USF Libraries
USF Digital Collections

Ecological studies and management of Missouri bats, with emphasis on cave-dwelling species

MISSING IMAGE

Material Information

Title:
Ecological studies and management of Missouri bats, with emphasis on cave-dwelling species
Series Title:
Terrestrial series ;
Physical Description:
1 online resource (53 p.) : ill. ;
Language:
English
Creator:
LaVal, Richard K
LaVal, Margaret L
Publisher:
Missouri Dept. of Conservation
Place of Publication:
Jefferson City, Mo
Publication Date:

Subjects

Subjects / Keywords:
Bats -- Missouri   ( lcsh )
Wildlife management -- Missouri   ( lcsh )
Cave animals -- Missouri   ( lcsh )
Mammals -- Missouri   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Bibliography:
Includes bibliographical references (p. 51-52).
System Details:
The content of this file may cause problems when attempting to access it through some versions of the Internet Explorer or Firefox browsers. Recommended browsers for accessing this file are Chrome, Opera, and Safari.
Statement of Responsibility:
by Richard K. LaVal, Margaret L. LaVal.
General Note:
Title from e-book t.p. (viewed Aug. 29, 2011).

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002494660
oclc - 748271931
usfldc doi - S62-00012
usfldc handle - s62.12
System ID:
SFS0036128:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ia 4500
controlfield tag 001 002494660
005 20110829120346.0
006 m o d
007 cr cn|||||||||
008 110829s1980 moua ob s000 0 eng
datafield ind1 8 ind2 024
subfield code a S62-00012
035
(OCoLC)748271931
040
FHM
c FHM
043
n-us-mo
049
FHMM
090
QL737.C5
1 100
LaVal, Richard K.
0 245
Ecological studies and management of Missouri bats, with emphasis on cave-dwelling species
h [electronic resource] /
by Richard K. LaVal, Margaret L. LaVal.
260
Jefferson City, Mo. :
b Missouri Dept. of Conservation,
1980.
300
1 online resource (53 p.) :
ill.
490
Terrestrial series ;
v #8
538
The content of this file may cause problems when attempting to access it through some versions of the Internet Explorer or Firefox browsers. Recommended browsers for accessing this file are Chrome, Opera, and Safari.
500
Title from e-book t.p. (viewed Aug. 29, 2011).
504
Includes bibliographical references (p. 51-52).
505
Abstract -- Introduction -- Study area -- Methods and materials -- Status and distribution -- Recaptures of banded bats -- Light-tagging -- Results of summer mist-netting -- Dietary preferences -- Myotis sodalis summer habitat -- Year-round activity at a Missouri cave -- Seasonal changes in weight -- Reproduction and growth -- The "squared-ear" anomaly -- Strategy for management of cave bats in Missouri -- Unpublished data -- Literature cited.
650
Bats
z Missouri.
Wildlife management
Missouri.
Cave animals
Missouri.
Mammals
Missouri.
700
LaVal, Margaret L.
773
t Natural Sciences Collection
776
i Print version:
LaVal, Richard K.
Ecological studies and management of Missouri bats, with emphasis on cave-dwelling species.
d Jefferson City, Mo. : Missouri Dept. of Conservation, 1980
w (DLC) 80623996
(OCoLC)6618904
830
Terrestrial series ;
#8.
4 856
u http://digital.lib.usf.edu/?s62.12



PAGE 2

ABSTRACT During a 4th-year period, ecological studies of bats were carried out over much of the state of Missouri. An important goal of these studies was to learn enough about the species' biology to insure that intelligent management programs could be formulated and carried out. Status: The population of the endangered Indiana bat is relatively stable in Missouri, although declines have been documented at certain hibernacula. The endangered gray bat has declined to one quarter of its former numbers, and its population continues to diminish. The little brown bat and the big broWn bat are probably both declining due to extermination efforts in the buildings where they raise their young. Keen's bat is more abundant than previously believed, and pipistrelles seem to be holding their own. Recaptures and movements: Loyalty to hibernacula was demonstrated in varying degrees for all six cave species. In ad dition, female gray bats were loyal to their maternity caves, and all gray bats were loyal to the caves of their colony area. Survivorship for Indiana bats was comparable to that reported for this species in Indiana. Movements from summering areas (as far away as Iowa and Illinois) to hibernacula were docu mented for Indiana, gray and little brown bats. Light-tagging and food habits: Male Indiana bats foraged in forests in southern Missouri and in riparian situations in northern Missouri. Gray bats foraged over streams and reser voirs, but may forage elsewhere during bright moon. Indiana bats and Keen's bats seem to be moth specialists, whereas gray bats typically selected insects of the aquatic orders. Little brown bats ate mostly Trichoptera, as did pipistrelles. Big brown bats were beetle strategists. Year-round trapping: At one cave, trapping was carried out monthly (except winter) during the course of the study. Although seasonal trends were evident in all six species, actual catch rates varied substantially from one year to the next. Seasonal changes in weight: In general, weight changes were minor during the spring and summer, except for preg nant females. All species gained weight rapidly in the fall for a few weeks prior to entry into hibernation. Total weight loss during hibernation depended on clustering behavior, roost temperature, relative size of the species and total time in hiber nation. Weight gain among juveniles was documented for gray bats and pipistrelles. Reproduction: Annual reproductive events were listed for all species, and the occurrence of the "squared-ear" anomaly was observed in four of the species. Management: The need for and means of protecting both summer habitat (caves for gray bats; streams and forests for all species) and winter habitat (caves) are discussed in some detail. Missouri has taken a leading role in the acquisition and protection of bat caves with gates, fences and signs. In the long run, it is probable that only an enlightened public can insure the recovery and stability of Missouri's cave bat populations. crerrestrial 8{o.8

PAGE 3

ECOLOGICAL STUDIES AND MANAGEMENT OF MISSOURI BATS, with emphasis on cave-dwelling species By Richard K. LaVal Margaret L. LaVal Missouri Department of Conservation photos Application of a chemiluminescent sphere to a Keen's bat (Myotis k eenii) at Twenty-three Degree Cave, Crawford County. The sphere is applied ventrally for ground obs ervation Terrestrial Series #8 MISSOURI DEPARTMENT OF CONSERVATION Copyright 1980 by the Conservation Commission of the State of Missouri

PAGE 4

Inside front cover 1 1 1 4 6 16 19 20 25 25 29 32 46 46 50 51 52 CONTENTS Abstract Int roduction Study area Methods and materials Status and distribution Recaptures of banded bats Light-tagging Results of summer mist-netting Dietary preferences Myotis soda/is summer habitat Year round activity at a Missouri cave Seasonal changes in weight Reproduction and growth The "squared-ear" anomaly Strategy for management of cave bats in Missouri Unpublished data Literatu re ci ted Acknowledgments 11

PAGE 5

INTRODUCTION There are more than 4,000 caves in Missouri (Vineyard, 1979), and not sur prisingly they serve as refugia for large of cave bats. Myotis lucijugus, brown bat), Myotis sodalis, (In dIana bat), Myotis grisescens, (gray bat), Myotis keenii, (Keen's bat), Pipistrellus subflavus (eastern pipistrelle) and Eptesicus fuscus (big brown bat) spend all or a portion of the year roosting in caves Three of these species, M. lucifugus, M. sodalis and M. grisescens, have been studied extensively in Missouri by Myers (1964). Recently Elder and Gunier (1978) reported on movements of M. grisescens in southwest Missouri and adjoining states. LaVal et al. (1977a) reported on the status of bats in the Meramec River area of eastern Missouri, and LaVal et al. (1977b) detailed the foraging behavior of M. grisescens and M. sodalis in same area. Clawson et al. (1980) studIe? the clustering and hibernating behavlOr of M. sodalis in eastern Missouri. Caire, et al. (1979) reported on certain aspects of the ecology of M. keenii in eastern Missouri. Earlier, Schwartz and Schwartz (1959) had il and discussed briefly the blOlogy and status of all species of Missouri bats, and Guthrie (1933) had carried out limited studies of the movements of some cave bats in central Miss?uri No other major ecological studIes of cave bats have been carried out in Missouri, even though, according to Humphrey (1978), 62 percent of the endangered M. sodalis and perhaps 20 percent of the endangered M. grisescens (our data) hibernate here. In other portions of their range, these cave bats have been studied to varying degrees. Earlier studies were summarized by Barbour and Davis (1969). Subsequently Humphrey and Cope (1976) have reported a major study of M lucifugus in Indiana and north central Kentucky. Tuttle (1975, 1976a, 1976b 1979a) and Tuttle and Stevenson (1977) have covered many aspects of the ecology, status and management of M grisescens in Tennessee and Alabama. Saugey (1978) discussed a detailed study of reproduction in Arkansas gray bats. Humphrey and Cope (1977), Humphrey et al. (1977a), Cope and Humphrey (1977) and Humphrey (1978) have discussed a number of aspects of the ecology, status and management of M sodalis, based primarily on research in Indiana. Kunz (1974) added significantly to the knowledge of E. fuscus, based on studies in Kansas, and Schowalter and Gunson (1979) studied the species in Alberta Numerous other studies of a more limited scope have appeared, deal ing mainly with M lucifugus and E. fuscus, both widespread in North America and commonly used as laboratory subjects. With the addition of the Indiana bat (M sodalis) to the federal endangered species list in 1973, and the gray bat (M flrisescens) 1976, the need for manag mg the habItat of these two species became apparent. Important papers on their status and management have already appeared (Humphrey, 1978; Tuttle, 1979a). The first version of the Indiana Bat Recovery Plan has been published (Engel et aI., 1976), and the Gray Bat Recovery Plan will soon be available Studies important to the management of these species were begun in eastern Missouri in 1975 by the University of Missouri (LaVal et aI., 1977a; Clawson et aI., 1980) and continued statewide from March 1977 to March 1980 by the Missouri Department of Conservation. Objectives were: (1) to learn as much as possible about the status and ecology of these bats in Missouri, in order to provide a sound basis for management decisions; and (2) to proceed with the acquisition and management of habitat considered important, in order to permit at least a preliminary evaluation of the manage ment techniques chosen. In the process, valuable supplementary data were obtained on the four other common species of bats that share caves in Missouri with the two endangered species (see LaVal et aI., 1977a, 1977b; Caire et aI., 1979; and Humphrey et aI., 1977b) In addi tion, some interesting data were obtained on rare cave bats and non-cave bats, a portion of which has been published (LaVal and LaVal, 1979). In general there are no nomen clatural or taxonomic problems relating to the species of bats occurring in Missouri, and names used follow Bar bour and Davis (1969). However, van Zyll de Jong (1979) suggested that the allopatric subspecies of Myotis keenii, M. k. keenii on the West Coast and M. k septentrionalis east of the Rocky Mountains were distinct species. In that case, the species that occurs in Missouri would be known as Myotis septentrionalis. Un fortunately, van Zyll de Jong based his analysis solely on Canadian specimens. For this and other reasons, it is not clear whether taxonomists will find van Zyll de Jong's conclusion acceptable. the name currently in use, M. keen'L'L, wIll be retained here. STUDY AREA The initial phase of the study (1975-76) was carried out in the 1 Meramec River area of eastern Missouri. This region was described by LaVal et aI., (1977a, 1977b) However, the Meramec Park Dam and Lake which would have seriously impacted local bat population, has since been halted by Congress, and bills for deauthorization are pending The final phase of our study (1977-80) occupied the entire state, although some areas received minimal study effort due to time limitations. Missouri north of the Missouri River, when first occupied by Euro peans, was a mosaic of tallgrass prairie and forest. Today it is devoted almost entirely to agriculture with little natural vegetation remaining Bats and other organisms requiring forest and edge habitat are restricted primarily to riparian strips, which still exist along many streams. The only caves are in an area near the Missouri River and another area near the Mississippi River. Habitat similar to that of northern Missouri extends southward along the Kansas-Missouri state line to within about 100 km of the Arkansas state line. This area lacks caves, as does the flat Mississippi River flood plain of southeast Missouri The remainder of southern Missouri is comprised of the Uplift, a region of rolling hills, heaVlly forested in areas of high relief, but often cleared for pasture where the gradient is less steep. The Ozarks are dissected by numerous forested stream valleys, whose flood plains are often cleared Streams are clear and unpolluted. Most of the summer caves used by M grisescens are here as are all the hibernacula used by M grisescens M. sodalis and M lucifugus from Missouri and portions of adjoining states. METHODS AND MATERIALS In this study, approximately 30 000 bats were banded with size XCL plastic bird rings obtained from A.C. Hughes London (Fig. 1) Of these, about 12,000 were placed on M. sodalis and 14,000 on M. grisescens. In addition, approximately 1,500 M. lucijugus, 1,000 M keenii, 1,100 Pipistrellus and 200 Eptesicus were banded Males were banded on the right wing, females on the left. Each species was identified with one or more colors restricted to that species. Nearly half the bats were tagged on the opposite wing with a two-color striped plastic band so that bats observed in hibernacUla could be identified as to site of banding, sex and species without handling. Bats were captured for band ing in a bat trap (Tuttle trap) as they flew in or out cave entrances during the spring, summer and autumn (Fig. 2). A

