., 1957, p. 622). This is roughly one-fourth to one-half the salinity of waters in the open Gulf. Mussels of the genus Brachiodontes are second in dominance only to C. virginica in Texas coastal lagoons (Ladd, et al. 1957:625). On the Atlantic coast, the modem mussels Mytilus edulis and Modiolus demissus also live in intertidal habitats (Morris 1975: 17) in association with C. virginica. Excavation at the quarry has also produced large numbers of the gastropod Hemisinus jaclcsonius. Palmer (1947:278) found only worn individuals of this species and suggested its presence was indicative of proximity to the mouths of local streams entering into the Eocene Sea. Modern species of Hemisinus are generally associated with fresh water habitats, but are able to tolerate the low salinities found in marginal estuarine areas (Or. Norman F. Soh!, U.S. Geol. Surv., pers. commun.). A number of recently acquired individuals of H. jacksonius display little evidence of transportational wear (Fig. 2), indicating they may have lived in association with C. a/abamensis close to or at the reef site. This association would have been enhanced during rainy seasons when salinities in the estuary were especially low. Figure 3 shows size ranges for whole individuals. This variability could not be determined for the type specimens of H. jacksonius (Palmer 1947:277), which are incomplete specimens. Hyposaline conditions are also indicated by the vertebrate taxa from Crow Creek Quarry (Westgate and Ward 1981). Table 1 shows modern habitat preferences of vertebrate genera found at the quarry. The habitat overlap implies deposition occurred in an estuarine environment. The diversity of taxa is likely due to seasonal turnover of more typically marine and freshwater forms as well as post-mortem transport of carcasses into the estuary. 1

1982 Westgate, Biological Sciences 51 \

Figure 2. Topotype specimens of Hemisinus jacksonius showing little transportational wear.

26 E _;_ 0::: w t:u 21 ::E <( 0 .· * 1640~·--~--4~5--~-5~0~~~5~5~~~6~0--~-6~5~~~70 HEIGHT(mm)

Figure 3. Size distribution of complete specimens of Hemisinus jacksonius from Crow Creek Quarry (circles), and approximate size of the holotype (PRI 4574) of H. jacksonius (star). '.I

Vol. 16 56 Transactions, Missouri Academy of Science tions can tolerance (Gaufin 1958). The severity and duration of damage at affecte_d sta f benthiC therefore be documented by measuring density, diversity, and composttion ° rnuni­ invertebrate communities and comparing these characteristics with unaffected corn ties upstream.

Description of the Study Area this The sections of Big River and Mill Creek, one of its tributaries, encornpasseddb~3 sin study, geologically and physiographically belong to the St. Francois Knob ~n formed Region (Sauer 1920). In this region, igneous rock knobs typically rise above basJ~~n in the by weak limestone and shale. The relief in this region is greater than any other ~egt es are state: and in some areas resembles mountainous terrain. Two major sot! y~ife and dommant: very infertile crystalline rock soils which have little value for crops or w\-ughland the Fredericktown group which contains the best residual soils of this Ozark D rninant Province. Most available acreage of the latter soil type is farmed (Sauer 1920). 0 terrestrial vegetation in this area is the oak-hickory climax forest (Collier 195 5~heastfor Mill Creek begins in Washington County, near Potosi, Missouri, and flows nof1!1ill creek about 10 miles before entering Big River, south of Blackwell, Missouri (Fig. 1) · dissolVed resembles most other Ozark streams in that it is normally alkaline, clear, and ha: rubble oxygen levels at or near saturation. Substrate is similar to other Ozark streams,_ 1· e~~eP (21 and grave~-sized_ fragments of limestone and sil~ca minerals. Stream gradient 15~ill creek feet per mtle). Dtatoms are the dommant aquatic vegetation. Some reaches _of rnoDthS· occasionally contain heavy growths of filamentous algae, especially during Wlr:' t~ ca tive of An earlier water quality survey revealed a benthic invertebrate community 10 ~ (J:.ZI..lester slightly polluted conditions resulting from intensive barite mining in the waters he 1964). . gtof"l, S~· Big River originates in western Iron County and flows north through Washll'yv1issol..lrtd Francois: an? Jefferso_n counties befo~e entering the M~ramec River near Eur~k~~g. 1 , aD). The portion mcluded m this study drams 718 square mtles at Desoto, Mtssoun ( eY 1. 91 S k has an average discharge of 682 cubic feet per second (U. S. Geological surv~he-r ozar d Physical characteristics of Big River are similar to thosefound in Mill Creek and.? L' ra.'-Jel an stream_s. Substrate in Big River upstream from Mill Creek is composed of s~ali ;.a.g~e~:~~ sand-sized fragments of limestone and Potosi druse with very few rubble-s1ze? 0 co.'-'1tl This substrate is a direct result of tailings deposition from upstream lead m i nl n g d that th~ near Leadwood, Missouri, in the early 1960s (Kuester 1964). Kuester suggeste ~3 fol..lnt unstable qualities of this substrate were responsible for reduced number of ~a. 50b~~a €. during his survey. Shortly after Kuester's study, lead mining ceased af"l -pia· 1. I w~, conditions improved by 1975. Substrate below the confluence of Mill Creek ~, ,, tie;S ot ~­ 1 similar to thatfound in Mill Creek and was not noticeably affected by mining a. en ~ 8 ,...-,. iD h early 1960s. Stream gradient is 5 feet per mile which is much more grad u a.1_ "t -p,;_ i .__,. e !" t a ~re~k. Diatoms were abundant and filamentous algae was less noticeable in Bl g t lV m M1Il Creek during winter months. .,_ ·n 8 t 0 85 Big River (upstream from Mill Creek) has a watershed three times as large L 0 1e "-''Jas Creek, resulting in a greater flow throughout the study. Dilution of the sed i rn ese s -rreal important variable considered when effects of sediment on the benthos in "the were assessed.

Materials and Methods Benthic invertebrates were sampled at three sites on Big River and two (Fig. 1). The stations on Mill Creek were located above (1) and below (2 1982 Duchrow, Biological Sciences 57

I

I N t I \ \ STATION II HIGHWAY~

TOWMQ3

SCALE IN MILES

JEFFERSON CO. ---

WASHINGTON CO. f T. FRANCOIS CO.

Figure 1. Invertebrate sampling stations (1, 2, 3, 4 and 5) on Mm Creek and Big River, 1915-1976.

sediment. The stations on Big River were established to monitor the invertebrate communities above (3) and below the mouth of Mill Creek (4), and at the lowest extent of the fish kill (5), about 8 miles downstream. Samples were collected bimonthly during August and September 1975 and monthly from October 1975 through August 1976. All samples were collected from permanent, stable riffle areas at each sampling site by disturbing the substrate with a 3-pronged digging tool to a depth of 4-6 inches. Dislodged organisms were captured in a Turtox No. 105T33 heavy nylon bottom net (20 mesh per inch) placed immediately below the sample site. Between 8 and 12 square feet of riffle substrate were sampled for each collection at each station throughout the study. Debris and invertebrates collected in the bottom net were placed in 2 screened pans for washing. The upper pan had hardware cloth screen (2 mesh per inch) and the lower pan a 58 Transactions, Missouri Academy of Science Vol. 16

stainless steel wire screen (40 mesh per inch). Debris remaining in the upper screen was checked for organisms and discarded. Organisms from the upper screened pan and all material from the lower screened pan were preserved in 10 percent formalin. Samples were transported to the laboratory where the preservative was changed to 70 percent ethanol. Samples to be sorted were washed with water in a U.S. No. 35 Standard Sieve to remove the ethanol. Most organisms were removed from the debris by the sugar flotation method described by Anderson (1959). Debris was also systematically hand sorted to assure removal of all invertebrates not suspended during sugar flotation. Organisms were then preserved in 70 percent ethanol before identification. Benthic organisms were identified to the following taxonomic levels: (1) Flatworms (Platyhelminthes), roundworms (Nematoda), and segmented worms (Annelida) were identified to class. (2) Flies (Diptera) were identified to family or genus. (3) Mayflies (Ephemeroptera), stoneflies (Plecoptera), caddisflies (Trichoptera), naiads (Pelecypoda), snails (Gastropoda), crustaceans, and other organisms were identified to genus or species. Water samples were collected with each invertebrate sample to measure turbidity. Analyses of these water samples were done with a Hach Model 2100 turbidimeter in accordance with procedures outlined in American Public Health Association (1971).

Data Analysis Benthic invertebrate data from each station were examined to determine extent and duration of damage to benthic communities. This examination included comparing benthic invertebrate community characteristics such as: number of mayfly and stonefly taxa, total number of taxa, invertebrate sample density, and species diversity index values between samples collected at affected and unaffected stations on each stream for the same date. Coefficients of similarity for invertebrate samples helped to document duration of damage in each study stream. Other methods of detecting damage at stations which received sediment included the presence, absence, or abundance of other benthos species such as caddisflies, and comparison of the above characteristics to criteria established for Missouri streams (Table 1).

Table 1. Water quality criteria for individual samples hom Missouri streams.1

Individual sample Species No. of mayfly Water quality diversity and stonefly designation index value taxa

Unpolluted >3.9 >9 Moderately polluted 2.2-3.9 5-9 Polluted <2.2 <5

1Based on work done in Missouri by Kuester (1964), Duchrow (1974, 1976a, 1976b, 1977), Ryck (1974, 1976b), and Dieffenbach and Ryck (1976) and regression analysis of 895 samples. 1982 Westgate, Biological Sciences 51

Figure 2. Topotype specimens of Hemisinus jackson ius showing little transportational wear.

26 E .E. 0::: LU tu 21 ~ <( 0 .· * 1640~·--~--45~~--5~0~~~5~5--~~60~~--6~5--~~70 HEIGHT(mm)

Figure 3. Size distribution of complete specimens of Hemisinus jacksonius from Crow Creek Quarry (circles), and approximate size of the holotype (PRI 4574) of H. jacksonius (star). 52 Transactions, Missouri Academy of Science Vol. 16.

Table 1. Modem habitats of common genera from Crow Creek Quarry.

Taxon Habitat*

Invertebrates Crassostrea E M Mytilus E M Hemisinus F E (marginal)

Vertebrates Galeocerdo E M Odontaspis E M Myliobates E M Pristis F E M Arius E M Trichiurus E M Sphyraena M Lepisosteus F E Crocodylus F E M Trionyx F Protoreodon T

*T-Terrestrial F-Fresh Water E-Estuarine M-Marine

Discussion The association of H. jackson ius with reef-forming C. alabamensis indicates deposition occurred under hyposaline conditions in a near-shore, marginal estuarine environment. Such proximity to late Eocene terrestrial habitats explains the presence of Protoreodon pumilus in the fauna. Remains of this artiodactyl were undoubtedly transported to the site by local currents. Transport of bloated carcasses into modern estuaries is relatively common. The fluvial influences indicated by the presence of H. jacksonius in the deposits at Crow Creek Quarry may have repeated this process during the history of reefal development. If so, the likelihood of finding additional terrestrial vertebrate remains in the deposit would appear to be good. The recent discovery of a second mammalian species at the quarry supports this prediction. The specimen (U.S. Nat. Mus. 299909) is a nearly complete right mandible of a weasel-sized species. The post-mortem loss of all teeth from the jaw has delayed taxonomic identification and the specimen is still under study. However, the discovery of this mandible in the bioherm indicates that deposition of the P. pumilus specimen was not due to unique circumstances. It remains to be seen whether additional mammals were also buried at the site, but continued excavation at Crow Creek Quarry is planned for the near future. 1982 Westgate, Biological Sciences 53

The large size and distinct morphology of H. jacksonius make it an ideal field indicator of deposits which formed under near-shore influences. It is suggested that exposures bearing large numbers of this species be carefully surveyed for associated mammalian remains.

Acknowledgments The Faculty Research Program, Southwest Missouri State University, provided funds for field work at Crow Creek Quarry. Mr. H. Woody Clark, Clark Excavation Co., allowed field crews to have access to the quarry site. Crew members included James Dilbeck, Larry Lowman, Daniel Snow, Karen Westgate, and the author. Dr. Peter R. Hoover, Paleontological Research Institute, arranged the loan of type specimens of Hemisinus jacksonius. Dr. James F. Miller, Southwest Missouri State University, critically read the manuscript.

References Cited Harris, G. D. 1946. Pt. 1, Bivalves and bibliography for parts I and II: in Harris, G.D. and K.V.W. Palmer, The Mollusca of the Jackson Eocene of the Mississippi Embayment (Sabine River to the Alabama River): Bull. Am. Paleont., 30 (117):564 pp. Ladd, H. S. et al. 1957. Environments and facies of existing bays on the central Texas coast: In Ladd, H.S., ed., Treatise on Marine Ecology and Paleoecology, vol. 2, Paleoecology: Mem. Geol. Soc. Am., 67:1077 pp. Morris, P.A. 1975. A field guide to shells ofthe Atlantic and Gulf Coasts and the West Indies: Houghton Mifflin Co., Boston. 330 pp. Palmer, K. V. W. 1947. Pt. 2, Univalves and index: in Harris, G. D. and K. V. W. Palmer, The Mollusca of the Jackson Eocene of the Mississippi Embayment (Sabine River to the Alabama River): Bull. Am. Paleont., 30 (117):564 pp. Palmer, K.V.W. and D.C. Brann. 1966. Catalogue of the Paleocene and Eocene Mollusca of the southern and eastern United States. Part 2, Gastropoda: Bull. Am. Paleont. 48, (218):471-1047. Westgate, J.W. 1981. First discovery of Protoreodon {Agriochoeridae) in Arkansas: Trans. Missouri Acad. Sci. 15:245. Westgate, J. W. and J.F. Ward. 1981. The giant aquatic snake Pterosphenus schucherli (Palaeophidae) in Arkansas and Mississippi: J. Vertebrate Paleont. 1 (2):161-164.

1982 Duchrow, Biological Sciences 55

Effects of Barite Tailings on Benthos and Turbidity of Two Ozark Streams 1

Richard M. Duchrow Missouri Department of Conservation 1110 College Avenue Columbia, Missouri 65201

Abstract: On August 15, 1975, the dam of a settling pond near Mill Creek in Washington County, Missouri, broke. Sediment from barite tailings in the pond was released, causing an extensive fish kill in 3 miles of Mill Creek and 9 miles of Big River. The mixture of clay and water from the tailings also increased turbidity to the mouth of Big River, a distance of 71 miles from the mouth of Mill Creek. The extent and duration of damage were determined by comparing benthic invertebrate communities found in affected and unaffected portions of each stream for 1 year. Initial damage to fish and benthos in both streams was apparent from the dead fish and reduced, dissimilar invertebrate communities. Duration of the damage varied between streams. Reestablishment of the invertebrate community in Mill Creek began after 38 days. However, similar invertebrate communities were not found at the control and affected stations until 264 days after the dam failure. Invertebrate communities in Big River, further downstream from the tailings pond, were severely reduced for only 14 days. After 14 days, similar invertebrate communities were found at the control and affected stations. Rapid recovery in these downstream reaches was attributed to the larger volume of flow and greater distance from the sediment source. Aesthetic quality of both streams was noticeably affected by increased turbidity in these normally clear Ozark streams. Turbidity values were consistently higher throughout the study in affected portions of both streams when compared to unaffected portions. Key Words: Missouri, Meramec river basin, turbidity, mining, water quality, pollution, fish kill, diversity, benthic macroinvertebrates.

Introduction Samples of benthic invertebrate communities in Mill Creek and Big River were used to document the extent and duration of damage caused by sediment inflow from a 90-acre barite tailings pond in northeast Washington County, Missouri. A mixture of water and red clay flowed through a dam break on August 15, 1975, and entered Mill Creek via Buss Branch near Tiff, Missouri (Fig. 1). The mixture continued down Mill Creek 3 miles and entered Big River in St. Francois County, south of Blackwell, Missouri. Turbidity increased downstream to the mouth of Big River, a distance of 71 miles from the mouth of Mill Creek. More than 85,000 fish worth approximately $12,000 were killed by suffocation in the 3 miles of Mill Creek and 9 miles of Big River immediately below the dam break (Ryck 1976a). The effect of barite tailings upon water quality in Big River and Mill Creek was investigated by sampling affected and unaffected areas. Because benthic organisms are relatively immobile and cannot quickly avoid harmful changes, their presence, absence or abundance reflects environmental conditions of the recent past (Chandler 1970). Population structure of the benthic invertebrate community also provides a measure of water quality conditions because different invertebrates have varying degrees of pollution

1This work was funded in part by Federal-Aid in Fish Restoration funds under Missouri's Dingell-Johnson Project F-19-R. 56 Transactions, Missouri Academy of Science Vol. 16 tolerance (Gaufin 1958). The severity and duration of damage at affected stations can therefore be documented by measuring density, diversity, and composition of benthic invertebrate communities and comparing these characteristics with unaffected communi­ ties upstream.

Description of the Study Area The sections of Big River and Mill Creek, one of its tributaries, encompassed by this study, geologically and physiographically belong to the St. Francois Knob and Basin Region (Sauer 1920). In this region, igneous rock knobs typically rise above basins formed by weak limestone and shale. The relief in this region is greater than any other region in the state, and in some areas resembles mountainous terrain. Two major soil types are dominant: very infertile crystalline rock soils which have little value for crops or wildlife and the Fredericktown group which contains the best residual soils of this Ozark Highland Province. Most available acreage of the latter soil type is farmed (Sauer 1920). Dominant terrestrial vegetation in this area is the oak-hickory climax forest (Collier 1955). Mill Creek begins in Washington County, near Potosi, Missouri, and flows northeast for about 10 miles before entering Big River, south of Blackwell, Missouri (Fig. 1). Mill Creek resembles most other Ozark streams in that it is normally alkaline, clear, and has dissolved oxygen levels at or near saturation. Substrate is similar to other Ozark streams, i.e., rubble and gravel-sized fragments of limestone and silica mineral$. Stream gradient is steep (27 feet per mile). Diatoms are the dominant aquatic vegetatibn. Some reaches of Mill Creek occasionally contain heavy growths of filamentous algae, especially during winter months. An earlier water quality survey revealed a benthic invertebrate community indicative of slightly polluted conditions resulting from intensive barite mining in the watershed (Kuester 1964). Big River originates in western Iron County and flows north through Washington, St. Francois, and Jefferson counties before entering the Meramec River near Eureka, Missouri. The portion included in this study drains 718 square miles at Desoto, Missouri (Fig. 1 ), and has an average discharge of 682 cubic feet per second (U. S. Geological Survey 1975). Physical characteristics of Big River are similar to those found in Mill Creek and other Ozark streams. Substrate in Big River upstream from Mill Creek is composed of small gravel and sand-sized fragments of limestone and Potosi druse with very few rubble-sized fragments. This substrate is a direct result of tailings deposition from upstream lead mining activities near Leadwood, Missouri, in the early 1960s (Kuester 1964). Kuester suggested that the unstable qualities of this substrate were responsible for reduced number of taxa found during his survey. Shortly after Kuester's study, lead mining ceased and substrate conditions improved by 1975. Substrate below the confluence of Mill Creek (Fig. 1) was similar to that found in Mill Creek and was not noticeably affected by mining activities of the early 1960s. Stream gradient is 5 feet per mile which is much more gradual than in Mill Creek. Diatoms were abundant and filamentous algae was less noticeable in Big River than in Mill Creek during winter months. Big River (upstream from Mill Creek) has a watershed three times as large as that of Mill Creek, resulting in a greater flow throughout the study. Dilution of the sediment was an important variable considered when effects of sediment on the benthos in these streams were assessed.

Materials and Methods Benthic invertebrates were sampled at three sites on Big River and two on Mill Creek (Fig. 1). The stations on Mill Creek were located above (1) and below (2) the inflow of 1982 Duchrow, Biological Sciences 59

Benthic invertebrate community characteristics were determined in the following manner. The number of mayfly and stonefly taxa and total taxa were simple counts of taxa in each sample. Density was expressed as the number of organisms per square foot for each sample collected at a station. Species diversity index values (d) were calculated for each sample using the equation derived by Margalef (1957) and discussed in detail by Wilhm (1967) and the U.S. Environmental Protection Agency (1973). In Missouri streams, diversity values have ranged from 0 (low) to 10 (high). Coefficients of similarity (C) described by Burlington (1962) were useful to determine the duration of damage caused by sediment discharged from the tailings pond by comparing invertebrate communities at two stations on the same date. In Missouri, coefficients of similarity greater than 50 are considered high, indicating that the 2 stations being compared have similar invertebrate communities. Values less than 50 indicate decreasing similarity in invertebrate communities being compared (Duchrow 1976a, 1976b, 1977, Dieffenbach and Ryck 1976). Information on invertebrate taxa other than mayflies and stoneflies that are pollution sensitive and typically found in unpolluted Missouri streams is published (Gaufin 1958, Roback 1962, Kuester 1964, Duchrow 1974, 1976a, 1976b, 1977, Ryck 1974, 1976b, Dieffenbach and Ryck 1976). Absence or reduction in abundance of these pollution­ sensitive taxa was used as further evidence of degradation in Mill Creek and Big River.

Results and Discussion

Mil/Creek Turbidity was significantly greater in the affected portion of Mill Creek for at least 4 days after dam failure (Fig. 2). Later, significant differences in turbidity (oc = 0.05) were noted following periods of increased discharge. Extremely high sediment concentrations, as indicated by turbidity values, caused initial loss of fish and aquatic invertebrates in the 3 miles of Mill Creek which received this sediment. The cause was determined using bioassay techniques with water samples collected from Mill Creek. Acute bioassay tests using bluegill (Lepomis macrochirus) showed immediate distress and death in 20 minutes at turbidity values of 75,000 JTU's.

75000 ~ I I~ I I \ ~ 30 I 1 \ ~Affected ,. I \ , 20 I \ ...e /·-- -e I '.-- ~ 10 ---- / ----- ...-=-Conl•ol 0 202 321 355 • •• Days after dam failure

Figure 2. A comparison of turbidity values determined from water samples collected at stations 1 (control) and 2 (affected) on Mill Creek during the study.

Analysis of invertebrate samples collected from the affected portion of Mill Creek (station 2) revealed a severe reduction in this community for the first 38 days following discharge of sediment. Coefficients of similarity (Fig. 3) comparing invertebrate community structure from the control (station 1) and affected stations on Mill Creek indicated very 60 Transactions, Missouri Academy of Science Vol. 16

7$

~

~ 50 ·e.." 0 <: "v 25 a; 0 u

0 14 31il 202 321 355 Days after dam failure figure 3. Coefficients of si:miiall"i.ty comparing invertebrate communities at stations 1 (control) and 2 (affected) on Mm Creek. dissimilar invertebrate communities for this 38-day period. Differences were primarily due to the reduced number of pollution intolerant mayfly and stonefly taxa in the affected portion (Fig. 4A). The number of mayfly and stonefly taxa at station 2 was much lower than criteria established for unpolluted Missouri streams (Table 1) and those found at station 1

A. Moylly and s!onelly taxa 20

"0.. E 0 Unpolluted "' min1mum ~ I 10 -- --- 0.."' a )( ....0

0 14• 311 83 144 202 264 321

0 B. Density 0 Station !(Control) 200 2 llliStation :!(Allee ted) ~ a ::> ~

g_ 100 :;:; ..s::; E z" ~ I I 0 14 3!11 83 144 202 264 3::!1 Days oiler dam failure figure 4. Biological characteristics of benthos samples collected at stations 1 (control) and 2 (affected) on Mm Creek. 1982 Duchrow, Biological Sciences 61

(Fig. 4A). A slight reduction in density (Fig. 48) was noted at the affected station for this time period when compared to the control. Species diversity as indicated by index values calculated for samples appeared unaffected in Mill Creek throughout the study since all values were above 4 and compared favorably with the control station. Reestablishment of invertebrates at station 2 began after 38 days. Coefficients of similarity comparing the control and affected stations on Mill Creek ranged from 35 to 45 for the next 226 days. The invertebrate community in the affected portion was becoming similar to the community found at the control station (Fig. 3). Additional evidence of recovery during this time period (83-264 days) was shown by comparing the number of mayfly and stonefly taxa between the control and affecting stations. Very little difference in this characteristic was noted between the 2 stations (Fig. 4A). Coefficients of similarity after 264 days showed similar invertebrate communities at stations 1 (control) and 2 (Fig. 3). This indicated that the invertebrate community in the affected area of Mill Creek had recovered from initial effects of sediment from the tailings pond. Values for the biological characteristics presented in Figure 4A-B also demonstrated this fact with the exception of mayfly and stonefly taxa 321 days after dam failure. Members of these two pollution sensitive insect orders may not have fully recovered from effects of the sediment deposited in the substrate at station 2. Different species of invertebrates in Mill Creek were affected differently by the sediment. Numerous researchers including Cordone and Kelly (1961), Hollis et al. (1964), Roback (1962), and Wilson (1960) have reported that adverse effects of sedimentation on aquatic life generally involve destruction of their habitat or their tolerance toward low oxygen concentrations. In other words, aquatic life with exposed gills or species that require a sediment free environment generally will be adversely affected by increases in the amount of sediment in a stream. Aquatic life with low oxygen tolerances, covered gills, or a preference towards silty environments will normally be affected positively if at all. Mayflies, stoneflies, and caddisflies are normally considered to be intolerant of pollution including sedimentation. A difference in abundance was observed for many of these groups between stations 1 and 2. The mayflies, Stenonema pulchellum, S. medio­ punctatum, Isonychia sp., Pseudocloeon sp., Ephemerella bicolor, and Stenacron interpunctatum were drastically reduced throughout the study, while Tricorythodes sp. and Caenis sp. appeared less affected. Other invertebrates such as: the Dobson-alderflies, Corydalus comutus and Nigronia serricomis; the caddisfly, Chimarra aterrima; the limpet, Ferrissia sp.; and crayfish, Orconectes sp. were also severely reduced (Duchrow 1978). Segmented worms (Class Oligochaeta) prefer silty substrates and normally tolerate low oxygen concentrations. These invertebrates appeared to do well in the affected portion of Mill Creek (Duchrow 1978). Invertebrates which showed no effects from the sediment were generally those with protected gills or those known to be silt tolerant, such as some Hydropsycid caddisflies (Roback 1962). Caddisflies (Cheumatopsyche sp., Symphitopsyche bifida, and Agraylea sp.), beetles (Stenelmis sp., Optioservus sandersoni, and Psephenus herricki), flies (Simuliidae, Chironomidae, and Tipula sp.), and an isopod (Asel/us sp.) were not reduced or enhanced (Duchrow 1978). In summary, the invertebrate community inhabiting the 3 mile portion of Mill Creek which received sediment from the tailings pond was drastically reduced for at least the first 38 days. The community did not fully recover from initial effects of sediment until321 days after dam failure. Aesthetic degradation from increased turbidity was apparent in the portion of Mill Creek which received sediment throughout the study. 62 Transactions, Missouri Academy of Science Vol. 16

Big River Turbidity in Big River was significantly different (oc = 0.05) between stations 3 (control) and 5 (affected) for the first 14 days (Fig. 5). After 14 days, differences were generally associated with periods of increased discharge. Initial turbidity in Big River was measured at 400 JTU' s. Elevated turbidity in Big River was documented by aerial photographs to its mouth, 72 miles downstream. Throughout the study, turbidity values in the reaches of Big River which received sediment consistently exceeded those found at the control station (Fig. 5). This caused a decrease in aesthetic value and fishability of affected reaches of Big River due to resuspended sediment during periods of increased flow (Personal communi­ cation 1976, Walter Klinkhardt, Conservation Agent, Missouri Department of Conserva­ tion, Farmington, Missouri).

400

0 4 14 3S 109 179 235 293

Days after dam failure "'

Figure 5. A comparison of turbidity values determined from water samples collected from Big River at stations 3 and 5, 1975-1976.

Analysis of invertebrate data showed that invertebrates at station 4, downstream from the mouth of Mill Creek, were noticeably affected by sediment for at least 14 days (Fig. 6A). The coefficient of similarity comparing this station with the control (station 3) showed dissimilar invertebrate communities at the 2 stations (C = 14). This effect was not noted initially further downstream at station 5 since the coefficient comparing stations 3 and 5 was similar (C = 51). After 14 days, the benthos community at station 5 gradually became dissimilar when compared to the control station. The greatest dissimilarity was noted 179 days (Fig. 6B). This could be an effect caused by sediment or a sampling artifact caused by differences in invertebrate density between the 3 stations (Fig. 7A). Sediment affected the invertebrate community at station 4 during the first 14 days by reducing the invertebrate density to only 6 organisms per square foot (Fig. 7 A). Total number of taxa at this station was reduced; however, mayfly and stonefly taxa (Fig. 7B) did not appear to be affected. Species diversity at the 3 stations on Big River was not affected throughout the study. No individual taxa collected from Big River clearly showed adverse effects as those found in Mill Creek. This further indicated that the degree and duration of damaging effects were Jess than in Mill Creek. Dilution of the sediment in Big River by greater flow and further distance from the tailings pond was the probable reason. In summary, the effects of sediment on invertebrates in Big River at station 4 immediately downstream from Mill Creek were severe for 14 days following dam failure. The invertebrate community at this site recovered rapidly after the 14 days as indicated by 1982 Duchrow, Biological Sciences 63

A. .:: b 75 "E

~ - 50 0

" 25 "v -iii 0 0 u 14 :n 109 179 235

El . .:: ~ 75 "E "'

i.: -~ 25 iii 0 0 u 14 38 179 Days alter dam failure Figure 6. Coefficients of similarity comparing benthic invertebrate communities at stations (A.) 3 vs 4 (B) 3 vs 5 on Big River.

D Station 3 (Conlrol) 1!11 Station 4(AIIected) A. Density 0 Station S(Aflected) - 200 0 .2 ~ 0 :>

..0 "E :> z 355

El. Mayfly and stonelly taxa D Station 3 11!111 Station 4 CJ Station 5 Unpolluted ii" E minimum ~ 0

"'"- 10 - "0. " 0 o-" :JJ

0 14 3!1 179 235 293 355

Days after dam failure figure 7. Biological characteristics of benthos samples collected at stations 3 (control), 4, and 5 on Big River. 64 Transactions, Missouri Academy of Science Vol. 16 coefficients representing similar communities for the remainder of the study (Fig. 6A). Recovery appeared delayed for 179 days at station 5 (8 miles downstream), however, this could be an artifact of sample since invertebrate density at the control station was unexplainably low. The most severe effect of sediment on Big River was the decreased aesthetics and fishability of this river from turbid water.

Conclusions Sediment that entered Mill Creek and Big River because of a collapsed tailings pond dam in 1975 caused noticeable damage to invertebrate communities of these streams. Severe damage was evident in Mill Creek for at least 38 days. Dead fish were observed initially, and the invertebrate community remained drastically reduced. Recovery of the invertebrate community in the affected portion of Mill Creek required at least 264 days. After that, no noticeable difference in invertebrate community structure between the control and affected station was found. Damage in Big River was evident during the first 14 days. Dead fish and a reduced invertebrate community at station 4 were observed during the initial 14 days. After that, noticeable differences were not observed in the aquatic life between the control and this affected station on Big River. The aesthetic value of Big River was reduced throughout the study (355 days), since water downstream from Mill Creek was consistently more turbid than that at the control, station 3.

Literature Cited American Public Health Association. 1971. Standard methods for the analysis of water and wastewater, 13th Ed. Am. Public Health Assoc., Washington, DC 874pp. Anderson, R. 0. 1959. A modified flotation technique for sorting bottom fauna samples. Limnol. and Oceanog. 4:222-225. Burlington, R. F. 1962. Quantitative biological assessment of pollution. J. Water Poll. Cont. Fed. 34:179-183. Chandler, J. R. 1970. A biological approach to water quality management. J. Water Poll. Cont. Fed. 42:415-423. Collier, J. F. 1955. Agricultural atlas of Missouri. Univ. Missouri Agr. Exp. Sta. Bull. No. 645. 75pp. Cordone, A. J. and D. W. Kelly. 1961. The influence of inorganic sediment on the aquatic life in streams. California Fish and Game 47:189-228. Dieffenbach, W. H. and F. M. Ryck, Jr. 1976. Water quality ofthe Elk, James, and Spring river basins, 1964-1965. Missouri Dept. Conserv. Aquatic Ser. 15. 25pp. Duchrow, R. M. 1974. Water quality of the North, Salt, and Cuivre river basins. Missouri Dept. Conserv. Proj. F-1-R-22, Study W-1, Job No. 3, Final Report. 33pp. -----· 1976a. Water quality of Prairie, Cowskin, and Beaver creeks, Douglas County, Missouri. Missouri Dept Conserv. D-J Proj. F-19-R-2, Study W-3, Job No. 3. 21pp. -----· 1976b. Water quality of Bryant and Hunter creeks. Missouri Dept. Conserv. D-J Proj. F-19-R-2, Study W-3, Job No. 3. lOpp. -----· 1977. Water quality ofthe Current, Jack's Fork, Eleven Point, Little Black, and Warm ForkoftheSpringrivers, 1974. Missouri Dept. Conserv. D-J Proj. F-19-R-2, Study W-1, Job No.1, Final Report. 80pp. -----· 1978. The effects of a barite tailings pond dam failure upon the water quality of Mill Creek and Big River, Washington County, Missouri. Presented at the 26th Annual Meeting of the North American Benthological Society, May 10-12, 1978, Winnipeg, Manitoba, Canada. 48pp. Gaufin, A. R. 1958. The effects of pollution on a midwestern stream. Ohio J. Sci. 58:197-208. Hollis, E. H., J. G. Boone, C. R. DeRose and G. J. Murphy. 1964. A literature review of the effects of turbidity and siltation on aquatic life. Dept. Chesapeake Bay Affairs, Annapolis, Maryland Staff Report, December 1964. 26pp. (Mimeo) 1982 Duchrow, Biological Sciences 65

Kuester, D. R. 1964. Part V. The benthos of the streams in the Meramec River basin as related to water quality. PP. 24-40, in Water quality in the Big, Bourbeuse, and Meramec river basins. Dept. Public Health Board, Jefferson City, Missouri. Margalef, R. 1957. La teoria de a! informacions en ecologia. Mem. Real. Acad. Ciencias y Artes de Barcelona 32:373-499. Roback, S. S. 1962. Environmental requirements ofTrichoptera. PP. 118-126, in Biological problems in water pollution, 3d Seminar, August 13-17, 1962. U.S. Dept. Health, Education, and Welfare, Public Health Serv. Ryck, F. M., Jr. 1974. Water quality survey of the southeast Ozark mining area, 1965-71. Missouri Dept. Conserv. Aquatic Ser. 10. 22pp. -----· 1976a. Missouri fish kill investigations, 1975. Missouri Dept. Conserv. D-J Proj. F-19-R-2, Study W-3, Job No. 2. 9pp. -----· 1976b. Water quality of Big Piney River. Missouri Dept. Conserv. Proj. F-19-R-2, Study W-4, Job No. 1. 36pp. Sauer, C. 0. 1920. The geography of the Ozark Highland of Missouri. Univ. Chicago Press, Chicago, Illinois. 245pp. U.S. Environmental Protection Agency. 1973. Biological field and laboratory methods of measuring the quality of surface water and effluents. Environmental Monitoring Ser. EPA-670/4-73-001, July 1973. 176pp. U. S. Geological Survey. 1975. Water resources data for Missouri. Water Year 1975. U. S. Geological Survey, Water data report M0-75-1. U. S. Geological Survey, Rolla, Missouri. 362pp. Wilhm, J. L. 1967. Comparison of some diversity indices applied to populations of benthic macroinvertebrates in a stream receiving organic wastes. J. Water Poll. Cont. Fed. 39:1673-1683. Wilson, J. 1960. Informal group discussion of, The effects of erosion, silt, and other inert materials on aquatic life, PP. 269-271, in Tarzwell, C. M. (ed.), Biological problems in water pollution. Trans. 2nd Seminar on Biological Problems, April 20-24, 1959. Robert A. Taft Sanitary Engineering Center, U. S. Department of Health, Education, and Welfare, Cincinnati, Ohio, Technical Report. W60-3.

1982 Hoyer & Reid, Biological Sciences 67

Seasonally flooded Missouri Wetlands: Early Winter Limnological Characteristics in Relation to Physiographic Regions1

Mark V. Hoyer School of Forestry, Fisheries and Wildlife University of Missouri-Columbia Columbia, Missouri 65211 and Frederic A. Reid Gaylord Laboratory University of Missouri-Columbia Puxico, Missouri 63960

Abstract: A limnological investigation of 10 Missouri man-made, seasonally flooded wetlands was conducted in December 1981. Conservative ions showed little temporal variability, while suspended solids, sulfate and phosphorus showed high temporal variability during the period of hydrophyte senescence and vertical mixing by wind. Statistical differences existed in the chemical composition of wetlands in different physiographic regions. Wetlands in Southeast Lowlands had lower ionic concentration than Dissected Till Plain of northern Missouri. A comparison of Mississippi and Missouri river wetlands, within the same physiographic region, revealed no differences. A comparison of these data and published data from other investigations reveal extreme variability among wetland types. Key Words: Moist soil management, water chemistry, Missouri and Mississippi river wetlands.

Introduction A large percentage of North American natural wetlands has been lost or greatly modified within the past century as a result of intense agricultural, urban and industrial developments. In Missouri, 50% of the wetlands along the Missouri River were lost by 1972 (Funk and Robinson 1974), 94% of lowland hardwood wetlands in southeast Missouri were lost between 1870 and 1975 (Korte and Fredrickson 1977) and 95% ofthe wetlands between Quincy and Hamburg, Illinois, along the Mississippi River were leveed or drained by 1982 (Reid, unpublished data). Although the ecological functions of natural wetlands cannot be duplicated, seasonally flooded, man-made impoundments have proven effective in recent wetland restoration activities. These management practices, known as moist soil techniques (Bellrose et a!. 1979), promote extensive animal usage through dynamic water regimes and establishment of emergent plant zones (Fredrickson and Taylor 1982). However, there is little information on the chemical or biological processes of these managed systems. Abundance and distribution of aquatic plants (Moyle 1945) and animals (Minshall and Minshalll978) are influenced by chemical characteristics of the water. Recent reviews of

1Contribution from Missouri Cooperative Wildlife and Fishery Research Units, Project funded by Missouri Department of Conservation-Federal Aid Project W-R-13, Gaylord Memorial Laboratory, Missouri Cooperative Wildlife Research Unit and Missouri Agricultural Experiment Station (Projects 176 and 183)---Journal Series 9129. 68 Transactions, Missouri Academy of Science Vol. 16 wetlands management (Weller 1978, Fredrickson and Drobney 1979) identify water chemistry investigation as a priority in research on the adaptions of aquatic plants and animals to dynamic wetland ecosystems. These reviews stress the need for research over broad geographic areas, and during little studied periods such as the period of vegetative senescence. The objectives of this study were to: (1) establish base-line water chemistry information on man-made, seasonally flooded wetlands of Missouri during early winter, and (2) determine if diffe~ences exist in water chemistry of wetlands in relation to physiographic regions of the state.