PAGE 6

relatively small number was banded after being captured in mist nets. In order to locate hibernacula and gray bat summer caves, a diligent search of information on Missouri caves was required Much of this came from the Missouri Speleological Survey, which produces a computer print-out listing all known Missouri caves and their localities, and maintains informa tion on each cave. In addition, many individual cavers, landowners, biologists and others were questioned about cave localities and bat use of caves. Only caves with known or suspected colonies were visited. Undoubtedly, some bat caves were overlooked. Several reported bat caves could not be located, and a few cave owners refused us per mission to enter. A total of 103 caves was examined during winter for hiber nating bats. About 188 caves were visited at other times of the year to search for gray bats or signs thereof. Since certain caves were checked both summer and winter, the total number of separate caves checked i s somewhat less than 291. Caves where trapping and censuses were carried out are shown in Figs. 16, 17, 18, 27 and 28. At Great Scott Cave, bat activity was sampled by trapping at monthly intervals (except winter) throughout the study. At other caves, trapping was more sporadic. The trap was erected at caves on 168 nights Netting was restricted to summer. During 1976, single nets were erected over streams in the Meramec area (see LaVal and LaVal, 1979). In the summers of 1977 and 1978, three mist nets stacked ver tically were erected over streams in northern Missouri, using a portable, pulley-operated device for raising and lowering nets. The high net apparatus also was used several times at sites south of the Missouri River Whether bats were trapped or netted, we re corded the following information about them: sex, reproductive condition and age (if known). Weights were also recorded, but not for all individuals. Young-of-the-year gray bats could be identified until the end of September by the reddish-yellow tint of the venter (gray in adults) Young of other species could no longer be identified after the phalangeal epiphyses ossified (late August in most species, late September in P. sUbflavus). Few bats were banded after 1976, except at gray bat summer caves in central Missouri and at an In diana bat hibernaculum (Pilot Knob Mine) in eastern Missouri. During the winter of 1975-76, In diana bat hibernacula in the Meramec area were censused monthly (see Clawson et al., 1980; Fig. 3). In following winters, a sing l e census was carried out at these and the other Missouri hibernacula, including new ones as they were discovered. Each of the three ma jor caves used as hibernacula by gray bats were censused at least once during the study (Fig. 4). Although bats in some small clusters were counted individually, most M. sodalis were estimated by multiplying the area of bats by the number of bats per unit area (3001ft.2 in this study; Fig. 5). Numbers of M grisescens were obtained by multiplying the area by 1828/m 2 (Tuttle, 1975). However, due to variations in packing observed in this (Fig 6), the result was modified as determined by notes made at eac h cave. Temperatures and relative humidities were recorded as described by Clawson et al. (1980). Throughout the study, as time per mitted, reputed gray bat caves were TABLE l .--Status of summer col onies of grisescens in Missouri. Total Number Number Smallest L a rgest Of Of X Colonies Bats Materni ty -Maximum Past l 44,248 2,000 250,000 41 1,814,200 te rn i ty -Minimum Past 2 30,431 2,000 250,000 41 1,247,700 Current 16,361 2,000 50,000 21 343,600 All other summer colonies Past 19,724 500 85,000 29 572 ,000 All other summer colonies n imum Past 11 ,583 500 45,00 0 30 347,500 All other summer colonies -Current 6,084 50 40,000 38 231,200 lEs timate based on area of all stains or old guano piles in cave. 2Estimate based on largest single stain or old guano pile in cave (Tuttle, 1 979). Note that bats frequently use more than one roost at a time, but rarely use all roosts simultaneous l y. 2

PAGE 7

visited in order to determine their status. However, a concerted effort was made in the summer of 1978 to visit all known gray bat caves in the state. The most important data recorded for each cave were: (1) current population, based on actual count or area of fresh guano; (2) past population, based on areas of ceiling stains (Fig. 7) and old guano piles; (3) col ony type (maternity, bache l or, transient, hibernaculum); and (4) apparent threat (if any) to continued existence of active colonies. Techniques followed Tuttle (1979a). A total of 48 active and 42 abandoned gray bat caves were examined. Among the active caves were many that we suspect were formerly maternity roosts, but are now only transient sites. Foraging behavior was observed by releasing bats tagged with chemilumi nescent spheres (LaVal et al., 1977b; frontispiece) In addition to the 400 light-tagged bats released in 1976, 501 were released in 1979. Of these, 179 were ventrally tagged for ground view ing, and 322 were dorsally tagged for helicopter viewing. Four hundred sixty light tags were affixed to M g risescens and 28 to M. sodalis. The 1976 observa-TABLE 2.--Status of Miotis Oldest Recorded Cave Census 1 Bat} 2,000 ( 1 959) Shannon Co. Chimney 175,000 ( 1966) Coffin 2 150,000 ( 1960) :>1arvel,) 130,000 (1960 ) TOTALS 457,000 Maximum population 19768 0 tions and some of the 1979 observations were made with no moon or one-quarter to one half moon. On three nights in 1979, observations were made at full moon to determine the effect of moon phase on behavior. The 1979 efforts were concentrated on five caves where insects and fecal samples had been or were being collected. Three of the five caves were in forested areas away from streams suitable for over-water foraging. At the time of our study, Beck Cave was ap proximately 1 km from the Pomme de Terre River. It is now 0.5 km or less from the edge of Truman Reservoir Holton Cave is about 3 km from an ap parently suitable stream. A man-made pond of about 0.5 ha is fed by the cave spring, but is large l y ignored by the emerging bats. Mauss Cave is 2 km from the nearest suitable stream. The other caves where bats were light tagged are adjacent to streams except for Moles, which is 1.5 km from the nearest body of water. Methods were essentially those outlined by LaVal et al. (1977b) All bats were tagged with glass spheres of 9 mm outside diameter attached with Davol #262 surgical appliance adhesive. During the summers of 1978 and 1979, data were collected on the food habits of M. grisescens, and to a lesser extent, on the other five cave species. These data were obtained through bimonthly collecting of insects and fecal samples at caves and foraging areas used by cave-roosting bats The caves included five gray bat maternity roosts and one bachelor roost used by Indiana and Keen's bats. Insects were collected in ultraviolet light traps suspended in foraging areas, and guano was collected from individual full-stomach bats. In sects and insect fragments were sorted by order. Techniques and limitations of such analyses were discussed by Belwood and Fenton (1976), Anthony and Kunz (1977), and Belwood (1979). Guano samples, insect samples and dead bats (when found) were sent to Patuxent Wildlife Research Laboratory for pesticide analysis [see Clark et al. (1978) for techniques and report on analysis of dead bats]. The bulk of the data from insect trapping and fecal analysis is still being compiled and will be presented elsewhere. grisescens hibernating populations in Missouri. Recent l Census 1975-76 1976-77 1977-7 8 1978-79 1979-80 25,000 (1972 ) 9,096 11 ,556 11 ,000 71,000 ( 1968) 53,905 120,000 ( 1 969) 250,000 4 8 ,400 (1 967) 3,400 264,400 318 861 1 Censused by R. Myers (pers. comm.) 2 Fenced by Missouri Department of Conservation 3Commercia1 3

PAGE 8

STATUS AND DISTRIBUTION Barbour and Davis (1969) indicated that the six most abundant species of Missouri cave bats are distributed throughout the state, except that M grisescens is missing from much of the northern half of Missouri. As shown by Figs. 8, 9, 10 and 11, those authors were largely correct. In general, Missouri cave bats hibernate south of the Missouri River in caves in the Ozarks The MississippI River flood plain of southeast Missouri and the prairie region of far western Missouri contain no caves and thus no hibernating cave bats. In the summer cave bats may be found throughout the state, except gray bats, which are restricted to major cave areas. No gray bats were encountered away from cave regions In summer, non-cave bats were netted at the same sites as cave bats throughout much of the state. Summer trapping and netting pro vide little clue to the population status of Missouri bats, except that relative numbers c aptured per unit netting ef fort may be compared Biases inherent in mist-netting and trapping have been discussed by many authors, including LaVal (1970) and LaVal and Fitch (1977) Using a pulley-operated, three tier, high net, numbers of bats per net hour were 2.90 in the heavily forested Ozarks, but only 1.15 in the relatively flat agricultural regions of northern Missouri. Obviously, fewer bats in north ern Missouri were flying where the nets were set, compared with southern Missouri, and we suspect that summer population density is lower in the north ern p o rtion of the state. Especially significant is the fact that in northern Missouri the riparian habitat we sampled is the only kind of extensive habitat remaining that we believe is suitable for bat use; small wooded areas exist but are patchy in distribution In the Ozarks, riparian habitat constitutes a small part of the total habitat available to and used by bats. h4yoUs grisescens Because gray bats occupy caves throughout the year, meaningful popula tion data for this species were obtained in both summer and winter. Table 1 shows the number of gray bats currently thought to occupy maternity colonies in Missouri as well as two different estimates of past numbers. Average col ony size is 16,000 with a range of 2,000 to 50,000. Estimated maximum past col ony size varied from 2,000 to 250,000 the mean being 44, 000 The actual population siz e w o uld be at least 50 per cent higher than the totals shown here Figur e 1. Application of a numbered plastic band to the for e arm of an adult male Myotis gris e c ens a t Holton Cave, Boon e County to account for males (assuming one-half the maternity population is adult females, the other half juveniles). The resulting total population figures are 515,000 current, 2,720,000 maximum past, and 1,870,000 minimum past. These figures suggest that the popula tion of M gris e scens has plummeted (72 percent to 81 percent decline), but the time frame over which this has occurred is not clear. Because virtually every abandoned cave was occupied at some time during the life of persons whom we interviewed, we suggest that most of the decline has occurred in the last 50 years The fact that of 27 known maternity caves which housed 238, 000 bats in the early 1960s (Myers, 1964) 16 have been abandoned and the remainder contain only 46, 500 bats total suggests that substantial losses have occurred within the past 20 years. Furthermore, the reduction in average colony size to at least half the former value may be ex tremely detrimental to the population as was the case in Tennessee and Alabama (Tuttle, 1975 1979a). In spite of this grim picture, certain caves that have been protected or otherwise left undisturbed have suffered little popula tion loss In a few cases, the populations have actually grown Population figures for transient, bachelor and unidentified colonies are not trustworthy as a measure of the total population, but nevertheless demonstrate well the extent to which the o verall population has declined 4 (Table 1). For various reasons few visits were made to the hibernacula during this study (Table 2). Hibernacula counts are exceedingly difficult to make, and may be off by as much as 50 percent (M.D. Tuttle, pers. comm.) The figures shown in Table 2 include some sudden and drastic population changes that seem improbable, but in general the figures seem to show an overall reduc tion in population in the last 15 years. Chimney and Marvel caves apparently have borne the brunt of the population decline, whereas the Coffin Cave popula tion has at least remained stable, assum ing differences in counting technique ac count for the sizable population increase shown The three major hibernacula are now protected and we hope their populations will stabilize or increase h4yotis soda/is Caves with numbers of hibernating M sodali s (Fig. 12) exceeding 1,000 are uncommon ; 13 are known in Missouri. Of these, two have been abandoned (Table 3). The remaining 11 account for about 350,000 bats, or 67 percent of all known Indiana bats, based partly on data presented by Humphrey (1978) An additional 15 caves accounted for a total o f about 1 000 Indiana bats (Table 4). Some of these, however, may have been major bat hibernacula in the past. Although Humphrey (1978) showed an 8.2 percent decline for the Missouri population between 1960 and 1975, cen sus figures for subsequent years do not

PAGE 9

Figur e 2. A bat tra p in use just inside the e ntrance of Holton Cave, Boone County, a gray bat bachelor a nd transi ent roost. 5 show a continuing decline The number of known hibernating M. sodalis actually has increased due to counts at newly discovered caves and significant increases at two of the largest hiber nacula. Of the hibernacula that have been censused annually, none has suf fered a steady decline. Because year-to year fluctuations in population seem to be the rule at individual hibernacula, it is difficult to conclude from these data that the total numbers of Indiana bats are either increasing or decreasing in Missouri, but it seems clear that no disastrous decline is in progress. Never theless, there is a real threat to the largest single hibernaculum, which contains nearly 40 percent of the Missouri population. A major collapse has largely blocked the entrance and further collapses could occur at any time We can only hope that the mine in which the bats hibernate eventually will stabilize without destroying its usefulness to the bats. Other species Most of our data on other species of Missouri bats were collected by trap and mist net sampling although hibernating bats of all species were recorded when seen Myotis lucifugus was never en countered in large numbers in our study. Seldom were more than 300 seen hiber nating in a single cave and they were uncommon in trap samples even at Pilot Knob Mine where Myers (1964) once reported 35,000. Although we have little comparative data, we suspect this species is in trouble in Missouri. Destruction of bats roosting in buildings as well as changes in architecture have been listed as causing decline elsewhere (Humphrey and Cope, 1976) Myotis keenii has long been con sidered an uncommon species, but it was trapped in sizable numbers at several caves in this study (see also Caire et al., 1979). At Great Scott Cave (Fig. 30) it was sometimes the most abundant species in the trap samples. Yet the most ever counted in hibernation was 21. These bats are usually observed deep in crevices, where they easily could be overlooked. Myotis leibii was encountered in small numbers (N = 20) only at Pilot Knob Mine, plus a single individual trapped at Great Scott Cave. Doubtless this species is either very rare in Missouri (see also Gunier and Elder 1972) or overlooked by our sampling methods We suspect that it is in fact rare here, because the species is commonly captured in the western portion of its range (Barbour and Davis, 1969). P ipistr e llus subflavus (Fig. 13) was s o metimes common in trap samples (Fig. 30) and was captured, at least in

PAGE 10

TABLE 3.--Major 1 hibernating populations of Myotis sodalis. YEAR Cave 19606 1974-75 2 1975-76 1976-77 1977-78 1978-79 19798 0 Bat-Shannon 71, 800 46,000 46,002 75,609 76,703 Bear 4 3,100 1,867 1,825 1,972 3,229 2,247 Brooks 19,461 19,363 Copper Ho 11 ow 4 21,000 12,618 9,039 12,040 8,862 9,295 Chimney 3,000 5 15 Grea t Scot t 4 81,800 46,606 59,515 45,596 69,387 58,465 Inca 2,000 3,9157 Martin 8 ,109 Onyx 4 12,850 10,130 12,683 10,817 11 ,094 8,994 Pilot Knob 100,000 139,000 3 139,000 3 Scotia 3,266 3,427 2,457 2,936 2,727 Tunne 1 4,000 Ryden 4 5,600 6,000 9,697 10,539 11,000 bats or more \oun ted by R in 1966 (pers. comm. ) 2Cen s use d by S.R. Humphrey (pers. comm. ) 6Reported by R. r 1yers, 1964 3Estimated from trapping data 7Counted by J.E. Gardner (pers COTTUll. ) 4Gated, see text small numbers, at all caves trapped. Bats of this species were observed, usually in small numbers, in almost every cave we entered in winter. A few caves contained as many as several hun dred individuals. For this reason, and because we believe summer colonies rarely come into contact with human ac tivity, (we found but two-see Hum phrey et aI., 1977b), we feel this species is fairly safe, at least in southern Missouri. Epte sicus fuscus (Fig. 14) was trapped in small numbers at a variety of caves. They were observed and heard in most of the caves we visited, at least in fall, winter and spring. Although they were rarely encountered by our sam pling methods in the summer (two maternity groups were seen in caves), most reports we received of bat colonies in buildings were attributed to this species, and at one summer roost we observed well over 1,000 emerge at dusk. In most cases the owners of buildings in which colonies were located wanted to get rid of the bats. We suspect that Eptisicus populations have suffered heavy losses in Missouri from extermination efforts. However, as in the case of Myotis lucifugus, we cannot be certain that current populations are lower than they were before the con struction of buildings, even though they are declining now. RECAPTURES OF BANDED BATS Introduction The effect of banding bats on mor tality is difficult to evaluate because there is no way to maintain a control group In this study, some bats were banded on each wing, some on one wing only If there is a significant difference in mortality between banded and un banded bats, we would predict a lesser but detectable difference between bats with one versus two bands. On some nights, we banded equal numbers of bats with one band only, and with two bands. From these data we tabulated recap tures of 1,016 M. sodalis and 1,002 M. grisescens banded with two bands, plus 832 M. sodalis and 931 M. grisescens banded with one band Recapture percentages for two-banded bats were 17.4 percent and 26.5 percent, respec tively, for the two species For one banded bats, the corresponding figures were 15.4 percent and 24.6 percent. In each case there was a 2 percent dif ference favoring two-banded bats. We 6 suspect that this represents a sampling error and that in fact there was no dif ference in mortality between oneand two-banded bats. It seems probable that mortality due to banding is slight, judg ing from these daJ;a and the fact that band injuries rarely were observed. Behavioral effects resulting from banding and other handling are also dif ficult to detect, but they undoubtedly oc cur, as noted by Tuttle (1976b). We often suspected that bats with previous ex perience at being trapped avoided the trap on future encounters. On 30 July 1976, only 7 percent of gray bats trapped at Twenty-three Degree Cave were recaptures Later in the evening, 86 bats were netted by hand from an ac tive cluster in a back room of the cave. Thirty-eight percent of these bats had been banded by us previously. Although these results may have been biased by our sampling technique, they do suggest that behavioral effects of banding and other disturbance must be taken into consideration in interpreting recapture data. Myotis soda/is Recapture rates: Of the approx imately 6,800 M. sodalis banded at five hibernacula in the Meramec area during

PAGE 11

Figure 3 View toward the entrance of Onyx Cave, Crawford County, an Indiana bat hiber naculum. late 1975 and 1976, 1,412 (20.9 percent) were recaptured at least once (Table 5). In addition, we recorded 3,346 sightings of hibernating bats with bands for which most numbers were not recorded, and 1,719 observations of bats with color coded cave bands, permitting us to iden tify the cave of banding. Although many individuals were probably counted more than once, we believe that we observed most of the surviving banded bats each winter and that a vast majority of the surviving banded bats returned to the Meramec area to hibernate each year after banding. If banded males and females were encountered with equal frequency, one would expect a sex ratio among recap tures proportional to the sex ratio of bats originally banded. As evident from Table 6, this is not the case. For reasons to be discussed later, females are rarely recaptured by trapping at cave en trances and thus the overall recapture percentage for females is lower. Males are captured more frequently by cave trapping during swarming activity. Although twice as many males as females were banded, 363 males were recaptured during swarming as opposed to only 14 females. On the other hand, among bats recaptured in the five hiber nacula, 74 percent were males, 26 per cent females (of banded bats, 64 percent were males, 36 percent females). These data might suggest higher mortality among banded females than banded males, although Humphrey and Cope (1977) found no difference in survival rates between sexes. However, the disparity we found might have resulted from sexual segregation in roosting; at four of the five caves many or most of the hibernating clusters were out of our reach. But at Onyx Cave, where we had equal access to all clusters, percentage of recapture for males was more than twice for females, suggesting that the observed differences in recapture rate represent a real situation rather than sampling error. Samples of hiber nating bats examined in early March 1976 had 50:50 sex ratios at all five caves. Since we have no evidence that females left the five caves to hibernate elsewhere, it appears that mortality was higher among banded females than banded males in our study. We can offer no hypotheses to explain such differen tial mortality, unless it relates to mechanical problems encountered by banded females in crowded maternity colonies under tree bark. This seems unlikely, however. We had long suspected that females arriving at hibernating caves in October entered hibernation immediately, whereas males remained active in order to copulate with females as they arrived. This would account for the prepon derance of males in trap samples during fall swarming, as well as sex ratios strongly favoring females among bats hibernating in late September and early October. Indirect evidence from this study, as well as from other works (Hall, 1962; Cope and Humphrey, 1977) sup ports this contention. Beginning on 1 October 1979, a trap was erected at Onyx Cave on alternate 7 nights over a two-week period. Myotis sodalis captured included 1,153 males and 289 females. On the first two nights, 626 males and 150 females were banded. On subsequent nights, total recaptures included 195 males (31.1 percent of those banded) and only 7 females (4.7 percent of banded). Many of the males returned to the cave on subsequent nights (24 returned twice) in spite of the trauma of being handled. In January 1980, large numbers of the fall-banded bats of both sexes were in hibernation inside the cave. Only a few of these bats were observed hibernating in other caves. These data provide the best evidence yet presented that individual male M. sodalis remain active for periods as long as two weeks at a single cave in early October, whereas individual females only make a single ap pearance at which time they presumably enter hibernation in the cave. One might predict that recapture rates for large samples of bats banded on single nights would approximate the overall recapture rates (Table 6), but they do not (Table 7). Instead, they vary widely. One factor that apparently af fects recapture rate is season of banding (Table 7). Fall-banded bats are recap tured with a higher frequency than springor summer-banded bats. Spring banded bats would be subject to greater mortality during the active months ahead than fall-banded bats, which are soon safely in hibernation where they are likely to be encountered by us. However, we can offer no explanation for the great differences in recapture percentage between falland spring banded males, as compared to females. Survivorship: The severe limita tions of the calculation of survival rates based on unaged cohorts were discussed by Humphrey and Cope (1977). Our data are further weakened because our cohort was banded partially in 1975 and partially in 1976. Thus overall mortality as indicated by numbers of un sexed banded bats counted in hibernacula must have been greater for the 1975 banded bats than for the 1976 banded bats. Fur ther, we have data for only four suc cessive winters after banding. Overall survivorship for bats banded at the five Meramec area caves was 54.4 percent for the first year after banding, 67.4 per cent for the second year and 68.2 per cent for the third year. Lower first-year survival was also observed by Humphrey and Cope (1977) who noted that the original banded cohort would contain many first-year bats that would suffer a much higher rate of mortality during the ensuing year than adults banded at the same time. Survivorship for the second and third years is close to that observed by Humphrey and Cope