Methods Man-made seasonally flooded wetlands, managed by moist soil techniques, are referred to as seasonally flooded wetlands in this paper. Surface waters of 10 Missouri seasonally flooded wetlands were sampled in December 1981 (Fig. 1, Table 1). Sites were located on 3 national wildlife refuges (NWR) and 6 state wildlife management areas (WMA). Wetlands were located in 3 of the 4 physiographic regions of Missouri (according to Fenneman 1938), Dissected Till Plain (DTP), the Osage Plains (OP) and the Southeast Lowlands (SEL). More recent physiographic divisions ofthe state (Thorn and Wilson 1980) increase physiographic regions to six, but such increased divisions do not change our wetland groupings within a region. So for clarity, we use Fenneman's terminology. The

DISSECTED TILL PLAIN

OUNTAIN GROVE (WMA). SWAN LAKE (NWR) • • LARKSVILLE (WMA)

OSAGE PLAINS

OZARK UPLANDS

Figure 1. Physiographic regions of Missouri (after Fenneman 1938) and location of man-made, seasonally flooded wetlands sampled in this study...... \0 Table 1. Chemical characteristics of man-made seasonally flooded wetland waters in Missouri. 00 1\:)

Dissected Till Plain Dissected Till Plain Osage Plains Southeast Lowlands Mississippi Missouri Ted Clarence Clarksville Marais Temps Fountain Swan Shell- Duck Mingo- Mingo- Total Sites Shanks Cannon Refuge Clair Grove Lake Osage Creek (Unit7) (Unit2) Combined

NumberofSamples (n) (3) (3) (3) (3) (3) 3) (3) (3) (3) (3) (30) Calcium Hardness 3 (mg/1 as CaC0 ) 47.7±1.5 47.3±9.1 149.3±82.1 63.0±4.6 48.0±5.6 34.0±14.8 48.7±6.3 51.3±16.8 9.0±4.4 13.3±8.1 51.2±38.4 :r:: Magnesium Hardness ~ (mg/1 as CaC03) 16.0±1.0 9.0±2.6 61.3±8.1 27.0±6.1 9.0±2.6 11.3±2.3 8.0±1.7 26.0±4.6 8.6±8.1 11.7±12.4 18.8±16.5 ....(tl Total Alkalinity R" (mg/1 as CaC03) 115.3±4.2 117.7±57.5 335.0±47.8 181.7±3.1 104.3±8.5 69.0±36.4 96.3±12.3 133.7±64.1 24.0±3.6 33.7±10.4 121.1±88.4 ~ pH 7.7±0.1 7.7±0.2 8.0±0.3 7.9±0.3 7.5±0.1 7.2±0.1 7.6±0.1 7.5±0.5 6.8±0.4 7.0±0.1 7.5±0.4 .?- !P. Specific Conductance §- (umho/cm2 at 250°C) 275±11 272±109 729±268 359±3 237±32 245±171 247±15 281±126 87±21 113±35 285±175 cs. () Turbidity (JTU) 18.7±2.1 25.7±2.3 47.0±39.6 4.7±0.6 62.7±50.3 126.0±49.4 176.7±15.3 14.3±10.0 140.0±88.5 49.0±9.0 66.5±60.0 5!!.. Chloride (mg/1) 9.8±2.4 15.8±11.4 19.7±20.3 7.2±3.0 8.2±0.8 8.3±4.8 5.8±0.6 6.3±1.0 4.0±0.0 7.3±2.8 9.2±4.8 (/l Q. (tl Sulfate (mg/1) 17.3±3.5 32.0±4.0 65.3±66.9 8.7±2.5 13.3±2.5 27.7±3.5 24.0±3.0 0.0±0.0 12.7±11.0 14.3±3.5 21.5±18.0 ;::! () Dissolved Silica !Jl (mg/1) 8.7±1.3 5.2±2.9 19.9±13.8 4.5±0.8 7.6±4.3 6.9±3.2 2.5±0.4 15.3±4.4 14.6±5.9 11.8±8.3 9.7±5.5 Total Phosphorus (mg/m2) 157.1±42.1 103.0±22.6 377.8±283.2 61.8±22.0 238.5±157.3 343.1±146.2 317.0±142.4 259.7±287.0 135.5±27.3 143.5±55.7 213.7±108.9 Inorganic Suspended Solids (mg/1) 22.6±1.4 39.6±11.7 103.1±93.0 5.8±2.2 195.4±71.3 164.1±54.1 40.0±37.5 15.6±14.3 124.3±136.9 80.9±72.4 79.1±65.9 Organic Suspended Solids (mg/1) 17.0±4.5 15.2±1.0 29.3±25.0 3.6±0.5 49.9±33.7 29.5±11.0 24.5±23.5 7.6±4.3 32.4±22.2 18.0±11.2 22.7±13.5 Organic Color (Pt Units) 20±01 22±3 26±10 18±6 32±8 42±20 38±3 85±82 68±26 78±8 43±25

Data Represented mean ± standard deviation a- \0 70 Transactions, Missouri Academy of Science Vol. 16

relation of these sites to major river systems and physiographic regions is shown (Fig. 1, Table 1). The lack of seasonally flooded wetlands within the fourth region, Ozark Uplands, precluded comparative sampling. Natural wetlands in that region consist of oxbows and small sink-holes (L. Fredrickson, pers. comm.) and were not sampled. Within the DTP, wetlands were classified as Mississippi or Missouri river marshes. Two of the sites, Swan Lake NWR and Fountain Grove WMA are directly related to the Grand River, but were classified as Missouri River wetlands according to their watershed location. All sampled wetlands were flooded at a depth of less than 70 em. Dominant vegetation consisted of grasses (Echinochloa sp. and Leersia sp.), smartweed (Polygonum sp.) and sedges (Cyperus sp. and Carex sp. ). Following 3 years of drought in Missouri, precipitation and surface water flooding were above mean levels in 1981. Although no direct data are available, rainfall, head water or backwater flooding sources differed both within and among sampled managed sites. Three water samples were collected from each wetland in acid-cleaned plastic containers on 3 separate dates. During December 1981, selected sites were flooded between 20 and 50 em. Chemical analysis was conducted on composite water samples within a single sampling date. Sampling was conducted in December because it is a period of peak usage by migratory mallards (Anas platyrhynchos) in Missouri (Bellrose 1976), a high priority waterfowl species for the state's wetland management program, and coincides with senesceooe-cf aquatic vegetation. Collection dates were separated by at least 4 days within a site. Samples were kept on ice until laboratory analysis 2-5 days later. Dissolved silica (mg/1), total and calcium hardness (mg/1 as CaC03), sulfate (mg/1) and chloride (mg/1) concentrations were determined using standard methods (Hach Chemical Company 1978). Magnesium hardness was assumed to be the difference between total and calcium hardness. Total alkalinity (mg/1 as CaC03) was determined by titration with 0.02 N sulfuric acid to a pH of 4.5, using a type 26 Radiometric Copenhagen pH meter. Turbidity (JTU) was measured using a Hach 2100 turbidimeter. Specific conductance J.Lmho/cm2 at 25°C) was measured using a type CDM2e conductivity meter. Organic color (Pt units) was determined using matched Nessler tubes and platinum-cobalt standards (APHA 1976). Concentrations of organic and inorganic suspended solids (mg/1) were determined by filtering a known volume of water through precombusted (550°C), preweighed 934 AH glass fiber filters. Filters were weighed after drying at 103°C for 1 hour and again after combustion at 550°C for 1 hour. Inorganic suspended solids were determined by the difference of the previous 2 measurements. Total phosphorus (mg/m3 ) was determined using the procedures of Murphy and Riley (1962) following persulfate oxidation (Menzel and Corwin 1965). After oxidation, samples were decanted to remove suspended particles. Prior to all statistical analyses these data were transformed to base 10 logarithms to meet assumptions required for parametric statistics. Statistical analysis was performed using a SAS computer package (Statistical Analysis System 1979). Statements of statistical significance imply P~O.OS.

Results and Discussion Results of the chemical analysis are summarized in Table 1. The mean specific conductance, total alkalinity, calcium and magnesium hardness, sulfate and chloride concentrations of all 10 seasonally flooded sites are similar to the mean values of 43 Missouri reservoirs (Jones 1977). Data from both studies, however, are extremely variable among sites. As an example, wetland values for total ion concentration (as measured by 1982 Hoyer & Reid, Biological Sciences 71 specific conductance), turbidity and total phosphorus range from 87 to 359 J,Lmho/cm2, 4.7-176.7 JTU and 61.8-343.1 mg/m3 respectively. To evaluate physiographic differences, we divided the Missouri sites by physiographic regions (Fig. 1) and statistically compared mean chemical characteristics between DTP and SEL sites using the T-test (Table 2). Because only one site was sampled in OP, statistical comparisons are inappropriate. SEL sites showed lower ionic concentrations than DTP sites. These data compare closely with data on Missouri rivers (Jones 1977), and both studies show that SEL surface waters have significantly less chloride, specific conductance, total alkalinity and calcium hardness than DTP sites. Because most Missouri wetlands are riverine in formation (Thorn and Wilson 1980) this close relationship is not surprising. The SEL wetland sites also had significantly more silica and color and lower pH than DTP wetlands. The chemical composition of Mississippi and Missouri river seasonally flooded wetlands, within the DTP, were not significantly different. This emphasizes the importance of physiographic location to the chemical characteristics of surface water. Regional differences in surface water chemistry are most commonly attributed to differences in geologic mineral composition of watersheds (Moyle 1956, Armstrong and Schindler 1971, Jones and Bachmann 1978). The DTP of Missouri is characterized by numerous stream valleys in recessional moraines of pre-Wisconsonian continental glaciation. Both the DTP and SEL are dominated by late Wisconsin outwash (Robertson 1938), while the OP, an unglaciated region, and the DTP contain loess deposits. We

Table 2. Comparison of means for chemical variables in physiographic regions-Dissected Till Plain and Southeast Lowlands.

Significant difference 1 Dissected Southeast by Student's Chemical variable Till Plain Lowlands T-test

Ca hardness (mg/1) 64.9 ± 11.62 24.6 ± 7.4 * Mg hardness (mg/1) 22.3 ± 4.6 15.4 ± 3.7 pH 7.7 ± 0.1 7.1 ± 0.2 Alkalinity (mg/1) 153.8 ± 22.3 63.8 ± 20.6 *

Conductivity (J.Lmho/cm2 ) 353 ±50 160 ± 37 Turbidity (JTU) 47.4 ± 11.7 67.8 ± 23.9 Chloride (mg/1) 11.5 ± 2.2 5.9 ± 0.7 * Sulfate (mg/1) 27.4 ± 7.1 9.0 ± 3.0 Silica (mg/1) 8.8 ± 1.8 13.9 ± 1.9 Color (Pt Units) 26.7 ± 2.8 77.2 ± 14.6 *

Total P (mg/m3 ) 214 ± 41 180 ±53 Inorganic solids (mg/1) 88.4 ± 20.3 73.6 ± 30.4 Organic solids (mg/1) 24.1 ± 5.0 19.3 ± 5.5

1 p~0.05. 2 Mean ± 1 SE. 72 Transactions, Missouri Academy of Science Vol. 16 suggest that watershed geologic minerals and associated ionic composition of flood waters are the probable source of physiographic water chemical differences of Missouri seasonally flooded wetlands. Fluctuating water levels, dilution rates and evaporation may also influence chemical composition of wetlands. Such a phenomenon was recorded during drought conditions of Oklahoma wetlands where regional differences in conductivity occurred (Heitmeyer and Vohs 1981). To further examine sources of variation in these data, we divided the total observable variance for all parameters into the following variance components, using a nested analysis of variance (SAS 1979): (1) variance among regions, (2) site within regions and (3) temporal variance within sites and regions. Table 3 shows the components expressed as a percent of total variance. Calcium hardness, pH, alkalinity, specific conductance, dissolved silica and organic color showed little temporal variance. Because these variables are closely related to geologic characteristics of the watershed, significant differences between regions (Table 2) may be expected. Sulfate, total phosphorus, inorganic and organic suspended solids showed a high temporal variance within sites and regions. Hydrophyte senescence and mixing by wind and rain are the probable source of such temporal variance within these shallow water bodies. Despite the small sample size and short sampling period of this investigation, the high temporal variance of nutrients and solids during early winter suggests future sampling programs to quantify chemical characteristics of wetlands should be intensive. While data from these 10 seasonally flooded basins demonstrate extreme differences for certain chemical characteristics (Table 1), wetlands with different geologic formations, flooding regimes and groundwater inputs may show even greater differences. Table 4lists

Table 3. Percent contribution of various sources to the total variance of each chemical. parameter.

Site Date Within Within Sites and Variable Region Regions Regions

Ca Hardness 70 14 16 Mg Hardness 18 59 22 pH 68 22 10 Alkalinity 64 27 9 Conductivity 57 23 20 Turbidity 21 62 17 Chloride 17 0 83 Sulfate 0 58 42 Silica 38 23 39 Color 77 8 15 Total P 0 61 39 Inorganic Solids 0 41 59 Organic Solids 0 39 61 ,_. '!:! Table 4. Chemical characteristics of certain non-forested wetlands in the Northern Hemisphere. 00 N

Ca Mg Alkalinity Conductivity Chloride Sulfate Total P Si Site Type of Wetland (mg/1) (mg/1) (mg/1 as CaC03) umho/cm2 ) pH (mg/1) (mg/1) (mg/m3 ) (mg/1) Source

Missouri, USA Emergent, moist 51.2±38.4 18.8±16.5 121.1±88.4 285±175 7.5±0.4 9.2±4.8 21.5±18.0 213.5±108.9 9.7±5.5 This study soil

Theresa Marsh, Emergent Riverine 56-168 23-73 268-420 519-942 7.4±8.2 17-54 15-99 110-690 Klopatek 1978 WI, USA Marsh

Holland Marsh, Emergent Riverine 435 6.4 10.5 Nicholls and ::r: 0 ONT, Canada MacCrimmon 1974 '< ~ Smartwood & FFA Emergent, moist 40-150 Cook and Powers go Marshes, NY, USA 1958 ::0 p:"' Oklahoma, USA Emergent Palustrine 100.00 159.0 6.0 Heitmeyer and OJ moist soil 30.0 125.0 5.0 Vohs 1981 i5' 0 Storkersen Emergent Shallow 74 730 8.7 Bergman et a!. 'fl. () Point, AK, USA Tundra Wetlands- 405-1370 8.5±8.9 1977 e>.. Coastal Uplands (fJ 0. Houghton Lake, Peatland 19.3±10.7 3.9±1.8 27.9±24.6 Richardson et a!. :l "'() MI, USA 1978

Dale Park & Bog 3.0±1.9 68±12 4.7±1.0 Gorham 1956 Rushland Moss, Lake Dis!., UK

Lake Agassiz, Bog 1.7±0.6 0.2±0.2 54±15 3.5±0.3 Heinselman 1970 MN, USA

Data represented as mean, mean ± S.D., or range --Jw 74 Transactions, Missouri Academy of Science Vol. 16 published data on several non-forested wetlands. These data suggest that direct comparisons of wetland chemical conditions consider input and functional characteristics as well as physiographic location. This survey indicates physiographic differences in certain water chemical parameters and suggests the need for further limnological research on seasonally flooded wetlands. Data are needed on chemical variability (spatial and temporal) within a functional marsh complex, and chemical differences related to basin morphology, depth and duration of flooding, source of water, macrophyte growth and senescence should be investigated. This information will increase our understanding of wetland water chemistry and potential biological responses, thereby aiding in productive moist soil management.

Acknowledgments Advice and comments, as well as laboratory facilities were provided by J. R. Jones. L. H. Fredrickson provided additional comments and insights that were essential to the project. Credit for assistance is given to management personnel on all the wetland areas and the Gaylord Laboratory, but especially to M. E. Heitmeyer, J. Kuykendall, T. L. LaRue, W. R. Leonard, T. Liefield, J. C. Ware, J. R. Wombwell and G. Wylie. Statistical assistance was graciously provided by J. Burroughs. Several reviewers, including J. E. Boyles, P. W. Brown, L. H. Fredrickson, M. E. Heitmeyer, D. D. Hum berg, J. R. Jones and D. J. Norris provided helpful suggestions. S. S. Clark and L. L. Hoyer also provided editorial assistance and typed the manuscript.

Literature Cited American Public Health Association. 1976. Standard methods for the examination of water and waste water. 14th ed., New York. APHA 1193pp. Armstrong, F. A J. and D. W. Schindler. 1971. Preliminary chemical characterization of waters in the Experimental Lakes Area, northwestern Ontario. J. Fish. Res. Bd. Canada 28:171-187. Bellrose, F. C. 1976. Ducks, geese and swans of North America. Stackpole Books, Harrisburg, PA 544pp. -----,F. L. Paveglio, Jr. and D. W. Steffeck. 1979. Waterfowl populations and the changing environment of the Illinois River valley. Ill. Nat. Hist. Surv. Bull. 32:1-54. Bergman, R. D., R. L. Howard, K. F. Abraham andM. W. Weller. 1977. Waterbirds and their wetland resources in relation to oil development at Storkersen Point, Alaska. U.S. Fish. Wild!. Serv. Resour. Publ. 129:1-38. Cook, A H. and C. F. Powers. 1958. Early biochemical changes in the soils and waters of artificially created marshes in New York. N. Y. Fish and Game J. 5:9-65. Fenneman, N. M. 1938. Physiography of Eastern United States. McGraw-Hill Book Co., New York. 714pp. Fredrickson, L. H. and R. D. Drobney. 1979. Habitat utilization by postbreeding waterfowl. Pages 119-129 in T. Bookhout (ed.), Waterfowl and wetlands-an integrated review. _____ and T. S. Taylor. 1982. A management handbook for seasonally flooded impound­ ments. U.S. Fish and Wild!. Serv., Resource Publ. 148. 29pp. Funk, J. L. and J. W. Robinson. 1974. Changes in the channel of the lower Missouri River and effects on fish and wildlife. Mo. Dept. of Conserv. Aquatic Series #11. Gorham, E. 1956. The ionic composition of some bog and fen waters in the English lake district. J. Ecol. 44:142-152. Hach Chemical Company. 1978. Water and wastewater analysis procedures. 4th ed. Hach Chern. Co., Ames, Iowa. Heinselman, M. L. 1970. Landscape, evolution, peatland types, and the environment in the Lake Agassiz Peatland Natural Area, Minnesota. Ecol. Monogr. 40:235-261. 1982 Hoyer & Reid, Biological Sciences 75

Heitmeyer, M. E. and P. A. Vohs, Jr. 1981. Environmental characteristics of Oklahoma wetlands. Environ. Series #5, Arts and Sciences Research, Okla. State Univ., Stillwater. 143pp. Jones, J. R. 1977. Chemical characteristics of some Missouri reservoirs. Trans. Mo. Acad. Sci. 11:58-71. _____ and R. W. Bachmann. 1978. Trophic status of Iowa lakes in relation to origin and glacial geology. Hydrobiol. 57:267-273. Klopatek, J. M. 1978. Nutrient dynamics of freshwater riverine marshes and the role of emergent macrophytes. Pages 195-216 in R. E. Good, D. F. Whigham and R. L. Simpson (eds.), Freshwater wetlands: ecological processes and management potential. Acad. Press, New York. Korte, P. A. and L. H. Fredrickson. 1977. Loss of Missouri's lowland hardwood ecosystem. Trans. N. Am. Wild!. Nat. Resour. Conf. 42:31-41. Menzel, D. W. and N. Corwin. 1965. The measurement of total phosphorus in seawater based on the liberation of organically bound fractions by persulfate oxidation. Limnol. Oceanogr. 10:280-282. Minshall, G. W. and J. N. Minshall. 1978. Further evidence on the role of chemical factors in determining the distribution of benthic invertebrates in the River Duddon. Arch. Hydrobiol. 83:324-355. Moyle, J. B. 1945. Some chemical factors influencing the distribution of aquatic plants in Minnesota. Am. Midi. Nat. 34:402-420. _____ 1956. Relationship between the chemistry of Minnesota surface waters and wildlife management. J. Wild!. Manage. 20:303-320. Murphy, J. and J. P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta. 27:31-36. Nicholls, K. H. and H. R. MacCrimmon. 1974. Nutrients in subsurface and runoff waters of the Holland Marsh, Ontario. J. Environ. Qual. 3:31-35. Richardson, C. J., D. L. Tilton, J. A. Kadlec, J. P. M. Chamie and W. A. Wentz. 1978. Nutrient dynamics of northern wetland ecosystems. Pages 217-241 in R. E. Good, D. F. Whigham and R. L. Simpson (eds.), Freshwater wetlands: ecological processes and management potential. Acad. Press, New York. Robertson, P. 1938. Some problems of the middle Mississippi River region during Pleistocene time. Trans. St. Louis Acad. Sci. 29:165-240. Statistical Analysis System. 1979. SAS user's guide. SAS lost., Inc., Raleigh, Northern Carolina. 494pp. Thorn, R. H. andJ. H. Wilson. 1980. The natural divisions of Missouri. Trans. Mo. Acad. Sci. 14:9-23. Weller, M. W. 1978. Management offreshwater marshes for wildlife. Pages 267-284 in R. E. Good, D. F. Whigham and R. L. Simpson (eds.), Freshwater wetlands: ecological processes and management potential. Acad. Press, New York.

1982 Norris & Elder, Biological Sci_ences 77

Distribution and Habitat Characteristics of the Painted Bunting in Missouri

Douglas J. Norris and William H. Elder School of Forestry, Fisheries and Wildlife University of Missouri-Columbia 112 Stephens Hall Columbia, Missouri 65211

Abstract: The approximate distribution of the painted bunting (Passerina ciris) in Missouri is described, based upon a number of sources. The distribution is "L" -shaped and encompasses a southern band of counties, and a western band of counties north to Kansas City. Characteristics of painted bunting territories and habitat are described_ Of 19 territories examined in Taney County, 18 were comprised of two distinct habitat types-a large old field component (x = 82% of total territory) and a smaller woodland component (x = 18%). Percent ground cover averaged 81% (old field component) and was fairly consistent between territories (C.V. = 24.0). Since birds foraged primarily within the ground cover, this suggests the importance of this variable in habitat selection. However, overall vegetative characteristics varied widely between territories, indicating rather broad habitat requirements. Key Words: Painted bunting, distribution, habitat, territory

Increased public interest in nongame birds calls for the consideration of these species' habitat requirements by resource management agencies. Adequate information to develop management schemes for nongame birds, however, is often lacking. A case in point is the painted bunting (Passerina ciris), a species of the southern United States, whose breeding range reaches southern and western Missouri. This study was undertaken to delineate the range of the painted bunting in Missouri and to determine its habitat requirements.

Acknowledgments This study was funded by U.S.D.A. Forest Service Cooperative Research Agreement No. 13-666 and by the Missouri Cooperative Wildlife Research Unit. Thomas Baskett, Frederic Reid and Jane Choromanski kindly reviewed the manuscript. We also thank John Faaborg and Charles Thompson for their comments.

DISTRIBUTION OF THE PAINTED BUNTING IN MISSOURI The Missouri range of the painted bunting was delineated by checking u.s_ Fish and Wildlife Service banding records (from the Bird Banding Laboratory at Laurel, Maryland), by consulting knowledgeable bird watchers throughout the state and by soliciting sightings through a notice in The Bluebird (publication of the Missouri Audubon Society). Also, pertinent literature was searched for past records of the bird within the state. The first recorded observation of breeding painted buntings in Missouri was in 1917 in Oregon county (Bennitt 1932). It is unclear whether the painted bunting extended its range into Missouri from Arkansas. Widmann (1907) suspected the bird's presence in southern 78 Transactions, Missouri Academy of Science Vol. 16

Missouri but had no observations. Other early records in Missouri were reported by Bennitt (1932) and Cunningham (1937). A number of additional sightings have been reported in The Bluebird since 1946 (Table 1). These sightings, banding records, personal communications and observations during this study (Table 1) provided the basis for the range map of the painted bunting in Missouri as shown in Fig. 1. Populations of painted buntings are low and irregularly distributed in Missouri, making it difficult to accurately delineate the species' range. Therefore, the map shows no exact boundaries. Factors responsible for the observed distribution are unclear, as it does not appear to follow any physical or biogeographical features within the state.

figure 1. Painted bunting records for Missouri. Records are keyed by number to those in Table 1. Squares indicate breeding records, circles indicate unspecified or other observations. Inset shows painted bunting range in contiguous states. 1982 Norris & Elder, Biological Sciences 79

Table 1. Painted bunting records for Missouri. Thi.s listillllg shows mnly the records compiled during this study and is undoubtedly incomplete.

County County* Name Number Record Source

Atchison 1 1 female banded (1966) H. & F. Diggs-pers. comm. USFWS Bird Banding Lab. Barry 2 5 sightings (1980-1981) present study 2 young reported Anderson 1966 numerous sightings D. Easterla-pers. comm. Clay 3 nesting pair observed D. Easterla-pers. comm. Dent 4 unspecified record (1940) unpublished records of R. Bennitt. Douglas 5 1 male banded (1977) J. Barnes-pers. comm. USFWS Bird Banding Lab. Dunklin 6 unspecified record Cunningham 1937 Greene 7 numerous past records C. Bonner, P. Galloway- (1965-1980) pers. comm. Hamilton 1969 Jasper 8 unspecified record Bennitt 1932 McDonald 9 unspecified record Bennitt 1932 Mississippi 10 1 male observed Hamilton 1967 Newton 11 unspecified record Bennitt 1932 Oregon 12 breeding pair, 3 young Bennitt 1932 (1917) Osage 13 1 male observed Bennitt 1932 Platte 14 1 male observed (1930) Bennitt 1932 Saline 15 1 male observed (1939) J. Wylie-pers. comm. St. Clair 16 male, female banded H. & F. Diggs-pers. comm. (1973) USFWS Bird Banding Lab. Stoddard 17 unspecified record Cunningham 1937 1 observed (1973) Anderson 1973 Stone 18 1 male observed (1972) Robbins 1972 Taney 19 numerous sightings-males, present study females, young numerous sightings (1972- Anderson 1972 1973) Anderson 1973 Vernon 20 1 male collected (1928) Bennitt 1932

*Keyed to county numbers in Fig. 1. 80 Transactions, Missouri Academy of Science Vol. 16

TERRITORY AND HABITAT CHARACTERISTICS

Study Area This portion of the study was conducted within a 40 km2 area surrounding the town of Protem in southern Taney County, Missouri, and in adjacent Boone and Marion counties, Arkansas. This locality was chosen because a suitable population of painted buntings was known to breed there (W. West., pers. comm. ). The topography and vegetation of this area is typical of the Ozark Region. Gently rolling hills predominate; however, areas of high relief (30-60 m) including vertical bluffs, are common near Bull Shoals Lake. Soils are generally thin and rocky, thus the principal agricultural land use is pasture and hay production. Most of the land is forested, comprised mostly of post oak (Quercus stellata), blackjack oak (Q. marilandica) and black hickory (Carya texana). Where the soil is thin, eastern red cedar (Juniperus virginiana) is the predominant woody growth and is associated with native grasses, primarily little bluestem (Andropogon scoparius) (Scrivener and Frieze 1953). More complete descriptions of the geology and vegetation of the region are given by Steyermark (1940), Kucera and Martin (1957) and Bretz (1965).

Methods Adult male painted buntings were located within the study area by driving county roads and listening for song. Once located, each male's territory was delineated through repeated observation and mapped to scale. The vegetation within each territory was then analyzed using 0. 04 ha (0.10 acre) circles centered around song posts or nest sites (James and Shugart 1970). A number of plots were measured in each territory, depending on the uniformity of the habitat. The habitat characteristics measured are shown in Table 2. Maximum canopy height was determined to the nearest 1.5 m (5 ft) by ocular estimation. Slope was also estimated ocularly to within 5 degrees. Aspect was recorded according to the 8 major compass points. Percent canopy cover was determined by sighting vertically through a tube 4 em in diameter and 10 em in length. Ten sightings were taken on alternate steps on each of 2 transects (N-S, E-W) through the plot. A "+" was recorded if any canopy vegetation was sighted. Percent ground cover was measured similarly by sighting through the ocular tube pointed atthe ground 1.2 m from the observer and held at arm's length. If green herbaceous vegetation was sighted a " +" was recorded. Readings were taken at the same points as for canopy cover estimation. Ground cover height was also measured at each of these points. Trees were classified according to species and size class. A Biltmore stick was used to measure dbh.

Results and Discussion

Territory Male painted buntings began to arrive on the study area during the first week of May; however, some sites were not occupied until late May or early June. Females were difficult to locate and their arrival dates could not be accurately determined. Males began to establish territories immediately upon arrival. Once established, territories remained relatively constant in size and location throught the remainder of the breeding season. Painted bunting territories appeared to be Type A, within which mating, nesting and feeding occurred (Nice 1941). Eight territorial male painted buntings were located in 1979 and 11 in 1980. Mean size of these 19 territories was 3.15 ha (range 0.64-6.66, Table 1). This is nearly three times 1982 Norris & Elder, Biological Sciences 81

Table 2. Charaderistics of 19 painted bunting territories in southern Missouri (49 sampiing plots).

Old Field Woodland Component Component Territory Characteristics x cv. x C.V.

Size of component (ha) 2.5 65.6 0.6 105.2 Percent of total territory 82.4 18.2 17.6 85.3 Canopy height (m) 9.5 40.5 13.9 23.9 Percent canopy cover 35.0 38.4 80.0 13.0 Percent subcanopy cover 29.9 63.0 39.7 51.8 Percent ground cover 80.9 24.0 69.7 30.2 Ground cover height (em) 32.4 57.1 27.0 74.6 Woody stems/ha < 2.54 em dbh 1362 55.4 1664 58.1 2.54- 6.35 289 110.6 200 57.1 6.35-15.24 135 106.1 288 94.0 15.24-22.86 23 110.4 123 64.7 22.86-30.48 13 151.1 70 51.9 30.48-38.10 4 182.6 20 107.5 >38.10 7 225.7 13 159.2 Number of tree species 4.3 35.1 7.0 27.0 Slope (degrees) 11.1 11.0 Aspect (most frequent) s s larger than a painted bunting territory measured in southern Oklahoma (approximately 1.13 ha) (Parmalee 1959) and nearly six times larger than indigo bunting (P. cyanea) territories on our study area (x = 0.55 ha, n = 11). The most likely explanation for such large territories relates to the density and distribution of painted buntings on the study area. Because of the low population levels, very few birds were found in one area and most territories were not contiguous, thus there was little or no constraint on male movements. By comparing territories that were adjoining or nearly adjoining (less than a mean territory width apart) with isolated territories (separated by more than a mean territory width), we found that isolated birds had significantly larger territories (x = 3.92 ha, N = 13) than birds with adjoining territories (x=1.44 ha, N=6) (Mann-Whitney U-Test, P<0.05). Nesting records of painted buntings in Missouri are rare. Two nests were located during this study, one in 1979 and one in 1980. The 1979 nest was situated in a red cedar tree, 125 em from the ground. The 1980 nest was located 30 em from the ground in a winged elm (Ulmus alata) shrub. Four young were fledged from the 1979 nest on July 7. Incubation lasted 11 days and fledging occurred 10 days after hatching. Three hatchlings in 1980 were victims of predation on June 10 when they were approximately 6 days old. 82 Transactions, Missouri Academy of Science Vol. 16

Habitat Characteristics The habitat used by painted buntings on the study area was mostly pasture or fallow fields in early to middle stages of succession, bordered by older, more mature woodlots. In some instances, territories were found on glades that showed cedar encroachment. All but one territory included two distinct habitat types-an old field component and a woodland component. The old field habitat type comprised about 82% of the total territory size with woodland making up the remainder (Table 2). The woodland component was distinguish­ able from the old field by its greater canopy height, percent canopy cover and number of trees in the larger size classes (Table 2). The large coefficients of variation for most of the characteristics measured indicate that painted buntings will apparently tolerate a wide range of habitat variability, for very few parameters were consistent between territories. Percent ground cover was one of the more consistent vegetative features, possibly signifying its importance in habitat selection. Limited observation revealed that painted buntings foraged mostly on or near the ground, lending support to this possibility. It is possible that other habitat characteristics not measured in this study are better indicators of painted bunting habitat. On the other hand, painted buntings are adapted to early successional habitat which may change significantly over a period of years, thus making it unlikely that they would select a narrow range of environmental values. If an early successional area is to be suitable for breeding for any length of time, a species must tolerate a wide range of vegetative characteristics. If this is true for painted buntings in southern Missouri, as our data seem to support, we can infer that range limitation and the observed low density of the bird in this region is probably not due to lack of suitable habitat. However, southern Missouri may represent poor painted bunting habitat in general when compared to habitat that is well within the painted bunting's range. Data for such a comparison are lacking. Other factors which may be responsible for range limitation and low density of Missouri painted buntings are climate, competition with other avian species and winter limitation (Fretwell 1972).

Conclusions The painted bunting in Missouri has a somewhat irregular distribution, within which population densities are both low and localized. Painted buntings appear to have rather broad habitat requirements. One essential feature is an edge situation where old field and woodland habitat types meet. Within these habitat types, a wide range of vegetative characteristics is tolerated. This, plus the fact that many areas of seemingly suitable habitat in Missouri remain unoccupied, indicates that painted buntings are not currently limited by habitat. Effective management in this instance seems to require more than just knowledge of the species' habitat requirements.

Literature Cited Anderson, R. 1966. Fall survey-August 16 to November 30. The Bluebird 33(1):10. -----· 1972. Summer survey-June 1 to August 15, 1972. The Bluebird 39(4):10. -----· 1973. Summer survey-June-July, 1973. The Bluebird 40(4):4. Bennitt, R. 1932. Checklist of the birds of Missouri. Univ. Missouri Studies VIII, No. 3. 81pp. Bretz, J. H. 1965. Geomorphic history of the Ozarks of Missouri. Missouri Geo. Survey and Water Resources Vol. 41. 2nd ser. 147pp. Cunningham, J. W. 1937. Records from southeast Missouri. Wilson Bull. 49:306-307. Fretwell, S. D. 1972. Populations in a seasonal environment. Monographs in Population Biology No. 5, Princeton Univ. Press, Princeton, N.J. 217pp. 1982 Norris & Elder, Biological Sciences 83

Hamilton, J. L. 1967. Summer survey-June 1st to August 31st. The Bluebird 34(2):16. -----· 1969. Spring Survey-March 1stthrough May 31st. The Bluebird 36(1):27. James, F. C. and H. H. Shugart. 1970. A quantitative method of habitat description. Audubon Field Notes 24:727-736. Kucera, C. L. and S. C. Martin. 1957. Vegetation and soil relationships in the glade regions of the southwest Missouri Ozarks. Econogy 38:285-291. Nice, M. M. 1941. The role of territory in bird life. Am. Midi. Nat. 26:441-487. Parmalee, D. F. 1959. The breeding behavior of the painted bunting in southern Oklahoma. Bird-Banding 30:1-18. Robbins, M. 1972. A week of birding in southern Missouri. The Bluebird 39(2):6. Scrivner, C. L. and J. A Frieze. 1953. The soils of Taney County, Missouri. Missouri Agr. Exp. Sta. Progr. Rep. No. 24. Steyermark, J. A 1940. Studies of the vegetation of Missouri. 1. Natural plant associations and succession in the Ozarks of Missouri. Bot. Ser., Field Mus. Nat. Hist. 9:349-475. Widmann, 0. 1907. A preliminary catalogue of the birds of Missouri. Trans. Acad. Sci. of St. Louis 17(1):288pp.

1982 Guyette & McGinnes, Biological Sciences 85

Fire History of an Ozark Glade in Missouri

by Richard Guyette and E. A. McGinnes, Jr. Research Assistant and Professor School of Forestry, Fisheries and Wildlife University of Missouri Columbia, Missouri 65211

Abstract: The past fire history of an Ozark glade was reconstructed using scars from trees of eastern redcedar (Juniperus virginiana L. ). Tree scars were dated using dendrochronological methods. Cross sections from 14 dead and 7live trees were used to establish a record of fire scars from 1730 to 1980. Fires were found to be more frequent before 1879, during presettlement times. Fires, as marked by tree scars, were present somewhere on the study area 2.59 km2 (1 mile2 ) every 3.2 years. After 1870, the frequency of scarring drops to one every 22 years. Key Words: Redcedar, dendrochronology, tree rings, fire history

Introduction Little is known about past fire history of the Ozark plateau. The glades in southwest Missouri Ozarks are undergoing a change in vegetation; steep open grasslands of the glades are being invaded by eastern redcedar (Juniperus virginiana L. ). The suppression of fire has been thought to be a factor which enabled redcedar and other woody vegetation to invade the open glades (Kucera and Martin 1957). The objective of this study was to determine the fire regime under which glade vegetation has evolved. Some idea of fire frequency is also necessary to give land managers a basis for making decisions on scheduling prescribed burning, suppressing fire and maintaining native vegetation, wilder­ ness areas and soils. The glades of southwestern Missouri are in a steep hilly region ranging in elevation from approximately 442 m (1450 ft.) to 244m (800ft.). The parent rock of the area is limestone and dolomite, which is exposed or just under the surface in most areas. There are approximately 200,000 ha (500,000 acres) of glades in southwestern Missouri. This area is a mosaic of oak-history forest and glade openings. Dominant species on the glades are grasses such as big bluestem (Andropogon scoparius Michx. ), little bluestem (Andropogon gerardi Michx.) and switch grass (Panicum virgatum L. ). Woody species common to the glades are post oak (Quercus stellata Wang.), eastern redcedar (Juniperus virginiana L. ), ash (Fraxinus L. spp.) and smoke tree (Cotinus obovatus Raf.) (Probasco 1978). Climate of the area is humid and continental with an average total annual precipitation of nearly 111.8 em (44 inches); May and June are months of maximum precipitation. Temperatures range from a mean of 1.05° C (33.9° F) in January to 25.3° C (77.6° F) in July. There are usually periods in early spring, summer or fall of nearly every year during which the right climatic conditions for a glade fire are present. 86 Transactions, Missouri Academy of Science Vol. 16

Materials and Methods Along the interface or ecotone between the oak-hickory forest and the glades grow old (> 2_00 years) eastern redcedar trees. Old redcedar also grows in rock outcrops along dry washes adjacent to the glades. Virtually all old eastern redcedars are to some degree protected from fire by rock outcroppings; none were found growing in the tall grasses of the glades. Nearly all old redcedars have signs of injury at the base of the bole and have some die back (usually extensive) in the crown. Redcedars in glades, with the exception of those growing on steep bluffs, have little sign of ever having had large (> 2 em in diameter) branches low on the bole. Few old redcedars have branches below 2. 74 m (9ft.). In con­ trast, young ( < 60 years) redcedars in glades grow on open, grassy regions and have foli­ age and branches close to the ground. Old redcedars in glades form the basis for this study.

Sampling Samples were collected from Buttram Hollow, Ava Range District, 36o 42' Lat., 92° 46' Long. Cross sections were taken from 7 live and 14 dead trees. Trees were selected for the duration of their ring record and the extent of scarring. Cross sections are necessary in doing fire history work with eastern redcedar because of the numerous false and locally absent rings associated with injuries, the lack of external signs for many scars and the characteristic convoluted form of the bole of redcedar which makes ring counting difficult. Cross sections were cut between 0.5 and 1.5 m high on the bole-wherever most scarring appeared. Fig. 1 shows a dated sample cross section.

Figure 1. Cross section of eastern redceda.r reveals a record of scarring more than 250 years long. Suspected fire scars, pith and most recent growth Iring are dated. Scars are associated with induced sapwood (light colored wood) resulting hom cambial injury. 1982 Guyette & McGinnes, Biological Sciences 87

Spatial distribution of sample trees in the study area was determined by necessity. The limited number of dead trees and a reluctance to kill old live trees made the sampling process one of taking samples wherever they could be found. Distribution of the sample trees used in this study is shown in Fig. 2.

OAK-HiCKORY !!l%11 OAK-RED CEDAR E::il GLADES .25 MilES

SAMPLE TREES CJ

Figure 2. Study area, showing location of the sample trees in relation to surrounding vegetation types. The sample trees, here represented by dots, are usually located on wck outcrops at the forest-glade interface.

Crossdating Dendrochronology is the study of the pattern of tree-ring width variation to date past events. Most woody plants in temperate climates grow an annual layer of wood in their stems which can be used as a time control. Events influencing a particular year's growth can be dated. There are problems, however: drought and injury can cause rings to be missing on sections of the bole, and false (non-annual) rings can occur in some species (particularly Juniperus spp.) (Kuo and McGinnes 1973). Crossdating is a dendrochronological 88 Transactions, Missouri Academy of Science Vol. 16 technique in which ring width patterns are matched to ensure that the temporal identity of a growth ring is correct (Stokes and Smiley 1968). The technique also enables dating of wood of unknown age. Crossdating will only work, however, where there is a common limiting growth factor, such as climate, influencing the trees in an area. Eastern redcedars in glades can often be crossdated. The common factor limiting their growth is drought. Approximately 50% of cross sections from dead trees sampled for this study were not datable, even with a full cross section. Tree rings in this study were crossdated by measuring each ring width and plotting a ring-width series for each cross section. Ring-width plots were compared and matched to assure the temporal identity of each ring. Many samples did not crossdate during certain periods in their ring-width series. Reliable ring counts, however, could often be made on sections which did not crossdate, but where rings were distinct. Although calendar years can be assigned to rings with great accuracy, scars are not always identifiable in this way. In general, only scars which occur during the period of cambial growth (May-July) can be assigned an exact calendar date. Scarring which occurs between annual periods of cambial growth can be assigned to either the year (ring) before there is any evidence of injury, as in the case of a late season (August-December) fire, or to the year (ring) following the last normally formed ring as in the case of an early season (January-April} fire. To confuse matters even more, actual dates of the beginning and ending of cambial growth are not necessarily the same for all trees in an area, thus the same fire might cause different types of scarring in different trees. Old trees, which put on very narrow rings (less than 10 cells per ring) may stop cambial activity long before younger, more vigorous trees. Because of these problems, scars were dated in several ways. Scars which occurred during the period of cambial reproducation were assigned the date of the ring in which the scar occurred. Scars which occurred when the cambium was not active were assigned two adjacent dates--the calendar year of the last normally formed ring before the scar and the calendar year of the first abnormal ring. When scars were compared between trees, adjacent dates were "lumped" into 1 date-the date when the most number of trees were scarred. This lumping introduces an unavoidable degree of error in the data.