PAGE 12

TABLE 4.--Minor1 hibernating populations of Myotis soda1is. YEAR Cave 1960 1974-75 1975-76 1976-77 1977-78 197 8-79 Bat-Crawford 800 3 Bruce 500 Carro1 2 600 Coffi n 3 250 Hami H on 122 94 ry L awson 600 Springs 33 Onyx-Pulaski 600 Piquet 600 Mi 112 1 50 50 Sal oon 47 154 Smitt1 e 458 Wet 30 flolf Den 11,000 bats or less 2Gated, see tex t 3Fenced (1977) for males, but below that for females. Bear in mind that two thirds of our banded bats were males. Survivor ship at individual caves during the sec ond and third years varied from 46.3 percent (Onyx Cave) to 85.7 percent (Scotia Cave). Although differences in recovery success at caves of different size and structure may account for much of the differences observed in survivor ship among the caves, the reasons for the unexpected low s urvivorship at Onyx Cave are not clear. If Onyx were omitted from the calculations, survivor ship would be 56.1 percent for year one, 73.3 percent for year two, and 73.7 per cent for year three. These values are reasonably high considering the ap parent high mortality rate for females indicated by recapture data. Longevity: Humphrey and Cope (1977) and Paradiso and Greenhall (1967) reported maximum longevity of banded M. sodalis in the neighborhood of 13 years after banding; note that the bats on which they reported probably were banded as adults of unknown age, as were ours. In this study we recap tured 19 bats banded by R. Myers be tween February 1958 and January 1971. Others were observed but not captured. Three of these bats were at least 6 years old, one was 7, five were 8 years old, one was 9, three were 11, one was 13, four were 14 and one was at least 20 years of age. The 20-year-old bat was from a mine that we have been unable to enter. Doubtless other old bats survive there from the large cohort originally banded by Myers. Now that Myotis lucifugus have been recaptured 30 years after banding (Keen and Hitchcock, 1980), it is reasonable to expect that older M. sodalis eventually will be recorded as well. Movements: Most of the movements rec orded in this study were among the five Meramec area hiber nacula where banding activity was con cent rated during 1975-76 (Fig. 15; Table 8). More than 350 bats (9.5 percent of those recaptured) were recorded as moving from one to another of the five caves. With only two exceptions, all possible combinations of movements among the five caves occurred; there was no observed movement between Bear Cave and Scotia Cave, and no bats moved to Copper Hollow Sinkhole from Scotia. All these caves were disturbed at 8 6 1 450 113 60 36 0 63 557 17 least once each winter by us. Beyond that, disturbance levels varied substan tially, a fact surely reflected in the number of bats that moved between caves. For example, Scotia Cave is remote and difficult to locate; we believe it is rarely disturbed in winter and note that it had the lowest percentage of movement by far. Onyx Cave was popular with amateur archeologists before gating in 1976, but has been relatively secure since then. Movement from Onyx was relatively low, but higher than movement from Scotia. Great Scott Cave, with only a slightly higher movement percentage than Onyx, was disturbed frequently by cavers until gated in 1978. Copper Hollow Sinkhole, also gated in 1978, has been relatively free of disturbance due to its location We feel an improperly designed gate may have led to a higher than average movement percentage at that cave Bear Cave, gated at the same time as Copper Hollow, had by far the highest movement percentage Bear Cave was formerly extremely popular with cavers, and the gate ha s been forced open several times by vandals. Most of the bats from Bear moved to

PAGE 13

Figure 4. Descent by rope into the entrance of Chimney Cave Shannon County, a gray bat hibernaculum. 9 Copper Hollow, which is only 1 km away. Only 10 Meramec-banded bats were recaptured in distant hibernacula (Fig. 15), even though annual censuses were conducted at most of them. Fig. 16 shows all movements we recorded outside the Meramec area. Some of these recaptures were made by us in hibernacula located in counties southeast and southwest of the Meramec area. Most of the others were from the public. Seven movements, based mainly on bats captured during the spring and fall, were to the north west, north and northeast, the direc tions we had anticipated M. sodalis would move to reach summering areas. Female and juvenile bats captured throughout northern Missouri (this study) and southern Iowa (J. Bowles, pers. comm.) suggest that these are ma jor summer areas for the species. However the capture of females and juveniles at two sites on and near the Mississippi River flood plain of southern illinois (Y. Brack, pers. comm.; W. Klimstra, in !itt.) suggested that some Missouri bats (probably from Pilot Knob Mine, the nearest hibernaculum) fly east and southeast into Illinois. The single bat banded in this study and recaptured in Iowa was a pregnant female banded in April at Pilot Knob Mine. She flew 463 km northwest to Marion County, where she was captured in a wooded area during June (J. Bowles, in litt.). Myers (1964), who banded mainly in Pulaski County southwest of the Meramec area, noted local movements similar to those we recorded. He also recorded two movements north of the Missouri River in Missouri, and one into Iowa. Subsequently one more Myers banded bat was recovered in Iowa (R. Myers, pers. comm.). Myers (1964) noted that 10 bats "changed hibernacula within the four months of the same winter season as banded". In our study, no such intercave movements occurred during the winter of 1975-76, when five separate censuses were carried out. It is probable that such movements resulted from disturbance, since Myers (1964) banded in the hiber nacula with the aid of groups of students. Myers himself suggested that disloyalty might be related to distur bance. Hall (1962) noted that "small numbers may move in or out of a cave throughout the winter," basing his statement on observed fluctuations in numbers censused on different dates during the winter. We also noted such fluctuations (Clawson et aI., 1980), but lacking a single recorded intercave movement of a banded bat during the winter, we attribute the apparent fluc tuations to intracave movement, which also included significant changes in

PAGE 14

TABLE 5.--Numbers of times banded Myotis sodalis were recaptured in the Meramec area. One Time Two Three Four Line Only Times Times Times Totals Males 812 205 37 19 1073 Females 294 38 7 0 339 TOTALS 1106 243 44 19 1412 x 2 x 3 x 4 --Total Recaptures 1106 486 132 76 1800* *This total does not include an approximately equal number of hibernating bats recorded only by presence of unnumbered cave bands. cluster size and location, as documented by Clawson et al. (1980). Loyalty: Myotis sodalis demon strate a strong loyalty (= philopatry, Tuttle, 1976b) for the cave of banding (Fig. 15; Table 8). The overall percen tage of recaptured bats that moved (9.5 percent) is probably much higher than would have been the case if disturbance had been eliminated, as suggested by the low movement rate from rarely dis turbed Scotia Cave Also, loyalty might have been higher had bats been banded in hibernation rather than during swarming, assuming disturbance was kept at a minimum A multiple tabula tion of recapture data showed that regardless of season of banding, season of recapture, or sex, banded bats were usually recaptured at the cave of band ing Neither sex was more loyal than the other. Myotis grisescens Recapture rates: Of the approx imately 5,000 M grisescens banded at 11 summer caves in the Meramec area dur ing late 1975 and 1976, 1,440 (28. 8 per cent) were recaptured at least once (Table 9). In addition, we made 580 observations of color-coded cave bands, half of them in hibernacula, half in summer caves. Total sightings of cave bands in hibernacula were relatively low because we only visited each of the two major hibernacula once during the study. As documented by Myers (1964) and Tuttle (1976a), and discussed by Tuttle (1976b), populations of M. grisescens ap pear to be divided into discrete colonies, each of which uses a definite set of caves in a well-defined geographic area and has little interchange with other colonies. Only one such colony was documented thoroughly in this study-the Meramec area colony, which uses the caves shown in Fig. 17. Banding undertaken later in the study gave us at least some indication of the range of two additional colonies. Recap ture percentage and sex ratio data were based entirely on bats banded in the Meramec area colony during 1975-76. As evident from Table 10, sex ratios among bats recaptured at bachelor and transient caves usually favored males to a greater extent than in the original banded sample. The banded sample reflected the sex ratio among bats cap tured by trapping (sex ratio among all bats banded: 57.2 percent males, 42.8 percent females; among all bats han10 died: 55.8 percent males, 44.2 percent females). Even at Saloon Cave, where about 700 of each were banded, males slightly outnumbered females among recaptures. Females, however, were relatively better represented than males among bats recaptured at caves other than the cave of banding. Among bats banded as flightless young at Roaring Spring Cave, males predominated among recaptures even though fewer were banded. However, we made little attempt to recapture bats at Roaring Spring Cave itself because it was especially susceptible to disturbance. Recapture percentages (Table 11) reveal a similar pattern. Adult females, regardless of banding site, were recap tured in low percentages from 17.3 percent to 31.5 percent (X = 22 percent). Male recaptures, in contrast, 25.0 percent to 63. 7 per cent (X = 38.3 percent). Juvenile males and females were recaptured at rates of 38.6 percent and 26.5 percent, respec tively. In almost every case, fewer bats of a banded cohort were recaptured at the cave of banding than elsewhere. This is as anticipated given the type of colonial behavior and movement described by Tuttle (1976b) and elsewhere in this