Results and Discussion All dated scars on trees from Buttram Hollow were combined into a single chronology similar to the composite fire chronology used by Dieterich (1980) and others. The injury composite for Buttram Hollow is shown at the bottom of Fig. 3. The height of the bar in this injury composite indicates the number of trees scarred in a particular year. The individual tree-ring series analyzed for scars are represented by a line for each tree. The bars on these lines indicate scarring at that year. The height of these bars on the individual sampled trees corresponds to the extent of wounding on the cross section; a taller bar means a larger wound. At the top of Fig. 3 are four trees from two other sites which are included for comparison. The Swan Creek site is approximately 30 km northwest, the Pinnacle site approximately 3 km north of the Buttram Hollow site. Only trees from Buttram Hollow are used in the injury composite shown at the bottom of Fig. 3 The interpretation of this injury composite is complex. There are many limitations inherent in the scar data as well as unavoidable sampling biases. Every scar recorded from the sample trees is not necessarily a fire scar. Scars caused by fire in eastern redcedar can be mistaken for scars caused by animals, man, floods or falling trees. Charcoal is not found in association with cambial death, but with the occurrence of repeated burns on the exposed 1982 Guyette & McGinnes, Biological Sciences 89

PtNN-122 163014%%~:=:=::::;::~::;::~======- PINN-124

SWAN-404 1692 1531 ===:;:::=====::::::!======--- SWAN-406 1601 BHM- 363 1657 BHM-607 1677 BHM -605 1710 BHM-621 ....,1680 BHE-303 ~1677 BHE-331 <1675 BHE-336 0 l63S BHE-342 r 1677 BHE -324 =1724 BHE-300 0..1730 1707 BHW-852 1736 BHW-855 1730 BHW- 856 1740 BHW-851 1698 BHW-854 1676--' BHW-130 1980 1730 1750 1800 IBSO YEAR 1900 1950

BUTTRAM HOLLOW INJURY COMPOSITE (NO. TREES SCARRED) BUTTRAM HOLlOW SAMPlE SILE

Figure 3. An mustratiioll of the extent and "pattem" of tree scars in the study area through time. Each horizontal Dine represents a particular tree's period of gwwth; each vertical bar on these lines indicates a scar. Height of the bars corresponds to extent of scarring on the cross section of the tree analyzed. The inju..uy composite below mustrates sample size am:l the nll.llmber of trees scarred in the study area. Height of the vertical bars indicates the number of trees scarred iMll a particular year. The light bars represent scius based only on ring count.

dry dead wood of the bole. Eastern redcedar has thin bark and is very susceptible to bole damage from heat alone (Starker 1932). There is, however, considerable evidence that the scars used in this study are fire scars. Because of the presence of charcoal and historic records of fire, we know that fire has been in the study area. Although animal- and man-caused scars can be mistaken for fire scars, these factors are usually very limited in comparison to fire damage. Nearly all the old redcedars in the study area are scarred extensively and have damaged crowns. These crowns, which in most cases consist of only a few live branches, are typical of fire damaged crowns in redcedar (Martin and Crosby 1955). Every scar may not be a fire scar, but on the other hand, every fire in the study area may not have left a scar on several or even on one of the sample trees. Fires on the glades are noted for "skipping around" and leaving some areas unburned. In fact, these areas which were left unburned because of their ground cover, aspect and slope are the rnicrosites for old eastern redcedars. This site factor coupled with a very limited sample tree distribution in Buttram Hollow (see Fig. 2) indicates that, if anything, our injury composite may have underestimated the frequency and extent of fire in the region. The age ofthe sample trees presents a bias overtime in this study. Young trees less than 5 feet in height are much more likely to be killed than older, taller trees, and trees that do survive are more likely to be unscarred. Thus there is a selective sampling bias in early growth of the sample trees which underestimates the number of fires. Very old trees may have a lower probability of being scarred by a fire because of the reduced amount of 90 Transactions, Missouri Academy of Science VoL 16 cambial area on their boles. The living cambium (usually confined to a narrow strip down one side of these old trees) is probably on the most protected side of the tree since it has survived many fires or other mishaps already. Thus, the older trees may not scar as easily as they once did. Also, one fire that is not recorded in the injury composite is the "killing fire" which leaves no scar, only a dead tree. With these limitations in mind, the injury composite can be interpreted meaningfully. The dated scars in the sample trees are assumed to represent a fire at a particular place at a certain time. The frequency of scarring of fires can be represented in many ways (Stokes and Dieterich 1980). Here the term "fire frequency" refers to the mean number of scars occurring during a given time period on a specified area. Table 1 shows several fire frequencies for the occurrence of a scar in at least one tree in Buttram Hollow during two time periods. The frequency of scarring was one scar year every 3.2 years before 1870 and a scar year every 22 years after 1870. The scar years after 1870 have no replication of scars between trees during the same year. Thus, there are no statistics for two of the categories. These categories show fire frequencies where two or more and three or more trees are scarred in any particular year. The more scars on different trees during a year, the more likely the fire was widespread and intense.

Table 1. Frequency of fires in Buttram Hollow as reconstructed from tree scars calculated as mean number of years with 1, 2, 3 or more trees scarred.

Fire Frequency

Pre-settlement Settlement Number of period period trees scarred 1730-1870 1870-1980

1 or more 3.2 22 2 or more 8.75 no occurrences 3 or more 20 no occurrences

The most striking feature of the injury composite in Fig. 3 is the change in scarring frequency that occurs around 1870. This change probably reflects less widespread and possibly less intense fires. One can only speculate these causes which may be many and complex. The associated cultural and land use practices are shown in Fig. 4. The use of fire by American Indians is well documented (Barrett 1980): they burned grasslands to attract game, aid in hunting and harass enemies; and forests were burned to aid in gathering acorns. Man was probably the most important ignition source in non-drought years. Very few lightning fires are on the records of the Ava Ranger District in which this study took place. Extreme drought years are different, however. In 1980, a year of extreme drought, high temperature and winds, at least 6 lightning fires were reported on the Ava Ranger District. The Osage Indians occupied the White River area in which the study site lies. They were probably the most important ignition source before settlement, particularly during years of little drought. Settlement of the glade region occurred between 1850 and 1870. The population density has never been great because of poor soils and steep topography. Grazing by cattle 1982 Guyette & McGinnes, Biological Sciences 91

CULTURE OSAGE EASTERN SETTLEMENT fOREST SERVICE T I E LAND USE HUNTING DOMESTIC GRAZING fiRE SUPPRESSION NATIVE GRAZERS lOGGING

INJURY COMPOSITE 1~1 0" I!! !I I!J t~mallii&ll I! i!dhdl .II .. 1.1 mhl Iii 1750 1800 1850' I '"' 1900 • ' •1950 • 1980

Figure. 4. A simplified diagram of cultural and land use practices in the glade region. Fire frequency as indicated by the injury composite drops off after 1870 with the advent of settlement and domestic grazing. was then, as now, the most important use of the glades. Rangelands (glades) were often burned to improve grass palatability and remove brush (Martin 1955). Severe overgrazing often resulted. Settlers also had reasons for suppressing fires, because wildfires endangered property and livestock (rail fence was particularly vulnerable to fire). Overgrazing was probably the most important factor affecting the change in fire frequency. Overgrazing causes fuel reduction and erosion, both which inhibit the spread and intensity of fire. Roads as firebreaks, fire suppression and a change in the use of fire by settlers were also important factors in changing fire frequency. Climate was probably not a factor in the change in fire frequency. Studies by the authors indicate that, if anything, the period 1800-1860 was wetter and cooler with less drought (Guyette 1981). Climate, however, seems related to the number of trees scarred in a particular year. In Fig. 5 the injury composite is compared with a ring width-index of eastern redcedar. This index, constructed by methods described by Fritts (1976), is composed of 69 samples from in and around the study area. No more than 2 samples per tree were used. The index has a mean of one and the values are on an arbitrary scale. The smaller the value, the narrower the width of the rings of the sample trees for that particular year. A narrow ring in Ozark eastern redcedar indices is correlated with severity of drought (Guyette eta!. 1980). Note that the drought years 1807, 1809, 1820 and 1855 in Fig. 5 are years when the injury composite shows scars on many trees. Thus, as one would expect, fires during drought years are more widespread and intense than those during wetter years.

("\ 0

~~~~~~~~~~~~~~~~------~----~~-----L[110~_ ,,, ...... Ji,.l.i a.• .1, [

YEAR

Figure 5. Four years in which many trees were scarred have vertical dotted lines drawn through them. These four years are associated with low ring-width index values from the glade region. Low index values represent narrow rings and are correlated with drought variables. 92 Transactions, Missouri Academy of Science Vol. 16

Condusion Eastern redcedar can be used to reconstruct the fire history of an area. Its heartwood is very resistant to decay and weathering, especially when its moisture content is lowered by die back caused by fire. Thus, this species of tree can provide evidence of fire for more than 700 years. Old trees are often found on sites which are climatically limiting to tree growth. These trees can be crossdated-an important consideration in doing tree ring studies. These are problems, however, in using eastern redcedar to reconstruct fire history. Until positive identification of individual scars can be made, some error must be accepted in the fire histories and results qualified accordingly. At the study area, Buttram Hollow, in the cedar glades of southern Missouri, a fire occurred somewhere in the area every 3.2 years between 1630 and 1870. More conservative, although not necessarily more accurate, estimates of fire frequency based on the scarring of 2 or more and 3 or more trees in a particular year were, respectively, 8. 75 years and 20 years for the period between 1730 and 1870. These estimates are in the range necessary to have kept redcedar from establishing itself on the glades. The percent kill on redcedar over 2m in height (~15 years of age) is 3% (Martin and Crosby 1955); although very severe fires could have killed a greater percent of even much larger redcedars, some larger trees would have survived (as they do at the forest-glade interface). Thus, fire frequency must have been high ( < 20 years) in order to keep older redcedar from becoming established, since there is no evidence of old redcedar (> 100 years) having grown on glade openings. There was a change in the frequency of fires around 1870. After 1870, fire scars are much less frequent. Much of this change in fire frequency is due to fire suppression, possibly by early settlers, and after 1940 by the U.S. Forest Service. Other possible reasons for the change in fire frequency are the removal of a major ignition source (the Osage Indians), the building of roads (firebreaks), severe overgrazing and erosion (reduced fuel loads).

Acknowledgments The authors would like to thank the North Central Forest Experiment Station, U. S. Forest Service, whose funding made this project possible, and the School of Forestry, Fisheries and Wildlife, University of Missouri. Also we would like to thank Garry Houf and Jerry Adams of the Ava Ranger District, Mark Twain National Forest, U. S. Forest Service, for their help in locating the study area. Figures 2, 4 and 5 are taken from ''Proceedings of the Ninth Cellulose Conference'', Journal of Applied Polymer Science, Symposium #37 (in press). Copyright 1983 by John Wiley & Sons. Used by permission of the publisher.

Literature Cited Barrett, S. W. 1980. Indians and fire. Western Wildlands 6(3):17-21. Dieterich, H. J. 1980. Chimney Spring forest fire history. USDA Forest Service Rocky Mountain Forest and Range Experiment Station, Res. Pap. RM-220. Fritts, H. C. 1976. Tree-rings and climate. Academic Press, London. 567pp. Guyette, R. P. 1981. Climatic patterns of the Ozarks as reconstructed from tree-rings. M.S. thesis, University of Missouri-Columbia. 70pp. Guyette, R. P., E. A. McGinnes, Jr., K. Evans and G. Probasco. 1980. A climate history of Boone County Missouri, from a tree-ring analysis of eastern redcedar. Wood and Fiber 12(1):17-28. Kuo, M. and E. A. McGinnes, Jr. 1973. Variations of anatomical structure of false rings in eastern redcedar. Wood Sci. 5(3):205-210. Kucera, C. L. and S. C. Martin. 1957. Vegetation and soil relationships in the glade region of the southwestern Missouri Ozarks. Ecology 38(2):285-291. 1982 Guyette & McGinnes, Biological Sciences 93

Martin, S.C. 1955. The placeofrangelivestockin the Missouri Ozarks. Range Manage. 8(3):105-111. Martin, S. C. and J. S. Crosby. 1955. Burning and grazing on glade range in Missouri. U. S. Dept. Agric., Tech. Paper 147. 13pp. Probasco, G. E. 1978. Bird habitat-woody plant relations on Missouri limestone glades. Midwest Prairie Conf. Proc., Iowa State Univ. 5:108-109. Starker, T. J. 1932. Fire resistance of the trees of the northeast United States. Forest Worker 3(3):8-9. Stokes, M. A and H. J. Dieterich. 1980. Proceedings of the fire history workshop. Rocky Mountain Forest and Range Experiment Station, U. S. Forest Service, Gen. Tech. Rep. RM-81. Stokes, M. A and T. L. Smiley. 1968. Introduction to tree-ring dating. Univ. Chicago Press. 73pp.

1982 Dunn, Biological Sciences 95

Problems in "Keeping-Up" With the Flora of Missouri

David B. Dunn Curator of the Herbarium University of Missouri-Columbia Columbia, Missouri 65211

Abstract: Nineteen taxa are cited for the flora of Missouri, 12 are new species to the state, three of which are in genera new to the state, six are considered rare but new locations are reported, and one represents the most difficult problem of misclassification or monographic revision. The taxa are, alphabetically: Aeschynomene indica, Aira caryophyl/ea, Aubrieta deltoidea, Bothriochloa bladhii, Bnmnera macrophyl/a, Ceratostigma plumbaginoides, Desmodium strictum, Dipsacus ful/onum f. a/bidus, Dipsacus /aciniatus, Draba apnea, Fimbristylis puberu/a, Lychnis coronaria, Physalis alkekengi, Scilla sibirica, Sedum sarmentosum, Veronica chamaedrys, V. hederaejo/ia, V. persica, and V. po/ita. The taxa represent a relict, grasses of unknown origin, old collections found in the herbarium, roadside and waste area invaders, introductions with lawn seed, plants escaped from gardens, plants introduced by migratory birds, and a change due to revised classification. Key Words: Herbarium, Missouri Rora

The verifications of the taxa cited were all made by the author at the University of Missouri herbarium, unless otherwise indicated. Most of the plants were keyed using Fernald (1950) and Gleason (1952). The names listed have been verified in Kartesz and Kartesz (1980) as those accepted by the authorities for the taxa recognized. Escaped cultivated taxa were keyed in Bailey (1949) and verification of the currently accepted epithet was determined by reference to Bailey's Hortus Third (1976). An Overlooked Native. Desmodium strictum (Pursh) DC., Fabaceae, (UMO# 129348, & 136874) was collected on Holly Ridge State Forest, Stoddard Co., by Leroy Korschgen. He considered the specimens as a small relictual population (personal communication). In verifying the species two possibilities were considered after use of the available keys. The leaf and seed joints were taken to Missouri Botanical Garden herbarium and matched with the two possible taxa. Both the leaf, hair-type and joint shape matched D. strictum quite accurately. New Grasses. The two species reported are probably naturalized since neither was reported in the Flora of Missouri by Steyermark (1961). Aira caryophyllea L., Poaceae (UMO# 131764) was collected on a sandstone glade 5 mi. SE of Walnut Grove, in Greene Co. by Tom Toney. Bothrioch/oa bladhii (Retz.) S. T. Blake, Poaceae (UMO# 148357) was collected in Howard Co. W of Brandsville, in a disturbed field by Charlene Bennett, a graduate student. The specimen was sent to Richard Pohl, Ames, Iowa, for determination. Lawn Seed Sources. Three species of Veronica have been identified, which inhabit lawns and probably originated from grass seed sources. Two of these have not been reported previously in Missouri, V. hederaefolia L., in Barry Co. (UMO# 160816) and V. persica Poir., Boone Co., (UMO# 160813). The former species was collected by Tom Toney, and the latter in Research Park, Univ. of Mo. campus by Douglas LeDoux and the author. The third species V. polita Fries., has been reported from six counties by 96 Transactions, Missouri Academy of Science Vol. 16

Steyermark, and it has been collected over a 20 year period on the University campus. It now appears in new grass plantings in Research Park (UMO# 152544) and is reported by Korschgen in his yard in Columbia. There are also recent collections from Jackson and Greene Cos., so it appears that it may become a widespread weed in Missouri. Garden Escapes. The several species reported are mostly in the stage of early invasion as adventives, prior to becoming fully naturalized. Brunnera macrophylla (Adams) I. M. Johnston, Boraginaceae, (Siberian Bugloss) escaped into an alley in Columbia, Boone Co., where it is competing with grasses such as Bromus tecto rum. Its capacity to migrate via the nutlets clinging to hair or clothes offers a potential for dispersal (UMO# 158635). Ceratostigma p/umbaginoides Bunge, Plumbaginaceae (UMO# 139710) has escaped into an alley along a fence in Columbia, Boone Co., and the colony has persisted for at least 10 years. It is a native to western China. The only previously known plants of this species were within the walled area at Missouri Botanical Garden. Lychnis coronaria (L) Desr., Caryophyllaceae (UMO# 116690) was collected along highway 160, 8 miles W of the junction of highway 65, in Stone Co., by George Kastler. The colony was growing on a gravel bank away from any dwelling. PhysalisalkekengiL. Solanaceae (UMO# 67878and 75754), Chinese lanterns, escaped in Boone and Callaway Cos. The colony in Boone Co. persisted in a fence row for nearly 15 years before trees shaded the area. The location in Callaway Co. was in a fence row near a feed lot. The strong perennial rhizomes produce extensive spreading of the colonies. Sedum sarmentosum Bunge, Crassulaceae (UMO# 148938) was collected along a bluff above Pulltite Camp on the Current River in Shannon Co., by Leroy Korschgen. The area was not disturbed and there were no dwellings nearby. Two locations were previously reported by Steyermark. The species is commonly grown in rock gardens but may be expected in glade habitats where it could become naturalized. Veronica chamaedrys L., Scrophulariaceae (UMO# 162384) was introduced ina garden in Columbia, Boone Co., from New England and has now spread into the adjacent oak-hickory woodland. Roadside and Waste Areas. Areas with disturbance are some of the most common places to locate species of plants invading by either seeds or fleshy fruits. Scilla sibirica Andr., Liliaceae (UMO# 41878, & 72134) has been growing in the same area in Columbia, Boone Co., for 27 years, in a fence row. It has spread to neighboring yards in lawns and under trees. The berry fruits are eaten by birds and seeds germinate like those of Asparagus, better after passing through the bird. Aubrieta deltoidea (L.) DC., Cruciferae (UMO# 158577) was collected by a student south of Columbia in an old field along the Hinkson Creek. The plants are perennials used as border plants in gardens but had existed for several years in the old field. It is a new genus and species for Missouri. Dipsacus laciniatus L., Dipsacaceae (UMO# 133878) was collected in a vacant lot in Columbia, Boone Co., by Ralph Baumgartner. At the time only one other collection had been made in Missouri. Several collections are now reported and it should be considered as an additional weed for the state. Leroy Korschgen reports locations at Providence and Turner St., Boone Co., and Rt. B., 1 mile E of Hwy 61, in Pike Co. Dipsacus fullonum L. forma a/bidus (Steyerm.) (UMO# 157365) was collected by George Kastler, of the Parks Department, in Wright Co. The white flowered form was not previously reported for Missouri. It was found on a gravel bank beside the highway. It should be noted that the species was formerly treated as D. sylvestris, but that both Kartesz and Kartesz (1980) and Bailey (1976) recognized the earlier Linnean name. Bird Migrations. This is one of the major means of plant migration for both plants with dry seeds and those with fleshy fruits. Long range and local migrations are caused by both 1982 Dunn, Biological Sciences 97

birds and omnivorous animals. The rapid invasion by some species such as Rosa multiflora, Morus rubra, Prunus serotina, and Juniperus virginiana should earn them consideration as weeds. Long range invasion by three species was reported by Dunn and Knauer (1975) as introductions to Mingo Wildlife Refuge, which was attributed to bird migration from the Gulf Coast, since this was the closest natural population on the migration route. Old Herbarium Collections. Two examples illustrate the location or records among herbarium specimens, although many more could be cited. Dunn and Knauer (1975) reported the introduction of a new genus, Aeschynomene indica L. to the Missouri Flora. Recently a second record was located, which had been collected by Steyermark in 1956, but had been misidentified and filed as Sesbania exaltata. It is Steyermark #83501 (UMO# 132725) from New Madrid Co., Mo. Draba aprica Beadle, Cruciferae, was discovered filed among specimens of Draba brachycarpa, which had been misclassified earlier. This resulted from verifying a specimen for an interested citizen who sent in a specimen of D. brachycarpa, which he hoped might be the rare D. aprica. The new locations found in the herbarium were Stone Co., Palmer 17221 (UMO# 131149), near Galena, and McDonald Co., Palmer 59607-A (UMO# 140457), near Powell. Errors or Changes in Classification. The problem of errors in manuals or changes from more recent monographs is illustrated by Fimbristylis puberula (Michx.) Vahl, Cyperaceae. In a study by a graduate student (David Bergmann) on some glade habitats in southern Missouri, specimens were brought to the herbarium for verification. In Steyermark' s Flora of Missouri (1961) this taxon and others were cited as synonyms of F. caroliniana. Correll and Correll (1975), in the revised edition of the two volume work on wetlands of the southwestern United States, treated F. puberula, F. caroliniana, and F. castanea all as separate species. Godfrey and Wooton (1979) and the latest information in Kartesz and Kartesz (1980) checklist on the Flora of North America confirmed that the authorities for the genus considered all three species as distinct. In short, none of the 34 specimens at the University of Missouri-Columbia, used by Steyermark to map the distribution of Fimbrystylis caroliniana proved to be F. caroliniana --all were F. puberula. F. caroliniana is rhizomatous, traveling in the coastal sands, not tuberous and caespitose. F. castanea is a deep-seated caespitose plant in sandy brackish areas along the Atlantic and Gulf Coastal areas. F. puberu/a ranges from Nebraska to Mexico and westward to Arizona. With specimens of the two other species collected from Georgia and Florida available for comparison it is now questionable whether either has ever been collected in Missouri.

Acknowledgments Appreciation is expressed to Leroy Korschgen and Tom Toney of the Missouri Department of Conservation for collection and contribution of plant specimens to the University herbarium. Appreciation is also expressed to George Kastler of the State Park Service, for contribution of specimens for verification at the University herbarium. Graduate students, Charlene Bennett, David Bergmann, and Douglas LeDoux, as well as several undergraduates brought in unknown plants for determination. Specimens were · also mailed or brought in by staff members of Agronomy and Horticulture or by Extension Agents, and numerous citizens of the state, requesting determinations. 98 Transactions, Missouri Academy of Science Vol. 16

Literature Cited Bailey, L. H. 1949. Manual of cultivated plants. Revised ed. The Macmillan Co., N. Y. Bailey, L. H. 1976. Hortus third. A concise dictionary of plants cultivated in the United States and Canada. Macmillan Co., N. Y. Correll, D. S. and H. B. Correll, Rev. Ed. 1975. Aquatic and wetland plants of southwestern United States. Stanford Univ. Press., Stanford, CA. Dunn, D. B. and D. F. Knauer. 1975. Plant introductions by waterfowl to Mingo National Wildlife Refuge, Missouri. Trans. Mo. Acad. Sci. 9:27-28. Fernald, M. L. 1950. Gray's manual of botany. 8th ed. Am. Book Co., N. Y. Gleason, H. A 1952. The new Britton and Brown illustrated flora. 3 Vol. The New York Botanical Garden and Hafner Press, N. Y. Godfrey, R. K. and J. W. Wooton. 1979. Aquatic and wetland plants of southeastern United States. 2 Vol. The Univ. of Georgia Press, Athens, GA. Kartesz, J. T. and R. Kartesz. 1980. A synonymized checklist of the vascular flora of the United States, Canada and Greenland. The Univ of N. Carolina Press, Chapel Hill, N.C. Steyermark, J. A 1961. Flora of Missouri. The Iowa State Univ. Press. Ames, Iowa. 1982 Powell et al., Biological Sciences 99

Maximum Sizes of Amphibians and Reptiles From Missouri

Robert Powell, Kenneth P. Bromeier, Nicholas A. Laposha, and John S. Parmerlee Department of Natural Sciences Avila College Kansas City, Missouri 64145 Brian Miller Division of Biological Sciences University of Missouri Columbia, Missouri 65211

Abstract: Maximum size records for amphibians and reptiles from Missouri are given with the counties in which the specimens were collected. Key Words: Maximum sizes, Missouri, amphibians, reptiles.

Maximum size records of amphibians and reptiles generate interest among both professional and amateur herpetologists. This interest increases time spent in the field, number of observations made, and our knowledge of natural history. Insight may be gained into the ability of local environments to provide adequate resources for growth. Records listed by state can provide information about geographic and dina! variations in growth. Because growth continues throughout life in amphibians and reptiles (Porter 1972), maximum sizes may prove relative to lifespans. Sizes may be further correlated with fecundity (Fitch 1970). The field guides most widely used in North America all include size records (Conant 1975, Smith 1978, Behler and King 1979, Smith and Brodie 1982); however, none cite the localities from which specimens were taken. Unfortunately many regional or state lists omit size records or fail to provide relevant localities. In Missouri, Hurter (1911) listed sizes of specimens, but did not note if they were from Missouri. Anderson (1965) made reference to the largest specimens of some reptiles he had observed, but omitted collection sites. Johnson (1977) omitted size records of amphibians altogether. We examined specimens held in collections at the University of Missouri-Columbia (UMC), the University of Kansas Museum of Natural History (KU), Lawrence, and at Avila College (BWMC), Kansas City, Missouri. Maximum sizes were recorded for each species, along with the county in which the animal was collected (Tables 1-5). If none of the specimens examined exceeded the size listed for that species by Anderson (1965), his record is also cited. If no animals of a species were examined, the lack of a record is noted. If specimens currently in collections not examined or captured in the future exceed the maximum sizes recorded, we ask that notification be sent to the senior author. Notification should include verification of the animal's identity, collection data, and a museum number (records must be deposited in a museum collection). Parameters measured are illustrated in Conant (1975). 100 Transactions, Missouri Academy of Science Vol. 16

Table 1. Maximum sizes of salamanders from Missouri. Measurements in millimeters.

Museum Snout-vent Total Species number County length length

Ambystoma annulatum - ringed salamander KU 088905 Stone 113 238* Ambystoma maculatum - spotted salamander BWMC 01609 Boone 100 213 Ambystoma opacum - marbled salamander BWMC 01414 Phelps 74 124 Ambystoma ta/poideum - mole salamander UMC 1981 C Stoddard 68 109 Ambystoma texanum - smallmouth salamander KU 089214 Pemiscot 94 168 Ambystoma tigrinum - tiger salamander UMC 117 C Camden 127 238 Amphiuma tridadylum - three-toed amphiuma UMC 1571 C Stoddard 650 811 Cryptobranchus al/eganiensis -hellbender UMC 1547 C Dallas 410 601 Eurycea /ongicauda - longtail salamander UMC 538 C Carter 57 150 Eurycea /ucifuga - cave salamander KU 028059 McDonald 65 177 Eurycea multip/icata - many-ribbed salamander UMC 1982 C Taney 51 102 Eurycea tynerensis - Oklahoma salamander UMC 1730 C Barry 41 73 Hemidactylum scutatum -four-toed salamander KU 115872 Wayne 49 81 Nedurus macu/osus - mudpuppy KU 089900 Morgan 222 312 Notophtha/mus uiridescens - eastern newt UMC 348 C Callaway 61 125* Plethodon dorsalis - zigzag salamander KU 089681 Ozark 42 81 Plethodon glutinosus - slimy salamander KU 089732 Butler 81 169 Plethodon serratus - southern redback salamander UMC 1667 C Iron 45 103 Siren interrnedia - lesser siren UMC 1541 C Bollinger 280 419 Typhlotriton spelaeus - grotto salamander KU 060779 Stone 59 109

*Exceeds U.S. record cited by Smith (1978). 1982 Powell et al., Biological Sciences 101

Table 2. Maximum sizes of frogs and toads from Missouri. Measurements in millimeters.

Museum Total Species number County length

Acris crepitans - northern cricket frog KU 090660 Phelps 33 Bufo americanus - American toad KU 090238 Jackson 105 Bufo cognatus - Great Plains toad KU 090107 Jackson 74 Bufo woodhousei- Woodhouse's toad KU 090533 St. Clair 96 Gastrophryne carolinensis - eastern narrowmouth toad KU 090985 Webster 36* Gastrophryne o/ivacea - Great Plains narrowmouth toad BWMC 00754 Ray 37 Hyla chrysoscelislversico/or - complex - gray treefrog KU 090910 Benton 51 Hyla cinerea - green treefrog KU 176195 Stoddard 58 Hyla crucifer - spring peeper KU 176204 Wayne 32 Pseudacris streckeri - Strecker's chorus frog KU 090705 Dunklin 38 Pseudacris triseriata - striped chorus frog BWMC 00357 Jackson 38* Rana areolata - crawfish frog UMC 1272 A Boone 99 Rana blairi - plains leopard frog KU 091189 Jackson 89 Rana catesbeiana - bullfrog KU 091082 Ray 151 Rana clamitans - green frog KU 091093 Miller 91 Rana palustris - pickerel frog BWMC 01879 Dade 66 Rana sphenocepha/a - southern leopard frog KU 091211 Miller 81 Rana sylvatica - wood frog USNM 58028** Stone 63 Scaphiopus bombifrons - plains spadefoot KU 090043 Jackson 52 Scaphiopus holbrooki - eastern spadefoot KU 090080 Dunklin 69

*Equals U.S. record cited by Smith (1978). **U.S. National Museum (information provided by Tom R. Johnson, Missouri Department of Conservation). 102 Transactions, Missouri Academy of Science Vol. 16

Table 3. Maximum sizes of turtles from Missouri. Measurements in miiUmeters.

Museum Carapace Species number County length

Chelydra serpentina - snapping turtle UMC 35 T Camden 326 Chrysemys picta - painted turtle BWMC 01821 Grundy 203 Deirochelys reticularia - chicken turtle UMC 33 T Stoddard 202 Emydoidea b/andingi- Blanding's turtle KU 091330 Miller 236 Graptemys geographica - map turtle KU 177222 Benton 227 Graptemys kohni - Mississippi map turtle KU 091332 Pemiscot 157 Graptemys pseudogeographica - false map turtle KU 088750 Holt 123 Kinostemon flavescens - yellow mud turtle KU 177109 Barry 131 Kinostemon subrubrum - eastern mud turtle KU 091317 New Madrid 95 Macroc/emmys temmincki - alligator snapping turtle UMC 36 T Stoddard 425 Pseudemys concinna/floridana complex - cooter KU 091342 Pulaski 300 Pseudemys scripta - slider KU 091345 Benton 251 Stemotherus odoratus - stinkpot BWMC 01815 McDonald 117 T errapene carolina - eastern box turtle KU 177217 Cole 142 T errapene ornata - western box turtle KU 088845 Davies 124 Trionyx muticus - smooth softshell UMC 34 T Boone 197 Trionys spiniferus- spiny softshell KU 091355 Dunklin 219 1982 Powell et al., Biological Sciences 103

Table 4. Maximum sizes of lizards from Missouri. Measurements in mmi.meters.

Museum Snout-vent Total Species number County length length

Cnemidophorus sex/ineatus - six-lined racerunner UMC 77 L Taney 78 250 Crotaphytus collaris - collared lizard UMC 44 L Boone 105 296 (322*) Eumeces anthracinus -coal skink BWMC 01782 Stone 63 166 Eumeces fasciatus - five-lined skink UMC 147 L Bollinger 72 186 Eumeces laticeps - broadhead skink KU 185893 Boone 108 268 Eumeces obsoletus - Great Plains skink KU 088556 Vernon 99 248 Eumeces septentriona/is - prairie skink KU 088559 Platte 64 185 Ophisaurus attenuatus - slender glass lizard UMC 233L Johnson 257 658 (702*) Phrynosoma comutum -Texas horned lizard ** ** ** ** Sce/oporus undulatus - eastern fence lizard BWMC 00105 Boone 73 167 (172*) Scincel/a latera/is - ground skink UMC 256 L Boone 49 121

*Anderson (1965), no county or museum number available. **No specimen examined, no record available. 104 Transactions, Missouri Academy of Science Vol. 16

Table 5. Maximum sizes of snakes hom Missouri. Measurements hn miiHmeters.

Museum Snout-vent Total Species number County length length Agkistrodon contortrix - copperhead BWMC 01428 Pettis 801 913 (1041 *) Agkistrodon piscivorus - cottonmouth KU 084341 Stoddard 921 1086 (1117*) Carphophis amoenus -worm snake KU 080948 Bates 279 317 Cemophora coccinea - scarlet snake UMC 72 S Camden 310 364 Coluber constrictor - racer KU 081154 Pemiscot 1015 1330 Crotalus homdus - timber rattlesnake KU 084437 Ray 1164 1257 (1550*) Diadophis punctatus - ringneck snake BWMC 01850 Jackson 359 419 Elaphe guttata -com snake KU 081981 McDonald 1029 1220 (1524*) Elaphe obsoleta -rat snake BWMC 01790 Grundy 1473 1762 CA 8134** Crawford 2133 Elaphe vulpina -fox snake KU 082077 Holt 970 1155 (1524*) Farancia abacura -mud snake UMC 567 S Stoddard 962 1124 Heterodon nasicus - western hognose snake KU 082089 Holt 425 521 ( 537*) Heterodon platyrhinos - eastern hognose snake KU 083065 St. Clair 722 842 Lampropeltis cal/igaster - prairie kingsnake BWMC 01906 Cass 1132 1306 Lampropeltis getulus - common kingsnake UMC 264S Stoddard 923 1074 (1522*) Lampropeltis triangulum -milk snake KU 082237 Jackson 711 835 ( 914*) Masticophis flagellum - coachwhip UMC 225 S Wayne 1383 1761 Nerodia cyclopion - green water snake KU 082394 Dunklin 693 911 Nerodia erythrogaster - plainbelly water snake KU 082418 Livingston 1023 1235 Nerodia fasciata - southern water snake KU 082978 Dunklin 796 1014 Nerodia rhombifera - diamondback water snake KU 082719 New Madrid 996 1272 (1370*) Nerodia sipedon - northern water snake BWMC 01901 Lawrence 867 1133 Opheodrys aestivus - rough green snake UMC 191 S Boone 570 735 1982 Powell et a!., Biological Sciences 105

Table 5. (continued)

Museum Snout-vent Total Species number County length length

Opheodrys vema/is - smooth green snake KU 083100 Cass 299 462 Pituophis melanoleucus -pine snake KU 083136 Vernon 1564 1782 (1978*) Regina grahami -Graham's crayfish snake KU 082597 Jackson 700 855 Sistrurus catenatus - massasauga KU 084563 Holt 681 764 Sistrurus miliarius - pigmy rattlesnake KU 084590 Wayne 437 550 Sonora semiannulata - ground snake UMC 568 S Taney 287 343 Storeria dekayi - brown snake BWMC 01263 Bollinger 274 342 Storeria occipitomacu/ata - redbelly snake KU 185933 Boone 250 303 T antilla gracilis - flathead snake KU 051707 Benton 199 249*** Thamnophis proximus - western ribbon snake UMC 331 S Boone 578 801 ( 920*) Thamnophis radix - plains garter snake KU 083895 Holt 620 787 ( 845*) Thamnophis sirtalis - common garter snake BWMC 01103 Ray 788 980 Tropidoclonion lineatum -lined snake BWMC 01120 Jackson 341 392 Virginia striatula - rough earth snake BWMC 01352 Phelps 237 286 Virginia valeriae - smooth earth snake BWMC 01714 Jackson 342 393****

*Anderson (1965), no county or museum number available. **Chicago Academy of Science (cited by Anderson, 1965). ***Exceeds U.S. record cited by Smith and Brodie (1982). ****U.S. record, Laposha and Powell (1982).

Acknowledgements Drs. Dean E. Metter and Joseph T. Collins allowed us access to collections and facilities at the University of Missouri and the Museum of Natural History, University of Kansas, respectively. Martin Jolley and John Wendt helped measure animals. Don Smith reviewed the manuscript.

Literature Cited Anderson, P. 1965. The reptiles of Missouri. Univ. Mo. Press, Columbia. 330 pp. Behler, J. L, and F. W. King. 1979. The Audubon Society field guide to North American reptiles and amphibians. Alfred A Knopf, Inc., New York. 720 pp. 106 Transactions, Missouri Academy of Science VoL 16

Conant, R. 1975. A field guide to reptiles and amphibians of eastern and central North America. Houghton-Mifflin Co., Boston. 429 pp. Fitch, H. S. 1970. Reproductive cycles of lizards and snakes. Univ. Ks. Mus. Nat. Hist. Misc. Pub!. No. 52: 1-247. Hurter, J., Sr. 1911. Herpetology of Missouri. Trans. Acad. Sci. St. Louis 20:59-274. Johnson, T. R. 1977. The amphibians of Missouri. Univ. Ks. Mus. Nat. Hist. Pub!. Ed. Ser. 6:1-134. Laposha, N. A., and R. Powell. 1982. Virginia valeriae -life history notes. SSAR Herp. Rev. 13:97. Porter, K. 1972. Herpetology. W. B. Saunders Co., Philadelphia. 524 pp. Smith, H. M. 1978. Amphibians of North America. Golden Press, New York. 160 pp. Smith, H. M., and E. D. Brodie, Jr. 1982. Reptiles ofNorthAmerica. Golden Press, New York. 240pp. 1982 Cary, Rhoades & Schmidt, Biological Sciences 107

Proestrus and Estrus of Missouri Coyotes

Diana Hallett Cary1 Missouri Cooperative Wildlife Research Unit School of Forestry, Fisheries and Wildlife, University of Missouri-Columbia Columbia, Missouri 65211 John D. Rhoades, D.V.M. 2 and Donald Schmidt, D.V.M. School of Veterinary Medicine University of Missouri-Columbia Columbia, Missouri 65211

Absrract: The onset of proestrus, estrus, and metestrus was determined in 6 coyotes over a 2-year period by the use of vaginal cytology. Proestrus, as indicated by a bloody discharge, persisted from 16 to 69 days and was not followed by estrus in subadult females (less than 1 year old). However, these same animals 1 year later did show smear patterns indicative of estrus. The mean onset of estrus was March 3. Key Wol!'ds: Coyote (Canis latrans), reproduction, estrus, Missouri.

The primary objective of this paper is to report the mean dates of onset of sequential events in the estrous cycle in Missouri coyotes (Canis /atrans). Duration of proestrus in subadult and adult coyotes was assessed. The breeding season and associated ovarian changes in the coyote were first described in 1938 (Hamlett 1938). Several years afterward, Whiteman (1940) discovered the long proestrum coyotes experience. Since that time reproductive anatomy (Gier 1960, Bekoff 1974) has been described for coyotes. Behavior and morphological and physiological changes which occur during the reproductive cycle have been inferred from dog-coyote hybrid studies (Kennelly and Roberts 1969, Silver and Silver 1969, Gipson eta!. 1975, Kennelly 1978). Exfoliative cytological changes occurring in vaginal epithelial cells during the cycle have also been described (Kennelly and Johns 1976).

Acknowledgments We thank Rollin D. Sparrowe for administrative support, Thomas S. Baskett for review of the manuscript, and University of Missouri-Columbia personnel and students who assisted with the penned animals. This is a contribution from the Missouri Cooperative Wildlife Research Unit (U.S. Fish and Wildlife Service; Wildlife Management Institute; Missouri Department of Conservation; and School of Forestry, Fisheries and Wildlife, University of Missouri-Columbia, cooperating), and from the Missouri Agricultural Experiment Station; Projects 182 and 184. The study was funded in part by Research

1Present Address - Missouri Department of Conservation, Columbia, MO 65201. 2 Present Address - School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA 70803. 108 Transactions, Missouri Academy of Science Vol. 16

Agreement #USDI 14-16-008-853 between the U.S. Fish and Wildlife Service and the University of Missouri; the senior author was an Edward K. Love Fellow.

Procedures The 6 female coyotes were captured in north-central Missouri at 2-7 weeks of age and were raised and maintained in outdoor facilities at the University of Missouri's Ashland Wildlife Research Area. They were housed 1 or 2 to 2.4 x 3. 7 m pens made with chain-linked fence and placed on a concrete pad. Male coyotes were also housed in these facilities. Weatherproofed but unheated shelters were also provided. Each animal was immunized against canine distemper, infectious canine hepatitis, leptospirosis, and rabies and dewormed as indicated by examination of fecal samples. The animals were fed a diet consisting of high protein dry dog meal supplemented with occasional horse meat scraps. Water was provided ad libitum. During the anticipated breeding season (January-April) in 1975 and 1976 6 subadult (less than 1 year old) and 5 adult females were observed twice a day for gross signs of proestrus and estrus. Onset of proestrus was determined by the appearance of a bloody vaginal discharge. A cotton swab was then used to collect representative cells from the vaginal mucosa. Samples were taken every 2 or 3 days from January 1 to March 19. Two slides were made for each animal. One slide was air dried and later stained with Wright's blood stain. The second slide was immersed immediately in a solution of 95% ethanol and ether in a 1:1 ratio and stained with Schorr's Trichrome stain. The smears were examined at 400X magnification. Onset of estrus was the first day vaginal smears showed characteristic cytological associations of cornified epithelial cells with pyknotic nuclei (Evans and Cole 1931, Gier 1960, Kennelly and Johns 1976).

Results Six female coyotes were observed for signs of proestrus and estrus during the anticipated breeding seasons (January-April) in 1975 and 1976. Onset and duration of the phases of the estrous cycle were determined for 5 subadult coyotes studied in 1975 and the same 5 animals plus 1 additional animal in 1976 Hable 1). All of these females showed sanguineous discharge from the vulva characteristic of proestrus in dogs and coyotes. The bloody discharge began early in the cycle, mean date of onset of proestrus was February 6 (SEM ± 7.82) for subadult and January 26 (SEM ± 4.43) for adult coyotes. The bloody discharge for Missouri females averaged 36.8 days (SEM ± 4.16) and ranged between 16 and 69 days. The duration of this phenomenon reported for dogs is an average of 9 days. Another early indicator of proestrus in dogs is vulval swelling. In all coyotes noticeable vulval swelling did not appear as early as did blood in the vaginal fluid. Vulval size (gross observation) increased most from February 10 to March 2 preceding estrus onset. These data corroborate information of Kennelly and Johns (1976) that vulval swelling occurred most during the weeks before onset of estrus. Slides of vaginal smears were examined under the microscope to determine typical changes in cell composition during the estrous cycle. Cellular associations observed were similar to those reported for the dog (Evans and Cole 1931) and the coyote (Kennelly and Johns 1976). Erythrocytes and leukocytes were common in proestrus. Epithelial cells with normal nuclei were oval or circular, non-cornified types. With the approach of estrus, fewer leukocytes appeared. Cornified epithelial cells with pyknotic nuclei predominated in slides indicative of estrus. The return of non-cornified epithelial cells marked the start of metestrus. Smears made during this phase of the cycle contained curled or angular-edged 1982 Cary, Rhoades & Schmidt, Biological Sciences 109

Table l. Dates of OIIUJet of proestrus, estrus, and metestms and end of proestrus in captive female coyotes hom north central Missouri., 1975-1976.

SUBADULTS ADULTS Animal Proestrus Proestrus Proestrus Estrus Metestrus Number Onset End Onset Onset Onset

1 Feb. 19 Mar. 18 Jan. 16 Feb.17 Mar. 2 2 Feb. 3 Mar. 20 Jan. 31 Mar. 4 Mar. 15 3 Feb. 3 Mar. 12 Jan. 31 Mar.l9 No data 4 Feb. 19 Mar. 20 Jan. 16 Feb.24 Mar. 12 5 Jan. 2 Mar. 12 Feb. 7 Mar. 9 Mar. 15 6 Feb. 22 Mar. 9

Mean Feb. 6 Mar. 15 Jan.26 Mar. 3 Mar. 11 (7.82)* (1.83) (4.43) (5.39) (3.08)

*Standard error of the sample mean. epithelial cells with nuclei denuded of cytoplasm. Other indicators of metestrus were debris and bacteria. Vaginal smears were taken for all (6) subadult females considered to be in proestrus. None of these showed smear patterns indicative of estrus. In 1976, all adult coyotes (20-21 months) had vaginal smear patterns indicative of estrus. The date of estrus onset ranged from February 17 to March 19. The mean date of onset as determined by vaginal cytology was March 3 (SEM ± 5.39). These data agree with breeding dates ranging from February 19 to March 6 as calculated for Missouri coyotes from whelping dates (Hallett 1977). Further checks by examination of ovaries were not performed, so whether these females experienced a true or false heat was undetermined. The variation of onset and duration of the reproductive phases among individual coyotes precludes direct comparison of these data with other studies. However, the mean date of estrus onset for this study was similar to the March 7 date reported for 41 southwestern coyotes (Kennelly and Johns 1976). During the first year of life, coyotes vary with regard to experiencing true estrus. The proportion of subadult coyotes that become sexually mature depends upon environmental conditions and the intensity of coyote control practices (Gier 1968, Knowlton 1972). Out of 19 subadult coyotes from Texas and Colorado 7 definitely ovulated, 4 could have ovulated according to their vaginal smear patterns, and 8 experienced only false heats (Kennelly and Johns 1976). Enders (1955) reported a semi-domesticated subadult coyote undergoing a false heat. All subadult females in this study failed to show signs of estrus; however, these same animals at 20-21 months of age did show vaginal smear patterns indicative of estrus. The incidence of true estrus in free-ranging subadult coyotes cannot be directly correlated with findings from pen studies. Behavioral response to human activity and intraspecific conflicts due to close proximity of other coyotes cause hormonal variations that affect the estrous cycle. However, during the same period of this study, Samson and Hill (1979) reported a lack of conception among subadult coyotes sampled from Missouri's free-ranging population. 110 Transactions, Missouri Academy of Science Vol. 16

Literature Cited Bekoff, M. 1974. Social play and play-soliciting by infant canids. Am. Zool. 14(1):323-340. Enders, R. K. 1955. False heat in the yearling coyote, Canis latrans. J. Mammal. 36(1):133. Evans, H.M. and H. H. Cole. 1931. An introduction to the study of the oestrous cycle in the dog. Memorial University, California 9(2):65-118. Gier, H. T. 1960. Estrus cycle in the bitch: vaginal fluids. Vet. Scope 5(2):2-9. -----· 1968. Coyotes in Kansas. Kansas Agr. Exp. Sta. Bull. 393. 118pp. Gipson, P. S., I. K. Gipson, and J. A. Sealander. 1975. Reproductive biology of the wild Canis (Canidae) in Arkansas. J. Mammal. 56(3):605-611. Hallett, D. L. 1977. Post-natal mortality, movements and den sites of Missouri coyotes, MS thesis, University of Missouri, Columbia, 144pp. Hamlett, G. W. D. 1938. The reproductive cycleofthe coyote. U.S. Dept. Agr. Tech. Bull. 616. 12pp. Kennelly, J. J. 1978. Coyote reproduction. Pp. 73-93 in Bekoff, M. (ed. ), Coyotes: biology, behavior, and management. Academic Press, New York NY. 384pp. _____ and J.D. Roberts. 1969. Fertility of coyote-dog hybrids. J. Mammal. 50(4):830-831. _____ and B. E. Johns. 1976. The estrous cycle of coyotes. J. Wild!. Manage. 40(2):272-277. Knowlton, F. F. 1972. Preliminary interpretations of coyote population mechanics with some management implications. J. Wild. Manage. 36(2):369-382. Samson, F. B. and B. D. Hill. 1979. Observations on coyote biology in north central Missouri. Prairie Naturalist 11(2):53-59. Silver, H. and W. T. Silver. 1969. Growth and behavior of the coyote-like canid of northern New England with observations on canid hybrids. Wild!. Monogr. 17. 4lpp. Whiteman, E. E. 1940. Habits and pelage changes in captive coyotes. J. Mammal. 21(4):435-438. 1982 Brown, Physical Sciences 111

Spin Glasses: A Two-Site Cluster Theory

H. A. Brown Physics Department University of Missouri-Rolla Rolla, Missouri 65401 USA

Abstract: A two-site cluster theory of spin glasses is presented. The Edwards-Anderson order parameters are calculated as functions of temperature, as are the susceptibility and heat capacity, and the phase diagram of the system is presented. The results are generally satisfactory except for-some dubious behavior at very low temperature where a transition from the spin glass to the ferromagnetic phase is predicted, which is unique and possibly spurious. Otherwise, second-order phase transitions are found for para/spin-glass/ferromagnetic states; there is a cusp in the susceptibility at the para/spin-glass phase boundary and a broad maximum combined with a discontinuity in the specific heat. Key Words: Spin Glass, Two-site cluster theory, Ising Model

Introduction Spin glasses are dilute magnetic alloys in which the magnetic moments are rather widely separated and have an interaction which is strongly oscillatory as a function of this separation, resulting in a peculiar magnetic phase which shows no conventional long-range order. The typical spin glass is a dilute (1 - 10%) solution of magnetic ions in a metallic lattice, for example, Fe in Au or Mn in Cu. Even though there is little or no long-range magnetic order, the local magnetic moments are fixed, independent of time. The fundamental interaction is assumed to be a long-range indirect exchange mechanism by way of the conduction electrons. The form of this RKKY 1 exchange interaction is

for two magnetic ions separated by R with kF the Fermi momentum of the conduction electrons of the metal. A schematic representation of the RKKY interaction and the magnetic moments in a spin glass is shown in Fig. 1. An alternative is a simply random interaction2. The important thing, apparently, is that the interactions conflict in their tendency to align and anti-align moments. The absence of long-range order can be shown by neutron diffraction3 but the occurrence of a phase transition is indicated by a cusp in the susceptibility, 4 and the presence of short-range order is indicated by small-angle neutron scattering5. A general Hamiltonian for the system is of the form 112 Transactions, Missouri Academy of Science Vol. 16

I

·' . '

J(R)

R

figure 1. A. schematic representation of the RKKY interaction and of the random magnetic moments in a spin glass (after K Binder35). where Jii is the exchange interaction between spins Si and Si and Ho is an external mag­ netic field. Because of difficulties associated with the random location of the magnetic spin sites and their essentially random interactions, most theoretical approaches have been limited to the Ising model in which Si = ± lJz = st In view of the randomness inherent in the system, it is not surprising that Monte Carlo techniques have been applied to evaluate the properties of spin glasses with some success6 , although results depend considerably on details of the various models. 1982 Brown, Physical Sciences 113

Numerous attempts have been made to determine distribution of the interactions in a self-consistent way using the Weiss molecular field theory. The best-known of these is the calculation of Edwards and Anderson7 They employed a symmetric distribution of the exchange integral magnitudes and introduced a new order parameter, q, which is the configuration average (average of all possible configurations of the spins in the random system) of the square of the thermal average of a spin. They were able to show the presence of cusps in the susceptibility and specific heat at-or near-the spin-glass/paramagnet phase boundary. Only the symmetry of the distribution function was needed to obtain an equation for q. Similar results were obtained by Sherrington and Kirkpatrick8 who treated a system of spins with a distribution function for the exchange integrals expressed as a Gaussian formula:

P (J .. ) "'exp - (J .. -J J2;2i 1 J lJ 0 .

These and other calculations suggest that the precise nature of the distribution is not critical. It should be noted that Monte Carlo methods also show a cusp in the susceptibility, but only a broad maximum, at a higher temperature, in the heat capacity. Another approach to the spin-glass problem involves the renormalization method of which there have been numerous applications with some degree of contradiction involved. Working in a d = (6 - E) - dimensional hyperspace the heat capacity shows a phase transition with a critical exponent a < - 1, whereas real-space normalization methods show a transition for d = 3 but not for d = 2 in one calculation; another calculation shows a transition ford = 2. A slightly different normalization approach has been used in which the distribution is

where p is a concentration parameter. This leads to a phase transition to a spin glass for all mixtures of ferro- and antiferromagnetic interactions, which is not observed. Finally, high-temperature series expansions of the susceptibility suggest that there will be no phase transition ford < 4, and other calculations indicate no phase transition ford~ 4 for the Heisenberg and X-Y models and the Ising model for d ~ 2, on the basis of thermodynamic inequalities. We note the inability of various calculations to agree on whether a phase transition exists in any given dimensionality and, if it does, what the order of the transition might be. In view of this, it seems to be more than a little over-optimistic to ask for the critical exponents of thermodynamic functions. Actually, doubts exist that the theories of spin-glass behavior are adequate for the definite prediction of a phase transition at al19 One thing that limited the accuracy of some of the aforementioned results was the unreliability of a calculational procedure known as the "replica trick." In forming the configuration average of the random but frozen-in spin arrangements, the proper quantity to average is the free energy or In Z, not Z, the partition function itself10. The replica trick consisted of using the algebraic identity xn_, ln X = ~.: (-n-) with the quantity X replaced by the partition function. There are some problems associated with extending the range of n from integers to real numbers and with interchanging the limit 114 Transactions, Missouri Academy of Science Vol. 16

n-o, and the thermodynamic limit, N-00 , where N is the number of spins. The limitations of the replica trick method have been discussed by van Hemmen and Palmer11 who explained, for example, why the calculated entropy terms out to have a physically unreasonable negative value for T __,... o. With or without the replica trick, it would be advantageous to have a theoretical approach that is somewhat more sophisticated than the molecular field theory and whose results did not strongly depend on the particular approximation used. Possible alternatives that have been considered include the use of mean-field theories that consider a cluster of two or more spins. Among these are the Oguchi pair approximation12·13; the BPW method14·15 which employs a cluster of a central atom and all its nearest neighbors; the Bethe lattice16·17 in which a central atom has n neighbors, each of which has n more neighbors and so on, all of these sets entirely independent. This is more exactly a tree than a lattice, but leads to mathematical simplifications. For an infinite Bethe lattice, all sites are equivalent.

Procedure The order parameter of Edwards and Anderson can be considered in two ways. We first define q as -2 2 q = <5 > /5 1 where S is the thermal average of a given spin and < > indicates an average over spin configurations. The usual order parameter for ferromagnets is m:

2 m = /5 .

In the spin-glass state, with magnetic moments randomly oriented up or down, the value of m is obviously zero. On the other hand, q is a correlation function. It can be redefined in the sense of a long-range order in time instead of space as18

3 52q = lim <5(t-O) 5(t-t')> = t .-+- 00

This correlation function, q, is a measure of the probability that at t = t' a given spin will be pointing in the same direction it was at t = o, in the limit t'-CX! The physical meaning of this correlation function can be seen by examining its extremes: q = 1 means a given spin maintains its direction for all time q = 0 means that this spin will have zero probability of doing so. Thus, the direction of any given spin is not important, only the configuration average of the square of its thermal average which has been replaced by the time correlation function, a replacement justified by the ergodic theorem. The spin-glass state is thereby defined as one in which the orientation of a given spin is fixed in time, i.e., frozen, without any spatial order. On the other hand, if the orientations of the spins do actually vary randomly in time, then m = o. The parameter q will also vanish and the system will be in a paramagnetic state. If there is a preferred direction along which the average spin moment is not zero, then both order parameters are non-zero and the system is in a ferromagnetic state. A little consideration shows that an antiferromagnetic state will consist of non-zero moments in two antiparallel directions, but with spatial order nevertheless. 1982 Brown, Physical Sciences 115

We consider an extension of the Oguchi pair approximation12 to the spin-glass problem. This offers some possible improvement over molecular field theories because, as is well known, the latter cannot include the effects of nearest-neighbor correlations, whereas a pair or cluster approximation in general can.

Results Details of the calculation leading to the partition function of the two-spin cluster and the thermal average of one of them are given in App. I. The results are given by: the partition function ' Z = L: e-sH 4 sk =±1 is evaluated using the two-spin Hamiltonian,

- sH = 2jl2 sls2 + ~ hlk sl + i h2L s2 +he (Sl+s2) , and

5 is the thermal average of spin S1. Here the reduced exchange integral (r3J) between the two spins is hz and the reduced field (r3Ho) includes the internal field, h1k due to the kth neighbor of spin 1 and the (reduced) external field, he. Assuming that the amorphous nature of the spin glass can be imitated by allowing the exchange and internal fields to vary randomly, a configuration average is taken (see App. II) and the results, though rather lengthy, can be abbreviated to increase clarity to

6

7

8

9

The summations are all over E = ±1, and k,l = 0 to n-1. The other abbreviations are found in eqtns II 11, 12. The derivative~~ will give the susceptibility, x• multiplied by the e absolute temperature divided by the Curie constant, and the derivative d~ is the heat capacity in units of the Boltzmann constant. The lattice coordination number is n. 116 Transactions, Missouri Academy of Science VoL 16

The equations for m and q can be considered as implicit functions F and G:

10

and

11

with m and q explicit functions of h0 , h, and he. The solutions for these order parameters are found by solving the pair of coupled

equations numerically after choosing the parameters like J 0 and J, H 0 and H, and, of course, n. Some examples of these results are shown in Figs. 2 and 3.

1

0 1

Figure 2. The ferromagnetic order parameter, m., vs. reduced temperature. (T c is the Curie temperature ofthe ferromagnet.) The curves shown

are for V :n J 0 /J = 2 and 3/2, top and bottom, resp.

We note that the ferromagnetic order parameter, m, behaves differently from an ordinary ferromagnet. For large values of Vn JJJ when J 0 ±J is always distinctly positive, the curve would show its usual monotonic approach to saturation. For lower values, such as Vn JJJ = 2, the anti-ferromagnetic (J 0 -J

1

q

0 1