PAGE 15

Figure 5 Hibernating Indiana bats in Onyx Ca v e, Crawford County Note that several of the bats are banded The predictable packing den sity of My o t is sodalis makes reliable estimates possible paper Since sex ratios among banded and handled bats were similar, one would expect that sex ratios among recaptured bats and recapture rates as shown in Tables 10 and 11 would be reasonably accurate. The lower recap ture rate among females is probably best explained by the following hypothesis: Females, especially at maternity caves, are more sensitive to disturbance than males, as observed in this study and reported by Tuttle (1976a and 1979a), and may therefore avoid recapture more frequently than males The higher recapture rate observed among females banded as inexperienced juveniles at the maternity cave lends support to this con tention, as does the relative infrequency of females recaptured more than once as compared with males. On 30 July 1976 at Twenty-three Degree Cave, 33 banded bats that refused to fly into the trap were caught by hand net Of these, 42 percent were males 58 percent females. However, among 239 bats that were caught in the trap, 63 percent were males only 37 percent females Obvious ly, a larger percentage of banded females than banded males refused to fly into the trap. Since this hand-netting was not normally done, it seems prob able that many banded females con sistently failed to enter or exit caves where the trap was erected. Overall recapture rate is also related to season of banding (Table 12), as in M. sodalis. Had a larger percentage of females been banded in the spring, the overall recapture rate would have been higher. However, this knowledge provides little in the way of explanation since we can not show why seasonal recapture rates vary so widely The relatively high recapture rates for bats banded at Saloon Cave reflect the Meramec colony structure. Saloon is a centrally located transient cave which is heavily used by both sexes, all ages, from April through October, excepting June and early July. No other area cave is used for such a long duration. Multiple recaptures were frequent, indicating that many, if not most of the colony bats made two or more visits to the cave an nually. One other recapture rate, that for males banded at Onyx Cave, stands out. Unfortunately, there is nothing in our data to suggest why Onyx-banded males are more likely to be recaptured than males banded at other caves. We suspect a sampling error ; i.e. perhaps by chance we banded a large transient group at Onyx consisting primarily of bats that summer at one or more of the other caves where our trapping ac tivities were concentrated. Longevity and survival: Richard F Myers (pers comm.) began banding large cohorts of M. grisescens in the Ozarks during the late 1950s. However, even though 61 gray bats banded by Myers and others were recaptured in this study, none was of advanced age. Of the older bats, the following ages are represented: six years, 2 bats; seven years, 4 bats; eight years, 12 bats; nine 11 years, 12 bats; ten years, 9 bats; eleven years, 3 bats Myers, M. D. Tuttle, (pers comm ) and W. J. Gunier (pers comm ) all. have recorded gray bats older than 11 years. The recapture data from this study do not allow us to calculate sur vivorship, since recapture effort has been minimal and sporadic the last three summers. However, it is interesting to note that when we last trapped at Saloon Cave (Meramec area) in late May 1979, we caught 186 males and 209 females; 10.8 percent of the males and 10.0 percent of the females were recap tured individuals banded in 1975-76. In late April 1978 the recapture percen tages were 15.6 percent and 8 1 percent respectively, whereas in April 1976 they were 33.6 percent and 28.1 percent. It would appear, therefore that survival has been fairly high in this population since 1975. Movements : The majority of the movements recorded for M grisescens during this study were among the 19 caves in the Meramec area (Fig. 17). Movements likewise were recorded among eight caves in central Missouri, and among the Meramec area and seven caves to the southeast and southwest (Fig. 18) A smaller number of movements were recorded between caves and stream netting sites A few were reported by the public. A review of the use pattern of Meramec area caves would be helpful at this point. Roaring Spring and Bat Cave #2 are the maternity caves used by the

PAGE 16

TABLE 6.--Sex ratios of j 1yotis sodalis recaptured in hibernation at cave of banding. Cave numbers from Clawson et al. (1980) correspond to caves listed on these tables as follows: Great Scott--Cave Scotia--Cave 031. 029-Great Scott M F Number Banded 1701 1123 % Sex Ratio of 60.2 39.8 Banded Bats % Sex Ratio of 60.7 39.3 Recaptured Bats 205 133 Recaptured 12. 1 11. 8 colony; both are on the northern edge of the area (Fig. 17). Bat Cave (Dent Coun ty), a maternity site to the south, is used by the Meramec colony only when in transit to and from hibernacula farther south. Saloon Cave, the major transient cave used by all colony bats, is centrally located. Onyx and Twenty-three Degree caves are bachelor roosts during the maternity season and transient caves usually dominated by male traffic the rest of the year. Both are near the main north-south migration route. Great Scott Cave has a similar use pattern, but is well east of the migration route and is used by even fewer females. All the other caves shown are used by comparatively few bats, almost entirely males during the maternity season and predominantly males at other times of the year. During the maternity period, male groups also contain nonreproduc tive females, many of which are yearlings (Tuttle, 1976b ; this study). Several of the minor transient caves contain large stains and guan o piles, suggesting that the use pattern of Meramec area caves has been altered substantially by disturbance. It is prob able that three centrally located caves, 029; Onyx--Cave 021; Copper Hollow--Cave 009; 8ear--Cave 017; CAVE OF BANDING 021-0nyx 009-Copper 017-Bear 031-Scotia All Caves HollO\-.I F M F F F F 1041 569 1130 563 356 123 111 104 4339 2482 64.7 35.3 66.7 33.3 74.3 25.7 51.4 48.6 63.6 36.4 81. 3 18.7 78.8 21.2 78.8 21.2 74.4 25.6 74.1 25. 9 413 95 93 25 89 39.7 1 6 7 8 2 4.4 25.0 Saloon (now transient use only), Fisher (abandoned), and Onondaga (commer cial, inaccessible to bats) originally served as maternity caves at a time when the Meramec area colony was several times its present size. Dispersal from a Meramec area maternity cave: During the maternity period of 1976, adult females and juveniles of both sexes were banded at Roaring Spring Cave. Dispersal movements from this cave are shown on Figs. 19, 20 and 21. Both adult and juvenile females most commonly moved to Saloon cave, and thence (we assume) to the hibernacula. We recorded few other moves for juvenile females, whereas some adult females moved to a variety of other caves. Juvenile males moved in nearly equal numbers to Saloon, Onyx and Twenty-three Degree caves. Otherwise their movements were almost as restricted as those of juvenile females. Dispersal from a centrally located transient cave in the Meramec area: We banded all bats trapped at Saloon Cave during the transient periods of 1975-76. Note (Table 10) that both sexes were banded in equal numbers. The 12 24 35 12 825 289 19.5 31. 5 11. 5 most commonly documented movements for females were northward to the two maternity caves (Fig. 22). Onyx and Twenty-three Degree caves both re ceived heavy female traffic, but Great Scott Cave was a relatively uncommon stop for Saloon-banded females. Other area caves were visited, but in small numbers. Males rarely moved north from Saloon, but flew to Onyx, Twenty three Degree and Great Scott in large numbers (Fig. 23). They visited more minor roosts than the females, and movements among caves were more fre quent. Dispersal of males from a male dominated transient cave in the Meramec area: Predictably, bats banded at Onyx Cave flew to Saloon, Twenty-three Degree and Great Scott in large numbers (Fig. 24). As with Saloon banded males, Onyx males dispersed in small numbers to numerous area caves, but fewer different movements were recorded. Use pattern of caves in Boone County (north of the Missouri River), and Camden and Hickory counties (Lake of the Ozarks area): Movements in these areas were not documented as

PAGE 17

Figure 6. Hibernating gray bats in Marvel Cave, Stone County Note the irregular dispersal of hibernating individuals, making estimates of winter populations unreliable. well in this study because effort there was concentrated on insect trapping, food habits and foraging behavior. The following numbers of bats were banded, mostly during 1978: Beck, Mauss and Moles caves (Lake of the Ozarks area)-775 males, 1,643 females; Holton and Hunter caves (Boone County) -555 males, 588 females. In the Lake of the Ozarks area, Mauss and Beck are the maternity caves; Moles is a major tran sient cave similar to Saloon A number of other nearby caves are used by gray bats, but movements were documented to only one of these. Myers (1964) presented good movement data for this area. We were not surprised to find movement among these caves (Fig. 25), but did not expect to find so many movements north to Boone County caves. Myers also noted such move ments. From his data we infer that recaptured bats which migrated north eastward to Boone County and to Fisher Cave in northeastern Missouri probably were banded during migratory stopovers at Mauss, Moles or Beck and subsequently recaptured in Boone County In Boone County, Holton is the only maternity cave. Hunter, Lewis and Clark, Devil's Icebox (not mapped) and Boone are transient caves. At one time, Boone Cave contained the major mater nity colony in the area. Movements oc cur among all these caves, but they are difficult to evaluate due to small sample size (Fig. 26). Southward movements to Beck Cave and to Coffin Cave (the major hibernaculum used by this colony) were noted, and one bat flew from the Meramec colony to Hunter. Movements between summer caves and hibernacula: As evident from Fig. 18, most of the Meramec area bats moved south about 90 km to Bat and Chimney caves Shannon County. A few moved about 150 km west to Coffin Cave, Laclede County. Bats from Boone County and the Lake of the Ozarks area flew mainly to Coffin Cave (further documented by the observation of 69 col ored bands not identified to cave of banding, but known to have been applied in the Boone County and Lake of the Ozarks areas) One bat banded in a storm sewer in Pittsburg, Kansas, was also recovered in Coffin. Some moved from the Boone County-Lake of the Ozarks areas to Shannon County (150 km from Lake of the Ozarks, 205 km from Boone County). Movement of bats from a single colony to several hiber nacula was documented by both Myers (1964) and Tuttle (1976b). One would also predict from the studies by Myers and Tuttle, as well as those of Elder and Gunier (1978), that some of our bats would have moved to another major hibernaculum-Marvel Cave in Stone County. No bats banded in this study were found in Marvel. However, the Marvel population has plummeted from 130,000 (1960) to 3,400 (1979-80), sug gesting that most of the bats that formerly used this cave are either dead or no longer use the cave. However, a bat banded at Marvel in 1970 by W. Gunier was captured at Beck in 1979, and many of the surviving bats hiber nating in Marvel bear metal tags of the 13 type used by Elder and Gunier (1978). As reported by those authors, some bats banded at Marvel later were found at other hibernacula. Elder and Gunier (1978) observed that banded bats from Marvel Cave dispersed widely, with a large majority of the recoveries from Oklahoma and Arkansas. There are six major gray bat caves in southwest Missouri (Figs. 27 and 28) that were not known to Elder and Gunier. We suspect that the colonies using these caves hiber nated at Marvel prior to the disap pearance of more than 95 percent of the Marvel population as documented in this paper. Rumors persist that there is a hibernaculum unknown to us in Oklahoma, a possibility also suggested by Elder and Gunier (1978). In the absence of such a hibernaculum, bats from southwest Missouri, northwest Arkansas, and northeast Oklahoma would have to travel distances of as much as 350 km to reach other known hibernacula in Missouri and Arkansas. Although a few populations of gray bats normally migrate as much as 500 km (Tuttle, 1976b), we suspect they would avoid such long, energy-consumptive flights if possible. Elder and Gunier (1978) also reported on bats recaptured at Marvel that had been banded at maternity col onies. Again they found that most of the bats had come from Oklahoma and Arkansas, where Gunier did most of his summer banding. However, the report (Elder and Gunier, 1978) of recovery of a gray bat banded at a maternity colony in Barber County in south-central Kan sas is almost certainly in error, as M. velifer rather than M. grisescens, occurs in that area (Barbour and Davis, 1969). J. Knox Jones, listed as the bander, stated that the bat in question certainly was not aM. grisescens (pers. comm.). Loyalty: The data collected in this study are inadequate to support conclu sions on loyalty comparable to those of Tuttle (1976b). No banding was carried out at hibernacula nor were recaptures made in subsequent winters at hiber nacula. The Meramec area maternity caves were not sampled annually after the year of banding. However, we did examine a few small samples and report the following : Of 10 lactating females banded at Roaring Spring during the maternity season and recaptured at maternity caves in subsequent materni ty seasons, eight were loyal, and two moved to Bat Cave #2. Of two banded at Bat Cave #2 during the maternity season and recaptured in subsequent maternity seasons, both were loyal. These data suggest a high degree of loyalty, small though the sample may be. On the other hand, we were unable to demonstrate loyalty among males