Figure 3. The spin-glass order parameter, q, vs. reduced temperature. Tf is the "freezing" temperature of the glass. The curves shown are for

Y n J 0 /J = 2, 1, and 0, top to bottom. spins are free to flip even though the rest of lattice is frozen. This, in turn, means that these spins are acted on by exactly cancelling fields and are completely free. The ground state of a spin glass shows a behavior, using Monte Carlo techniques19, that is in agreement with this interpretation. That the simple two-state model used here can show this effect on ground state is encouraging. It indicates that the present model is, in this aspect, superior to molecular field theories which always predict saturation for q because they do not consider the possibility of competing interactions and the problem referred to as "frustration" to which they give rise (see below).

Depending on the values chosen for J0 and J (and n), thermal behavior of the system can show considerable differences as we have seen. A phase diagram of the systems will show that paramagnetism always obtains at high enough temperatures. At lower tempera­ tures either ferromagnetic or spin glass (or antiferromagnetic if we include the case

J 0 < 0) will take over, depending on the broadness of the distribution of exchange integrals for a continuous distribution or on J0 /J in our case where Jpq = J0 ± J. For very narrow distributions or J0 > > J the system should go from the paramagnetic to the ferromagnetic phase as the temperature is lowered. For broader distributions or smaller J0 /J the transition should be to the spin-glass phase. In Fig. 4, we find the phase diagram of the system as defined by the simultaneous solution of eq'ns (7) and (8) with no external field (he = 0), noting that h0 and h (in the arguments of the sinh and cosh functions s1,2 and c1.2 ) are themselves functions of m and q 0114, 15). 118 Transactions, Missouri Academy of Science Vol. 16

2

2 kT \7;}

0 2 Vn Jo/J figure 4. The phase diagram of an amorphous ferromagnet. A more typical result is shown in the inset on a reduced scale. (These are molecular field theory results (ref. 23), the dashed line indicating a first-order transition.)

We first note thateq'n 7 always has the solution m = 0. In that case, the pair of coupled equations reduce to a single implicit equation for q whose solution indicates the existence of the spin-glass phase. The boundary between this phase and that of a paramagnet is found by allowing q to approach zero (q - 0 with m = 0 since in the paramagnetic phase both m and q are zero). The explicit method of doing this is to expand (8) to linear terms in q

(quadratic terms in h) to find the expression (depending only onj0 andj) which defines the phase boundary. The ferromagnetic phase is defined as the region where both parameters, m and q are non-zero. To find the phase boundary, both (7) and (8) are expanded to linear terms in m and q giving a pair of linear homogeneous equations for the two parameters. Coefficients in the equations are functions of io and j and the existence of non-trivial solutions for m and q depends on the vanishing of the determinant of these coefficients and this determines the phase boundary. More precisely, at the ferromagnet-spin-glass phase boundary, q remains finite while m- 0. Atthis boundary, we need to find functional relationships amongj0 , j, and q. Done directly, this would require considerable algebraic manipulation because of the compli­ cated dependence of the two equations (7) and (8) on m and q. To avoid this, the equations were solved numerically for various values of j 0 and j and the phase boundary found from the condition mentioned before. 1982 Brown, Physical Sciences 119

The phase diagram obtained in this way (Fig. 4) shows the boundaries between paramagnet (m=q =0) and ferromagnet (mT'O, qtO) and spin glass (m=O, q ,toO). The transitions from one phase to another are all second order. The para/ferro and para/spin-glass boundaries are qualitatively similar to those obtained in several other theories7•14•15•20•21 Some disagreement arises from results predicting first-order transitions19•22•23 instead of second. However, there is general disagreement with the behavior shown in Fig. 4 at low temperature. The present method predicts that a ferromagnetic phase takes over at sufficiently low temperature so that a given system which is a paramagnet above T1 and a spin glass below, would become a ferromagnet at very low T. All other methods show the spin-glass phase persisting to absolute zero. Behavior of the heat capacity and susceptibility at very low temperature, as discussed below, suggests that the spin-glass to ferromagnet transition may be spurious. The ferromagnet would have a lower entropy and would satisfy Nernst's third law, but a spin-glass state could very easily be metastable because of the absence of a mechanism capable of flipping the frozen-in spins. Cluster methods, such as the one used here, do in general overestimate the possibility of a transition. This is because of the assumption that the average spin moment of the cluster is the same as the magnetization of the system, making it impossible for these methods to distinguish between finite and infinite clustering. Some experiments24•25 suggest that a low temperature spin-glass phase is characterized by the presence of randomly oriented ferromagnetic clusters. The Oguchi method used here, as well as other similar approximations, might not distinguish between that kind of a clustered spin-glass and the ferromagnetic phase. Because of the random occurrence of both positive and negative exchange interac­ tions, it is clear that the tendency of a spin and its neighbors to be aligned or anti-aligned cannot always be satisfied. Such an effect has been given the name ''frustration,'' and those interactions that cannot be satisfied are referred to as frustrated bonds. Since frustrated bonds raise the energy of a system, the ground state should have a minimum number of them. Nevertheless, there will be a large number of spin arrangements with the same energy so that ground state will be highly degenerate19·26. If the dominant exchange

interaction is positive (J 0 >J) then some residual ferromagnetism might be expected at low temperatures because it would then require less energy to frustrate a negative bond than a positive one. Returning to the average magnetization shown in Fig. 2, we note the increased tendency toward ferromagnetic ordering at low temperatures. The inflection point seen in the curves may be associated with spin reorientations which preferentially frustrate the negative exchange bonds, producing a state of lower energy. In Fig. 5, we show the calculated values of the (reduced) susceptibility as a function of temperature which, from II 10, is found from

12

The shape of the curves around T = T1 is similar to that found elsewhere, both theoreticalll·20•27-31 and experimentally32·33 We note there is a pronounced cusp which is largerfor larger values ofYnJ 0 /J forT =T 1somewhatlikethe behavior of an antiferromagnet near the Nee! temperature. For low enough temperatures, our susceptibility diverges as the spin-glass/ferromagnet phase boundary is approached, as it should if there were such a transition. Almost all other results, experimental and theoretical, show a simple monotonic decrease of x with T below the cusp. Only one other theory27 shows a low-temperature 120 Transactions, Missouri Academy of Science Vol. 16

II II II I I

X

0 2

Figure 5. The reduced susceptibility, x/(N~J-2/2J) vs. reduced temperature for Vn JJJ=0.6 and 1.0, lower and upper curves, resp. divergence similar to ours and this is difficult to reconcile with the fact that the phase diagram they find is the usual one, with the spin-glass phase persisting down to T = 0. The magnetic part of heat capacity of a spin glass has an uncertain temperature dependence from various theoreticaf·20•28•34•35 and experimental results32•36. Some of these results show a broad maximum and others a cusp around T1. Our results in Fig. 6 show a gentle decrease in heat capacity below Tt. not unlike the broad maximum results. However, there is a discontinuity at T1 (this is a second-order transition) and at higher temperatures, the heat capacity decreases roughly as 1 2. The discontinuity in the specific heat curve at T1 is a product of the Oguchi approximation. In other theories, no such break is predicted, only a discontinuity in the slope17•20•27•28. Heat capacity is proportional to the negative temperature derivative of the short-range order parameter, <51 · 52>, and this short-range order persists in the paramagnetic region, decreasing about as predicted. At very low temperatures there is another problem: as the ferromagnetic phase boundary is approached, heat capacity appears to increase without limit. While such a singularity for the susceptibility is expected at this phase boundary, it is not for heat capacity. Actually, any such singularity in heat capacity must be weak enough so that the entropy, c S = T dT, does not exceed k In 2 per spin. 1982 Brown, Physical Sciences 121

I I

C/k

0 1

Figure 6. The reduced heat capacity, Cv/(k!VnJ), vs~ reduced temperature.

The results are rather insensitive to Vn J0 /J. For both Figs. 4 and 5, n = 6 was chosen.

It must be concluded that, although the model is generally successful, its very-low­ temperature predictions are suspect and that the part of the phase diagram referring to the spin-glass to ferromagnet phase change in this region may be incorrect. The author acknowledges with gratitude the debt owed to Dr. Mark Jursich for his contributions to this work. 122 Transactions, Missouri Academy of Science Vol. 16

APPENDIX

The two-spin Hamiltonian is

Il

In the Ising model, S = ± 1/2. The fields Hk' are composed of two parts,

the internal field on spin~' Hk, and an external field, He' acting on all spins.

12

With obvious abbreviations, using s = 1/kT,

I3

The partition function is

14

From this we obtain

I5 s - afnZ- [e!j12 sinh '(h +h +2h ) + e-tj12 sinh t 1 -atil- 2 l 2 e

I7 s = t ~~nz = [e!jl2sinh t (h 1+h 2+2h 2)]/Z = t (s1+s2) e

5=5 = .1 afnZ = [e!j12 cosh t (h '+h ') - e-tj12 cosh t (h '-h ')]/2Z IS 12 2ajl2 1 2 1 2

19 ~~e = [etj12 cosh t (h 1+h 2+2he)]/Z-[ejl2 sinh2t(h1+h 2+2he)];z2 .

The last is proportional to the susceptibility, actually kTx, to order he 2

We now turn to the problem of considering the fields H1 and H2 in detail. We note that n-1 n-1 IlO H' = z Hl k , H' = z H2t 1 k=l 2 f=l 1982 Brown, Physical Sciences 123

where Hlk and H2l are the fields produced by individual neighboring

atoms: k,l = 1 ,2---(n-1). The modifications that follow are shown below. We now write H as

Ill

!12

!14

and, finally,

The Oguchi assumption13 relates the internal field to the average spin by

!16

APPENDIX II

The amorphous nature of the material is introduced through the rather artificial assumption that the exchange integral and the internal fields vary randomly according to

Ill and

I I 2

with equal probability for the+ or- sign. A configuration average is then taken by summing over all possible choices for Jij and Hij and dividing by total number of configurations

considered; 22n-l = 22(n-l)+l. In the sum, each term is weighted by the 124 Transactions, Missouri Academy of Science Vol. 16

binomial factors giving its likelihood of occurrence. In forming the sum, we note that

a) the maximum of I hlk +I h2£ + 2he = 2(n-l) h0+2(n-l)h+2he

and b) the minimum of E hlk + E hu + 2he = 2(n-l) h0 -2(n-l)h+2he whi 1e

c) the maximum of I hlk + E h2£ = 0 + 2(n-l) h d) the minimum of 1: h1k + l: hu = 0- (2n-l) h Hence,

and n-1 II 4 cosh t (I h1k-I h2£) = I cosh (k-£) h k,£=0 since the arguments change by integral steps. Similar relations hold for the two sinh functions. The proper binomial factors are, of course, (nkl)(n[1). The configuration averaging process can be described by

II 5

with j 12 = j 0 + Ej.

Thus

1 II 6 = -----1 IEI (nk- 1 )(n~ 1 )tn{et(jo+Ej)cosh[(n-1)h +(n-1-k-!)h+h] 22 n- Eki "- o e

+e-t(jo+sj)cosh(k-!)h} + const. and

II 7

>.(j +E.) e• o J sinh [(n-1)h o+(n-1-k-!}h + h]e

(The denominator in the curly brackets is tZ).

We note that = = and 1982 Brown, Physical Sciences 125

Further,

- 2 1 (n-l)(n-1 IIB 4 = ~ ~ k t ) 2 c:,k,t

since the cross terms ~t (n~ll("~ 1 )sinh[(n-l)h 0 +(n-l-k-t)h+he)sinh(k-t)h=O and finally

(as it must by definition) and

as1 a<51> 1 ~ (n-l)(n-ll II 10 2 = 2 ~ = ~ €:' k ,t k t e e 2

et(jo+c:j)cosh[n-l)h +(n-1-k-t)h+h] (H) e(jo+c:j)sinh2[(n-l)h +(n-1-k-t)h+h] - o e ( 2 ) 2 } "' kT X + 0 h (tZ) e If we let

1 (n-l)(n-1) !Ill ckt = ~ k t 2 and

1 ( • + ) 2 I I 12 c1 = e Jo "'j cosh[(n-l)ho +(n-1-k-t)h+h]e

.1. ( • + . ) s e 2 Jo "'J sinh[(n-l)h +(n-1-k-t)h+h] 1 = o e

then

II 13 = ~ Ckt tn (c1+c 2) + const c:,k,t and the rest of the equations 6 to 9 also follow. 126 Transactions, Missouri Academy of Science Vol. 16

In view of the assumption (I 15), the values of hand h0 are expressed

by

II 14

and

References 1. C. Kittel, Solid State Physics 22, 1 (1968), orlntroduction to Solid State Physics, Wiley 1976, p 554. 2. K. Binder and R. Schader, Solid State Comm. 18, 1361 (1976). 3. A. Arrott, J. Appl. Phys. 36, 1093 (1965). 4. V. Canella and J. A. Mydosh, Phys. Rev. B6, 4220 (1972). 5. A. P. Murani, Phys. Rev. Lett. 37, 450 (1976). 6. K. Binder, J. de Phys. 39 C6-1527 (1978). 7. S. R. Edwards and P. W. Anderson, J. Phys. F5, 965 (1975). 8. D. Sherrington and S. Kirkpatrick, Phys. Rev. Lett. 35, 1972 (1975). 9. W. Kinzel and K. H. Fischer, J. Phys. C 11, 2115 (1978). 10. R. Brout, Phys. Rev. 115, 824 (1959). 11. J. L. van Hemmen and R. G. Palmer, J. Phys. A 12, 563 (1979). 12. T. Oguchi, Prog. Theor, Phys. 13, 148 (1955). 13. M. Jursich and H. A. Brown, J. Phys. Soc. Jpn. 47, 1411 (1979). 14. S. Katsura and S. Fujiki, J. Phy. Soc. Jpn. 12, 1087 (1979). 15. M. W. Klein, L. J. Schowalter, and P. Shukla, Phys. Rev. B 19, 1492 (1979). 16. S. Katsura and M. Takizawa, Prog. Theor. Phys. 51, 82 (1974). 17. T. Oguchi andY. Ueno, J. Phys. C 10, 2981 (1977). 18. S. Kirkpatrick and D. Sherrington, Phys. Rev. B 18, 6270 (1978). 19. S. Kirkpatrick, Phys. Rev. B 16, 4630 (1977). 20. S. Kirkpatrick and D. Sherrington, Phys. Rev. B 17, 4384 (1978). 21. B. W. Southern, J. Phys. C 9, 4011 (1976). 22. J. F. Fernandez and D. Sherrington, Phys. Rev. B 18, 6270 (1978). 23. T. Kaneyoshi, J. Phys. C 6, 3130 (1976). 24. G. J. Niewenhuys, B. H. Verbeck, and J. A. Mydosh, J. App. Phys. 50, 1685 (1979). 25. B. R. Coles, B.V.B. Sarkissian, and R. H. Taylor, Phil. Mag B 37, 489 (1978). 26. S. Alexander and P. Pincus, J. Phys. A 13, 263 (1980). 27. S. Katsura and S. Fujiki, J. Phys. C 12, 1087 (1978). 28. E. M. Soukoulis and K. Levin, Phys. Rev. B 18, 1439 (1978). 29. D. A. Smith, J. Phys. F 5, 2148 (1975). 30. J. R. David and T. J. Hicks, J. Phys. F 9, 753 (1979). 31. D. J. Thouless, P. W. Anderson, and R. G. Palmer, Phil. Mag. 35, 593 (1979). 32. G. Heber, J. Magn. Magn. Mat. 2, 47 (1976). 33. K. H. Fischer, Physica 86-88 B, 813 (1977). 34. K. H. Fischer, Solid State Comm. 18, 1515 (1976). 35. K. Binder, Festkorperprobleme XVII, 55 (1977). 36. D. L. Martin, Phys. Rev. B 20, 368 (1979). 1982 Austin & Dowell, Physical Sciences 127

Compactness of Missouri's Counties

Robert F. Austin and Tom Dowell Department of Geography University of Missouri-Columbia Columbia, Missouri 65211

Abstract: Compactness, an aspect of spatial shape, is measured for Missouri's 115 counties. Contrary to expectations based on the nature of township and section survey methods, these counties exhibit marked, statistically significant rectilinear elongation. Key Words: Spatial shape, shape indices, compactness

The theories of central places and regional space packing have, in combination, given rise to sets of expectations regarding 1) hierarchical functions and 2) two-dimensional shapes of economic and political ''service areas.'' The degree to which the first of these sets of expectations is realized in a given region is easily determined. The second of these expectations is more difficult to examine, however, for two reasons. First, the lattice of regular hexagonal service areas proposed by classical (that is, non-random) location theory assumes an isotropic surface for its formation. However, the earth's surface is, with minor local exceptions, anisotropic. Second, even if the lattice of polygons is generated on an isotropic surface, we cannot describe the shapes of the polygons with precision. Many efforts have been made to define a single, unambiguous, easily interpreted measure of shape. There is as yet no such measure, although recently initiated studies using Fourier analysis appear promising (Blackwell et al. 1982). For a general review of shape analysis in geography, see Boots and Lamoureaux (1972), Whittington et al. (1972), Frolov (1975), Moellering and Rayner (1979) and Austin (1983). Two factors recommend the application of a widely used, although somewhat restricted, methodology to the question of the shapes of Missouri's counties. First, although we cannot measure shape per se, we can measure certain aspects of shape including compactness (a critical component of location theory). Compactness is measured here in two ways: by means of a linear dimension compactness ratio and by means of a radial vector shape index. The methodology has been presented in detail and discussed by Austin ( 1981, 1982a, b, 1983) and others. A study of Missouri's counties using this methodology will constitute an additional case for empirical analysis. The second reason for examining the shapes of Missouri's counties is related to the basis of their delineation. Whereas classical location theory assumes an "organic" or ''natural'' development of boundaries, Missouri's counties were surveyed and, at least in theory, delimited using the township and section method. This method of surveying and delimiting boundaries should result in a set of square regions. Deviation from this pattern would require some explanation, however trivial. In other words, we will apply a methodology first outlined for the study of a regular hexagonal lattice to the study of another of Keppler's regular tesselation~ lattice of squares.

Linear Dimension Compactness Ratio Because the circle is the most compact two-dimensional polygon it has been used as the reference polygon for measuring compactness in several different expressions. The 128 Transactions, Missouri Academy of Science Vol. 16 simplest of these may be written: C = A,IAc (1) where C is the compactness ratio, A, is the area of a given region or polygon, and Ac is the area of the smallest enclosing circle (circumcircle) of that region. Austin (1982a, b) suggested that such a measure should be considered a special case of Stoddard's (1965) ellipticity index which in turn may be profitably· considered in the context of a subset of conic sections. In a slightly different context, Rademacher and Toeplitz (1957) discussed such measures in terms of the [minimum] "spanning circle of a finite set of points" (d. Chrystal 1885). In practice, C is generally calculated using the reduced expression: C = 1.2732 A,/D2 (2) where Dis the diameter of the circumcircle. Allowing for the effects of rounding the value of pi (implicit in rephrasing from equation 1 to equation 2) to four decimal places, equilateral triangles, squares, and regular hexagons (those polygons generating regular tessellations) have C-values of 0.413, 0.637, and 0.827, respectively. A rectangle of dimensions L-by-2L has a C-value of 0.509, while a circle has a value of 1.0. The range of values for Missouri's counties was from 0.197 to 0.789, with a mean of 0.535 and a standard deviation o£0.110. A simple student's t-test [ (x-1-L}/(sY n-1 )] ofthese values permits us to say with a level of confidence in excess of 99.999 that Missouri's counties do not in aggregate resemble equilateral triangles, squares, regular hexagons, or circles in terms of compact­ ness, but that they do resemble the L-by-2L rectangle. We wish to emphasize, in response to a reviewer's query, that we are making inferences with regard to a subset of the universe of equiangular polygons -those which are "space-filling" -rather than with regard to the universe of U. S. counties. The choice of Missouri's counties was based on regional interest and availability of data.

Radial Vector Shape Index A second method of measuring compactness is based on the definition of a circle as the set of points equidistant from a given point p. Let the distance from these points to p be written r. We may construct a set of n line segments of length r such that they are equally spaced (that is, each pair of radial lines forming an angle of 360°/n}. For example, 161ine segments or radials would be drawn in such a way that each adjacent pair of radials formed an angle of 22. 5°. We may sum the lengths of the radials and determine the proportion that each radial contributes to the sum. The proportions will be equal-with 16 radials, 6.25%. For polygons other than circles, some or all of the radials will contribute different proportions to the sum of radial lengths. Boyce and Clark (1964) suggested that we may measure the deviation of the shape of a given polygon from the shape of a circle by comparing lengths of corresponding radials. This is done by summing absolute values of the difference between the proportion each radial ri contributes to the sum of lengths and the proportion that the corresponding radial of a circle would contribute. The index R may be expressed mathematically as: i=n i=n R = 2: I (100r/ 2: ri) - (100/n) I (3) i=1 i=1 This index is sensitive to the number and orientation of the radials (Cerny 1975, Clark and Gaile 1973), but a standard method of application has been developed: the center of gravity (barycenter) of the polygon is substituted for the point p and at least 16 radials are used, with one oriented due north. Under these conditions, and with n = 16, the range of 1982 Austin & Dowell, Physical Sciences 129 values is from 0 for a circle to 175 for a straight line. If an infinite number of radials is used, the maximum value approaches 200. For Missouri's counties, index values range from 8.394 to 71.599, with a mean of 17.363 and a standard deviation of 9. 722. With 16 radials the value for an L-by-2L rectangle is approximately 30.556 and the value for a square is approximately 11.767. These values assume that at least one radial is drawn to a vertex of the polygon, hence the qualifier. Although the value for a square is closer to the mean value for Missouri's counties, we may not (using the test and significance criterion discussed above) reject the statistically significant possibility of correspondence with either reference shape. In other words, this index also reveals substantial linear elongation.

Conclusions The township and section survey system that prevailed through most of the U. S. west of the Appalachian Mountains is in theory a survey system that employs a grid of square townships, each of which is subdivided into square sections. In practice, of course, the projection of a spherical surface onto a planar surface does result in some distortion for which a correction must be made. In theory, then, the shapes of counties west of the Appalachians surveyed after 1785 should reflect the underlying geometry of this survey system. In fact, however, the counties of Missouri reflect in their configurations (at least in aggregate) a remarkably high degree of elongation according to both of the measures used here. The nature and degree of this "distortion" exceeds that which might reasonably be expected from a simple spherical projection for a small area. Rather, we must conclude that some physical or cultural feature (or set of features) has influenced the delineation of Missouri's counties in, it should be noted, a remarkably consistent manner. Although this linear distortion is surprising in light of the survey methods, it is less so when similar results of analyses of other areas are considered (Austin 1982b). One interesting question this raises is the determination of the agent or agents responsible for this similarity. We have considered several possible simple explanations. The first possibility considered was the relationship between frontage along the Missouri and Mississippi rivers and elongation, either parallel or orthogonal to the river. Correlations between frontage and elongation were very weak and not significant. A second possible explanation that we considered was related to the fact that some of Missouri's counties might have been delimited, at least in general terms, on the basis of French and other early settlement patterns. "Long-lot" counties are not significantly different in terms of compactness from township and section counties. We also found that counties on the state borders are not significantly different from non-border counties. Finally, an explanation in terms of macro­ and meso-physiographic regions was attempted. This too was unsatisfactory. We are of the opinion, currently being tested, that a terrain index may be the best single correlate for compactness. Further examination will be necessary.

References Austin, R.F. (1981), 'The Shape of West Malaysia's Districts," Area, 13:2, 145-50. Austin, R.F. (1982a), "A Definition and Estimate ofTemporal Area," Professional Geographer, 14:2, 297-304. Austin, R.F. (1982b), "On Contiguity, Contact Numbers and Compactness: Reply to Boots and Blair," Area, 14:2, 127-31. Austin, R.F. (1983) "Bibliography of Measures of Spatial Shape," Vance Bibliographies. In press. 130 Transactions, Missouri Academy of Science Vol. 16

Blackwell, P., Johannsen, J., Chang, S. and Austin, R.F. (1982), Unpublished research on polar coordinate, discrete Fourier transforms and shape analysis. Boots, B.N. and Lamoureaux, L.F. (1972), Working Notes and Bibliography on the Study of Shape in Human Geography and Planning, Council of Planning Librarians, Exchange Bibliography No. 346. Boyce, R.R. and Clark, W.A.V. (1964), "The Concept of Shape in Geography," Geographical Review, 54, 561-72. Cerny, J.W. (1975), "Sensitivity Analysis ofthe Boyce-Clark Shape Index," Canadian Cartographer, 12:1, 21-27. Chrystal, G. (1885), "On ,t.he Problem to Construct the Minimum Circle Enclosing n Given Points in a Plane," Journal of Tfie Edinburgh Mathematical Society, 3, 30-33. Clark, W.A.S. and Gaile, G. (1973), "The Analysis and Recognition of Shapes," Geografiska Annaler, 55B:2, 153-63. Frolov, Y.S. (1975), "Measuring the Shape of Geographical Phenomena: A History of the Issue," Soviet Geography Review and Translation, 16, 676-87. Moellering, H. and Rayner, J.N. (1979), Measurement of Shape in Geography and Cartography, Department of Geography, Ohio State University, Numerical Cartography Laboratory (NSF SOC77-11318). Rademacher, H. and Toeplitz, 0. (1957), The Enjoyment of Mathematics (trans. Zuckerman, H.), Princeton: Princeton University Press. Stoddard, D. (1965), "The Shape of Atolls," Marine Geology, 3:5, 369-83. Whittington, G., Beavan, K.S.O. and Mabin, A.S. (1972) Compactness of Shape: Review, Theory and Application, Environmental Studies: Occasional Paper No. 7, Johannesburg: University of Witwatersrand. 1982 Fairweather & Kurtz, Social Sciences 131

Determining Forest Management Dispositions of Missouri Eastern Ozark Nonindustrial Private Forest Landowners1

John R. Fairweather and William B. Kurtz School of Forestry, Fisheries and Wildlife University of Missouri-Columbia Columbia, Missouri 65211 David C. Baumgartner USDA Forest Service East Lansing, Michigan 48823

Abstract: Forest management decisions by nonindustrial private forest (NIP F) owners in Missouri are important in determining the future availability of timber from their lands since this category of owner controls 84 percent of the State's commercial forestland. Sixty NIPF owners in the Missouri eastern Ozarks sorted a series of statements along an agree-disagree continuum (Q-sort) to place themselves into groups with common forest management motivations and objectives. An analysis of stated forest management motivations and objectives by type of NIPF owner provides valuable insight into owners' decision making processes relative to the adoption of timber management practices. Key Words: nonindustrial private forestry, Q-sort, forest management, Ozarks, public programs

Forest management decisions by nonindustrial private forest (NIPF) owners play a crucial role in determining the future availability of timber and nontimber products from their lands. With society's needs becoming more diverse and input costs continually rising, decision-making guidelines to help the forest manager attain a higher level of production efficiency and also to contribute to increased public satisfaction will be of crucial importance. The availability of such guidelines is vital to the forest industry of Missouri since some 84 percent of the state's 12.4 million acres of commercial forestland is controlled by farmers and other private owners (Spencer and Essex 1976). Our purpose in this paper is to examine the rationale for producing change in owners' attitudes toward management for timber production. Management decisions by NIPF owners are complex since these landowners typically have different mixes of motivations and objectives for their forests. Such mixes reflect an owner's disposition relative to management for timber production. However, it is generally recognized that for the majority of NIPF owners timber production is not a primary land-management objective. For both private industry and public agency foresters to be more effective in encouraging forest management for timber production by this group it is essential that forest landowner objectives and motivations be recognized and related to their forest management strategies.

1Funding for this study was provided by the USDA Forest Service, North Central Forest Experiment Station, Carbondale, Illinois, under Cooperative Agreement Number 13-601, Classification of Private Nonindustrial Forest Landowners by Management Objectives and Motivations. 132 Transactions, Missouri Academy of Science Vol. 16

NIPf Studies in the Missouri Eastern Ozarks In an earlier study of NIPF landowners in the eastern Missouri Ozarks, Kurtz and Lewis (1981) described four landowner types identified as the Timber Agriculturalist, Timber Conservationist, Forest Environmentalist and Range Pragmatist. A non-random sampling technique consistent with the suggestions of Stephenson (1953) was used in selection of the 60 subjects interviewed (Lewis 1979). Use of the non-random selection process required that another study of the population be undertaken to determine whether lack of random sample selection had any effect on results. In the restudy (Fairweather 1979) a different group (N =58) of landowners in the same geographical area was surveyed two years later using the same method. Three types of landowners were identified in the second study and have been named Practical Forester (n = 17), Dedicated Farmer (n = 20) and Concerned Ecologist (n = 8) (Fairweather 1979). Thirteen landowners could not be clearly typed. The restudy results generally match results of the original study. The Concerned Ecologist type identified by Fairweather (1979) is comprised ofthe Timber Conservationist and Forest Environmental­ ist types described by Kurtz and Lewis (1981). The Practical Forester type (Fairweather 1979) is similar to the Timber Agriculturalist (Kurtz and Lewis 1981), and the Dedicated Farmer type (Fairweather 1979) is similar to the Range Pragmatist (Kurtz and Lewis 1981 ). Taken together the results of both the original study and the restudy provide a detailed, valuable and coherent basis for the improved formulation of resource management policy. The fact that only three types were described for the restudy reflects, in part, an arbitrary methodological decision. It was found that use of three types maximized the explanatory value of interpretations within each type. Although similar implications for policy can be drawn from combined results of both the study and the restudy, interpretation of the three restudy types provides additional understanding of landowner motivations and objectives. Landowner type identification in both studies was accomplished through use of the Q-sort technique (Stephenson 1953). With Q-sort, each landowner interviewed sorts a series of 60 selected statements relative to forest management. Three statements-­ positive, neutral and negative-were developed for each of the 20 cells shown in Table 1. Statements were obtained from previous interviews with NIPF landowners, professional foresters and a review of relevant literature. Each'landowner places the statements along an agree-disagree continuum which characterizes their attitude towards a particular topic. Landowners are then grouped on the basis of their common sorting patterns, and their forest management attitudes are described.

Management Dispositions of NIPF Owners The pattern of emphasis of motivations and objectives for each of the three types is shown in Table 1. The importance of a given combination of management motivation with objective for the respective owner types is measured by the numbers within each cell. These numbers are raw scores from the subject Q-sort and reflect degree of respondent approval or disapproval for the Q-sort statements. The respective subtotals shown in the margins, indicate the relative importance of the particular management motivations and objectives. For example, timber production is the most important management objective for the Practical Forester type while preservation is most important to the Concerned Ecologist type. Distinctive patterns of objectives and motivations emerge to clarify forestland management dispositions of the three types. From a detailed interpretation of data within 1982 Fairweather & Kurtz, Social Sciences 133

Table 1. Importance of management motivations and objectives by landowner type (* - highest score).

OBJECTIVES

Timber Wildlife- T y Production Recreation Grazing Preservation p e Subtotal 1 11 6* -4 2 15 Financial 2 13* 3 5* 1 22* Return 3 8 5 -4 9* 18

1 12* -2 -1 6* 15 M Investment 2 9 8* 5* 3 25* 0 T 3 8 3 -5 5 11 I v 1 9* 5 -4 9 19* A Aesthetics 2 6 4 5* 2 17 T I 3 3 9* -2 13* 23* 0 N 1 5 2 0 -2 5 s Residence 2 5 3 7* -3 12* 3 1 3 1 1* 6

1 3* 3* -6 7 7* Social Re- 2 -1 2 -3* 4 2 sponsibility 3 1 -1 -5 9* 4

1 40* 14 -15 22 Subtotal 2 32 20* 19* 7 3 21 19 -15 37*

NOTE: Type: 1) Practical Forester; 2) Dedicated Farmer; 3) Concerned Ecologist

Table 1, and an analysis of other data derived from the same source, a more extensive description of each of the types is possible. The first type, the Practical Forester, has the primary goal of continued management and production of timber. He harvests timber and carefully manages his woodlot to ensure future supplies because of concern about the future of woodland. Grazing cattle is incom­ patible with his forestry operations so he opposes forest grazing. He has a modest interest in wildlife because the welfare of wildlife is linked to the welfare of the forest resource. The enjoyment experienced from owning a woodlot is derived from an attraction to the forest, the security of growing a valuable product and the challenge of managing resources properly. The Dedicated Farmer (Type 2) is characterized by an ever present business attitude combined with a strong interest in cattle farming. Strong feelings about the need to produce, and to manage resources to efficiently achieve one's goals are emphasized. Although he is a:ware that excessive conversion jeopardizes future supply, he is tempted to 134 Transactions, Missouri Academy of Science Vol. 16 convert woodland to pasture if it is profitable. Cattle grazing operations are given first importance-woodland is seen as useful for forest grazing rather than long-term timber management. Wildlife is appreciated in the woods for hunting enjoyment rather than aesthetic appreciation. The Concerned Ecologist (Type 3) is oriented to preservation and conservation of forestland with a long-term perspective regarding its future. He enjoys the non-economic benefits of ownership and aesthetic returns from the resource. Beauty, pleasure derived from walking through the woods and the joy of seeing wildlife are key facets of ownership. This type of landowner is most likely to encourage wildlife by growing food plots and discouraging hunters. Although timber is harvested, this is not a primary management objective. Instead, he is more concerned about broad issues of the future of forestland and the conservation of wildlife.

Encouragement of Timber Management by NIPf Owners The Practical Forester needs encouragement to implement correct management practices and to keep up to date. Excessive contact is unwarranted because he is already dedicated to the cause of forest management. The Dedicated Farmer responds well to the economic aspect of forest management so information on its profitability will be of interest to him. He is business oriented and could be expected to respond to changes in stumpage price. Valid knowledge on the effects of forest grazing is also necessary to ensure the perpetuation of his woodlots. It may be necessary to encourage fencing off woodlots to protect regeneration. He is concerned about the future of forestland so an appeal to achieve this objective may be effective in encouraging him to manage his woodlot in a productive way. To encourage the Concerned Ecologist to harvest timber and manage for this goal, requires that valid information on the effects of timber management practices on wildlife and amenity values, be made available. There will always be a problem with the strong aesthetic appreciation element of this type. Harvesting may always be seen as deleterious to beauty no matter how carefully executed. It should be noted that profitability is important to all three types, especially the Dedicated Farmer and the Practical Forester. For most, the increase in profitability from a stumpage price increase is sufficient to increase their awareness of the value of woodland because they are more likely to protect a valuable asset than one that is marginally profitable. Knowledge of these three types of forest landowner allows professional foresters to better understand their clients' decision-making process. Although any one individual may not exactly fit the above descriptions, and although some landowners will exhibit characteristics of all three types, these descriptions allow greater understanding of the landowner's perception of forestry. This knowledge is important for the development and delivery of more effective public assistance programs to encourage timber production while also assisting NIPF owners in achieving the respective goals for use of their forests.

Literature Cited Fairweather, J. R. 1979. Types of forest landowners in the Missouri eastern Ozarks. Unpublished M.A. thesis, Department of Rural Sociology, University of Missouri-Columbia, 114 p. Kurtz, W. B. and B. J. Lewis. 1981. Decision-making framework for nonindustrial private forest owners: An application to the Missouri Ozarks, Journal of Forestry 79(5): 285-288. 1982 Fairweather & Kurtz, Social Sciences 135

Lewis, B. J. 1979. Private nonindustrial forest landowners in the Missouri Ozarks: a Q-methodological establishment of types. Unpublished M.S. thesis, School of Forestry, Fisheries and Wildlife, University of Missouri-Columbia, 251 p. Spencer, J. S., Jr. and B. L. Essex. 1976. Timber in Missouri, 1972. USDA Forest Service Resource Bulletin NC-30, 108 p. North Central Forest Experiment Station, St. Paul, Minnesota. Stephenson, W. 1953. The study of behavior: Q-technique and its methodology. University of Chicago Press (Midway reprint), Chicago, Illinois. 376 p.

1982 Masters & Haemmerlie, Social Sciences 137

Effects of a Regular Exercise Program on Geriatric Residents in a Nursing Home

Mary Masters and Frances M. Haemmerlie Department of Social Sciences University of Missouri-Rolla Rolla, Missouri 65401

Abstract: The positive effects of a regular exercise program, with and without music, on the psychological well-being of the elderly residents in a nursing home was assessed. Fifteen residents in a medium care nursing home (ages 59-89) were randomly selected and assigned to one of two experimental groups receiving the exercise program, or to a no treatment control group which did not receive the program. One experimental group had music accompanying their sessions while the other did not. Two life satisfaction measures and a nurse's rating of subjects' physical and psychological well-being were assessed one week prior to and immediately after the exercise program in all three groups. Significant differences were obtained between the two experimental groups and the control group on one measure of life satisfaction and the measure of psychological well-being. However, the two experimental groups did not differ significantly from each other suggesting the exercise program had a positive effect on the elderly subjects but that the music did not have an additional facilitative effect. Key Words: psychological well-being, music and exercise, reduced deterioration.

Exercise or movement is recognized as valuable for maintaining health at any age. Low levels of activity are frequently observed in the frail elderly and especially in those living in residential environments such as nursing homes and long term care facilities. Comstock, Mayers & Folsom (1969) have shown that low rates of motor behavior adversely affect respiratory and circulatory systems in elderly nursing home residents, and Bonner (1969) found reduced levels of motor behavior to produce bladder and bowel dysfunctions. On the other hand, the positive effect of physical exercise on the physical health (deVries, 1970) and cognitive functioning (Powell, 1974) of institutionalized elderly patients has been established. However, the possible positive benefits of regular exercise on the general psychological well being of such individuals living in instiutionalized settings or the facilitative effects of adding music to such a program have not received much attention. In an effort to demonstrate the effect of exercise on the institutionalized elderly, a regular exercise program was instituted with residents in a medium care nursing home. An assessment of effects of the project was based on residents' responses to two life satisfaction questionnaires and nurse's ratings of residents' status before and after a four week exercise program. One of the exercise groups additionally had music accompanying all exercise sessions. It was hypothesized that the exercise program would improve the elderly residents' psychological well being and that the use of music would have an additional facilitative effect. 138 Transactions, Missouri Academy of Science Vol. 16

Method

Setting The Rolla Mannor Care Center where the study took place is a medium care, one story nursing home in Rolla, Missouri, that maintains 60 beds for geriatric residents and 40 for mentally retarded individuals. The head administrator was an RN licensed adminis­ trator. Twenty-four hour supervision was provided by an RN, LPND, or trained nurse aide. Residents were seen by private physicians or a house physician.

Subjects Because most of the residents were physically or psychologically handicapped, 30 that were the most mobile and coherent were selected for participation. These subjects were randomly assigned to either one of two exercise groups, one with and one without accompanying music, or to a no treatment control group. A substantial number of subjects in all three groups either failed to show up to a majority of their scheduled exercise sessions or dropped out of the program before it was completed. Subjects in the exercise group that did not attend% of their exercise sessions were not included in the data analyses. A total of five subjects were thus dropped from each group and data analyses were conducted with five subjects in each group. Average ages in each group were: Control group average age = 76 years (SO = 8. 7), range = 60-86 years; Exercise With Music group average age = 79 (SD = 8.5), range = 76-89; and Exercise Without Music group average age = 72 (SO = 9.98), range = 58-88. The first two groups included two males and three females and the latter group had three females and two males.

Assessment Subjects in all groups were assessed one week prior to, and immediately after subjects in the exercise groups underwent the exercise program. Forms A and B of Neugarten, Havighurst & Tobin's (1961) Life Satisfaction Index were given to all subjects on an individual basis. Additionally, a member of the nursing staff, unaware of the nature of study, rated subjects on a 1-9 point scale for health status and Zest for Life. On the health status form, a 1 = perfect health and 9 = extremely ill, while on the Zest for Life scale, 1 = extremely enthusiastic about life and 9 = completely hopeless about life.

Treatments Subjects in the two exercise groups were given 15 minute exercise sessions three times per week for a total of four weeks. Exercises were done while subjects were seated and consisted mainly of rotations, extensions and stretching of arms, legs and fingers. Subjects in the music group did the same exercises but had marching music accompanying their sessions to encourage more enthusiastic participation. All exercises were led by the first author who also made sure that all participants completed all the exercises.

Results Means and standard deviations for all subjects in the 3 groups on all measures taken in this study are shown in Table 1. A series of 3 x 2 analyses of variance with repeated measures (group x pre- to post-exercise program change) were performed on these measures. Significant interactions of the group by pre- to post-exercise program change 1982 Masters & Haemmerlie, Social Sciences 139

Table 1. Means and Standard Deviations for Subjects in Experimental and Control Groups at Pre- and Posttreatment Assessment.

Control Group Music - Exercise No Music - Exercise

Pretest Pastiest Pretest Posttest Pretest Posttest

M SD M SD Change M SD M SD Change M SD M SD Change

Life Satisfaction - Form A" 13.4 2.3 10.4 5.1 -3.0 9.4 2.2 10.6 2.1 + 1.2 9.2 2.2 11.8 1. 7 + 2.6

Life Satisfaction - Form B" 24.4 4.2 23.8 6.1 - .6 31.6 2.9 32.8 3.1 + 1.2 28.0 5.8 28.0 5.1 0 Health Rating by Staff" 3.0 .7 3.2 .9 + .2 5.6 2.1 4.2 1.1 -1.4 6.0 . 7 4.2 1.1 -1.8 Zest for Life Rating by Staffb 4.8 1.1 5.0 1.0 + .2 5.0 1.8 3.0 1.1 -2.0 4.8 1.9 3.6 2.0 -1.2

"Higher score means feeling better. bLower score means feeling better.

factors occurred on two of the four measures: Life Satisfaction-Form A, F(2, 11) = 5. 08, p < 0.05, and theZestfor Life rating, F(2,11) = 13.13, p < 0.001. The interaction factor on the other two dependent measures did not show any significant differences present in the amount of improvement for subjects in the three groups: Life Satisfaction-Form B, F = 1.39; Physical Health Rating, F = 0.151. Subsequent simple effects analyses carried out on the Life Satisfaction-Form A and the Zest for Life data revealed the same pattern of results: the exercise groups differed significantly from the control group but the two exercise groups did not differ significantly from one another. On the first measure, the control group differed from the music group at thep < 0.05level of significance, F(1,11) = 5.28, and from the no music group atthep < 0. 025level, F(1, 11) = 9. 38. The Zestfor Life rating showed the control group differing from the music group atthep < .001 significance level, F (1,11) = 23.4, and from the no music group at the p < 0.01level, F(1,11) = 9.69. The F values for the subsequent statistical comparisons of the two experimental groups was 2.01 for the Life Satisfaction-Form A questionnaire and 2.33 for the Zest for Life measure. An examination of Figure 1 shows the nature of the significant interaction for the Life Satisfaction-Form A measure. Although the two exercise groups showed slight improve­ ment on this measure, the control group deteriorated and this suggests that effects of the exercise program were to prevent deterioration as well as to foster improvement in residents' degree of life satisfaction. An examination of the nature of significant interaction on the Zest for Life scale shows a similar pattern of pre- to post-exercise program change for elderly residents (Figure 2). 140 Transactions, Missouri Academy of Science Vol. 16

15

10

e = Control Group • = Music Group

.& = No Music Group 5

PRETEST MEAN POSTTEST MEAN

aHigher score means feeling better

Figure 1. Pretest to Posttest Change x Group Interaction on Life Satisfaction

Fonn A Measure8 •

Discussion The results of this study indicated that the exercise program had a positive effect on the psychological well being of subjects who participated in the program as measured by the Life Satisfaction Index-Form A and by Zest for Life ratings. The program itself involved relatively simple exercises that most any staff member in a residential setting could administer, and it took little time. Thus, the exercise program employed may represent a practical as well as an effective method for maintaining or increasing patient satisfaction in nursing homes for the elderly. A rather substantial drop out rate was experienced in the present study. Although this degree of lack of participation is not unusual for elderly subjects in an institutional setting (Konarski, Johnson & Whitman, 1980), it does lead us to consider what factors might promote greater utilization of such programs by larger numbers of clients. In this regard, if subjects had volunteered on their own initiative for participation in the program, instead of being selected by the nursing staff, perhaps the drop out rate would not have been as high. Also, participation might have been increased if a more individualized exercise program had been used, although such a program would have required more staff time and effort. 1982 Masters & Haemmerlie, Social Sciences 141

8

7

6

5

4 3 ~• • = Control Group

2 • = Music Group

• = No Music Group 1

PRETEST MEAN POSTTEST MEAN

aLower score means feeling better

Figure 2. Pretest to Posttest Change x Group Interaction on Zest for Life Rating Measure a.

Given that activity or interaction control groups were not used, design of the present research effort does not allow one to ascertain exactly which aspects of the program caused positive results. It may be the case that any activity, whether it involves exercise or not, could provide the stimulation necessary that in turns leads to improved (or less deteriorated) well-being. Additionally, social interaction with the exercise leader or with other exercisers may have been factors contributing to positive results obtained with the exercise program. To examine the possible role of these factors in an exercise program further research is needed. The present results do, however, indicate that such a program can lead to improved well-being for the institutionalized elderly.

References Bonner, C. D. Rehabilitation instead of bed rest? Geriatrics, 1969, 24, 109-118. Comstock, R. L., Mayers, R. L., & Folsom, J. C. Simple physical activities for the elderly. Hosp. Commun. Psychiat., 1969, 20, 377-380. deVries, H. A Psychological effects of an exercise training regimen upon men aged 52-88. J. Gerontology, 1970, 25, 325-336. Konarski, E. A, Jr., Johnson, M. R., & Whitman, T. L. A systematic investigation of resident participation in a nursing home activities program. Behaviour Therapy and Experimental Psychiatry, 1980, 11, 249-257. Neugarten, B. L., Havighurst, R. J., & Tobin, S. S. The measurement of life satisfaction. J. of Gerontology, 1961, 16, 134-143. Powell, R. P. Psychological effects of exercise therapy upon institutionalized geriatric mental patients. J. Gerontology, 1974, 29, 157-161.

1982 Irvin, Isbell & Typpo, Social Sciences 143

Caregivers~ Attitudes Toward the Aged and Levels of Authoritarianism

Carol B. Irvin, M.S. Instructor Child and Family Development University of Missouri 35 Stanley Hall Columbia, Missouri 65211 Lou Isbell, M.S. Assistant Professor State Child and Family Development Specialist University of Missouri Extension 14 Gwynn Hall Columbia, Missouri 65211 Marion H. Typpo, Ph.D. Assistant Professor Child and Family Development University of Missouri 14 Gwynn Hall Columbia, Missouri 65211

Abstract: Questionnaires were completed by 119 white, female administrators and caregivers, from nursing homes and hospitals in southeast Missouri, about their attitudes toward the aged and authoritarianism. Positive attitudes toward the aged were related to caregivers' age, educational level, type of facility where worked, and size of the community. A significant relationship between higher levels of education and less emphasis on the importance of authority was also found. Administrators who had been employed on their present job for at least five years held more negative attitudes toward the aged. All occupational groups reported aging to be more difficult for men than for women. Attitudes and beliefs of caregivers might affect not only their performance with aged patients, but may also affect their attitudes toward their own aging. Key Words: Aging, authoritarianism, nursing home, nursing staff attitudes.

Social context of the elderly in nursing homes is dependent in part upon the relationship established with the caregiver. As Rosow (1975) noted, it is the younger age groups which determine status and position of the older person in the social order. For the aged in the nursing home, social context is their relationship with caregivers and other patients within the home. How caregivers view the elderly in general and how they view elderly health care needs are of particular importance, since individuals with more unfavorable attitudes may be rather impervious to real needs of their patients (Linn & Linn, 1980). Nursing homes and hospitals are the primary institutional settings in which nurses and nurses aides interact with the aged. These institutions often lack geriatric expertise or sensitivity to realities of aging (Almquist et al., 1981). Behaviors of caregivers and aged patients are influenced by the attitudes and behaviors of one another. Caregivers' behaviors reflect their attitudes, knowledge, and the influence of previous experiences with 144 Transactions, Missouri Academy of Science Vol. 16 the elderly (White, 1977). Attitudes generally reflect one's values, and values determine one's perceptions of and subsequent reaction to social interactions (Cyrus-Lutz & Gaitz, 1972). It is generally thought that attitudes are triggers for behavior. An individual's feelings and knowledge in response to a particular stimulus or situation will cause the individual to behave in a particular way. In situations where negative attitudes are held, individuals may behave in one of two ways: exhibit behaviors which explicitly reflect negative attitudes or exhibit behaviors which appear appropriate but covertly convey negative or conflicting messages (White, 1977). Larson eta!. (1969) found that nursing home patients preferred nurses who, when tested for attitudes about older people and patients in nursing homes, held more favorable attitudes. Researchers have repeatedly found negative attitudes and stereotypic thinking among personnel in nursing homes and other institutional settings (Kosberg & Gorman, 1975; Gillis, 1973; Kahana & Coe, 1969; Coe, 1967). Keith (1971) noted that professionals' attitudes not only affect social policy but also quality of care that clients in geriatric settings receive. Futrell and Jones (1977) found that nurses with longer work experiences with the elderly portrayed more favorable attitudes. Devine (1980) studied the degree of old age stereotyping of nursing staff in a geriatric hospital and a private nursing home. Results indicated no statistically significant differences in staff attitudes toward the elderly. For caregivers in this study type of facility where worked was not associated with differences in staffs' perceptions. Hatton (1977) conducted the only available study ofthe relationship between hospital nurses' (N = 7) expressed aging related attitudes and their observed behaviors with patients. Nurses with more favorable attitudes were involved in a higher percentage of positive interactions with patients. A problem exists in comparing studies by different authors on different sample groups. Because of variations in objectives, differences in methodology, and use of scales which lack reliability and validity data, valid comparisons between them cannot be made (Green, 1981; Thorson et al., 1974; McTavish, 1971). The elderly in institutional settings have been described as the most powerless, voiceless, invisible, and uncared for group in this country (Kosberg, 1973). Life in institutional settings often represents a loss of control or power to effect even minor changes in patients' lives. The range of choices regarding even the most basic daily activities are restricted. This diminishing of personal control and choice can result in helplessness, hopelesness, and rapid deterioration in health and behavior (Gamer & Mercer, 1980; Mercer & Kane, 1979). Long-term care is often seen as nonproductive, dull, and tedious. As a result, the elderly patient can become lost in depersonalized treatment by staff (Almquist et al., 1981; Goffman, 1961 ). The good patient may exhibit learned helplessness (Seligman, 1975) while the bad patient may show psychological reactance or resistance as a reaction to loss of control (Brehm, 1966). Authoritarianism, or belief in the importance and legitimacy of power, would seem to be an important issue for two reasons. One, the aging process itself can be viewed as a gradual but inexorable ebbing of power with the aged, therefore, being seen as a particularly powerless group. Secondly, loss of personal control in institutional settings would seem to increase this feeling of helplessness and loss of power. Administrators and caregivers possess authority and power which can be used legitimately or illegitimately. If caregivers act out of positive attitudes toward the aged in general, authority and power may be used quite differently than in situations where caregivers see, as their right, power to make the elderly conform. 1982 Irvin, Isbell & Typpo, Social Sciences 145

Are attitudes toward old people related to authoritarian tendencies? Kogan (1961) gave undergraduates an authoritarian scale (Adorno et al., 1950) and a Likert scale he had developed to study attitudes toward old people. When the data were analyzed Kogan found no consistent relationship between scores on his "Old People" scale and authoritarian tendencies. However, the question as to whether or not caregivers' attitudes toward the aged would be related to authoritarianism has yet to be answered. The present study examined the effect of background characteristics and the level of authoritarianism on attitudes toward the aged of caregivers in nursing homes and hospitals in southeast Missouri.

Hypotheses 1. It was expected that negative attitudes toward the aged would be related to selected demographic variables: age, education, length of time on present job, and size of the community. 2. It was expected that marital status and type facility where worked would not be related to attitudes of caregivers. 3. It was also expected that nurses would be more positive in their attitudes toward the elderly than would be either the nurses aides or administrators. 4. It was expected that lower levels of authoritarianism would be related to higher levels of education and more positive attitudes toward the aged. 5. Finally, attitudes were expected not to be related to whether aging was perceived as being more difficult for men than for women.

Methods and Materials ' Sample. Individuals attending a University of Missouri Extension Validation Therapy training conference in southeast Missouri in March 1980, served as subjects. The training session was for the purpose of learning new ways to better relate to disoriented elderly persons; however, this study was not related to the training topic. A questionnaire was administered to subjects in a group setting at the beginning of the training session. Those individuals who did not wish to respond were asked to turn the questionnaire face down until the rest of the group was finished; however, all individuals participated in the study. Subjects consisted of administrators, ombudsmen, nurses, nurses aides, and students. To include only full-time caregivers in nursing homes and hospitals, students, ombudsmen, and caregivers working in facilities other than nursing homes and hospitals were dropped. Since 91% of the subjects were female and 97% were white, it was decided to consider only these as the sample. The final sample, therefore, consisted of 119 white, females: 32 administrators, 36 nurses, and 51 nurses aides. Table 1 presents the personal, job, and community characteristics.

Instruments. The first section of the questionnaire included personal and family information, as well as questions about job and community. The second section of the questionnaire was an adaptation ofTuckman and Lorge's "Attitudes Toward Old People" scale (1952). This was renamed "Attitudes Toward Aged" (Table 2). Subjects' responses were scored on a scale from strongly agree (1) to strongly disagree (7). The subjects were also asked to respond to an additional item: Aging is more difficult for (check) __ men __ women. The third section of the questionnaire contained an authoritari­ anism scale (Adorno eta!., 1950) which was renamed "Attitudes About People". This section attempted to measure the subjects' attitudes about power and its use (authoritarian­ ism). The average time required to complete the questionnaire was 30 minutes. 146 Transactions, Missouri Academy of Science Vol. 16

Table 1. Personal, Job and Community Chanu:teristicsa.

Administrators Nurses Nurses Aides Variables N % N % N %

Age 18-25 6 19 6 17 6 12 26-35 7 22 12 33 14 27 36-45 12 37 8 22 18 35 46-55 5 16 6 17 6 12 56-65 2 6 4 11 7 14 Marital Status Single 4 13 2 6 6 12 Married 25 78 26 72 34 67 Divorced 3 9 3 8 7 14 Widowed 5 14 4 7 Educational Level Grade School 6 12 Attended High School 1 3 14 27 High School Graduate 15 47 1 3 17 33 Nurses Training 5 16 19 53 3 6 Attended College 7 22 5 14 8 16 Community College Graduate 6 16 1 2 College or University Graduate 4 12 5 14 2 4 Facility Where Worked Nursing Home 28 88 17 47 29 57 Hospital 4 12 19 53 22 43 Length of Job Less than 1 year 7 22 9 25 8 15 1-5 years 17 53 16 44 26 51 6-10 years 5 16 9 25 11 22 11-15 years 3 9 2 6 5 10 21-25 yearsb 1 2 Size of Community Less than 1, 000 3 9 1 2 1,001-2,500 5 16 5 14 8 16 2,501-5,000 5 16 6 17 7 14 5,001-10,000 6 18 13 36 17 33 10,001-25,000 7 22 5 14 5 10 25,001-50,000 6 19 7 19 13 25

•n = 119 bno cases 16-20 years 1982 Irvin, Isbell & Typpo, Social Sciences 147

Table 2. Attitudes Toward Aged Questionnaire Statements Classified Into 13 Categories.

Category Statement

Physical Need glasses to read. Are hard of hearing. Spend much time in bed because of illness. Have poor coordination. Feel tired most of the time. Mental Deterioration Repeat themselves in conversation. Cannot learn new things. Are better off in old age homes. Cannot manage their own affairs. Cannot concentrate, even on simple tasks. Insecurity Prefer to be alone. Are easily moved to tears. Feel sorry for themselves. Are afraid of dark. Like to be waited on. Are bad patients when ill. Personality Trails Are grouchy. Are hard to get along with. Are touchy. Are selfish. Like to gossip. Family Expect their children to support them. Are a burden to their children. Conservatism Are old-fashioned Prefer old friends rather than make new ones. Dislike any changes or interference with established ways of doing things. Financial Are unproductive. Hide their money. Attitude Toward Future Believe in life after death. Are afraid of death. Best Time of Life Would like to be young again. Never had it better. Love life. Activities and Interests Likes religious programs on the radio or TV. Spend most of their time reading or listening to the radio or TV. Sex Should not marry. Have no interest in the opposite sex. Interference Meddle in other people's affairs. Are a nuisance to others. Cleanliness Are careless about their table manners. 148 Transactions, Missouri Academy of Science Vol. 16

Data Analysis. Mean scores on the Attitude scale and the Authoritarianism scale were calculated, with higher scores indicating more negative attitudes toward the aged and higher degrees of authoritarianism. Raw scores for each of the 13 categories on the Attitude scale were also calculated. Correlations between the authoritarian total score, the attitudes total score, and scores on the 13 categories of the Attitude scale are shown in Tables 3 and 4. One-way analyses of variances, chi-squares, and Fisher's exact tests (two-tailed) were the primary analytic strategies used. Duncan's multiple range test was used to examine differences between means:

Results Caregivers' Attitudes Toward the Aged. Responses of nurses and nurses aides were combined to give a total mean score for primary caregivers. Data analysis indicated that educational level, subjects' age, type of facility where worked, and community size were important factors in terms of caregivers' attitudes toward the elderly. 1. Level of education. Significant differences in attitudes toward the aged, based on the caregivers' level of education were found. Those caregivers whose educational level was beyond high school were more positive in their attitudes toward the aged, F (1, 83) =

Table 3. Conelations Between Scores on Categories of Attitudes Toward the Aged Scale and Authoritarianism.

Authoritarian Category Total Score

Physical 0.50**** Mental Deterioration 0.59**** Insecurity 0.62**** Personality Traits 0.49**** Family 0.35** Conservatism 0.28* Financial 0.43**** Attitudes Toward Future 0.16 Best Time of Life 0.05 Activities and Interests 0.20 Sex 0.29* Interference 0.37*** Cleanliness 0.40*** Attitudes Total Score 0.59****

*p<0.05. **p

Table 4. Conelations Between Scores on Categories of Attitudes Toward the Aged Scale and Attitudes Total Score.

Attitudes Toward Aged Category Total Score

Physical 0. 78**** Mental Deterioration 0.80**** Insecurity 0.87**** Personality Traits 0.84**** Family 0.65**** Conservatism 0.63**** Financial 0. 73**** Attitudes Toward Future 0.45**** Best Time of Life 0.40*** Activities and Interests 0.54**** Sex 0.57**** Interference 0.66**** Cleanliness 0.65****

*p<0.05. **p

4.81, p < 0.05. Caregivers with lower educational levels accepted to a greater degree negative stereotypes which emphasized the physical, mental, and financial dependence of the aged (Table 5).

Table 5. Caregivers' Mean Scores on Physical, Mental Deterioration, Insecurity, financial, and Sex Categories By Levels of Education.

High School Beyond Category or Less High School F (N = 36) (N = 49)

Physical 5.32 4.70 5.66* Mental Deterioration 3.46 2.97 4.15* Insecurity 3.96 3.33 4.70* Financial 3.97 3.26 5.27* Sex 2.72 2.08 4.91 *

*p<0.05. 150 Transactions, Missouri Academy of Science Vol. 16

2. Age. Caregivers 36-45 had significantly higher scores on the physical, F (3, 81) = 2. 76, p < 0.05; and mental deterioration subscalesF (3, 81) = 3.23, p < 0.05 than did the caregivers 26-35 years of age. Caregivers 26-45 scored significantly higher on the conservatism subscale than did the 18-25 year olds, F (3, 81) = 4.23, p < 0.01. Older caregivers tended to have more negative attitudes toward the aged than younger caregivers. Older caregivers are likely to have had more work experience with the elderly and may be more familiar with some of the negative aspects of aging. In addition, older caregivers may also be having to deal with their own aging. What accounts for the less negative attitudes of younger caregivers? It is possible that these subjects may have a different image of old age based on limited work experience with the elderly, as well as positive relationships with their own active parents. Younger caregivers were also more likely to have higher levels of education. 3. Type of Facility. Whether caregivers worked in nursing homes or in hospitals also influenced caregivers' attitudes toward the aged. Caregivers who worked in nursing homes perceived the aged more negatively on mental deterioration, personality traits, and sex subscales than did caregivers working in hospitals (Table 6).

Table 6. Caregivers' Mean Scores on Mental Deterioration, Personality Traits, and Sex Categories By Type of facility Where Worked.

Nursing Home Hospital Category Mean Mean F (N = 45) (N = 40)

Mental Deterioration 3.46 2.86 6.37* Personality Traits 4.13 3.39 6.27* Sex 2.64 2.02 4.67* *p

The length of stay, medical prognosis, and kind of care required would differ from hospitals to nursing homes and might have an impact on caregivers' attitudes. Additional research is needed to explore these interrelationships. 4. Community Size. Size of community was found to be significant only for the subscale best time of life. Caregivers from communities of less than 2,500 had mar~ negative views of the aged than did those caregivers from communities with populations of 25,000-50,000, F (2, 82) = 7.80 p < 0.01. Many small communities do not have adequate resources for the elderly. Nursing homes in small communities may not only care for the elderly with physical problems but may also have to provide care for senile elderly as well as those who are mentally ill. Inadequate resources might have an impact on caregivers' ability to care for the elderly as well as influence caregivers' attitudes. 5. Marital Status and Time on Job. Marital status and length of time on the present job were not significantly related to the endorsement of stereotypic statements. There was some indication that widows seemed to have the least positive attitudes toward the aged, but the difference was not significant Caregivers who had worked less than a year tended to have more positive attitudes toward the aged than did caregivers with more experience; however, this difference was not significant 1982 Irvin, Isbell & Typpo, Social Sciences 151

Nurses and Nurses Aides View of Aged Compared. A two-way analysis of variance was done to examine differences between nurses and nurses aides. Nurses who worked in nursing homes were found to have more negative attitudes toward the aged than did the nurses aides, F (3, 81) = 7.14, p <0.001. Nursing home nurses also scored significantly higher than did nurses aides on the physical, mental deterioration, and insecurity subscales (Table 7). However, the reverse was found for caregivers in hospitals. It was the hospital nurses aides who scored significantly higher on the physical, mental deterioration, and insecurity subscales. Hospital nurses aides also had more negative attitudes toward the aged than did the hospital nurses.

Table 1. Nursing Home Staff's Mean Scores on Physical, Mental Deterioration, and Insecurity Categories of Attitudes Toward the Aged Scale.

Nurses' Nurses Aides' Category Mean Mean F (N = 36) (N =51)

Physical 5.17 4.72 19.51 **** Mental Deterioration 3.46 2.86 10.63** Insecurity 3.64 3.55 5.27* *p

The level of education was significantly different between nurses and nurses aides, l (1) = 41.76, p < 0.0001. Ninety-seven percent of the nurses had training beyond high school as compared to 28% of the nurses aides. Forty-four percent of caregivers had educational levels of high school or Jess. Nurses aides comprised 43% (N = 37) of this total. While 56% of the caregivers had educational training beyond high school, nurses aides comprised only 16% (N = 14) of this group. Overrepresentation of nurses aides in high school or less group may have accounted for this difference, or the difference may be reflection of attitudinal change as a result of increased experience in educational training. Yet to be explained is the influence of type of facility on acceptance of negative statements about the physical, mental deterioration, and feelings of security. Why did nurses in nursing homes and nurses aides in hospitals hold more negative views? Perhaps this is a function of the type of experiences these caregivers have. Do the requirements of their jobs determine the type of experience, or are differences a reflection of type of training? Do patient characteristics influence the type and quality of interactions for these caregivers? Caregivers' Level of Authoritarianism. Caregivers' authoritarianism scores were related to their total attitude scores as well as to scores on the 13 subscales (Table 3). Coefficients ranged from 0.05 to 0.62 with 11 ofthe 14 correlations reaching significance at or beyond the 0.05 level. 152 Transactions, Missouri Academy of Science Vol. 16

The level of education was found to be an important influence on the level of authoritarianism. As educational level increased, the authoritarian score decreased. Caregivers who had higher levels of education (beyond high school) endorsed significantly fewer authoritarian statements than those caregivers with lower educational levels, F (1, 72) = 16.23, p < 0.001. Higher levels of education were associated with more positive attitudes toward the aged and less emphasis on the importance of authority. Only through understanding the nature of caregivers' negative attitudes and beliefs about the importance of authority can attempts be made to change these attitudes. It would seem that training and educational programs which focus on caregivers' attitudes toward the aged, as well as their sensitivity to the importance of authority in patient-caregiver relationships would have the potential to create a supportive climate for optimum patient care. Administrators' Attitudes. In general, administrators responded in much the same way as did the nurses and nurses aides. The higher the level of education the more positive attitudes toward the aged, F (1, 30) = 6.90, p < 0.05. Length of time on present job was also an important factor. Administrators who had worked five years or more had significantly higher scores on the mental deterioration, attitudes toward future, and cleanliness subscales (Table 8).

Table 8. Administrators' Mean Scores on Mental Deterioration, Attitudes Toward the Future, and Cleanliness Categories By length of Time on Job. -

Less Than One To More Than Category One Year Five Years Five Years F (N = 7) (N = 17) (N = 8)

Mental Deterioration 3.06 3.14 4.10 3.51* Attitudes Toward Future 3.41 4.29 4.94 3.38* Cleanliness 2.57 3.41 5.00 4.00*

*p<0.05.

Analyses of mean scores indicated that those administrators with higher levels of educa­ tion were less authoritarian. Administrators' authoritarian scores were related to scores on the financial, r = .39, p < 0.05; interference, r = .51, p < 0.01; and cleanliness subscales, r = .52, p < 0.01. Differences Between Administrators and Caregivers. Administrators and caregivers did differ significantly based on type of facility, where they worked, length of time on present job, and size ofthe community. Compared to caregivers (52%), more administrators (88%) worked in nursing homes than in hospitals, x2 (1) = 11.93, p < 0.001. Administrators who worked in nursing homes scored higher on the activities and interest subscale, F (5, 111) = 6.30, p < 0.01, than did caregivers working in either hospitals or nursing homes. Administrators who had been in their present job for over five years had a significantly higher score on mental deterioration, F (5,111) = 3.56, p < 0.05; attitudes toward future, F(5,111) = 3.25, p < 0.05; and cleanliness subscales, F (5,111) = 3.20, p < 0.05, than did caregivers. 1982 Irvin, Isbell & Typpo, Social Sciences 153

Length of time on the job was an important factor for administrators only. Why did these administrators after a five year period hold more negative views? Are administrators forced to give more attention to regulations and requirements which deal with maintaining a facility and thus lessen their interactions with caregivers and patients? Do administrators see issues of cleanliness, mental deterioration, and fear of death as potentials for management problems? For Whom Is Aging More Difficult? Subjects generally reported aging to be more difficult for men than for women. However, administrators were more likely to say aging is more difficult for men than were the caregivers, i (1) = 4.13, p < 0.05. The type of facility where they worked was an important influence on nurses' responses. Those nurses who said aging is more difficult for men were more likely to work in nursing homes, while nurses working in hospitals were more likely to say aging is more difficult for women, Fisher's Exact test, (2), p < 0.01. Caregivers who said aging is more difficult for women tended to be less authoritarian, but the difference was not significant, F (1, 72) = 3.79, p < 0.06. Caregivers in this study generally saw aging as being more difficult for men than for women. Again, the type of facility was associated with differences in nurses' perceptions. Is there a difference between male and female patient behaviors in hospitals and nursing homes? Or, is this finding a reflection of the fact that the majority of patients in nursing homes are female? Does the fact that all caregivers in this study were female also make a difference?

Conclusion The findings of this study must be considered on the basis that the study was limited to a seven county area, so findings cannot be assumed to apply to the general population. Although the majority of subjects came from southeast Missouri, some came from St. Louis and Illinois. The sample was not a randomly selected group, and some bias may exist in that subjects were those caregivers who took time off from work, made time after work, or were given release time to attend the training session. If attendance does reflect the supervisors' judgment for potential growth, or the caregivers' expressed interest in improvement, this group may not be typical of the general population of caregivers. The nation as a whole is aging. Missouri continues to hold fifth place in the nation in terms of the percentage of its population being over 65. According to the 1980 Missouri census data on aging, Missouri counties range from 10 to 31% of their population over 60 years of age (Missouri Gerontology Institute, 1980). As more and more people live longer, the number of elderly and frail elderly will continue to increase. Caregivers themselves will be older, with perhaps greater reliance for care for the frail elderly placed on nursing homes and other institutional settings. Several interesting observations may be made from the data analyzed. That all categories of caregivers included in this study demonstrated stereotypic attitudes toward aging is consistent with previous research (Brown, 195 7; Gillis, 1973 ). This study suggests that level of education is an important influence on caregivers' attitudes toward the aged and their beliefs about the legitimacy of authority. The importance of stereotypes of the aged lie in the fact that caregivers could be expected to reflect their beliefs when they interact with the aged. Often the aged respond to these stereotypes by reacting in the behavior patterns expected of them. Not only might the attitudes of caregivers affect their performance with aged patients, but they may also affect their attitudes toward their own aging (Wolk & Wolk, 1971). Additional research is needed for further insight into the effect 154 Transactions, Missouri Academy of Science Vol. 16

of caregivers' attitudes and belief systems on caregiver behavior. What effect do patient characteristics have upon caregivers' interactions? Do patient characteristics differ in nursing homes and hospitals? Or, is it the characteristics of the institution, length of patient stay, and/or the requirements of the job which exert more influence upon caregivers' attitudes and behaviors?

Acknowledgements Special thanks are extended to Edith Lusk, Southeast Extension Area Child and Family Development Specialist and the Southeast Area Validation Therapy committee. Thanks are also extended to Dr. Jean Sebaugh for statistical consultation and to Gwen Hendrix for help in data preparation. Data collection was supported by a grant from the University of Missouri Extension Aging Back-Up Funds, Spring, 1980.

References Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J. &Sanford, R.N. The Authoritarian Personality. New York: Harper, 1950. Almquist, E., Stein, S., Weiner, A, & Linn, M. W. Evaluation of continuing education for long-term care personnel: Impact upon attitudes and knowledge. Journal of the American Geriatrics Society, 1981, 29, (3), 117-122. Brehm, J. W. A Theory of Psychological Reactance. New York: Academic Press, 1966. Brown, M. I. Nurses' attitudes toward the aged and their care. Annual Report to Gerontology Branch USPHS, June 30, 1966 to July 1, 1967. Coe, R. M. Professional perspectives on the aged. The Gerontologist, 1967, 7, 114-120. Cyrus-Lutz, C. & Gaitz, C. M. Psychiatrists' attitudes toward the aged and aging. The Gerontologist, 1972, (2), 163-167. Devine, B. A Old age stereotyping: A comparison of nursing staff attitudes toward the elderly. Journal of Gerontological Nursing, 1980, 6, (1), 25-32. Futrell, M. & Jones, W. Attitudes of physicians, nurses, and social workers toward the elderly and health maintenance services for the aged: Implications for health manpower policy. Journal of Gerontological Nursing, 1977, 3 (3), 42-46. Garner, J. D. & Mercer, S. 0. Social work practice in long-term care facilities: Implications of the current model. Journal of Gerontological Social Work, 1980, 3 (7), 71-77. Gillis, Sr. M. Attitudes of nursing personnel toward the aged. Nursing Research, 1973, 22, 517-520. Goffman, E. Asylums. New York: Anchor, 1961. Green, S. K. Attitudes and perceptions about the elderly: Current and future perspectives. International Journal of Aging and Human Development, 1981, 13 (2), 99-119. Hatton, J. Nurse's attitude toward the aged: Relationship to nursing care. Journal of Gerontological Nursing, 1977, 3 (3), 21-26. Kahana, E., & Coe, R. M. Self and staff conceptions of institutionalized aged. The Gerontologist, 1969, 9, 264-267. Keith, P. An exploratory study of sources of stereotypes of old age among administrators. Journal of Gerontology, 1971, 32 (1), 163-169. Kogan, N. Attitudes toward old people: The development of a scale and an examination of correlates. Journal of Abnormal and Social Psychology, 1961, 62, 44-54. Kosberg, J. Differences in proprietary institutions caring for affluent and non-affluent elderly. The Gerontologist, 1973, 13, 229-304. Kosberg, J. & Gorman, J. F. Perceptions toward the rehabilitation potential of institutionalized aged. The Gerontologist, 1975, 15, 398-402. Larson, C. E., Knapp, M. L., Zuckerman, I. Staff-resident communication in nursing homes: A factor analysis of staff attitudes and resident evaluations of staff. Journal of Communication, 1969, 19, 308-316. 1982 Irvin, Isbell & Typpo, Social Sciences 155

Linn, M. W. & Linn, B. S. Realities of institutional care that affect outcome. Aged Care and Services Review, 1980, 2 (3), 1, 3-13. McTavish, D. G. Perceptions of old people: A review of research methodologies and findings. The Gerontologist, 1971, 11, 90-101. Mercer, S. 0. & Kane, R. A Helplessness and hopelessness in the institutionalized aged: An experiment. Health and Social Work, 1979, 4 (1), 71-77. Missouri Gerontology Institute, 1980 Missouri census data on aging. Census of population and housing, 1980: Summary Tape File 1 A (Missouri). Bureau of the Census, Washington, D. C., 1981, 3(1), March, 1982. Rosow, I. The aged in post-affluent society. The Gerontologist, 1975, 1 (4), 3-16. Seligman, M. E. P. Helplessness, San Francisco: Freeman, 1975. Thorson, J., Hancock, K. & Whatley, L. Attitudes toward the aged as a function of age and education. The Gerontologist, 1974, 14, 316-318. Tuckman, J. & Lorge, I. Attitudes toward the older worker. Journal of Applied Psychology, 1952, 36, 149-153. White, C. M. The nurse-patient encounter: Attitudes and behaviors in action. Journal of Gerontologi­ cal Nursing, 1977, 3 (3), 16-20. Walk, R. L. & Walk, R. B. Professional workers attitudes toward the aged. Journal of the American Geriatrics Society, 1971, 19, 624-639.

1982 Annual Meeting 157

The 1982 Annual Meeting of the Missouri Academy of Science April 23-24, 1982 School of the Ozarks Point Lookout, Missouri

The meetings of the 1982 sessions of the Missouri Academy of Science are listed as the meetings of the Senior Division of the Missouri Academy of Science and the Collegiate Division of the Missouri Academy of Science.

SENIOR DIVISION Biology

Kruse, K. J., and M. N. Rogers, 1-40 Agriculture, Horticulture Departmen~ University of Missouri, Columbia, MO 65211. FACTORS AFFECTING REGENERATION OF BEGONIA X H/EMAUS LEAF CUTTINGS. Regeneration of plantlets from Begonia x hiemalis leaf cuttings varied depending upon the season. Summer is the most difficult time of propagation due to the high temperatures and light intensities. 'Renaissance' stock plants were grown at 21 oc and 31 oc in growth chambers under 5400, 10800, and 21600 lux. Leaf cuttings taken from stock plants grown at 21 oc produced more shoots. The leaf size ofthe cuttings became a critical factor at 31 °C, with the larger leaf cuttings producing more shoots. Leaf cuttings of 'Renaissance', 'Schwabenland Red', and 'Balaleika' were sprayed or dipped in solutions of cytokinins (kinetin or benzyladenine) dissolved in ethanol to stimulate bud development. In all cases the ethanol controls produced the highest fresh weight of shoots. Varietal differences also affected the regeneration ability. Janzow, F., Southeast Missouri State University, Cape Girardeau, MO 63701. SOCIAL BEHAVIOR IN THE LARGEMOUTH BASS (MICROPTERUS SALMOIDES). Social behavior in the largemouth bass consists of several social stages. Qualitative and quantitative analyses of juvenile groups indicate that group organization is complex and dynamic. Observations revealed that group dissolution and appearance of adult-like behavior results from increasing agonistic behavior.

Wallin, J. R., and D. V. Loonan, USDA-SEA-ARS and Plant Pathology Department, University of Missouri-Columbia, Columbia, MO 65201. AIR SAMPLING FOR ASPERGILLUS FLAVUS SPORES AT FIVE LOCATIONS IN 1981. Sampling for air borne spores of A flauus was initiated In April at two locations and May and June at three locations and continued through August at one location and through October at four. Using the Rotorod sampler, sampling 5 mins twice a week, spores were trapped 5 of 7 months at Bradford and 7 of 7 months at Rollins plots in Columbia. At Novelty spores were trapped 4 of 5 months; at Portageville, 3 of 4 months; and at Spickard, 4 of 5 months. The greatest number of spores were aloft at Bradford May and October and at Rollins May and August Three samplers were used at Novelty i.e. the Rotorod, Burkard and the Andersen (quantitative). Spores were trapped weekly throughout the growing season. The greatest number was trapped by the Burkard, weekly sampler, 35 in September. Stefan, S., T. Gowin and D. F. Millikan, Department of Plant Pathology, University of Missouri-Columbia, Columbia, MO 65211. FURTHER INVESTIGATIONS ON BROWNING ACTION IN BLACK WALNUT EXPLANTS. Progress on the shoot tip meristem culture of black walnut, Jug/ans nigra L., has been hampered by the lethal enzymatic browning action following removal of the dome and primordia from the buds. We have introduced some modifications which minimize this problem. Soaking the tips iri 0.01 M DIECA for 4 hours prior to removal is effective in reducing the browning. Reducing the sucrose content of the medium also is effective. Finally altering the Ca.. , ~ + and NOs content of the medium as proposed by McCown and Uoyd (Unpublished data) is beneficial in protecting the explant from the lethal browning action and appears to be superior to the MS formulation for the I stage. Elliott, D., Central Methodist College, Fayette, MO 65248. DlVERSl1Y, DOMINANCE AND AREAL SPECIES RICHNESS OF VASCULAR PLANTS IN OLD FIELDS. Four similar but chronologically different old fields were carefully chosen in Howard County, Mo. One 50m x lOOm plot was selected from

Wallin, J. R., C. A. C. Gardner, and D. V. Loonan, USDA-SEA- ARS and Plant Pathology Department, University of Missouri-Columbia, Columbia, MO 65201. A METHOD FOR TESTING CORN GENOTYPES FOR RESISTANCE TO ASPER­ GILLUS FLAVUS AND LOW AFLATOXIN PRODUCTION. Kernels of exotic corn genotypes, primarily flints, were decapped, placed on moist filter paper in petri dishes, then inoculated with an aqueous suspension of 20,000 spores/cc of A. flavus. The plates were incubated for 28C for 7 days then removed to record the number of infected kernels and "blacklight" (ultra long wave UV light) the kernels for bright green yellow fluorescence (BGYF), the presumptive test for aflatoxin. Those kernel remnants which fluoresced wlil be analyzed for aflatoxin while those kernels which did not and developed seedlings, those seedlings were planted in steamed soil, placed in growth chambers and exposed to nine hr light periods. The survivors will be planted in the field in the spring and the ears of the mature plants will be inoculated.

Key, D. P., Southwest Missouri State University, Springfield, MO 65802. MODERN POLLEN SPECTRA IN POND SEDIMENTS IN SOUTHWEST MISSOURI AND ITS RELATIONSHIP TO THE CONTIGUOUS FOREST. Five pond sediments in southwest Missouri were sampled and analyzed for pollen content. The contiguous forest was sampled by the quarter method and the resulting values were compared on the basis of percent basal area of the tree species in the forest to the percent arboreal pollen recovered in the sediments, the coefficient of community and the percentage similarity. The results ofthis study indicated thatthere is an existing relationship between the pollen spectra in small pond sediments and the woody vegetation which surrounds it.

Jafar, S. S., and G. D. Sells, Northeast Missouri State University, Kirksville, MO 63501. EFFECTS OF WATER STRESS ON PROLINE OXIDATION BY MITOCHONDRIA ISOLATED FROM SOYBEAN HYPOCOTYLS. Rate of oxidation of proline by mitochondria isolated from water -stressed etiolated soybean hypocotyls was determined by use of an oxygen electrode. Range of stress was,between -5 and -16 Bars. Both State 3 and State 4 rates were observed to decrease exponentially with stress from a value of 30 to 6 nmoles of oxygen per mg protein per minute. Respiration rates for normal substrates (succinate, malate + pyruvate and exogenous NADH) were also measured and found to decrease only slightly in this stress range. Results from this study support an hypothesis that a highly specific inhibition to the mitochondria occurs at the onset of stress, and that this inhibition is a major factor in proline accumulation in plants. Mokry, L. E., and R. G. Rhodes, Department of Life Sciences, Southwest Missouri State University, Springfield, MO 65802. PRELIMINARY STUDIES OF FRESHWATER RED ALGAE IN OZARK STREAMS AND RIVERS. Freshwater red algae including Asterocystis sp., Audouinella sp., Lemanea sp., Acrochaetium sp., Batrachospermum moniliforme and B. vagum have been observed in several Ozark streams and rivers. Morphological and life history studies were conducted on Acrochaetium and Batrachospermum spp. Results from studies on B. moniliforme and B. uagum carried out both in situ and in uivo were compared and shown to be similar.

Craig, M. M., Southwest Missouri State University, Springfield, MO 65804. ULTRASTRUCTURE OF CYTOPLASMIC LOCALIZA­ TIONS IN GASTROPOD POLAR LOBES. A brief survey of the occurrence of polar lobes and cytoplasmic localizations possibly residing within them has been made for an array of freshwater prosobranch gastropods from Michigan and Missouri. A number of species from the prosobranch families Hydrobildae, Paludomidae, and Pomatiopsidae possess eggs which bear small polar lobes; however, no lobes were observed on the egg of a valvatid. Preliminary ultrastructural examination of polar lobes in the hydrobiid Amnicola limosa reveals the presence in the first lobe of a localization containing 500-lOOnm diameter vesicles, dense plaques, and dense bodies. The second lobe contains few small vesicles and no dense plaques or dense bodies, suggesting a dispersal of the first lobe localizations into the C and D blastomeres. Nelson, C. J., K. M. Zarrough, D. D. Randall, and D. A. Sieper, University of Missourt-Columbia, Columbia, MO 65211. PHOTOSYNTHESIS, PARTITIONING OF ASSIMILATES AND YIELD OF TALL FESCUE. There is genetic variation in both leaf photosynthesis and forage yield of tall fescue (Festuca arundinacea Schreb.) yet the two are not correlated strongly. A decaploid genotype, that is about 50% higher in photosynthesis than normal hexaploids, was crossed with a hexaploid to give octaploid progeny. We found the artificial octaploids were higher in photosynthesis than the midparent mean and had a top:root ratio similar to the hexaploid parent In contrast, the decaploid had a less favorable top: root ratio and had a larger amount of carbohydrate stored in the stem bases than did the other genotypes. Photosynthesis was highest for the decaploids. In order to have high yield it is apparent that plants must have a favorable dry matter distribution to utilize a high photosynthate rate. Partially supporied by U.S.D.A. Competitive Grant no. 5901-0410-9-0366-0. Jakubowski, T.. W. W. Zagaja and D. F. Millikan, Institute for Pomological Research, 96-100 Skiemiewice, Poland, and Department of Plant Pathology, University of Missouri-Columbia, Columbia, MO 65211. BREEDING WINTER HARDY PEACHES IN POLAND. Three clones of peaches, Prunus persica L., from North China which survived -38 C are being used as source of winter hardiness in peach breeding program in Poland. When crossed with high quality western cultivars several F1 seedlings retained the winter hardiness of the Chinese parent. When selected seedlings were intercrossed or backcrossed to high quality cultivars only 11 of 456 F2 population possessed the hardiness of the original Chinese clones. Selected seedlings from this population are being used to develop the F3 generation which is expected to have acceptable size and quality as well as winter hardiness.

Babrakzai, N., Central Missouri State University, Warrensburg, MO 64093. KARYOTYPE ANALYSIS OF ASHMUNELLA RHYSSA COMPLEX; GASTROPODA: PULMONATA. The diploid chromosome number in A. rhyssa complex is 2n =58 and the haloid number is n=29. Statistical analysis of chromosomes of two groups point out small or insignificant differences, due to homozygous translocations, pericentric inversions, different rates of contraction of chromosomes, or artifacts of measuring technique etc. Comparisons of Giemsa-banded and Centromeric-banded karyotypes did point small, but cytologically significant differences in three groups of A. rhyssa complex. The differences appear to have risen by homozygous translocations and inversions, with consequent loss or gain in constitutive heterochromatin. It is therefore, hypothesized that actual genetic differences in the groups of the A. rhyssa complex might be greater than the observed cytological ones. There are at least three cytologically different but morphologically confusing sibling spedes in the complex. 1982 Annual Meeting 159

Bio-Medical & Oncology Roberts, M., G. Resch, University of Missouri-Kansas City, Department of Biology, Kansas City, MO 64110. IN VITRO STUDY OF GONADAL INFLUENCE ON ADOLESCENT RATS. The pituitary portion of the hypothalamic-pituitary axis was examined to determine if as previously suggested programming of Prolactin (PRL) secretion by 30 day old rats was due to gonadal influence. Pituitaries from intact and gonadectomized (Gx) groups of male or female rats were examined for differences in PRL secretory response patterns by using dispersed pituitary cells in continuously perfused Bio-Gel, polyacrylamide bead, columns. Maximal inhibition of PRL secretion by Dopamine (DA) occurred at a cone. of 5X10·6M.. Comparisons of slope of DA dose response curves between the four groups revealed no change in response to DA due to gonadal influence thus suggesting that neuroendocrine programming is at the hypothalamic lev!'! rather than at the pituitary. Gx elicited prolongated Haloperidol (H) antagonism of DA. While Gx resulted in an alteration of DA response following H antagonism, it did not influence the pituitaries overall response to DA, as revealed by DA dose response curve characteristics for the four animal groups. Sheffield, L. G., and R. R. Anderson, University of Missouri-Columbia, Columbia, MO 65211. SIMULATION STUDY OF THE RELATIONSHIP BETWEEN MAMMARY GLAND WEIGHT, NUCLEIC ACID CONTENT AND MILK PRODUCTION IN GUINEA PIGS. Previously published equations were used to predict values for mammary wet weigh~ dry fat-free tissue weight (DFFT), DNA conten~ RNA content and milk production for each of the first 21 days of lactation. Regression equations were developed to relate wet weight, DFFT, DNA and RNA to milk production. Coefficients of determination were: .52 for wet weigh~ .03 for DFFT, .84 for DNA and .20 for RNA when the entire lactation was considered. Coeffidents of determinations were larger when the lactation period was divided into ascending and descending phases and these phases considered separately. Anderson, R. R., and L. G. Sheffield, Department of Dairy Science, University of Missouri-Columbia, Columbia, MO 65211. REGRESSION OF GONAD WEIGHTS ON BODY WEIGHTS IN MALE AND FEMALE GUINEA PIGS. Common albino Hartley guinea pigs were weighed and sacrificed at ages 20 to 150 days in males and ages 20 to over 365 days in females. Body weights were 152 to 1078 gin males and 181 to 1059 gin females. Numbers of animals were 100 males and 105 females. Gonads were dissected, trimmed, and weighed on a Sartorius electro balance to the nearest0.1 mg. Testes were 104.1 to 6370 mg. and ovaries were23.6 to 296.8 mg. Data were analyzed by computer for best fit of regression according to various programmed models such as linear, quadratic, cubic, and exponential. For males, the best fit was a quadratic equation: Gonad Wt (mg) = 12.69 + .215 Body Wt (g) + .00529 BW2 (R2 = .952). The exponential equation was also good: G.W. (mg) = .00598 BW (g)l.99 (R2 = .920). Forfemalesand best fit was an exponential equation: GW (mg) = .445 BW (g)·88 (R2 = .825). Stitzer, P. W., Northeast Missouri State University, and 0. B. Mock, Kirksville College of Osteopathic Medicine, Kirksville, MO 63501. ORGAN WEIGHTS IN THE LEAST SHREW. Total body and selected organ weights were recorded for known aged, laboratory raised least shrews, Cryptotis parva. The animals were weighed, anesthetized with ether and sacrificed by exsanguination. A total of 18 males and 16 females were examined. Ages ranged from 21 to 630 days. Mean body weight was 4. 91 g. No significant differences in weights of the selected organs based on sex or age were obsetved. The mean weight and standard deviation for each organ are listed in milligrams as follows: liver, 261 ±57; heart 42 ± 9; lungs 48 ± 10; spleen 26 ± 15; gastroenteric tract from the diaphragm to the anus, 192±92; kidneys, 78±14; and brain, 98±15. The absence of significant age-related variation in these organs indicates that any weight changes obsetved as a result of the autumnal phenomenon in wild least shrews can be accounted for as a consequence of environmental factors. Supported by AOA grant 81-04-306. Leslie, C. J., Northeast Missouri State University, and 0. B. Mock, Kirksville College of Osteopathic Medicine, Kirksville, MO 63501. AGE-RELATED CHANGES IN BROWN ADIPOSE TISSUE OF THE LEAST SHREW. Body and brown adipose tissue (BAT) weights were recorded for known aged, laboratory raised least shrews, Cryptotis parua. The mean weight of weanling males, 20-23 days, was 4.16 g with BAT representing 12% of the total body weight. Young adults, 100-199 days, displayed the highest mean body weight,.5.39 g. Older males showed a gradual age-related weight loss with changes in the amount of BAT explaining most of the observed weight changes. The mean BAT weight of pubertal males, 24-40 days, was significantly (P<. 01) less than weanling or young adults. The same general pattern of age-related distributions of weights was noted in females with weanlings averaging 4.41 g and young adults averaging 8.12 ± 2. 76 g and a mean BAT weight of 3.16 g. The large standard deviation reflects the extreme individual variations in BAT among this group. Supported by AOA grant 81-04-306. Imboden, R. T., Northeasi Missouri State University and 0. B. Mock, Kirksville College of Osteopathic Medicine, Kirksville, MO 63501. AGE DETERMiNATION IN THE MALE LEAST SHREW. Age-determining criteria were established from the examination ofthe testes of 101 known aged, laboratory raised least shrews, Crypototis parua. Greatest testis lengths of over 3f4 the adult measurements of 4.0 mm were recorded in weanlings (21 days). Adult testis size was observed in animals of 30 days of age. The most advanced stage of spermatogenesis lists by days of age was: 20 days-primary spermatocytes; 22 days-spermatids (no spermiogenesis); 30 days­ spermatozoa; 35 days-epididymides packed with spermatozoa. Lipofuscin accumulations were noted in the interstitial cells of all animals over 35 days. Gross testis color ranged from pale yellow in pre-pubertal shrews to dark olive gray in animals over five months of age. This research was supported by AOA grant 81-04-306. Cornell, R. P., Northeast Missouri State University, Kirksville, MO 63501. EFFECT OF GUT-DERIVED ENDOTOXIN ON BASAL AND GLUCOSE-STIMULATED PLASMA INSUUN AND GLUCAGON IN RATS. Recent studies by this laboratory have described the enhanced pancreatic secretion of insulin and glucagon in response to a low dose (331J.g/l00g) of intravenously administered bacterial endotoxin in normal fasted rats. The possible role of endogenous gut-derived endotoxin in modulating pancreatic hormonal secretion was investigated by blocking normal endotoxin absorption from the gut into the portal venous circulation. Rats which were treated for nine days by gavage with neomycin sulfate to sterilize the gut or with lactulose and cholestyramine to bind nonabsorbed endotoxin or by injection with endotoxin to produce endotoxin antibodies or with polymixin B to bind absorbed endotoxin all had reduced basal plasma portal and systemic venous levels of insulin but not glucagon compared to normal control animals. McClellan, C., and R. P. Cornell, Northeast Missouri State University, Kirksville, MO 63501. INFLUENCE OF PANCREATIC HORMONES ON MACROPHAGE PHAGOCYTOSIS IN VIVO AND IN THE ISOLATED PERFUSED UVERS OF RATS. Recent 160 Transactions, Missouri Academy of Science Vol. 16 studies by this laboratory have described macrophage hyperphagocytosis of intravenously administered particulate materials including collodial carbon in rats with experimental diabetes mellitus. Because hypoinsulinemia as well as hyperglucagonemia and hypersomatostatinemia occur in diabetes, the influence of pancreatic hormones on macrophage phagocytosis was investigated. Both in vivo and in isolated perfused livers of normal rats insulin significantly depressed colloidal carbon phagocytosis while glucagon and somatostatin significantly stimulated carbon uptake by macrophages. These findings suggest that the pancreatic hormonal imbalance of diabetes may be responsible for the enhanced macrophage phagocytosis in diabetic rats.