PAGE 18

banded at Twenty-three Degree Cave during the maternity season (at which time it hosts the largest bachelor colony in the Meramec area) and recaptured during subsequent maternity seasons. Only 13 such recaptures were recorded, and of these, only one was at Twentythree Degree. Tuttle (1976b) and Myers (1964) both observed a high level of loyalty at hibernacula and maternity caves. Except for migratory movements between summer areas and hibernacula, bats in the Meramec area colony and in the combined Boone County-Lake of the Ozarks colonies were loyal to their respective colony areas. One bat banded as a juvenile male at Hunter Cave (Boone County) on 27 July 1977 was recorded at Saloon Cave (Meramec) on 22 May 1979 Tuttle (1976b) noted that some recoveries of yearling males were outside the colony area. Timing of movements: The timing of gray bat m ovements observed in this study is not unlike that observed by Tut tle (1976b). Males were first active in the colony's summer area in early March, emerging to forage on warm nights. However, it was not clear if these in dividuals were early arrivals or had wintered nearby. Never more than a few hundred were encountered at anyone cave during March. By late March, a few females were present, usually inter mixed with males During April, large groups of both sexes, sometimes numbering in the thousands, arrived at the transient caves. Males were en countered more frequently, but at times sex ratios were nearly equal or groups were largely females. During early May, numbers continued to increase. Many samples at the major transient caves were predominantly female. By mid May, pregnant females began to congregate in the maternity caves with maximum buildup corresponding with parturition in early June. After the birth of the young in June, maternity colonies remained relatively stable until mid-July. Meanwhile, males were widely dispersed in bachelor groups, most consisting of only a few hundred bats although a few were much larger. In late July, most adult females and juveniles deserted the maternity caves. Some of the maternity caves served as transient caves thereafter; others were abandoned until the next year. During late July and August, bats of mixed ages and sexes could be found at any cave in the colony area, with fre-quent intercave movement. However, both adult and juvenile females were more commonly encountered in the major transient caves-Saloon, for example (Figs. 19 and 22) Males still predom inated in the smaller groups seen in peripheral transient caves. During September, female congregation in creased at major transient caves and by the end of the month, most had moved south toward the hibernacula. On 4 October 1976, a large sample sexed at the Bat Cave (Shannon County) hiber naculum indicated that the 27,000 present were females. Males remained in the summer colony area in October, gradually moving south toward the hibernacula By early November, only a few males were active in the summer colony area. In the Meramec area, a few hundred males remained behind to hibernate in Great Scott and Onyx caves; virtually all other summer caves were deserted by gray bats by mid November. Tuttle (1976b) noted that females emerged from hibernation before males and arrived before males in the summer area. His data are based mainly on April observations, although he stated that emergence of females begins the last TABLE 7.--Summary of recapture percentages from 20 large samples of Myotis sodalis banded at five Missouri caves during 1975-76; tabulated by cave and by season. Range of Sample Sizes -X Sample Size Range of Recapture Percentages X Recapt ure % Number Banded Number Recaptured Recapture % 029-Great Scott M F 33-415 38-509 130 151 8.8-34. 2 4.0-24.1 14.7 12.5 FALL-BANDED M F 2098 1071 596 165 28. 4 15.4 CAVE OF BANDING 009-CoQQer Hoilow M F M F 30-391 36-184 53-432 38-128 161 107 177 9 1 2041. 4 5.6-26.6 15.1-36 2.6-25.7 34.9 18.2 25.6 10. 8 SPRING BANDED M F 404 1057 65 127 16. 1 12.0 14 Oll-Bear 031-Scotla M 164 36 F M 70 22,59 25.7 18.2,27.1 SUMMER BANDED M 134 16 11.9 F 31,72 9.7,16.7

PAGE 19

Figure 7 Stains in ceiling domes at Boone Cave, Boone County Such stains result from the roosting of thousands of gray bats and provide a reliable means of estimating current or past populations. week of March Our data show that some males (for example, over 500 at Saloon Cave 15 March 1976) were ac tive in the Meramec area during March, whereas virtually no females were observed in most samples. However, at Bat Cave in Dent County (Fig. 17) on the southern edge of the Meramec area, 200 M. grisescens, mostly females, were pre sent on 23 March 1976. Our data for April show that large numbers of both sexes were present (for example, on 2 April 1976, a mist-netted sample at Onyx Cave contained 23 males and 26 females), but that intercave movement was so frequent we had no way of know ing if we ever succeeded in capturing an unbiased sample of the population It is also possible that many of the gray bats active in the Meramec area in March and early April might have been the ones hibernating in Meramec caves the previous winter Myotis lucifugus Of the approximately 1,600 M. lucifugus banded in the Meramec area during the 1975-76 period, 228 (14 per cent) were recaptured. Recapture rates at the five major banding caves varied from f1 percent (Great Scott) to 22 6 percent (Bear). Trapping at these caves yielded mainly males. Although females predominated in April trapping samples, relatively few M. lucifugus were cap tured at that time of year, so few females were banded Not suprisingly, only five of the recaptured bats were females. The number of bats banded far exceeded the number we observed in hibernation in these caves, a fact which may explain partially the low recapture percentage. This may result from secon dary dispersal (Humphrey and Cope, 1976), but the phenomenon was not documented in our study. We did not find the missing banded bats in other Meramec area caves. Eighty-eight per cent of the bats we recaptured were at the cave of banding. Myers (1964) noted that 92 percent of his M. lucifugus recaptures were also at the cave of banding. The remaining bats in our study moved mostly among the other five banding caves, but a few moved to more distant sites (Fig 29). All of these movements were to buildings and other non-cave locations; one was from a known maternity colony. These few recapture records suggest a wide sum mer dispersal of M. lucifugus hiber nating in the Meramec area similar to that reported for bats hibernating in Pilot Knob Mine by Myers (1964) However, small numbers of male M. lucifugus are active at cave entrances during May, June and July. Of bats banded in the spring or fall, only 26 were recaptured in the May-July period. All but four of these were recaptured at the cave of banding. Only two bats actually changed hibernacula; this apparently rare event has been documented by several authors (see data and citations in Humphrey and Cope, 1976). No inter cave movements within a single winter were documented by us. 15 Myotis keenii Our recapture data for M. keenii through August 1977 were summarized briefly by Caire, et al. (1979). Only 17 ad ditional bats were recaptured during 1978 and 1979, and all but one were recaptured at the cave of banding. The one movement noted was short be tween two banding caves. Eptesicus fuscus Of 194 E fuscus banded in the Meramec area 27 (13.9 percent) were recaptured at least once. Two bats were netted over streams and subsequently trapped at the nearest cave. The remain ing individuals were recaptured at the cave of banding. As noted by Barbour and Davis big brown bats are relatively sedentary. The low recapture rate probably reflects the typical behavior of this species, crawling into in accessible holes and crevices, rather than movement away from the cave of banding. Pipistrellus subflavus Only 34 (3.1 percent) of the approx imately 1,100 pipistrelles banded were recaptured. All but one were at the cave of banding. The single movement was between two caves only 1 km apart. The loyalty to a swarming cave demon strated by these data is remarkable in a species that is widespread in Missouri and which must be migratory, since it hibernates in caves, most of which are in the Ozarks It is more remarkable since relatively few pipistrelles were observed hibernating in our banding caves, and banded P. subflavus were observed in hibernation only twice. Thus it appears that our recaptured bats are, for the most part, individuals that hibernate elsewhere, but may return in subsequent years to the cave of banding-mainly during August swarming. Yet no banded pipistrelles were recaptured while hibernating or observed in hiber nation at Meramec area caves other than the caves of banding, suggesting that the banded bats were transitory and moved entirely out of the Meramec area to hibernate. Banding and recap ture dates support this hypothesis. All bats were banded during late July, August, September early October, April and May. Except for the two hibernating bats, all bats were recap tured during the same periods. Some bats originally caught in the fall were recaptured in the spring and vice versa. Other authors also have had poor suc cess in recapturing banded pipistrelles (Barbour and Davis, 1969). Since sum mer roost sites of this species have been encountered rarely (see Humphrey et aI., 1977), movements have been poorly