Twining, L. C., Division of Science, Northeast Missouri State University, Kirksville, MO 63501. THE EFFECTS OF MITOGENS ON PlASMODIUM BERGHEI INFECTION IN THE AIJ MOUSE: PATENCY, COURSE OF INFECTION, AND MORTALITY. Concana· valin A (Con A), pokeweed mitogen (PWM), and phytohemagglutinin P (PHA-P) were administered to AIJ mice who were sub­ sequently infected with 5 X 105 Plasmodium berghei infected red blood cells. Dally blood films were made to monitor onset of infection and parasitemia levels. Treatment with PWM delayed the onset of infection with 100% patency occurring on day 7 of the infection as compared to day 3 of control groups. All three mitogens affected the course of infection, shifting the peak in infection levels to a later point in time (Con A and PHA-P), and showing, in the early part of the infection, lower parasitemia levels (Con A, PWM and PHA-P). Con A and PHA-P decreased mortality, extending the mean date of death by 5 days. Severn, D. J., 608 W. ScottSt, Monett, MO 65708. PARENTAL RECESSNEALLELEFREQUENCY AND RECESSIVE SURVIVAL RATE IN HUMAN GENETIC DISORDERS. Undetected prenatal losses of homozygous recessives suggest a need for an approach to estimation of parental recessive allele frequency (q) that avoids the observed filial recessive frequency (R) in calculation. This need is met by the equation q = H/(2D + H), in which D and H represent observed filial homozygous dominant and heterozygote frequencies, respectively. Survival (w) of recessive offspring between conception and time of detection can then be estimated with the equation w = 4DRIH2. Hampton, C. M., R. R. Walsh, W. J. Leverich, Department of Biology, St Louis University, StLouis, MO and L. M. Dunkle, School of Medicine, St Louis University, St. Louis, MO 63103. IN VITRO TRANSFER OF AMPICILLIN RESISTANCE FROM HAEMOPHILUS PARAINFLUENZA£ TO HAEMOPH/LUS INFLUENZA£ TYPE B. Two hundred and twenty isolates of H. parainfluenzae and seven isolates of H. influenzae type b were obtained from three hundred healthy children less than five years of age in the St Louis area. Isolates of H. Parainfluenzae, an infrequent pathogen, was shown to have a high incidence of B-lactamase activity (72%). All isolates of H. influenzae type b, a causative agent of meningitis and other invasive diseases, were from children that were cocolonized with H. Parainfluenzae. Ampicillin resistant H. parainfluenzae and a spontaneous mutant of H. influenzae type b to rifampin were incubated as a mixed culture and selectively plated to explore a possible mechanism for spread of ampicillin resistance to H. influenzae type b. Marx, R. J. and D. D. Isaak, Northeast Missouri State University and Kirksville College of Osteopathic Medicine, Kirksville, MO 63501. MULV-FRlEND SUSCEPTIBILITY OF LYMPHOID CELLS FROM YOUNG VERSUS OLD MICE. Murine leukemia virus-Friend (MuL V-F) reportedly induces fatal lymphoid leukemias within several months postinfection of new-born mice but infection of adult mice induces only transient viremia and mild splenomegaly but no fatal disease. We have analyzed the spleens and thymi of young versus old mice with an in vitro infection system employing lipopolysaccharide and concanavalin A as a probe for susceptible B and T cells. Surprisingly, equivalent numbers of susceptible B cells were detected in spleens of young and old mice; old mice appeared to have more susceptible splenic T cells but fewer susceptible thymus cells. Mice infected as newborns experienced a higher level of viremia and died earlier but because adult-infected animals uniformly experienced fatal lymphomas after a long latent period, our results suggest that adult mice may not be as resistant to MuLV-F as previously reported. Isaak, D. D., Kirksville College of Osteopathic Medicine, Kirksville, MO 63501. CELLULAR INTERACTIONS IN IMMUNE SYSTEM ACTIVATION. Specific immune responses result from two basic processes, antigen-independent and antigen-dependent differentiation of immunocytes. The first involves the genetically programmed events which lead to the differentiation of pluripotent bone marrow stem cells into the principal cells of the immune system, the thymus-derived or T lymphocytes, the bone marrow-derived or B lymphocytes, and macrophages. While macrophages may interact with antigens nonspecifically, during the process of antigen-independent differentiation, lymphocytes acquire surface receptors which allow them to interact with antigen very specifically (ie. they become clonally restricted, even in the absence of antigen). Antigen-dependent differentiation of immunocytes begins when introduced antigen is phagocytized, specifically processed, and presented by macrophages to lymphocytes. This activates antigen-committed (by virtue of their surface receptors for antigen) regular T lymphocytes to either help or inhibit antigen-committed T effector of B lympl]ocytes to result in initiation or suppression of cell or antibody-mediated immune responses respectively.

Conservation Topping, M.S., and D. M. Sullivan, Southwest Missouri State University, Springfield, MO 65802. SECONDARY PRODUCTION IN /SONYCHIA BICOLOR (WALKER). Life history of Isonychia bicolor was studied in Finley Creek, Missouri from April1979 until June 1980. These data were reconsidered to estimate secondary production. Emergence occurs primarily during spring and fall, but whether this is due to two distinct generations or. a single, asynchronous generation is not certain. Estimates of annual production from length-frequency data using Allen curve, instantaneous growth, and Hamilton-Hynes methods were 1.1-1.4, 1.6 and 1.6 g drwt m-2, respectively. Although asynchrony would lead to underestimate by either Allen curve or instantaneous growth methods, their general agreement with theHamilton-Hynes method (assuming a single generation) implies the species is characterized by a single asynchronous generation. Standing crops averaged 0.33 g dry wt m·2 and turnover ratios ranged between 3.3 and 5.0. Kerns, H. A., University of Kansas, Lawrence, KS 66045. THE LIFE HISTORY OF THE TOPEKA SHINER, NOTROPIS TOPEKA (GII..BERT). The distribution of the Topeka shiner, Notropis topeka (Gilbert) in Kansas centers on the Hint Hills tributaries of the Arkansas and Kansas Rivers. Life history of N. topeka was studied on the South Fork of the Cottonwood River (Arkansas drainage), Butler County, Kansas from June 1979 through May 1981. Information on habitat. associated species, growth, longevity, fecundity, reproductive age, and other population characteristics on the South Fork population is reported. 1982 Annual Meeting 161

Taber, C. A., alld B. A. Taber, Southwest Missouri State University, Springfield, MO 65802. REPRODUCTIVE BIOLOGY AND AGE AND GROWTH OF THE MISSOURI SADDLED DARTER, ETHEOSTOMA TETRAZONUM. Populations of Missouri saddled darters in the Niangua River and the Pomme de Terre River of southwest Missouri were compared with respect to age and growth, reproductive cycles and fecundity. These darters typically produced clutches of 40-50 eggs spawned periodically in a season that extended from mid-March to late July. Reproductive activity was interrupted in late April when heavy rains produced high stream flows. Darters in the smaller Porn me de Terre River were larger at a given age and in better condition than those in the Niangua River. Maximum longevity for both sexes was found to be four years with sexual maturity attained at one year in females and after two years in males. flora, S. J., and M.S. Topn>ing, Southwest Missouri State University, Springfield, MO 65802. POPULATION DYNAMICS AND PRODUCTIVITY OF THE PROSOBRANCH SNAIL, GONIOBASIS. Abundance, growth, and production of Goniobasis sp. have been studied since March 1981 at three locations in the upper James River. The species is univoltine and individuals live from 12 to 16 months. Reproduction appears to occur during late spring to early summer and young appear by mid-summer. Abundance averaged 870m-2 and ranged 17 to 5962. Biomass, computed from length~frequency data using an empirically determined length-weight relationship, average 4.0 g ash free dry wt of live matter m-2 and ranged from 0.14 to 6. 7. Preliminary estimates indicate annual production to be about 16.5 gash free dry wt of living matter m·2.

Crunl

Haefner, R. The Nature Conservancy (TNC), P.O. Box 176, Jefferson City, MO 65102. MISSOURI NATURAL HERITAGE INVENTORY. One of 28 in the nation, the Heritage Inventory is an integral part of TNC's systematic approach to preseroing natural diversity. Authorized by the Missouri Natural Areas Committee (DNR and MoDoC), the Inventory is an information management method designed to identify, analyze, and store information on significant "elements" of natural diversity: vegetative communities, special plants, special animals, and geologic features. Data about the life history and occurrences of significant elements is stored in an interactive system of files which includes a computer and a map file. This information will guide TNC and the state in setting preseroation priorities and in designing preseroes. The Inventory will also be useful in the environmental review process and scientific research. The system will be fully operational by 1983.

Durhom, J F., W. IS. Kurtz, H. E. Gorrert ami M. D. Brown, School of Forestry, Fisheries and Wildlife, University of Missouri, Columbia, MO 65211. THE SILVICULTURALAND ECONOMIC FEASIBILITY OF THINNING MIXED STANDS OF SCARLET AND BLACK OAK IN SOUTH MISSOURI. A silvicultural and economic analysis was conducted in 1980 following 1960 and 1963 thinnings of 40~year old mixed black and scarlet oak stands in south Missouri. Trees were sampled from residual basal areas of 50, 60 and 70 square feet per acre on areas with site indexes of 60 and 70. Diameter growth of released trees was significantly greater (p ~ 0.05) than nonreleased trees. There was no significant difference in diameter growth between the three residual stocking levels or the two site indexes. At all residual stocking levels and both site indexes, diameter growth was greater after the thin than before for periods of up to 16 years. Thinning black and scarlet oak was found to be economically feasible for stumpage prices of $50.00 and $70.00 per MBF on both sites at residual stocking levels of 60 or 70 square feet basal area over the 60-year rotation.

!Hough, R. D., Southwest Missouri State University, Springfield, MO 65802. LITTLE SAC WOODS METRO FOREST: A LAND MANAGEMENT METHODOLOGY AND PLAN. This study was done to develop a method for resource evaluation and to assess suitability of specific locations within the 312 ha area for development. Some 320 grid cells were evaluated with respect to vegetation and woodland type; soil type; slope; aspect; position; water resource and management; frequency and duration of flooding; depth to bedrock; cultural resources; and ratings for camping, picnicking, building paths and trails, and making shallow excavations. Each location was rated for each attribute and an average score for all attributes was computed for each grid cell using an SPSS program. Results were then plotted using SYMAP routines to facilitate interpretation. Methodology, criteria used, and specific recommendations will be presented and discussed.

Schmidt, T. L., and A. R. Everson, Associate Professor, School of Forestry, Fisheries and Wildlife, University of Missouri, Columbia, MO 65211. OUTDOOR RECREATION AS A LAND USE IN MISSOURI. The five objectives of this study were 1) to determine the past and present patterns of outdoor recreation land use for four public agencies, 2) to compare past to present for the four agencies, 3) to compare present patterns of the four agencies to total present use of land for outdoor recreation, and 4) to compare outdoor recreation land use to total land use within Missouri, and 5) to project outdoor recreation as a land use in Missouri. Results showed that the four agencies now provide 81 percent of present outdoor recreation land, and that these lands had increased notably since 1969. Seven percent of all Missouri acres are now used for recreation, the vast majority located south of the Missouri River. Changes are expected near urban areas and along major transportation conidors.

Tansil, J. E., Department of Physics, Southeast Missouri State University, Cape Girardeau, MO 63701. WHITEWATER RECREATIONAL POTENTIAL OF THE ST. FRANCIS RIVER IN MISSOURI. The St. Francis River west of Fredericktown, Missouri, provides the only whitewater of any significance for potential users within a 300 mile radius. It will be shown that the St. Francis is a unique recreational resource for those interested in whitewater sports. Factors affecting river use such as water level, temperature, access, and river safety will be discussed. The impact on river use due to the recent purchase by the Missouri Department of ConseJVation of the best whitewater section on the river (Millstream Gardens) will be discussed. 162 Transactions, Missouri Academy of Science Vol. 16

Meyer, D. and A. R. EversoiDl, Associate Professor, Schoo! of Forestry, Fisheries and Wildlife, University of Missouri, Columbia, MO 65211. RECREATIONAL USE OF THE MISSISSIPPI RIVER. The purposes of this study (between Saverton, Missouri and Cairo, Illinois) were to 1) identify characteristics of Mississippi River recreationists (residence, mode of travel, visitation time), 2) determine parties' primary recreational activity, and 3) determine what improvements or facilities were perceived as being needed by the party. Questionnaires were used to inteiVi.ew all persons present during 4~hour intervals at randomly selected recreation access sites. Results indicated that three-quarters of 500,000 total users were from Missouri, travelled in a car or truck, and recreated on the river for 4-8 hours. Boating (58 percent} was the largest activity with fishing {40 percent) being second, and with waterskiing (26 percent), picnicking (24 percent} and swimming (21 percent) being tied for third. The three most repeated suggestions from users concerned too much fluctuation of water levels, too few accesses, and too few toilet facilities.

Zarrick, P., Research Assistant and A. R. Everson, Associate Professor, School of Forestry, Fisheries and Wildlife, University of Missouri-Columbia, MO 65211. ASSESSING LATENT DEMAND FOR OUTDOOR RECREATION ON THE MISSISSIPPI RIVER THROUGH PUBLIC INTERVIEWS. The purpose of this study was to assess latent demand for recreation on the Mississippi River; the objectives were: 1) to determine the percentage of recreation users and non-users, 2} to identify conditions which could be improved to increase recreation use of the river, and 3} to project future use levels on the river. Using a telephone survey, 4000 people were randomly selected and interviewed within a specified area (between Saverton, MO and Cairo, IL). Results indicated that 17% of the sampled population used the river within the last year. In addition, increasing riverfront development, cleaning the river, and providing more information on existing recreational areas could expand opportunities for recreation. For the coming year, recreation use on the river will increase by nearly 15%.

Pernhall, R.. B., and A. R. Everson, Associate Professor, School of Forestry, Fisheries and Wildlife, University of Missouri, Columbia, MO 65211. ASSESSING PRIVATE LAND USES AS ALTERNATIVES TO PUBLIC STRIP MINE RECLAMATION NEAR COLUMBIA, MISSOURI. Thousands of acres of old strip-mined lands exist near Columbia, Missouri. These create environmental and economic problems for the region. Public funds are inadequate to reclaim these strip-mined areas, therefore private investment is explored as an alternative avenue for reclamation. The specific purpose of this study is to investigate two specific tracts of strip-mined land in regard to their potential for selected private use. A matrix is developed and employed to evaluate each study site as regards fourteen physical site factors involved in five different private land uses of reclaimed land. The economic feasibility of reclaiming the two study sites is explored. Private reclamation does seem possible, although only some kinds are economically feasible; private and public enterprise could share costs of reclamation.

!lllcCormicl<, F. H., W. J, Resetarits, N. Aspinwall and J. E. Thomerson, Saint Louis University, St. Louis, MO 63103 and Southern Illinois University, Edwardsville, IL 62025. A NOTE ON THE DISTRIBUTION OF FISHES IN THE MERAMEC RIVER, MISSOURI. The Meramec River in east-central Missouri possesses tremendous habitat diversity. A distributional analysis of the fishes of the Meramec River was made from collections at four sites. The collection of 58 species from one locality near Grey Summit, Franklin County, Missouri is a record for a single locality in North America. The collection of the skipjack herring, Alosa chrysochloris from two locations represents a range extension for the species.

Entomology Mei•er, R. F, !Ill, F. Haskins, and J. H. Shaddy, Northeast Missouri State University, Kirksville, MO 63501. HARVESTMEN ACTIV!lY IN AN ECOLOGICALLY MANIPULATED OLD FIELD. Harvestmen activity was measured in burned, mowed, and plowed fields and compared to that of a control field. Significant differences in Harvestmen activity were observed in fields that had been annually manipulated for 4 years as compared to their succession counterparts. Hadrobunus and Leiobunum genera were identified. White and clear pit traps were used during an 8-month period as a sampling device and the number of Hatvestmen captured in each compared.

Haskins, M. F., R. F. Meiser, J. H. Shaddy, Northeast Missouri State University, Kirksville, MO 63501. THE EFFECTS OF HABITAT MANIPULATION ON GROUND INHABITING SPIDERS (ARANEAE) IN AN OLD FIELD ECOSYSTEM. The effects of burning, mowing and plowing on spider activity were measured and compared to that of a control field. Significant differences in spider activity were obsetved in annually manipulated fields as compared to manipulated fields undergoing succession and the control field. Spider activity was correlated with vegetation data. Samples were taken from April-November 1980 using white and clear pit traps as the sampling device. Significantly more adults were captured in white traps than dear traps. Spiders in the families Lycosidae and Gnaphosidae were the most frequently captured. Schizocosa a vida {Lycosidae) and Zelotes laccus (Gnaphosidae) exhibited the highest active density during the sampling period.

Feir, D, and R. Veras, Biology Department, St. Louis University, St. Louis, MO 63103. CORPORA ALLATA VOLUME CHANGES IN THE LARGE MILKWEED BUG, ONCOPELTUS FASC/ATUS. Surface areas and volumes ofthe corpora allata were calculated from measurements of tissue sections of 4th, 5th, and adult instar milkweed bugs. The volume changes were similar to literature reports of juvenile hormone titer patterns in other Hemiptera. The application of an exogenous juvenile hormone mimic four hours after ecdysis caused changes in the corpora allata volumes. The significance of these changes will be discussed.

Gerdes, C. F, Division of Science, Northeast Missouri State University, Kirksville, MO 63501. PHENETIC AND CLADISTIC COMPARISONS OF NEOTROPICAL LITTER THRIPS (THYSANOPTERA: PHLAEOTHRIP!DAE: GL YPTOTHRIPINI) Neotropical litter thrips in one tribe were compared by phenetics and cladistics to determine to what extent the groups agreed with traditional classification. Specimens were measured for 22 continuous characters which were used in phenetic analysis. Eight non-continuous characters, each with 2 states, were used in cladistics of 15 genera. With phenetics specimens of Eurythrips and Terthrothrips were separated into similar groups as with traditional non-numerical characters. Cladistics applied to the genera produced a few patterns that reinforced geographical and evolutionary peculiarities, such as a possible African origin, general increase in the female/male ratio, and highly elaborated cuticle in more derived genera. 1982 Annual Meeting 163

Environmental Science Brown, 0. R., R. Seither, D. Perez, and H.. Amash, Dalton Research Center, University of Missouri, Columbia, MO 65211. MECHANISM OF TOXICITY OF HYPERBARIC OXYGEN, PARAQUAT AND SOME REDOXACTIVE DRUGS. Using Escherichia coli, the toxicities of hyperbaric oxygen, paraquat, adriamycin, and nitrofurantoin have been examined at the biochemical level. The most data have been obtained with E. coli tested with paraquat and hyperbaric oxygen. Specific enzymes required for the synthesis of specific amino acids (AA) are selectively poisoned and provision of these AA leads to significant protection against growth inhibition. In the absence of the AA, genetic stringency is induced via ppGpp production and leads to inhibition of the many cell processes so regulated. Quinolinate phosphoribosyl transferase, required for de novo synthesis of pyridine nucleotide coenzymes is also poisoned and leads to severe decreases in these coenzymes unless niacin, which enters the biosynthesis pathway below the poisoned step, is provided. Thiamine also declines in cells poisoned by oxygen and paraquat. Nitrofurantoin also induces stringency, but adriamycin does not.

Resch, G. !E., and C. W. Simpson, Department Biology/Medicine, University of Missouri-Kansas City, Kansas City, MO 64110. REWARMING IN THE HELIUM-COLD HYPOTHERMIC HAMSTER One classic distinction between experimental hypothermia and natural hibernation has been the inability of the experimental preparation to spontaneously rewarm. We have undertaken the study of rewarming in Helium-Cold hypothermic (Hpc) Golden Hamsters (M. auratus). Hamsters made Hpc by the method of Fischer & Musacchia (1968) were implanted with carotid cannulas and stereotaxically positioned guide tubes lmm above Preoptic Anterior Hypothalamic (AHPOA) thermosensitive sites. Colonic temperature (T c) was measured with a YSI thermistor probe. Carotid NE (2mg/ml) injections elicited a 20% increase in heart rate but no detectable changes in Tc. ACH injected at the AHPOA in animals near but below the 11.5°C threshold for spontaneous rewarming elicited sufficient thermogenesis to initiate spontaneous rewarming. This suggests the activation of a possible nonshivering thermogenic mechanism not typical of the hibernator. To the best of our knowledge this is the first controlled initiation of rewarming an experimental Hpc animaL (Suppmted by FRG2401-2100 & FRG 10498) Simp•on, C. W., a"d G. E. Re•ch, Department of Biology/School of Medicine, University of Missouri-Kansas City, 5100 Rockhill Rd., Kansas City, MO 64110. CENTRAL PHARMACOLOGICAL STIMULATION OF THERMOGENSIS IN THE HELIUM-COLD HYPOTHERMIC HAMSTER. Models of central excitation of thermogensis developed in normothermic animals have rarely been studied in hypothermic depressed metabolic states. Central injections of Ach(Beckman & Satinoff) stimulated rewarming in the hibernating animals, however experimental hypothermic animals do not spontaneously rewarm. This obseiVation may suggest alternations in the central mechanisms. Prostaglandin (PGE2i, Serotonin (5-HT) and Acethylcholine (ACH) injections in AHPOA thermosensitive sites through stereotaxically implanted guide tubes in He-cold hypothermic hamsters elicited increases toT c in the range above 9oC. Below this range however the same pharmacological stimuli elicited reduced or no response in Tc compared to saline control injections. The difference in response to the pharmacological stimuli in the temperature range 7-9° C suggests to us a cold block of central receptors. These results suggest the circuits in the AHPOA are the same however the functional properties are altered in deep hypothermia. Supported in part by FRG 10498 & FRG 2401-2100

Melethil, S., W. Cowden, S. Reynertson and T. Baker, Schools of Pharmacy and Medicine, University of Missouri, Kansas City, MO 64108. KINETICS OF ABSORPTION OF CADMIUM CHLORIDE FROM RAT LUNG. The disposition of intratracheally administered cadmium, an environmental pollutant, was examined in rats at three dose levels (10, 100 and 1000f,Lg of cadmium). Each dose was spiked with about 0.25J.LCi of carrier free Cd109. The disappearance of cadmium from the rat lung was biphasic at all3 doses. Over an 8 hour period, 60-70% of the administered dose was absorbed. During the rapid initial phase (first 1.5 hours), 50% ofthe dose at the 10 and 1001J.g levels was absorbed and 30% at the 10001J.g leveL The majority of absorbed cadmium was found in the liver (about 50% of the dose). Only trace amounts (<0.5% of dose) appeared in the heart, spleen, blood and testes. The terminal half-lives for cadmium loss were 9.04, 11.50 and 6.30 hours for the 10, 100 and 10001J.g doses, respectively. Preliminary studies show that at equimolar doses, selenium does not influence the disposition of the cadmium at the 100J.Lg dose level.

Gan, 0., S. Hsu, and D. Yourtee, Toxicology Research Program, University of Missouri, Kansas City, MO 64108. DETERMINATION OF THE NUCLEOPHILIC PROPERTIES OF ENVIRONMENTAL CARCINOGENS. In this study the nucleophilic reactivity of the potent toxins/carcinogens aflatoxin 81 and ethionine was evaluated. These substances were dissolved with divalent copper in aqueous solutions and their ability to form coordination complexes was measured by microcalorimetry and ion selective potentiometry. Aflatoxin B1 and ionic copper in methanol-water (2:3) formed weak complexes IK1 ~ 2.453x103 at 24°C, L677x104 at 37°C). Computed Ll.H was endothermic. Ethionine and ionic copper yielded both solution and crystalline copper complexes; the formation of solution complexes was an endothermic process. Potentiometric and thermodynamic nucleophilicity of ethionine resembled methionine and was different than cysteine. Carcinogenicity is usually equated with electrophilic centers in the active agent. The role of nucleophilic sites in the agent is often overlooked although, ultimately, the action of such chemicals should be dependent on their total reactivity.

Kirk-Yow-tee, C. and D. M. Yow-tee, Toxicology Programs, University of Missouri, Kansas City, MO 64108. ROLE OF AMINO ACIDS IN THE ANTAGONISM OF AFLATOXIN MUTAGENESIS. Studies are being continued to evaluate apparent reductions in mutagenicity of aflatoxins caused by certain amino acids. In this work the effect of amino acids on the dose mutagenicity response caused by aflatoxin B1 in the Salmonella microsome test is explored. Previously (Trans. Mo. Acad. Sci., Jan. 1982) methionine suppressed the microsome dependent mutagenic effect of 81 at some of the doses studied. These experiments have been extended and the effects of other sulfur containing amino acids explored in the test system including ethionine, glutathione, cysteine, and cystine. These substances had only marginal effects on the Salmonella spontaneous mutation, with or without liver microsomes in the test. Since the discovery of the potent carcinogenic effect of the aflatox.ins, a number of studies have revealed potential for protection against these agents through dietary alterations and/or supplementations. A perspective relating these and the current Salmonella mutagenicity work will be given.

Elashl

Forensic Science Nienhouse, J., Southeast Missouri Regional Crime Lab, Cape Girardeau, MO 63701. ON THE INCIDENCE AND AUTHENTlCllY OF POISONINGS IN CONTEMPORARY DETECTIVE FICTION. The presentation deals with study and research on the incidence and authenticity of poisonings in contemporary detective fiction. Certain authors, using accurate chemical, toxicological, and state of the art forensic analyses, have created a new standarQ of the mystery novel involving a poisoning. In addition to identifying a general scheme of the mystery novel involving a poisoning, the speaker will point out numerous examples of poisonings in contemporary literature which reveal a high degree of sophistication on the pari of the author. The presentation will also incorporate the structures and pharmacological data on the compounds used in the poisonings dted as well as the role forensic scientists played in their determination. A critique on the authenticity of the reported mode of action and choice of the route of administration of the poison will be provided. Crlppln, J. B., Missouri State Highway Patrol, Troop H Satellite Laboratory, St. Joseph, MO 64506. EXPLOSIVE RESIDUE ANALYSIS. In a case concerning explosives, more often than not there is a limited amount of actual residue left to identify. By examining the debris and explosive device fragments the explosive may be divided into a high or low explosive class. Further analysis is done by chemical spot tests, thin layer chromatography, liquid chromatography, infrared spectrophotometry, gas chromatography/mass spectrum, and x-ray power diffraction to identify the explosive involved. Smith, D. E., H. Chriswell, and G. Gietzen, Forensic Science, 321 E. Chestnut Expressway, Springfield, MO. GAS CHROMATOGRAPHIC ANALYSIS OF ARSON ACCELERANTS. A presentation identifying various components of petroleum hydrocarbon accelerants and others will be demonstrated. The ability to identify these components is valuable in characterizing various accelerants. Pape, B., Pathology-Toxicology, Health Sciences Center, University of Missouri-Columbia, Columbia, MO 65212. FORENSIC TOXICOLOGY: QUALllY ASSURANCE OF ANALYSES WHICH ARE INFREQUENTLY PERFORMED. Analyses which are infrequently performed are often subject to greater variance and less accuracy compared to "routine" analyses. This variance and accuracy can be indirectly translated into decision-making errors in forensic cases (causation,.....[analyte]....,.. association). In quantitative analyses, variance and accuracy can be grossly modeled as a function of analyte and concentration, biological sample, analytical procedure, method of standardization, technical skill, instrument function, quality assurance criteria, and clinical-forensic interpretation of analytical (case) results. The effects of these variables are discussed, and approaches to providing "in-laboratory" or "out-of-laboratory" analyses are discussed. Johnson, C., P. Cary, P. Whitter, and B. Pape, Pathology-Toxicology, Health Sciences Center, University of Missouri, Columbia, MO 65212. FORENSIC TOXICOLOGY: REVIEW OF THE RESULTS OF A NATIONAL QUALilY ASSESSMENT SURVEY. Results of the Forensic Toxicology Laboratory Proficiency Testing Research Program (Dept. of Justice Grant No. 80-1J-CX-0072) are reviewed in terms of qualitative analyses (methods, false ± ), quantitative analyses (method, all-laboratory mean, coefficient of variation), and consequences related to decision-making models (case and analytical results). Design factors incorporated in quality assessment surveys are discussed. Corcoran, R. D .. Jefferson County Crime Laboratory, Jefferson College, Hillsboro, MO 63050. MOCK CRIME SCENE - A TRAINING SESSION. The Jefferson County Law Enforcement Training Center will be discussed with emphasis placed on the Evidence Technician Training Section, which culminates in the processing of a Mock Crime Scene. Slides of the Mock Crime Scene will be shown and discussed. Russell, S., Department of Biology, Southeast Missouri State University, Cape Girardeau, MO 63701, and R. C. Briner, Southeast Missouri Regional Crime Lab, Southeast Missouri State University, Cape Girardeau, MO 63701. CHEMICAL METHODS OF ANALYSIS FOR SEMINAL FLUID INDICATORS. Presumptive color tests will be presented and discussed for acid phosphatase, choline and spermine. Acid phosphatase is a usual marker in seminal fluid identification. The presence of choline and spermine can be indicated in the same battery of tests. This is presently being used as a routine method here at our laboratory.

Stockam, J., and P.R. Whittle, MSSC Regional Crime Laboratory, Joplin, MO 64801. ESTIMATE OF THE TIME OF DEATH VIA THE POTASSIUM LEVELS IN THE VITREOUS HUMOUR: MAN AND BEAST. The post-mortem increase of potassium in the vitreous humour of humans was reported as early as 1959; subsequent studies have correlated the level of potassium with the time since death. Analytical methods will be discussed along with selected case studies conducted by this Laboratory since 1974. Preliminary results from a study of the potassium levels in the vitreous humour of deer as a function of time after death will be presented. Freedman, H. J., and B. Baker, University of Missouri-Kansas City, MO 64108. SISTER CHROMATID EXCHANGES (SCEs) IN NORMAL HUMAN LYMPHOCYTE LINES IN HUMAN VS. BOVINE SERA: A POTENTIAL IN VITRO TEST FOR CARCINOGENS. Scoring of chemically induced SCEs in either Chinese hamster ovary (CHOP) cells or fresh human lymphocytes has some value in screening chemicals for carcinogenicity. However deficiencies include unknown validity of hamster-to-human extrapolation, need for human donors, or inter-sample lymphocyte variability. Therefore, we are attempting to simulate the human condition whtle abolishing sampling needs. Normal permanent human lymphocyte lines (RPM!-1788, B-85) having a stable chromosome number were sought and grown in human and bovine sera at different % strengths-both sera being effective. Their stained metaphases and SCE rates are being examined in the light of fresh donor cells and CHOP cells. It is suggested that SCE induction in permanent human lymphocyte lines (RPMI-1788) adapted to human serum supplemented media, could serve as a screening test for suspect carcinogens or leukemogens. 1982 Annual Meeting 165

Geography

Roark, M., Department of Earth Science, Southeast Missouri State University, Cape Girardeau, MO 63701. LEAD MINERS IN EASTERN MISSOURI. Lead mining was a critical activity in forming the economy and culture of Southeast Missouri. There were four identifiable periods of mining, each with distinct culture groups. The first miners, the French, were followed by Anglo-Americans from the Upper South of Virginia and from Pennsylvania. The first European immigrants in the nineteenth century were English and Irish. The next migrants were Germans, and lastly Eastern and Southern Europeans migrated to the Lead Belt. The result of the migrations was a complex culture region, quite distinctive from the surrounding Upper Southern Ozarks.

Phillips, R. I..., Geography, Central Missouri State University, Warrensburg, MO 64093. PINGOES: PROCESSES AND AREAL EXTENT. One of the more unique landforms of the cold climate Arctic regions of the world is the pingo ice-mound. In addition to the intriguing question of origin, one is immediately impressed by the areal distribution of these unusual geomorphic features. Pingoes are not found uniformly distributed across the Arctic landscape, but are instead highly localized in their occurrence. !tis the intent of this paper to investigate the apparent relationship between process and spatial distribution of pingoes.

Balcenak,l'., and A. C. Robertson, SOUTHEAST MISSOURI STATE UNIVERSITY, CAPE GIRARDEAU, MO 63701. URBAN HEAT ISLAND OF CAPE GIRARDEAU, MISSOURI. Cape Girardeau exhibits an urban heat island similar to that commonly associated with large metropolitan areas of the world. Temperature data were recorded during November and early December at six sites located along a linear route from the central business district westward into the rural environment. Supportive weather information was also obtained from the FAA at the Cape Girardeau Municipal Airport located south of the city. The findings were compared with urban heat island studies conducted in cities of comparable size.

Driever, §., Department of Geosciences, University of Missouri-Kansas City, Kansas City, MO 64110. OIL AND THE ECONOMIC DEVELOPMENT OF MEXICO. In the late 1970's, drawing on newly discovered petroleum reserves, Mexico again became a major oil producer. Mexico has exported most of her oil production in order to acquire the capital, technology, and services needed for rapid economic development. The developmental priorities of the government are continued exansion of the oil industry and decentralization of industrial growth. Resulting socio-economic and environmental dislocations, though evident in several regions, have received scant official attention. According to Hubbert's Curve, Mexico's oil production (and oil-dependent rapid economic development) may well peak in the late 1980's.

Geology-Geophysics Fraunlelter, G., Southern Illinois University, Carbondale, IL 62901. GEOLOGY OF THE PALENQUE, MEXICO, AREA-SOME ASPECTS. During the summer of 1976 clay samples were collected in the Palenque area in order to determine the sources of the clays utilized by the Mayans to make their pottery and other ceramic objects. In addition, extra clay samples for microfossils and megafossils from adjacent beds were collected for age determinations. This was done because the area had not been mapped geologically since the late 1800's, the area is structurally complex, and many new roadcuts were available for study. Most of the megafossils have now been processed and identified. The fauna consists of Bivalves and Gastropods with some Corals, Echinoids, Bryozoans, Crabs, Barnacles and vertebrate remains. The megafossil study indicates that much of the bed-rock area mapped as Oligocene is Miocene in age, but most of the area mapped as Eocene is of Eocene age. Closer age determinations were not possible because many of the megafossils were preserved as molds and casts and the microfossils have not yet been studied. Thus, the age of the clay samples collected has only grossly been ascertained.

Sumner, H. S., University of Missouri-Rolla, Rolla, MO 65401. APPLICATION OF HECKEL'S CYCLOTHEM MODEL TO INTERPRET DEPOSITION ENVIRONMENTS OF THE MULKY CYCLOTHEM OF NORTH-CENTRAL MISSOURI. University of Missouri-Rolla, Rolla, MO 65401. Interpretation of depositional environments of the Pennsylvanian Mulky cyclothem in an area of North-Central Missouri indicates that a transgressive-regresive sequence of strata compatible with Heckel's cyclothem model is present. Environmental criteria imply that the basal bed of the Mulky cyclothem represents the interval of maximum regression or subaerial emergence. Stratigraphically higher beds represent a single transgressive pulse of the sea with maximum marine transgression occurring during deposition of an anoxic black phosphatic shale. Beds above this shale represent a single regressive withdrawal of the sea until subaerial emergence begins a new cycle. Deltaic construction during the initial, subaerial stages of transgression introduces local variability into the stratigraphic sequence but does not hamper the applicability of the cyclothem modeL Westgate, J. W., Department of Geography and Geology, Southwest Missouri State University, Springfield, MO 65804. HEM/SINUS JACKSON/US PALMER (GASTROPODA) AS AN AID IN THE PALEOENVIRONMENTAL RECONSTRUCTION OF A LATE EOCENE ESTUARINE DEPOSIT Recent quarrying operations in marl deposits of late Eocene age (upper Jackson Group) exposed near Madison, Arkansas have produced unexpected finds of mammalian remains. No prior records of terrestrial mammals in Paleogene deposits of the Mississippi Embayment exist, and few have been discovered in equivalent strata throughout the Atlantic and Gulf Costal Plain. Remains of oysters and mussels in life position indicate deposition occurred under estuarine conditions. Large numbers of unworn Hemisinus jacksonius shells are also present. Modern relatives of this snail live in fresh to slightly brackish waters. Apparently, deposition of the marls and terrestrial mammals occurred on the landward margin of the estuary. Conspicuous size and preference for low salinity habitats make H. jacksonius an excellent field indicator Of deposits having a high potential for yielding terrestrial mammalian remains. Weaverling, 1'. H., University of Missouri, Columbia, MO 6521 L FACIES TRANSITIONS IN UPPER CAMBRIAN STRATA OF THE MIDCONTINENT REGION-TECTONIC IMPLICATIONS. Regional subsurface data in Missouri, Illinois, Kentucky and Arkansas is being studied in an effort to document lateral and vertical facies changes in Upper Cambrian strata in this region. The distribution of these facies suggests the tectonic evolution of the North American midcontinent exerted major controls on Cambrian sedimentation. Incipient rifting during the late Precambrian or early Cambrian led to the development of a graben system in the midcontinent. This system included several major extensions such as the Rough Creek Graben recognized in Kentucky and another in western lllinois and Missouri. Basinal facies usually are thicker and shalier than associated shelfal sandstone and carbonate facies. The basinal facies display abrupt 166 Transactions, Missouri Academy of Science Vol. 16

pinch-out relationships and truncations on the shelf edges. Thus, the existence of a relatively deep-water graben basin in the midcontinent region must alter our previous concepts of early Paleozoic tectonics and the subsequent effects on paleogeography and sedimentation. Hou§e,, K., H. Hutchison, and D. Houseknecht, University of Missouri-Columbia. Columbia, MO 65211. PETROGRAPHY AND DIAGENESIS OF THE MUDDY SANDSTONE, POWDER RNER BASIN. WYOMING. Classic depositional models representing this Lower Cretaceous sandstone are not consistent with the petrography seen in seven subsurface cores. A conspicuous volcanic component is present, suggesting source areas besides the Black Hills. Environment of deposition exerts a major control on the detrital composition of the Muddy Sandstone, and therefore affects source area interpretation. The vertical and lateral variations in debita] composition determine the diagenetic mineral distribution. Both detrital and authigenic components affect the porosity-permeability of the sandstone; for example, primary porosity is preserved near chert grains and the facies containing the highest percentage of detrital monocrystalline quartz exhibit the greatest amount of silica cement. Detrital compositional facies and superimposed diagenetic facies can be correlated to characteristic electric log kicks. Based on knowledge of the environment of deposition and the detrital composition, porosity-permeability predictions can be made. lewis, J., J. M. Sharp, Jr., Geology Department, University of Missouri-Columbia, Columbia, MO 65211. HYDROGEOLOGY: DISSECTED TILL PlAINS PROVINCE. A characterization of the hydrogeology of shallow deposits in Northern Missouri provides a base for assessing the effects of strip mining. Units overlying minable coal include Pennsylvanian shales with interbedded sandstone and limestone and the overlying clayey till. The only aquifers are a few Pennsylvanian channel sands and smal1 sandy lenses and channels within the till. Using a computer program to model flow in the till and a water budget to determine net groundwater inflow and outflow, it was found that 1) overall hydraulic conductivity of the till is 10-4 -10·5 em/sec, 2) infiltrating water moves generally vertically through the till and then horizontally along the nearly impermeable bedrock boundary and is discharged as streamflow or as seepage along the till/bedrock contact, and 3) leachate from spoil piles can and does travel through the clayey till matrix to reach sandy aquifers at an approximate rate of 0.03 km/year. Stewart, C. A., and J. MI. Sharp, J,., Department of Geology, University of Missouri-Columbia, Columbia, MO 65211.FLOW OF WATER BETWEEN THE MISSOURI RIVER AND THE SURROUNDING ALLUVIAL AQUIFER. The water table in the Missouri River alluvial aquifer fluctuates in response to changes in river stage. Data obtained from piezometers in the alluvial aquifer near McBaine, Missouri, suggest that the rate and paths of flow in the aquifer are influenced by sedimentary inhomogeneities within the alluvium. Other factors affecting the interchange of river water and groundwater are sedimentologic properties of the river banks and channel bottom. A two-dimensional numerical model of the river-alluvial aquifer system provides only a rough approximation to the three-dimensional response of the aquifer to a· change in river stage.

Seifert, G., University of Missouri-Columbia, Geology 101, Columbia, MO 65211. HYDROGEOCHEMISTRY OF COAL MINE SPOIL PILES OF VARIOUS AGES IN MACON COUNTY, MO. Ground waters from spoil piles of various ages and undisturbed areas were studied to determine the effects of coal strip mining on ground water in north-central Mo. Comparison of chemical analyses of water from piezometers in spoils and wells in undisturbed aquifers, and investigation of rock-water equilibrium relationships, indicated differences in groundwater composition in the two areas. Ground water in undisturbed areas is classified as Ca-Mg, HC03 or Ca-Mg, HC03-S04 water. Spoil waters are of Ca-Mg, S04 composition. Ground water from the spoils has a higher TDS content (1700-2200 ppm vs. 400-800 ppm for undisturbed) due to increased concentrations of S04, Ca, and Mg. No significant differences in groundwater composition between different age spoil piles was found. Spoil piles are extremely inhomogeneous and a wide range of hydraulic conductivities were observed. Weathering of pyrite, precipitation of iron oxides, and carbonate equilibrium are thought to be controlling the composition of spoil waters.

Olde,, K., University of Missouri, Columbia, MO 65211. EFFECTS OF SPOIL MATERIAL ON GROUND WATER GEOCHEMISTRY. A mineralogically inhomogeneous mixture of refuse material or spoil results from strip mining. Ground water chemistry in the vicinity of strip mines in Macon and Randolph counties, Missouri has been affected by such coal mining activity. It is possible that geochemical reactions occurring in the spoil material are at least partially responsible for chemical changes in ground water. It may be possible to devise a method of predicting ground water geochemistry by documenting mineralogic changes with age and/or with depth of spoil and comparing such changes with obsetved water quality data. D'aney, D., Department of Geology, University of Missouri-Columbia. Columbia, MO 65211. GEOTECHNICAL PROPERTIES OF STRIP MINED AREAS IN NORTHCENTRAL MISSOURI. Spoil piles of three distinct ages, as well as unmined a'eas in southern Macon and northern Randolph Counties, Mo. were studied to find variations in geotechnical properties. Premined and postmined geotechnical properties remained unchanged except for permeability. Postmined permeability greatly increased. Atterberg Limits and soil color showed a break between oxidizing and reducing conditions within the tilL Average values of the liquid limit and plastic index for oxidized till were 36.8% and 20.3%, respectively. Those for the reduced till were 29.3% and 13. 7%. Hydraulic conductivities determined by consolidation theory and constant head permeameter were 10-8 to 10-9 em/sec. Slug tests gave results of 10-4 to 10-5 em/sec. Average weight percent of calcite and dolomite were 5.5% and 4. 7%, respectively. Kortenhof, M. H., and J. M. Sharp, Jr.. University of Missouri, 101 Geology, Columbia, MO 65211. NATURAL HYDRAULIC FRACTURING IN A SEDIMENTARY BASIN. The occurrence of fluid pressures greater than hydrostatic in compacting sedimentary basins is well documented. If the excess pressure is great enough and if other basin conditions are correct, hydraulic fracturing can occur as a means of fluid release. A Mohr failure diagram, using a two-part failure envelope, is used to predict when fracturing would occur and the orientation of the fractures. The coupling of this model with a numerical model for excess pressure build-up allows the study of the effect of the entire range of sedimentary basin parameters on fracturing. A variety of stress conditions are considered, resulting in both horizontal and vertical fractures. Other significant factors controlling fracturing include the sedimentation rate, the sand versus shale ratio of the sediments, and the tensile strength and hydraulic conductivity of the shale.

Pignolet, §., and R. D. Hagni, 208 Norwood Hall, University of Missouri-Rolla, Rolla. MO 65401. CHARACTERIZATION OF THE COBALT -NICKEL MINERALIZATION AT THE MADISON MINE. FREDERICKTOWN, MISSOURI. Renewed inte,est in the recovery of cobalt from southeast Missouri Pb~Zn-Cu ores and the probable reopening of the former Madison Pb-Cu mine for its Co content has 1982 Annual Meeting 167 prompted a detailed ore microscopic and chemical study of the Co-Ni minerals in the Madison ore body. Ore microscopy has shown that siegenite [(Co,Nil3S4] is the most abundant thiospinel and that nickelean carrolite [Co,Ni, Cul3S4] and polydymite (Ni3S4) are present. Other Ni minerals include bravoite ((Ni,Fe)S2l and millerite (NiS). Siegenite forms thick coatings on chalcopyrite, bravoite usually occurs as disseminated crystals, millerite is present locally as tiny prisms crystallized in small vugs and polydymite pseudomorphically replaces millerite. Paragenetic and microprobe data on the sulfide minerals at Madison are currently under study.

Staples, M. E., Jr., University of Missouri-Rolla, Rolla, MO 65401. INTERPRETATION OF THE STRUCTURAL HISTORY OF THE WEST LIME HILLS AREA, LUNA COUNTY, NEW MEXICO. The West Lime Hills area in south central New Mexico has been studied by field mapping, magnetic surveying, and geochemical sampling. The West Lime Hills consists entirely of an Early Cretaceous stratigraphic section, about 1535 feet thick. These Cretaceous rocks are anomalous to the area and contrast to the principal rocks of the area to the east, which are Tertiary volcanics and a quartz monzonite intrusive, and sedimentary rocks of late Paleozoic age. The structt1ral history of this complex area has been interpreted, and potential for mineralization of these structures has been evaluated.

!Fridley, M. S., University of Missouri-Columbia, Columbia, Missouri 65211. MIDDLE PENNSYLVANIAN FORAMINIFERA OF MISSOURI. The Middle Pennsylvanian (Desmoinesian) marine shales of Central Missouri yield a diverse fauna of non~Fusulinid foraminifera. Twenty species representing fourteen genera are described. Both calcareous and agglutinated forms are present. The Paleotextularids construct their tests with an outer coarse, agglutinated layer and an inner calcic ''fibrous'' layer which possibly represents the inner zone of pseudopores observed in modern T extularids.

McKay, R. H., and G. Fraunfelter, Union Oil Company, Oklahoma City, OK 73112 and Southern Illinois University, Carbondale, IL 62901. TRACE FOSSILS AND ENVIRONMENTS OF DEPOSITION OF THE AUX VASES FORMATION AND THE JOPPA MEMBER OF THE STE. GENEVIEVE FORMATION (MIDDLE MISSISSIPPIAN) IN SOUTHEASTERN MISSOURI AND SOUTHERN ILLINOIS. The trace fossils from the Aux Vases and Joppa were identified, described and figured. These trace fossils along with the megafossils, lithologies, and sedimentary structures of the containing beds were used to determine environments of deposition involved. The trace fossil fauna consists of the resting traces Sagittichus, Lockeia, and unidentifiable resting traces; dwelling structures Brookvalichnus, Conostichus, and unidentifiable vertical burrows; crawling traces Gordia, Gyrochorte, Protichnites, Psammichnites, Scolicia, and Torrowangea; feeding traces Chondrites, Phycodes, and an unidentifiable feeding probe; and a deposit feeding pattern Cosmorhaphe. Eight facies containing traces were recognized; alluvial channel and shallow marine barrier island, barrier bar, bay, lagoon, tidal channel, washover fan and destructional bar.

Gerontology Irvin, C. B .. L. Isbell, M. H. Typpo, Child and Family Development, University of Missouri-Columbia, Columbia, MO 65211. CAREGIVERS OF THE ELDERLY: ATTITUDES TOWARD AGING AND AUTHORITARIANISM. How caregivers view the elderly in general and how they view the elderly health care needs are of particular importance, since individuals with more unfavorable attitudes may be rather impervious to the real needs of their patients. This study examined the effect of background characteristics and the level of authoritarianism on the attitudes toward aging of caregivers in nursing homes and hospitals in Southeast Missouri. Findings and discussion will focus on the multidimensional nature of these attitudes and the implications that are relevant to the quality of care for the elderly in institutional settings. This research was supported by a grant from the University of Missouri Aging Back-Up Funds, Spring, 1980.

Norman, M., Lincoln University, Jefferson City, MO 65101. THE BLACK ELDERLY &SUPPORT SYSTEMS. The prevalentthinking is that Black elderly face a double jeopardy because they are old and Black. It is also thought that the Black elderly do not participate in the services provided for the elderly because they are either not available or the Black elderly do not know about the setvices. This paper is a literature review to test the theory that the Black elderly are not participating in the services for the elderly. The literature review indicated that although there is not as much participation as white elderly, there may be more participation on the part of the Black elderly than previously thought. There is also some indication that other minority elderly such as Asian-Americans and Hispanics, participate even less than the Black elderly.

Buckner, R. R., Missouri Veterans' Home, St. James, MO 65559. ASSESSMENT OF ELDERLY NURSING HOME RESIDENTS. A critical analysis of the purpose and technique of mental status evaluation of elderly persons entering a nursing care facility. The importance of understanding the complexity of late~ life personality development is stressed, and the probability of variance in reality bases is advanced. Parameters for defining a final stage of personality development are presented and a paradigm for assessing life status is proposed. The most effective evaluative procedure was found not to be a diagnostic technique, but rather a clinical interpretive assessment to facilitate a major change in lifestyle.

Herpetology lResetarits, W. J., Jr., and R. D. Aldridge, Biology Department, Saint Louis University, St. Louis, MO 63103. ECOLOGY AND REPRODUCTION OF CAVE ASSOCIATED RANA PALUSTRIS. Rona palustris are present from July-April at the study cave in Boone Co., Missouri. The cave supports an overwintering population of ca. 300~400. They are found throughout the cave in July-Oct., moving to the more thermally stable central portion of the cave wlth decreasing temperatures. Few remain in open areas, most spend the inactive period in thermally and physically sheltered "hibernacula" within the cave. Both sexes appear healthy, with fat stores present in fall and decreasing through the winter. Food last appears in stomachs in Nov. and reappears in Mar., indicating cessation of feeding in the winter months. Females contain a full complement of eggs in Oct. Oviducts showing size increase over the winter. Testis size changes little through the overwintering period. All R. pa{ustris leave the cave in late Mar.-early Apr. and breeding occurs outside of the cave. Frogs found in the cave in July and August are all males. Mortality in winter is greatest among small juveniles.

Minesky, J. J., and R. ID. Aldridge, Saint Louis University, St. Louis, MO 63103. THE MALE REPRODUCTIVE CYCLE OF THE QUEEN SNAKE (REGINA SEPTEMVITTATA) IN WESTERN PENNSYLVANIA. Queen snakes (Regina septemuittata) were collected from several localities in western Pennsylvania between May and August 1981. Histological examination of the testes was performed to 168 Transactions, Missouri Academy of Science Vol. 16

determine the spermatogenic cycle. Data will be presented on size at maturity and the seasonal changes in fat body mass, the sexual segment of the kidney and testis size and weight. The correlation between mating and the sexual segment of the kidney will be discussed.

M\o&on, C. P., D. A. Kanga§, and A. A. Schreiber, Northeast Missouri State University and Louisiana State University. THE STATUS OF THREE ENDANGERED AQUATIC VERTEBRATES IN MISSOURI. During 1981 we investigated the ecology, natural history and distribution ofthe Ulinois Mud Turtle (Kinostemonflavescens spooneri) and the Blanding's Turtle (Emydoidea blandingii) in Missouri. Goose Pond, a patch of remnant marsh in Clark County, is the only remaining natural habitat in Missouri that contains populations of these two turtles. Also at this locality we discovered a population of the Central Mudminnow (Umbra limi), which is a fish new to the fauna of Missouri. Population parameters, natural history, and critical habitat of the three species are discussed.

Aldridge, It D,, Saint Louis University. St. Louis, MO 63103. THE REPRODUCTIVE CYCLE OF THE CROWNED SNAKE TANTILLA CORONATA. Sexual maturity in females is attained at a snout-vent length (SVL) of 150 mm. Maximum length ~f females is 246 mm. Vitellogenesis begins in spring and ovulation occurs from early May to mid-July. Based on 10 individuals (SVL 156-220), the mean litter size is 1.8±0.92 (range 1-3). Utter size is not correlated to SVL (r~ 0.51, df~ 9, p>0.05). Sexual maturity in males is attained at SVL of ca 145 mm. Maximum SVL is 223 mm. Spermatogenesis begins in May and reaches a peak in August. The diameter of the sexual segment of the kidney is lowest in May and highest in August and September.

Smith, D. D., University of Kansas Medical Center, Kansas City, KS, R. Powell, Avila College, Kansas City, MO, H. 1... Gregm:y, Kansas City Zoo, Kansas City, MO. THE MISSOURI ACADEMY AS A CLEARING HOUSE OF HERPETOLOGICAL DATA, PART II. The authors, acting as a committee authorized by the Herpetology Section in 1980, will present a mechanism whereby the Missouri Academy of Science may function as a center for collection and distribution of statewide herpetological information. This information will consist of records and observations, insufficient in themselves for individual publication, but collectively providing significant insights into the natural history of Missouri amphibians and reptiles.

Linguistics

Gulstad, I>, E., University of Missouri-Columbia, Columbia, MO 65211. PHONOLOGICAL DIFFERENCES BETWEEN TWO FRENCH IDIOLECTS OF THE 16TH CENTURY: A DISTINCTIVE FEATURE EXPLANATION. In the second quarter of the 16th century, two Frenchmen, Louis Meigret and Jacques Peletier, published some of their respective works in phonemic orthographies of their own invention. Most of these works were also dedicated to the defense of their respective orthographies. Since Peletier had more thoroughly adopted the usage of the Parisian court, and Meigret was more conservative in retaining elements of his native Lyonnese, together they provide a wealth of data on early 16th-century pronunciation, a time when French phonology was in transition. This comparison of the tvJo idiolects in distinctive feature terms is an effort to describe more formally the phonological system of the day, both synchronically and diachronically.

Sample, W. A., Department of English, Central Missouri State University, Warrensburg, MO 64093. NOTES ON THE VARIETIES OF SWAHILI SPOKEN IN WESTERN KENYA. Although Swahili is the official language in most of East Africa, there seems to be, at first glance, two forms of Swahili spoken in Western Kenya by the Luyia people, speakers of a related Bantu language. Closer investigation, however, reveals that there are several varieties of Swahili ranging along a continuum of increasing complexity from a highly simplified form, Kihind.i, to a form that is fairly close to Standard-Swahili, Kiswahili. There seems to be a correlation beteen a particular speaker's position on this continuum and the amount of education and experience outside the Luyia speaking areas the speaker has had.

Cohen, G. 1..., University of Missouri-Rolla, Rolla, MO 65401. ORIGIN OF THE RUSSIAN NAME RASKOLNIKOV. Raskolnikou (in Dostoyevsky' s Crime and Punishment) is known to derive from raskol 'schism', but a satisfactory explanation has not yet been given as to why Raskolnikov's name contains this root. The answer seems clear to me, however: Raskolnikov is a schizophrenic.

Jei!eclt, 1J. B,, Southwest Missouri State University, Springfield, MO 65802. WHAT WE KNOW ABOUT "KNOW." The institutions of linguists analyzing presupposition and {activity have not performed well on ''know,'' for it has been regularly considered a factive verb in spite of the fact that it does not meet any test for £actives. More precise analysis of the differences in meaning betv.reen £actives and non-factives suggests that factive verbs are predicating something about the speaker's emotions or attitude toward a given, presupposed condition, whereas non-factives give information about a speaker's state of mind, perceptions, belief, etc. Thus, "know" rig~tly follows the pattern for non-factives.

Pulley, 1\'1, Southwest Missouri State University, Springfield, MO 65802. STYLISTIC SHIFTING IN THE MISSOURI OZARKS. The discussion will be based on research compiled from an investigation of certain linguistic features of twelve Missouri Ozark natives. The investigation was patterned from William Labov' s Sociolinguistic Patterns, with attention to "The Isolation of Contextual Styles." The research deals with four phonological variables which Wolfram and Christian studied in Appalachian Speech; the coalescence with Wolfram and Christian is logical since much of the Ozark settlement was from Appalachia. The results of the study showed that, in one degree or another, some type of stylistic shifting occurred with the informants interviewed, and the research also revealed an extension of a phonological feature that Wolfram and Christian did not observe. Methodology and interview techniques will also be discussed.

Keating, J. 111., University of Missouri-Columbia, Columbia, MO 65201. WOMEN AND LANGUAGE: A CURIOUS RELATIONSHIP. The changing social position of the American woman during the last tv.renty years has led many sociologists, psychologists, and linguists to speculate and hypothesize about ''how the American English female speaks.'' The research on this topic reveals a profile that is unclear and inconstant as a result of the various aspects of and attitudes towards female speech; such is the emphasis of this discussion. l\'lasl, A, University of Missouri, Columbia, MO 65211. SEX-TYPED LANGUAGE IN THE VILLAGE VOICE PERSONALS. This paper analyzes female and male "personal markers" in personal ads from the Village Voice, a New York metropolitan area "radical" magazine, for evidence of what Robin Lakoff calls "women's," "men's," and "neutral" language. By comparing the kinds of modifiers heterosexual and homosexual women and men use to describe themselves and each other, the study reveals some striking contrasts between women's and men's assumptions about their ''marketable" physical, emotional, intellectual, and financial assets, as reflected in their use and choice of specific nouns, adjectives, and metaphor. 1982 Annual Meeting 169

Psychology Johnson, M., Missouri Western State College, St. Joseph, MO 64507. THREE GENERATIONS OF MISSOURIANS: A STUDY OF SIMILARITIES AND DIFFERENCES IN VALUES. The present study assessed the nature of intergenerational differences in values. 111 Missouii Western State College undergraduates, 193 of their parents, and 148 of their grandparents filled out questionnaires containing 24 scales of attitudes and values. A multivariate analysis of variance with all 24 scales indicated significant generational (F = 4. 90); df=52; p<.OOOl) differences. Generational differences were greatest for: acceptance of authority, patriotism, secular morality and religious morality. Differences betvveen the generations were sequential with children and grandparents at the ends and parents in the middle. It appears that significant generational differences continue to exist and are not unique to the present parent -child generation but also exist for the parent and grandparent generation. Puckett, J. Ill., Southwest Missouri State University, Springfield. MO 65804. SHORT-TERM MEMORY AND AGING. Correlations were found beh.veen performances on tv.ro short-term memory tasks in young and elderly adults. One task was memory span. The other was Sternberg's scanning task. A prior survey of experiments had found an almost perfect correlatiori between task performance across material types: as memory span improved from nonsense syllables (e.g., CVCs) to digits, scan time per item shortened proportionately. The present study attempted to ascertain whether the correlation held up for individual subjects and to extend the investigation to aged adults. Digits, words, and CVCs were used in both tasks for each subject. Experiment 1 found that half of the individuals (all young) and the group as a whole had statistically significant correlations. The second experiment included old adults, replicating the first experiment for the young adults, but finding few significant correlations for senescent adults. In experiment 3, elderly adults were run for twice as many sessions, and in later sessions their data began to resemble those of young adults.

Hun11termark, J., Missouri W¢stern State College, St. Joseph, MO 64507. THE EFFECTS OF EVALUATIVE INSTRUCTIONS ON THE SHORT-TERM RETENTION OF COLLEGE STUDENTS AS A FUNCTION OF SEX OF EXPERIMENTER AND SUBJECT. The short term retention of college students as related to three variables was assessed. Subjects given evaluative instructions recalled more nonsense syllables than subjects given standard instructions. Also, males had higher retention scores than females due to the lowered performance of females when tested by a female experimenter. More syllables were recalled at a 5 sec. as compared to a 15 sec. retention intervaL No differences between males and females were found at the 5 sec. interval but males had higher retention scores at the 15 sec. interval. Overall, sex of experimenter had no effect at the 5 sec. interval but at the 15 sec. interval more syllables were recalled when the experimenter was male.

Wann, P. D., Missouri Western State College. and D. C. Mcintyre, Carleton University. DIFFERENTIAL EFFECTS OF HIPPOCAMPAL EPILEPTIC FOCI AND LESIONS ON AVOIDANCE LEARNING IN RATS. Bilateral epileptic foci induced by penicillin injection into the ventral hippocampus did not affect acquisition of Y maze avoidance learning by rats. while nonepileptic hippocampal lesions severely retarded learning of the avoidance response. Neither treatment interfered with the ability of the animals to learn a brightness discrimination to escape shock in the maze. The avoidance results indicate that hippocampal seizure activity does not produce a functional lesion which is behaviorally equivalent to hippocampal destruction. In this task, hippocampal lesions appear to interlere with an animal's cognitive mapping capability, while periodic epileptic discharge does not.

Johnson, M., J. Terry, and§. Radmacher, Missouri Western State College, St. Joseph. MO 64507. THE EFFECT OF AEROBIC ACTIVITY ON SELF-CONCEPT. The purpose of this study was to investigate the effect of aerobic conditioning on self-concept Fifty students from aerobic classes were compared with 63 students from general studies and bowling classes. The experimental group (aerobics) was then divided into 3 groups according to the changes in the students' cardiovascular fitness and compared to see how the physiological changes related to self-concept changes. Aerobic conditioning was provided by walking, jogging, or running during 32 sessions. Self-concept measures were derived through factor analysis of the pretest data of a series of items relating to body image and self-concept. Pre and post data were collected from all students, but only aerobics received physiological testing. The experimental group showed significant improvement in the major self-esteem factor; however, this improvement was not related to change in cardiovascular fitness.

Masters, M. and F. Haemmerlie, University of Missouri-Rolla. Rolla, MO 65401. EFFECTS OF A REGULAR EXERCISE PROGRAM ON GERlA TRIC RESIDENTS IN A NURSING HOME. The positive effects of a regular exercise program, with and without music, on the psychological well-being of elderly residents in a nursing home was assessed. Fifteen residents in a medium care nursing home (ages 59-89) were randomly selected and assigned to one of two experimental groups receiving the exercise program or to a no treatment control group which did not receive the program. One experimental group had music accompanying their sessions while the other did not. A variety of measures were assessed one week prior to and immediately after the exercise program on all three groups of residents. Significant differences were obtained between the two experimental groups and the control group on a measure of life satisfaction. However, the two experimental groups did not differ significantly from each other suggesting the exercise program had a positive effect on the elderly subjects but that the music did not have an additional facilitative effect

Davis, J. 0., Southwest Missouri State University, Springfield, MO 65802. ASSESSING COMPLIANCE WITH DIETARY AND SLEEP REGIMENS. There is increasing concern over the health habits of Americans and the increasing costs of health care that some say can only be ameliorated through positive health regimens. The willingness of a patient to follow a prescribed regimen is obviously important while the variables that affect compliance remain somewhat abstract. What for example is the effect of didactic instruction and rationales? Will explanations for health regimens result in higher compliance? This project investigated dietary and sleep regimen compliance as a function of instruction in the relationships of these practices and physical and mental well-being. The results indicate the difficulty of establishing compliance is in part a function of information and the complexity of the prescribed regimen.

Hestand, R. E., MBA, PhD, Breckenridge R-1 School, Breckenridge, MO 64625. COUNSELOR INTERVENTION IN SCHOOL DISCIPLINE: THREE STYLES. An investigation into the comparative value of counselor intervention in school discipline will be discussed, comparing Reality Therapy, Client Centered, and Rational Emotive Therapy. The results, while non-significant, lean toward a greater effect of Rational Emotive Therapy. The group make-up was small, and the period of follow-up was admittedly short, which denigrates the absolute value of the study, but indications of further study, using much larger groups than available to the author are indicated. Implications and possible further studies are discussed. 172 Transactions, Missouri Academy of Science Vol. 16

noted was a slight increase in right and left lateral flexion and a decrease in cervical extension. The findings suggest that a neurological mechanism may be involved in balancing the cervical musculature through spinal manipulation.

Mathes, 1.., Maryville College, St. Louis, MO 63141. A SURVEY OF PLANKTON AND BOTTOM VEGETATION IN A SMALL MANMADE POND IN WARREN COUNTY MISSOURI. Small ponds in the Midwest have received considerable study, however the plankton and aquatic vegetation of Warren Co. have not been surveyed despite their importance in several prominent studies of Missouri amphibians. Regular sampling was undertaken from May to August 1981. Although the pond is only 1/16 acre in extent, a wide range of plankton and bottom vegetation was observed; a full report is available from the biology department, Maryville College, 13550 Conway Road, St. Louis, Mo.

Ho.,lihan, T., Missouri Southern State College, Joplin, MO 64862. A SMALL MAMMAL SURVEY OF PRAIRIE MANAGEMENT AREA FOUR AT GEORGE WASHINGTON G:ARVER NATIONAL MONUMENT. A live trapping study of small mammals was conducted on p[airie management area four at George Washington Carver National Monument. Movements in linear distance, species composition and population estimates were determined over 192 trap nights. The various species found were Sigmodon hispidus, Peromyscus /eocopus, Peromyscus maniculatus and Cryptotis parva. The average linear distances traveled were: Sigmodon hispidus, 38 meters; and Peromyscus leocopus, 10 meters. When the Lincoln Index was applied the following relative population estimates were made for the 30 sq. m. trapping area: Sigmodon hispidus, 16.3 individuals; Peromyscus feocopus, 4 individuals; and Peromyscus maniculatus, 5 individuals.

Brush, 1\. W., and S. H. Mills, Central Missouri State University, Warrensburg, MO 64093. OXYGEN CONSUMPTION IN ACETONYLACETONE TREATED RATS. Rats treated ad libitum in their drinking water with the metabolite of various ketones, Acetonylacetone, have been tested to examine their effectiveness of temperature regulation by measuring oxygen consumption at 9, 14, 19, 24 and 29°C. In control rats there were no differences in oxygen consumption at 24 and zgoc (p>O.OS) although oxygen consumption was significantly elevated at l9°C {p0.05) by exposure to the experimental temperatures, central temperature regulation appears functional.

Whited, L. D., Norihwest Missouri State University, Maryville, MO 64468. THE NATURE OF THE EFFECT OF ANTIHISTAMINES ON EUGLENA'S ABILITY TO ABSORB VARIOUS WAVELENGTHS OF LIGHT. Antihistamines are common chemicals used by humans in relief of allergy and cold symptoms, however other effects are possible. In a previous experiment, it was shown that antihistamines affected the way Euglena respond to different wavelengths oflight.ln treated Euglena cultures, the Euglena preferred pink and yellow over clear, and green over red. In general the antihistamines affected Euglena's ability to absorb useful wavelengths of light. This experiment attempts to explain the nature of this effect. Twenty petri dishes (10 control dishes and 10 treated dishes) containing Euglena were covered with colored emulsions to regulate the wavelengths of light admitted to the Euglena. Those treated cultures behaving differently than the control cultures were subcultured in the absence of the antihistamines to see if the effect was reversible or irreversible. The outcome would determine whether the nature is enzymatic or genetic.

Chemistry & Geology

Spalding, T. G., and. M.D. Glascock, University of Missouri Research Reactor, R. W. Gale, and S. E. Manahan, University of Missouri-Columbia, Department of Chemistry, Columbia, MO 65211. A TRACE ELEMENT STUDY IN COAL AND COAL-DERIVED SOLIDS BY NEUTRON ACTIVATION TECHNIQUES. Various neutron activation analysis methodologies have been applied to determine as many elemental concentrations as possible in coal and coal-derived solids. The fate of various elements from a simulation of burning Western coals for the purpose of underground in-situ coal gasification is discussed.

Heubel, K. C., Department of Chemistry, University of Missouri, Columbia, MO 65211. LIQUID CHROMATOGRAPHIC SEPARATION OF TRANSITION METAL COMPLEXES WITH 1, 10-PHENANTHROLINE. This study concerns the high performance liquid chromatographic separation of 1, 10-phenanthroline complexes with iron(ll), nickel(ll), cobalt( II>, zinc ill) and cadmium(ll) using the Hamilton-PRP1 (polystyrene divinylbenzene bead) column, focusing on the effects of 1, 10-phenanthro\ine and lithium perchlorate concentrations in the mobile phase of acetonibile and water. Furthermore, the study demonstrates some quantitative aspects of this separation technique for the determination of metals of trace concentration levels.

Snow, D. D., Department of Geography and Geology, Southwest Missouri State University, Springfield, MO 65804. A CLAY MINERAL ANALYSIS OF SEDIMENTS FROM A LATE ECOCENE DEPOSIT IN THE EAST-CENTRAL ARKANSAS REGION OF THE MISSISSIPPI EMBAYMENT. Previous studies of late Eocene marls exposed in the area near Madison, Arkansas have reported the presence of glauconite in the strata. X-ray diffraction analyses of samples taken from the "Crow Creek Quarry" have not indicated the presence of glauconite. Mixed layer clays, montmorillonite, kaolinite, and illite constitute the major clay minerals of the sample. X-ray analyses of shark teeth from the site have indicated a minor alteration, causing discoloration.

Boeger, P., and V. F. Hoffman, Division of Science, Northeast Missouri State University, Kirksville, IviO 63501. BROMINATION OF DIMETHYL MALEATE AND DIMETHYL FUMARATE IN THE PRESENCE OF QUATERNARY AMMONIUM BROMIDES. In an organic solvent series of varying polarity (as measured by dielectric constant) bromine with quaternary ammonium bromide salts reacted with dimethyl maleate and dimethyl fumarate to give partial stereospecific, anti-addition. These results differ from those observed when bromide ions were introduced in bromine addition reactions with cis- and trans- stilbene, with cis- and trans- methyl cinnamate, with cis­ and trans- crotonic acid, and with tiglic and angelic acids. 1982 Annual Meeting 173

Callahan, C. L., and H. D. Anderson, Stephens College, J. S. MORRIS, Missouri University Research Reactor. THE DETERMINATION OF SELENIUM IN HUMAN MILK, MARKET MILK AND COMMERCIAL INFANT FORMULAE. Human Milk samples from 21 mid-Missouri women were analyzed via Instrumental Neutron Activation Analysis. Total and bound Se levels were determined and compared to Market Milk, and Commercial Infant formulae. The Se levels with respect to days since birth will be discussed. In addition, data from a fractional collection for one "feeding" in three volunteers will be given. The average Se level in whole milk is 0.020 fLg/ml. Green, M., D. Johnson, L. Godwin, L. Wood, and G. Baker, Drury College, Springfield, MO 65802. COMPARISON OF THE ACTIVITY COEFFICIENTS OF BIVALENT HALIDES IN WATER AS WELL AS STRONG HYDROHALIC ACIDS AT 25°C. Activity coefficients of several bivalent halide-hydrohalic acid-water mixtures (such as HBr + CaBrz + HzO, HCI + SrCiz + HzO, etc.) have been computed at a large number of total ionic strengths (in the range of 0.1 to 5.0 mol kg-1) at 25°C. The electromotive-force measurements of cells containing the hydrogen and silver-silver halide electrode were used to determine the variation of the logarithm of the activity coefficient of the acid and the bivalent salt with the change in the amount of salt contained in the solution. The results have been interpreted in terms of Harned's Rule and are compared with those of similar systems. Flynn, V. K., and J. S. Morris, University of Missouri-Columbia Department of Chemistry and the Missouri University Research Reactor, Columbia, MO 65211. THE DETERMINATION OF SELENIUM IN THE PITUITARY GLAND AND ITS HORMONES. Se levels were determined in the anterior and posterior lobes of bovine pituitary glands, five bovine pituitary hormones (FSH, GH, TSH, LH and prolactin) and bovine corpus luteum tissue from six animals. Se extractability and binding studies were conducted on the anterior and posterior lobes. Davidson, R.. and L. L. Lambing, Missouri Western State College, St Joseph, MO 64507. SYNTHESIS OF 1,1,1,­ TRIDEUTEROTETRADECANE. Two methods of synthesizing long chain deuterated alkanes have been investigated for the synthesis of C-14 trideutero compounds. The lithium dialkyl copper reagent as suggested by Corey and House was used, employing 1-bromododecane and trideuterated ethyl iodide. The second method utilized the production of the tosyl chlorides followed by reduction with lithium aluminum deuteride. Kanatzar, C. J., and H. D. Anderson, Stephens College, and J. S. Morris, Missouri University Research Reactor. THE DETERMINATION OF Na, Mg, Cl, K, Ca and Br VIA NEUTRON ACTNATION ANALYSIS IN HUMAN MlLK, MARKET MILK AND COMMERCIAL INFANT FORMULAE. This study involved the simultaneous determination of Na, Mg, CJ, K, Ca and Br in whole, two-percent and half-percent Market inilk, commercial Infant formulae and samples of human milk collected from five mid-Missouri women. The elemental levels of each will be compared and the human milk values will be given for samples collected starting immediately after birth and continuing for 170 days. The elemental levels in fractional collections of "one feeding" will be discussed for four volunteers.

Neumann, W. L., and S. Searles, University of Missouri-Columbia, Columbia, MO 65211. HYDROGEN-DEUTERIUM EXCHANGE REACTIONS OF ORTHO- AND PARA-NITROTOLUENES. The HID exchange of ortho andpara-nitrotoluenes does not proceed readily with DzO and base, and thus is unlike HID exchange with other active methylene compounds, such as aldehydes, ketones, and malonic esters. The present studies indicate that this difficulty is not due to low acidity, as the nitrotoluenes have, in fact. higher acidities than aldehydes, ketones, and malonic esters, but to a slow rate of ionization in ortho-nitrotoluene and a pronounced tendency for condensation side-reactions in para-nitrotoluene. An improved method employing phase transfer catalysis has been developed, which has thus far given about 50% exchange in both ortho- andpara-nitrotoluene, using benzene-DzO biphasic mixtures. There are unexpected differences in the reaction of the two compounds with regard to the development of colored intermediates and the effect of heat on the biphasic reaction. Albin, T., and R. G. Baughman, Northeast Missouri State University, Kirksville, MO 63501. STUDY OF NITROSOCARBONYL SYSTEMS USING MIND03. Structural analyses of simple nitrosocarbonyls using the quantum mechanical computer program M1ND03 will be presented. Distances and angles between atoms and the effects on the bond orders by different functional group substitutions are determined. Sigman, M., and C. Thompson, Southwest Missouri State University, and J. Leffler, Aorida State. INVESTIGATION OF A PROPOSED INTERMEDIATE TO THE DECOMPOSITION OF DIACYL PEROXIDES. Reaction of m-chloroperoxybenzoic acid with a-naphthoyl chloride gave n-naphthyl-m-chlorobenzoyl carbonate instead of the expected n-naphthoyl-m-chlorobenzoyl peroxide. The product can be explained in terms of a proposed intermediate in the decomposition of diacyl peroxides. This proposed interme­ diate has been Investigated from a theoretical standpoint by use of an Extended Huckel Molecular Orbital Calculation. Osborne, D., and J. O'Brien, Southeast Missouri State University, 901 S. National, Springfield, MO 65802. STUDIES OF THE INFLUENCE OF UCI CONCENTRATIONS ON THE VISCOSITY OF DMSO-WATER SYSTEMS. Viscosities and densities of a full range (0.02-2.5M) of LiCI concentrations in five different DMSO-Water mixtures were determined at 25, 35, and 45°C. The approximately 150 viscosity measurements were conducted with an Oswald viscometer. Flow time ranging from 87 seconds to 4,546 seconds (75.8 minutes) were recorded. The relative viscosities were successfully fitted to an extended Jones-Dole plot.

Physics & Mathematics Von Hatten, M. A., Parks College, Cahokia, IL 62206. TRANSPORT PROPERTIES OF IONS IN A NEUTRAL GAS. Measurement and theory of the transport properties of ions in a neutral gas are described. A computational process utilizing a "two-temperature" theory is used to obtain values which are then compared to experimental results. Kohring, G., University of Missouri-Rolla, Rolla, MO 65401. FURTHER INVESTIGATIONS OF ALTON'S NEGATNE SPUTTER ION SOURCE. (This work was performed at the Oak Ridge National Laboratory under the supervision of G. D. Alton.) Furiher 174 Transactions, Missouri Academy of Science Vol. 16 investigations have been conducted with Alton's modified, refocus Middleton-Adams geometry negative sputter ion source. The beam perveance was found to differ by a factor of 1/6 from previous theoretical values and the effective area of the emitter is shown to be less than 5% of the total emitter surface. Additionally, we have determined the Vex, To pairs necessary to produce space charge limited positive ion currents.+ ll\rooks, H. E., Physics Department, William Jewell College, Liberty, MO 64068. CUPROUS OXIDE/COPPER PHOTOVOLTAIC THIN FILMS. Cuprous oxide/copper thin films have been produced by baking and annealing copper sheets at temperatures above lOOOoC. The resultant films are photovoltaic. Experiments concerning the effect of baking time on the sensitivity of the films have been performed. The results as well as the method of production will be discussed. The theoretical basis for the phenomenon and possible applications in the area of solar cells will be mentioned.

Watldns, S. E., University of Missouri-Rolla, Rolla, MO 65401.FEASIBILITY OF A FIBER OPTIC MAGNETIC FIELD SENSOR USING INTENSITY MODULATION. Preliminary investigation indicates that it is possible to develop a fiber optic sensor for magnetic fields. The essential components of the sensor are a bifurcated fiber and a reed relay. The intensity of light reflected from the reed's surface to the fiber varies as flux causes the reed to bend. This intensity was found to be proportional to the magnetic field. With further study into reed characteristics and noise reduction, this type of probe could be reliable, versatile, and economicaL Buzzard, IK. E., am! M. Elick, Missouri Southern State College, Joplin, MO 64801. ILLINOIS OR BOARDWALK?? An interesting survey of the probabilities associated with the board game Monopoly, and the statistics derived from them. A flow chart of the computer program which generated the data will be included in the presentation, as well as an interpretation of the program itself. Emphasis will be on the probability of landing on certain properties and their respective rents when compared against the expenses of ownership. The various groups will be compared to determine which are the most economicaL

Graessle, D. E., University of Missouri-Columbia and A. C. Lind., McDonnell-Douglas Research Laboratories. NUCLEAR MAGNETIC RESONANCE STUDY OF POLYMERS. Pulsed proton techniques were employed to study relaxation times in polymers. Acetylene-terminated sulfone {ATS) presented interesting data with respect to the spin-lattice relaxation. The null time, which is proportional to the spin-lattice relaxation time, can be crudely determined using a 180° pulse and a subsequent delayed sampling pulse, the latter yielding zero induced voltage if the time of delay is equal to the null time. This and other data were taken at intervals as the ATS samples were being heat cured. The null time data were not expected to yield striking results, but surprisingly the data began to change markedly, well after the samples were thought to be completely cured. These findings strongly indicate a need for further, more detailed study of the curing process for ATS.

Liu, D. K., and B. DeFacio, University of Missouri-Columbia, Columbia, MO 65211. A SMALL-ANGLE NEUTRON SCATIERING STUDY OF NUCLEPORE FILTERS. Nuclepore filters are thin membranes used to filter extremely small particles. The small pores are produced by chemically etching collimated nuclear tracks from U-235 + neutron fission in a nuclear reactor. The Small Angle Neutron Scattering {SANS) facility at the Missouri University Research Reactor in Columbia was used to measure the scattering intensity patterns from several nuclepore filters. The SANS scattering pattern contains structural features ranging from 10 to 100 Angstroms and shows average features of the whole filter. An approximate scattering amplitude containing a background and a diffraction term is shown to yield an excellent least-squares fit to the data. The cylindrical diameter is measured as d = 384.5 A, to be compared to the nominal value d -=-300 A given by the manufacturer. The good quality of the fit also indicates that the individual pores have nearly the same diameter.

Wass, R., and J. O'Brien, Southwest Missouri State University, Springfield, MO 65802. VERIFICATION OF LIGAND SUBSTITUTION MECHANISMS BY EXTENDED HUCKEL MOLECULAR ORBITAL CALCULATION. Utility of the Extended Hucke! Molecular Orbital Calculation in distinguishing Ia and Id mechanisms has been investigated. Experimentally it 2 + has been shown that the solvent exchange reaction of Mn {H20)2 + occurs by an la mechanism while that of Ni(H20)2 + is ld. The capability of Extended Hucke! Molecular Orbital Calculations to discern this mechanistic changeover has been tested.

Schaffer, B., University of Missouri-Rolla, Rolla, MO 65401. ION IMPLANT DAMAGE PROFILE BY ELLIPSOMETRY. Ion implantation of silicon effects ice nucleation on the surface. The damage profile can be quantified by el\ipsometric measurement The result is compared to that of Hale/Sadowski. The result explains the high dose effect on ice nucleation.

Huff, T. C., University of Missouri-Rolla, Rolla, MO 65401. THE DESIGN AND CONSTRUCTION OF A WORKING TTL COMPUTER This is the story of a first-year undergraduate who began work on his computer before microprocessors became inexpensive. It demonstrates the reliability of large TTL systems and the potential for the layman to construct a complicated piece of equipment by learning how solely through books and the successes and failures in construction. Applications of the computer to simple games and elementary number theory will be discussed.

McCoy, B. M., University of Missouri-Rolla, Rolla, MO 65401. THERMAL DECOMPOSITION STUDIES OF IODINE DOPED POLYACETYLENE. By means of electrical conductivity measurements the thermal decomposition of iodine doped polyacetylene films, (CHiy)x, was examined at different temperatures from room temperatures to 125oC and different dopant concentrations (y- 0.14 toy -0. 0035). The relationship of the intrinsic conductivity to the specific temperature values was also investigated. The intrinsic value gave an energy gap of 1.49 eV compared to 1.48 eV determined experimentally by others. Results of the decomposition show it does not follow simple kinetics. The strong effects of the intrinsic value on the drop in conductivity at lower dopant concentration are also discussed.

Reago, D., University of Missouri-Rolla, Rolla, MO 65401. BET ANALYSIS OF THE WATER/SILICON INTERFACE. Equipment was constructed to provide calibrated partial pressure of water vapor for BET isotherms on silicon. Ellipsometric measurement of the film thickness allows analysis of the critical parameters c and d 0 • Preliminary results are in good correlation with the theoretical modeL Social Science Bruun-Oisen, K, And B. Meyer, Northeast Missouri State University. Kirksville, MO 63501. LOCUS OF CONTROL AND THE PLACEBO EFFECT ON THE INCIDENCE OF STATE DEPENDENT LEARNING. The manipulation of placebo was used to test the 1982 Annual Meeting 175 effect of locus of control on the occurrence of state dependent learning. Subjects, numbering 160, were either adminiStered a caffeine placebo or were part of the control group. Subjects were tested through the use of semantic word processing by a recognition test over a 24-hour period. The subjects' internal and external locus of control was also measured in each condition.

Barger, R. R.. and D. Shackelford, Lincoln University, Jefferson City, MO 65101. PERSONALllY AND CULTURE. This paper will explain individual personality and adjustment from cultural influences. The special focus will be on psychophysiological reactions, behavioral disturbances, and social adjustment problems. Also included will be subtopic areas of concern to specific cultures, such as initiation rites, parental relationships, and physical and mental health.

Bosch, S. J., and D. Shackelford, Lincoln University, Jefferson City, MO 65101. THE CULTURAL DISASSIMILATION OF GERMAN IMMIGRANTS IN CENTRAL MISSOURI 1830-1866. This paper uses primary letters, documents, and memoirs of nineteenth century German immigrants and secondary material to explore the basic questions of German emigration to Central Missouri. Special focus wUl be on the prime motivating factors of migration and emigration.

Bushur, K. A., C. H. Lipscomb, and N. A. Marlin, University of Missouri-Rolla, Rolla, MO 65401. CONTEXTUAL ASSOCIATIONS IN CONDITIONED FEAR RESPONSES IN RATS. Recently attention has been levied on the importance of the context as one element of a compound stimulus in classical conditioning studies. The present experiment investigated if interelemental associations are formed between the context and a tonal CS. In two different experiments measuring fear responses in rats, the fear of the tone was manipulated by changing the fear of the context It was hypothesized that increasing the fear of the context would result in an increased fear of the tone and that decreasing the fear of the context would decrease the fear of the tone. This hypothesis was substantiated by the results found in both of the experiments.

Raines, M., and D. Shackelford, Lincoln University, Jefferson City, MO 65101. AN ANALYSIS OF POLITICAL OPINIONS FROM CENTRAL MISSOURI COMMUNITY LEADERS. An analysis of political opinions and economic concerns of central Missouri elected officials and community leaders. Special focus will be on taxation, reassessment, block grants, and other issues of concern affecting state and local government, rural communities, and school districts.

Bums, F. R., and D. Deeds, Drury College, Springfield, MO 65802. EFFECTS OF TETRACYCLINE ON MEMORY FOR APPETITNE CONDITIONING. Hungry male rats were tested in a 12-choice maze for their ability to find a reward of food after being injected subcutaneously near the spine with tetracycline. Injections of tetracycline, a protein synthesis inhibitor, in a dose of 75 mg/kg body weight resulted in rats running the maze significantly slower and with more errors than rats injected with diluent The lack of learning in the tetracycline rats seemed to represent a lack of LTM formation due to lack of protein synthesis in the CNS.

Errata

The paper entitled "A Study of Sister Chromatid Exchanges in Human Lymphocytes Exposed to Various Drugs" cited on 254 o!Vol. 15 (1981) was authored by A P. Hammond, C. Kirk and H. J. Freeman.

1982 Index to Authors 177

Adams, E. .... 171 Driever, S ...... 165 Albin, T...... 173 Dunkle, L M...... 160 Aldridge, R D. ... 167, 168 Dunn, D. B ...... 95 Amash, H ...... 163 Duchrow, R. M...... 55 Anderson, H. D...... 172, 173 Dunsdon, R. L .. 170 Anderson, R. R...... 159 Durham, J. F. .... 161 Arnold, T...... 171 Elashkar, S. ..163 Aspinwall, N...... 162 Elder, W. H. .... 77 Austin, R. F...... 127 Elick, M...... 174 Babrakyai, N...... 158 Elkins, L...... 171 Baker, B...... 164 Elliott, D ...... 157 Baker, G...... 173 Everson, A. R...... 161, 162 Baker, T...... 163 Fairweather, J. R...... 131 Balcerzak, P...... 165 Fashing, E. M ..... 170 Barger, R. R...... 175 Feir, D...... 162 Baskett, T. S...... 5 Flora, S. J ...... 161 Baughman, R. G...... 173 Flynn, J. J ...... 171 Baumgartner, D. C. .. 37, 131 Flynn, V. K...... 173 Bell, M...... 171 Fraunfelter, G...... 165, 167 Bolger, P ...... 172 Freedman, H. J. .... 163, 164 Bosch, S. J...... 175 Fridley, M. S ...... 167 Bromeier, K P ...... 99 Fritzen, E. K ...... 43 Brooks, H. E...... 174 Fussell, L. K...... 170 Brown, H. A...... 111 Gan, 0 ...... 163 Brown, M. D...... 161 Gardner, C. A C.. ··················· ...... 158 Brown, 0. R...... 163 Garrett, H. E...... 161 Brush, A W...... 1 72 Gerdes, C. F...... 162 Bruun-Olsen, K ...... 174 Gietzen, G. .164 Buckner, R. R...... 167 Glascock, M. D. ... 172 Burns, F. R...... 175 Godwin, L. ... 173 Bushur, K A ...... 175 Gordon, A R...... 171 Buzzard, K. E...... 174 Gowin, T...... 157 Callahan, C. L ...... 1 72 Graessle, D. E...... 174 Cary, D. H...... 107 Green, M...... 173 Cary, P...... 164 Gulstad, D. E...... 168 Choromanski, J. F...... 43 Guyette, R...... 85 Chriswell, H...... 164 Haefner, R...... 161 Clawson, M. E ...... 5 Haemmerlie, F. M. .137, 169 Cohen, G. L...... 168 Hagni, R. D...... 166 Cook, J. C...... 17 Hampton, C. M...... 160 Corcoran, R. D...... 164 Harless, S. P ...... 171 Cordts, M...... 171 Haskins, M. F...... 162 Cornell, R. P ..... 159 Hestand, R. E ... .. 169 Cottrell, J. S...... 170 Heubel, K. C...... 172 Cowden, W...... 163 Hoffman, V. F...... 172 Craig, M. M...... 158 Hoggard, F...... 170 Crunkilton, R...... 161 Holloran, J. G...... 171 Crippin, J. B...... 164 Hough, R. D ...... 161 Dallali, A ...... 157 Houlihan, T...... 172 Davidson, R...... 173 Houseknecht, D...... 166 Davis, J. 0...... 169 Houser, K...... 166 Deeds, D...... 175 Hoyer, M. V...... 67 DeFacio, B...... 174 Hsu, S...... 163 Denney, L...... 171 Huff, T. C...... 174 Dowell, T...... 127 Huntermark, J...... 169 Draney, D...... 166 Hutchison, H...... 166 178 Transactions, Missouri Academy of Science Vol. 16

Imboden, R. T...... 159 Nelson, C. J ...... 158 Irvin, C. B...... 143, 167 Newman, W. L...... 173 lsaah, D. D...... 160 Nienhouse, J ...... 164 Isbell, L...... 143, 167 Norman, M...... 167 Jackson, J ...... 171 Norris, D. J ...... 77 Jafar, S. S ...... 158 O'Brien, J...... 173, 174 Jakubowski, T...... 158 Older, K...... 166 Janzan, F...... 157 Osborne, D...... 173 Jellech, V. B ...... 168 Oshlo, D...... 171 Johnson, C...... 164 Othic, S. T...... 37 Johnson, D ...... 173 Pape, B...... 164 Johnson, M...... 169, 170 Park, C. V...... 25 Kanatzar, C. J ...... 173 Parmerlee, J. S ...... 99 Kangas, D. A ...... 168 Perez, D...... 163 Keating, J. B ...... 168 Pershall, R. B...... 162 Kerns, H. A ...... 160 Peterson, A ...... 171 Key, D. P ...... 158 Phillips, R. L...... 165 Kirk-Yourtee, C...... 163 Pignolet, S ...... 166 Kiska, A L...... 170 Powell, R...... 99 Kofron, C. P ...... 168 Puckett, J. M...... 169 Kohring, G...... 173 Pulley, M...... 168 Kortenhof, M. H...... 166 Radmacher, S ...... 169 Kruse, K. J ...... 157 Raines, M...... 175 Kurtz, W. B ...... 37, 131, 161 Randall, D. D ...... 158 Lambling, L. L...... 173 Reago, D ...... 174 Laposha, N. A ...... 99 Reid, F. A ...... 67 Leslie, C. J ...... 159 Resch, G. E...... 159, 163 Leverich, W. J...... 160 Resetarits, W. J., Jr...... 162, 167 Lewis, J ...... 166 Reynertson, S ...... 163 Lind, A C...... 174 Rhoades, J. D ...... 107 Lipscomb, C. H...... 175 Rhodes, R. G ...... 158 Lire, D. K ...... 174 Roark, M...... 165 Loonan, D. V...... 157, 158 Roberts, M...... 159 Marlin, N. A ...... 175 Robertson, A C...... 165 Marx, R. J ...... 160 Rogers, M. N ...... 157 Masi, A ...... 168 Roy, P ...... 170 Masters, M...... 137, 169 Russell, S ...... 164 Mathes, L...... 1 72 Ryan, S. T...... 171 McClellan, C ...... 159 Sample, W. A ...... 168 McCormick, F. H ...... 162 Schaffer, B ...... 174 McCoy, B. M...... 174 Schmidt, D ...... 107 McGinnes, E. A ...... 85 Schmidt, T. L...... 161 McKay, R. H...... 167 Schreiber, A A ...... 168, 171 Melethil, S ...... 163 Searles, S...... 173 Meiser, R. F...... 162 Seifert, G ...... 166 Meyer, B ...... 174 Seither, R...... 163 Meyer, D ...... 162 Sells, G. D ...... 158 Miller, B ...... 99 Severn, D. J ...... 160 Miller, I. E...... 170 Shackelford, D...... 175 Millikan, D. F ...... 157, 158 Shaddy, J. H...... 162 Mills, S. H...... 172 Sharp, J. M., Jr...... 166 Minesky, J. J ...... 167 Sharp, R...... 170 Minter, K...... 170 Sheffield, L. G ...... 159 Mock, 0. B ...... 159 Sigman, M...... 173 Mokry, L. E...... 158 Simpson, C. W...... 163 Morris, J. S ...... 173 Sieper, D. A ...... 158 1982 Index to Authors 179

Smith, D. D...... 168 Typpo, M. H. .... 143, 167 Smith, D. E...... 164 Veras, R...... 162 Snow, D. D...... 172 Von Hatten, M. A...... 173 Spalding, T. G...... 172 Wallin, J. R...... 157, 158 Staples, M. E., Jr.. .. 167 Walsh, R. R. ········· ...... 160 Stefan, S. T. ... 15 7 Wann, P. D...... 169 Stitzer, P. W. . ..159 Wass, R...... 174 Stockam, J. . . .. 164 Watkins, S. E. .... 174 Stombaugh, T. A...... 17 Weaverling, P. H ...... 165 Sullivan, D. M...... 160 Weichinger, T...... 170 Summer, H. S...... 165 Westgate, J. W...... 49, 165 Taber, B. A .161 Whited, L. D. .... 172 Taber, C. A...... 161 Shitter, P ...... 164 Tansil, J. E...... 161 Whittle, P. R...... 164 Terry, J...... 169 Woldow, N ...... 170 Thomerson, J. E. .162 Wood, L.. .. .173 Thompson, C. C...... 171, 173 Yourtee, D ...... 163 Thompson, S. E...... 171 Zagaja, W. W...... 158 Topping, M. S...... 17, 160, 161 Zarrick, P ...... 162 Twining, L. C...... 160 Zarrough, K. M...... 158 Information For Authors Authors are expected to comply with the following guidelines in their preparation of abstracts of papers submitted for presentation at the annual meeting, and manuscripts submitted for publication in Transactions of the Missouri Academy of Science.

Abstracts: 1. Type the abstract as a single spaced paragraph with a 61/z X 3 inch space using a fresh ribbon. 2. Type the name of the author and the affiliated institution, using appropriate capital and lower case letters. If co-authors have different institutional affiliations, follow each author's name with their affiliation. Type the title in capital letters. Continue the paragraph with the main body of the text. Underline generic and specific names. Acknowledgment of support may be included as the last sentence of the text. See the section on abstracts of the annual meeting for examples. 3. Submit abstracts to the appropriate section chairperson by 1 March of the year of the meeting.

Manuscripts: 1. Editorial policy. The Transactions publishes several types of original contributions from the disciplines within the Academy: research papers, research notes, reviews and annotated bibliographies. Manuscripts must be authored or sponsored by a member of the Academy. An associate editor in the appropriate subject area is responsible for initial review and acceptance and rejection of the manuscript on the basis of originality and scientific merit. The Editor has final authority for acceptance or rejection. Manuscripts should be submitted to the Editor, John R. Jones, Missouri Academy of Science, 112 Stephens Hall, University of Missouri, Columbia, MO 65211. 2. Manuscript preparation. Type all material double spaced, on one side of standard sized bond paper with a minimum margin of 2.5 em. Submit 3 copies of the manuscript and illustrations for review purposes. Retain the original typescript and illustrations in your files. Manuscripts accepted for publication will, if appropriate, be returned to the author for revision and editing. At that time an original copy of the text and glossy photographic prints of the illustrations will be requested. Assemble the parts of each copy in this order: title page, abstract, key words, text, acknowledgments, literature cited, tables, figure legends, figures. Number all pages, Authors should refer to the style manual appropriate to the discipline for details on style, format, and citation of references. If doubt exists, consult the appropriate associate editor (see inside front cover). Abbreviations should agree with those used in the appropriate abstracting service for the discipline. Use the binominal Latin name of an organism when first mentioned. Subsequently the genus or common name may be used. Names of taxa should be underlined. Each paper must incluqe an informative abstract which records succinctly the essential findings, followed by a short list of key words for abstracting purposes. Each table must be typed on a separate page and be suitable for photoprinting. Number tables consecutively and provide a short title at the top of each table. All illustrations should be drawn and lettered using India ink and a lettering set or professional press-on materials. Hand-written or typewritten lettering or symbols are normally not acceptable. 3. Galley Proof and Charges to Authors. Authors will be charged $25.00 per printed page. Galley proofs will be sent to the senior author for final corrections. Authors will be liable for actual charges for alterations of galley proofs due to authors' negligence.