PAGE 20

documented, and our study has done lit tle to alleviate this situation. LIGHT-TAGGING Ground observations Myotis grisescens: Relatively small numbers of light-tagged bats were observed from the ground at the three maternity sites studied for the first time in 1979 (Table 13). The results differ from each other and from 1976 results (Table 14) because of differences in siting among the caves. The caves studied in 1976 were adjacent to rivers; the 1979 caves were 1 km to 3 km from the nearest potential foraging area, as previously noted. At Beck and Mauss caves, we followed many of the bats to the nearest stream. At Mauss, observations were from a distance, making it difficult to determine if foraging behavior occurred. Furthermore, many bats were lost from view without any useful observations being made. At both caves, the bats moving "cross-country" may have been heading toward over-water foraging areas. On 16 May at Beck, the emerging bats divided into four columns which disappeared over the forest, flying in different directions. Plotted on a map, compass bearings taken from the column movements would intersect the Pomme de Terre River within 2 km. The evidence suggests these bats were mostly foraging over the water or adjacent riparian strip and supports similar con clusions reached by LaVal, et al. (1977 b). At Holton Cave, bats could not be followed to their foraging area. Observers on the nearest suitable stream saw no lighted bats over the stream, although they saw two pass in the distance, heading south. Bats continuing in that direction would en counter the Missouri River in about 17 km, or in 12 km if they flew southwest. Twenty-one of 25 bats released flew either south or west, so the Missouri River might well have been their even tual destination. Alternatively, the bats may have foraged in some other situa tion. Based on our observations at other caves, we find this doubtful. Sixteen bats released at Moles Cave dispersed in equal numbers in all direc tions. Since this is a transient cave, and the date was 17 May, it seems likely that many of these bats were headed in the direction of other caves instead of toward the nearest potential foraging area. Bats released at Twenty-three Degree and Onyx caves during bright moon conditions (Table 14) flew cross country from the release site in a much higher percentage than in 1976 during dark moon conditions. Whereas 62 per cent of the 1976 bats (dark moon) were observed foraging over water or in riparian strips, only 23 percent of the 1979 bats (bright moon) were seen foraging in these situations. However, 20 percent of the bright moon bats were observed foraging in ridge and hillside forest (a difficult observation to make!), while no dark moon bats were seen foraging in these forests. Previously, in 1976, we had noted that far fewer bats were mist-netted over water during bright moon than during dark moon. Furthermore, on 3 July 1979 (> one half moon), bats released at Mauss Cave did not fly to the nearest over-water foraging area, although they had done so on 18 May 1979 (dark moon). Some of tne effects of moonlight on bat activity were reviewed and dis cussed by Fenton et al. (1977). Those authors observed that bat activity in open areas was reduced during bright moon. Merlin Tuttle (pers. comm.) observed a fivefold increase in gray bat TABLE 8.--Loyalty of banded Myotis sodalis based on percentage of bats recaptured at caves other than cave of banding. Data are tabulated in three different ways. PERCENT THAT MOVED Cave of Banding 029-Great 021-0nyx 017-Bear 009-Copper 031-Scotia Scott Hollow % N* % N* % N* % N* % Recaptures of Species-Banded Bats l 4.9 489 3 .6 669 22.9 231 15.6 302 4.2 Observations of Cave-Banded Bats 2 10.0 269 10.6 539 46.4 97 9.0 424 3. 1 Recaptures of Bats Previously Recorded in Hibernation 3 15.5 58 5.6 107 8.0 25 30.0 20 2.8 X % 7.4 6.6 28.3 12.2 3.2 Overa 11 X % 9.5 % *N i s total number of banded bats recaptured or observed, including those in cave of banding. 1 Based on actual recaptures of bats with numbered bands. 2Based on visual observations of hibernating bats with color-coded cave bands. 3 Based on recapture of bats that had been recorded at least once in hibernation in cave of banding. 16 N* 48 385 36

PAGE 21

activity over water from full moon to new moon. Terry Zinn (pers. comm.) observed light-tagged M. austro riparius (a species with over-water foraging habits similar to those of M. grisescens) in Florida. He noted that during bright moon the bats foraged in densely wooded swamps instead of over open water. Our data suggest strongly that while gray bats normally forage over water as noted by Tuttle (1976a) and LaVal et al. (1977b), they may forage in the forest canopy during bright moon conditions. We noted in 1978 and 1979 that lactating females with full stomachs returned to their maternity caves throughout the night regardless of moon conditions. Fenton et al. (1977) suggested that response to moonlight in some bats may be related to danger from predators. Although owls are common in Missouri and were sometimes heard near bat caves (one roosted in the entrance to a gray bat cave at times), there is no evidence to confirm or deny the hypothesis that owls are a major predator in our area. However, our insect trapping and fecal analysis data from 1978 suggest that in sect abundance or distribution may be a contributing factor. Insect biomass per trap hour (early sample) in ultraviolet light traps over water near Roaring Spring Cave (a gray bat maternity cave) was 442 g at quarter-moon in June, dropped to 41 g at full moon, and rose again to 468 g during last quarter (dark). At these three ,stages of the moon, in sects of aquatic orders predominated in the diet of gray bats at Roaring Spring Cave. If insect biomass over water drops by a factor of 10 during bright moon, then bats either must devote more time to foraging or forage where biomass is higher. Alternatively, as Fenton et al. (1977) suggest, fewer insects may be at tracted to light during bright moon. Our insect trapping and diet data do not reveal the answer to this dilemma, but we believe they do support our conten tion that relatively few gray bats forage over water during full moon. Further studies are needed to determine what in sects are available to gray bats in the various foraging habitats during bright moon and to learn how bats exploit these food resources. Are hatches of aquatic insects reduced during bright moon? Or do aquatic insects retreat to the forest to avoid excessive predation by nightjars, small owls and bats? Myotis 8odalis: Ground observa tions of Indiana bats were limited to 11 males released at Great Scott Cave on 23 May 1979 (dark moon). All these bats flew cross-country (Table 14), instead of flying toward the nearby creek. Two followed the stream valley up or .-.. _-. i i r'-' Key to symbols Myotis sodalis acti .... e Myotis soda/Is hibernating a Figure 8. Map showing counties in which we captured one or more Myotis sodalis. r--' Key 10 symbols Myotis IUcjfugus active Myotis Iucifugus hibernating [] Myotis keeni; activ e Myolis keen;; h ibernating 0 Myotis leibjj activ e 1tf. Figure 9. Map showing counties in which we captured one or more individuals of each of the species listed. 17

PAGE 22

downstream; the others flew north or west into the forest, where four were clearly foraging. These results do not differ drastically from those reported earlier (LaVal et aI., 1977b), although a notably larger percentage of individuals flew cross-country in 1979. Other species: On 5 August 1979 observations of seven M. keenii and four M lucifugus under bright moon condi tions revealed the following: Five male M keenii vanished into the forest above the cave, flying low in the trees, often within a meter or two of the ground. Most of these bats appeared to be forag ing The two females milled around in trees near the cave then, in beeline flight, flew high above the trees. One was lost over the forest above the cave; the other was last seen flying south over a distant ridge The observations of the males do not differ greatly from those reported by LaVal, et al. (1977b). At this time, we have no hypothesis to explain the behavior of the females. Of the four M lucifugus, one male and one female flew erratically through forest near the cave, at about three-fourths tree-top height, until they could no longer be seen. The other two males milled around in the trees near the river, drank and eventually flew downstream over the water. One was watched for a total of six minutes. The observations of M. lucifugus are not very useful in generalizing about their foraging behavior in Missouri. We recommend that more extensive light-tagging studies of this species be undertaken. Observations from aircraft Mgotis grisescens: During 1979 helicopter observations were made on four nights at three caves (Table 15). At Bat Cave #2 on 1 June with ideal weather conditions, only 21 different bats were observed, although we made 40 sightings. A sizable, but uncounted number of lighted bats returned to the cave. Other reasons for failing to observe lighted bats after release were detailed by LaVal et al. (1977b). A ma jority of the observations (67.5 percent) were of bats foraging over water Of these, 20 were foraging near wooded bluffs, whereas only six were foraging near the broad flood plain, which along this river is cleared for agriculture. In general, these results match closely those of 1976 (LaVal et aI., 1977b) and support the same generalizations The maximum upstream dispersal distance of 20 km exceeds any other recorded in this study. The maximum downstream dispersal distance of only 4.8 km pro bably reflects the approach of the river to a large town with its attendant lighting, noise, pollution and riverbank development. Key 10 .ymbol. Eptes/cus Iuscus -octive Eptes/cus fuscus -winter C Nyctlce/us humerolls -active Plplstrellus subflovus -active Plplstrellus 5ubflovU5 -hibernating i iN!W'ON i "",,,,., 6. r_ Figure 10. Map showing counties in which we captured one or more individuals of each of the species listed ._._ . . :--. $CCH
PAGE 23

Mauss Cave was the object of two light-tagging experiments using helicopter observation. The first, on 3 July, produced relatively little data. Mauss is 2 km north of a suitable stream (the Little Niangua River), but is 7.5 km south of the Lake of the Ozarks. Ground observations on 18 May indicated the bats flew south to the Little Niangua. However under bright moon conditions on 3 July, six bats were observed in the forest south of the cave, two in flood plain forest near the river and only two over the river. Subsequently, the cave owner reported to us that a friend who lived to the north on the lake had observed lights flying low over the water surface about an hour after we had released the bats. On 29 July we made a second attempt, this time flying north toward the lake. Bats taking this route must cross several hundred meters of open pasture and a busy highway. Although the bats flew along a wide front to the north, making observations difficult, we succeeded in recording 19 individuals. Nine were over pasture and forest, flying north Eleven had reached the vicinity of the lake and were forag ing at widely scattered sites as far as 24.8 km from the cave. Many of these sites were adjacent to wooded bluffs However, no bats were observed south of the cave, either in the forest or along the river. Based on these data, it ap pears that some or all the gray bats from a maternity cave may switch from one to another foraging area during the course of a season, even when it involves flying the opposite direction and over pastures to a different body of water Although Beck Cave is not located on the Pomme de Terre River, our observations from the ground, stated earlier in this report, led us to believe that all emerging bats flew toward that river and that many, if not most, for aged there. On 16 July, the river and riparian trees were blanketed with fog It was possible to see some of the light tags, but difficult to determine details of the bats' location and behavior. Never theless, most of the bats seen were over or near the water. The 12 km down stream stretch was partly free flowing river, partly standing water which, following heavy rains, had recently backed up behind the newly completed Truman Dam No bats were observed upstream, whereas some of the bats were seen flying in that direction in May. Myotis sodalis: Our previous helicopter observations of lighted In diana bats were made on 30 September 1976 well into the fall swarming season; we were able to follow only six bats that night (LaVal et aI., 1977b) On 30 July 1979, we captured males from the sum mer resident bachelor colony at Great Scott Cave. Sixteen of the 17 released individually were followed until lost. Two bats were resighted after being given up for lost, resulting in 18 obser vations. Eight of the bats foraged in the flood plain forest, where two of them seemed to have established foraging ter ritories in which they remained during the observation period The other six eventually were lost in the forest where they foraged below tree top level, as did all bats observed on this date. The re maining 10 observations were of bats that flew west, up a steep, densely forested hill above the cave. Five were lost foraging in the ridge forest. The other fiv e crossed a road on the ridge and foraged in a wooded hollow extend ing northwest from the ridge. Two of these were followed until, still following TABLE 9.--Numbers of times banded Myotis grisescens were recaptured in Meramec area. One Time Two Three Four Line Onli Times Times Times Totals Adult Males 553 128 22 5 708 Juvenile Males 169 26 9 2 206 Adult Females 355 50 7 0 412 Juvenile Females 102 9 3 0 114 TOTALS 1179 213 41 7 1440 x 2 x 3 x 4 --Total Recaptures 1179 426 123 28 1756* *This total does not include hibernating bats recorded only by observation of unnumbered cave bands. 19

PAGE 24

the hollow (by then a stream valley), they crossed a highway approximately 5 km from the cave. LaVal et al. (1977b) suggested that most male M. soda lis forage in ridge and hillside forest in Missouri, even though they reported some bats foraging in flood plain forest and edge. Our 1979 data suggest that more individuals may forage in flood plain forest than previously suspected However, our previous contention (LaVal et aI., 1977b) that male Indiana bats do not forage over water in riparian situations in southern Missouri (at least at the sites where we have observed them) is strengthened by these data. RESULTS OF SUMMER MIST-NETTING Summer netting was initiated with the primary objective of learning more about the summer distribution of M. soda li s and M grisescens in Missouri. As evident from Figs. 8, 9, 10, 11 and Table 16, a variety of species was captured, including all those whose presence might have been predicted from existing distribution data (see Hall and Kelson, 1959) Except for Lasiony cteris noctivagans, all these species are summer residents of Missouri. Two individuals, caught in May, were undoubtedly tran sients. We did not encounter this species even in extreme northern Missouri, although it is a common summer resi dent in central Iowa (Kunz, 1971) and in summer occurs as close to the Missouri line as southwest Iowa (Easterla and Watkins, 1970). Mist-netting over streams yields a biased sample, since those species that forage over streams would be more likely caught than those that use streams as a flyway or only come to drink. Bats that drink and forage elsewhere obviously are even less likely to be captured Species which are colonial may be common in one sample and absent from others, depending on whether nets are erected in the colony's foraging ter ritory The netting data (Table 16) are correct in showing Lasiurus borealis to be the most common species statewide, as red bats were commonly observed in a variety of situations (LaVal and LaVal, 1979). Lasiurus cinereus, on the other hand, is relatively rare throughout the state, an observation in agreement with the species' status east of the Great Plains (Barbour and Davis, 1969). Gray bats appear common, at least in southern Missouri, but this observation reflects three facts: (1) M. grisescens live in large colonies; (2) M. grisescens colonies have large foraging territories; (3) M. grisescens forage over water (Tuttle 1976a; LaVal et aI., 1977b; this paper). Myotis lucijugus, Eptesicus and Nyc ticeius colonies typically are found in buildings (Barbour and Davis, 1969). Since most of our netting sites were as remote as possible from such structures, these species are probably more common than our netting samples indicated. The small amount of data available (see, for example, Humphrey et aI., 1977a; Humphrey et aI., 1977b; Cope and Hum phrey, 1972) suggests that M. sodalis, M. keenii and Pipistrellus subflavus roost in small colonies in various situa tions. It is perhaps not surprising that these three species were caught in relatively small numbers. It seems clear, however, that the latter two species are more abundant in the Ozarks of southern Missouri than they are north of the Missouri River. Except as just noted, our data do not show significant differences between relative abundance of species north and south of the Missouri River. However, it does seem clear that TABLE 10.--Sex ratios among recaptured grisescens at five Missouri caves. Total Recaptured at Recaptured Banded Cave of Banding Elsewhere Cave and Type % Sex % Sex % Sex of Colonz Sex N Ratio N Ratio N Ratio Great Scott Male 526 81. 3 84 93.3 56 72.7 (Bachelor, Transient) Female 121 18.7 6 6. 7 21 27.3 Onyx Male 386 67.6 82 98.8 148 78.3 (Transient) Female 185 32.4 1 1.2 41 21. 7 Saloon 1 e 703 50.0 99 75.6 205 51. 9 (Transient) Female 704 50.0 32 24.4 190 48.1 Twenty-three Degree Mal e 539 60.1 7 58.3 128 69.2 (Bachelor, Transient) Female 358 39.9 5 41. 7 57 30.8 Roaring Spring Female Ad. 207 35.8 5 38.5 32 22.2 ( Ma te rn i ty) Female Juv. 196 33.9 3 23.1 49 34.0 1 e Juv. 176 30.4 5 38.5 63 43.8 20

PAGE 25

more bats are captured per net per hour in southern Missouri. Since suitable foraging habitat in northern Missouri is restricted mostly to riparian situations, whereas forest and forest edge habitat covers extensive areas in southern Missouri, it seems obvious that bat den sity would be greater south of the Missouri River. Observations made at dusk at many sites throughout the state lend support to this statement North of the river a few bats can be observed at dusk in good riparian habitat; elsewhere few or none can be seen. South of the river bats are commonly in view at dusk along roads, streams, in forests, around buildings and in small towns. It appears also, from our data, that diversity is lower in northern Missouri than in southern Missouri, although there were fewer nights of netting in the north. The H' values obtained are comparable with those given by Humphrey (1975) for Missouri, Iowa and Indiana. DIETARY PREFERENCES Myotis soda/is As mentioned in the section on light tagging, feces from netted or trapped bats were collected at several sites in southern Missouri from June through early August 1979; also for M. grisescens only during June-August 1978 Unfortunately, analyses of these results are not complete. Preliminary results are interesting, however. Analysis of 1,256 pellets from 140 male M. sodalis plus 47 pellets from 7 female M sodalis (11 samples from five nights) showed that Lepidoptera comprised the bulk of the diet in all samples, ranging from 60 percent (by volume) to 95 per cent. Coleoptera, Trichoptera and Diptera accounted for a relatively small portion of the food items identified. Belwood (1979) found only 57 percent Lepidoptera in the diet of female M. sodalis at Webster, Indiana, with Diptera accounting for 18 percent. However, she noted that during lacta tion, Lepidoptera accounted for 70 per cent of the food. Most of the fecal samples from males were obtained by us during the calendar period when females were lactating. It appears that male M sodalis at our study site (Great Scott Cave), forag ing in the forest, concentrate on moths regardless of what other insects are available For example on 7 June Lepidoptera comprised only 15 percent of the sample from a light trap in the forest canopy where the bats were foraging Yet the bats' diet that night was nearly 85 percent moths. If M sodalis males are dietary specialists, they would be unusual among U.S bats studied thus far, which appear to be food Figure 12. Newly banded Myot is s o dalis taking fl. ight from a tree near Pilot Knob Mine, Iron County generalists during at least part of the year (Ross, 1967; Black, 1974, 1979) Belwood (1979), using food preference indices, showed that lactating females also selected moths from among available food items. Myotis keenii The 477 pellets collected from 75 adult male M keenii (six samples, three nights, June and early JUly) showed that Lepidoptera predominated (55 percent to 90 percent) in five of six samples with Trichoptera (60 percent) heavily used in the last sample. Plecoptera were also a significant dietary item (5 percent to 28 percent), with Coleoptera, Homoptera and Ephemeroptera in small quantities A mixed sample of 11 adults and juveniles of both sexes (92 pellets) cap tured on 1 August had switched their emphasis to Coleoptera (55 percent) as well as Trichoptera (15 percent) and Lepidoptera (30 percent). It is perhaps significant that the diet of male M. k e enii trapped at Great Scott Cave is similar to that of male M. sodal is trapped at the same cave on the same 21 nights. According to our observations of light-tagged individuals (LaVal et al., 1977b; this report), both species feed in the forest around the cave However, as we noted, the two species tended to occupy vertically stratified foraging areas (although not exclusively separate). M sodalis forage around the crowns of trees; M k e en i i foraged well below the crowns. Thus these species may not be in direct competition even if they are eating the same kinds of insects Myotis grisescens The diet of female M. grisescens undergoes complex changes through the course of the season; unfortunately, we are not yet prepared to discuss them in detail. The 6,272 pellets examined from 685 bats on 34 nights show that insects of the aquatic orders Plecoptera Ephemeroptera, and especially Trichoptera, are extremely important in the diet, accounting for as much as 98 percent in some samples However, dur ing late summer small Coleoptera (oak weavils) were commonly eaten, account ing for as much as 50 percent of the diet.

PAGE 26

On occasion, Lepidoptera were taken in sizable numbers; food preference included more than 50 percent moths at one site on two nights. At two of five sites, Diptera became important in late June and early July, with up to 55 per cent consumed on one night. Merlin Tut tle (pers. comm.) has informed us that the diet of gray bats foraging over a reservoir in Tennessee consists primarily of Ephemeroptera. Earlier (1976a) he stated that mayflies and other aquatic insects seem to be preferred. The only generalization we can make at this time is that gray bats seem to be oppor tunistic feeders, concentrating on the aquatic insects available where they forage, but taking advantage of other insects-especially moths and beetleswhen they are abundant in the bats' foraging areas. We assume that gray bat diets would differ during bright moon, when they apparently forage more in the forest, but almost all samples were taken during dark moon (see previous discussion of foraging and moon phase, this paper). Figure 19. Pipistrellus subflavus typical of those captured in much of the state. This particular individual was photographed in Mississippi. Myotis /ucifugus Only 266 pellets, collected on six dif ferent nights from 30 adult male M. lucifugus, were examined. Trichoptera accounted for the bulk of four samples (60 percent to 95 percent). One sample was half Trichoptera and half Plecoptera. The remaining sample was 75 percent Lepidoptera, with most of the rest being Trichoptera. Buchler (1976) found that nine M. lucifugus on Long Island, New York, ate mainly Ephemeroptera. Sixteen bats examined from Indiana (Whitaker, 1972) had eaten 22 percent Lepidoptera, 13 per cent Trichoptera, 20 percent Homoptera and 19 percent Diptera, among other things. Anthony and Kunz (1977) ex amined feces from 62 little brown bats in New Hampshire. From results expressed in frequency of occurrence, Diptera, Lepidoptera, Ephemeroptera and Hymenoptera, in sequence of descending frequency, were all more common in the diet than Trichoptera Belwood and Fenton (1976) found that TABLE ll.--Recapture percentages of Myotis grisescens at five Missouri caves; multi pl e recaptures included. Total Recaptured at Recaptured Total Cave and Type Banded Cave of Banding El Reca(:ltures of Colony Sex -N-N % N % N % Great Scott Mal e 526 84 16. 0 56 10.6 140 26.6 (Bachelor, Transient) Female 121 6 5.0 2 1 17.4 27 22.3 --------------------TOTAL 647 90 13.9 77 11.9 167 25.8 Onyx Ma 1 e 386 82 21. 2 148 38.3 230 59.6 (Transient) Female 185 1 0.5 41 22. 2 42 22.7 ---------------------TOTAL 571 83 14.5 189 33.1 272 47.6 Saloon Male 703 99 14. 1 205 29.2 304 43.2 (Transi ent) Female 704 32 4.5 190 27.0 222 31. 5 ---------------------TOTAL 1407 1 3 1 9.3 395 28.1 526 37.4 Twenty-three Degree f1a 1 e 539 7 1.3 128 23.7 135 25.0 (Bachelor, Transient) Fema 1 e 358 5 1.4 57 1 5.9 62 17. 3 --------------------TOTAL 897 12 1.3 1 8 5 20.6 197 22.0 Roaring Spring Female Ad. 207 5 2.4 32 15.5 37 17. 9 (Maternity) Female Juv. 196 3 1.5 49 25.0 52 26.5 Male Juv. 176 5 2.8 63 35.8 6 8 38.6 ---------------------TOTAL 576 13 2.3 144 24.9 157 27.1 22

PAGE 27

TABLE 12.--Differential recapture rates among fall, spring-, and summer-* banded bats. "N banded" represents only large samples banded at major caves. liN recaptured" is based on bats recaptured at least once. FallSpring-Summer-* Banded Banded Banded M F M F M F Number Banded 869 327 384 562 795 697 Number Recaptured 328 61 114 130 179 126 Recapture Percentage 37.7 18.7 29.7 23.1 22.5 18.1 *June-July TABLE 1 3.--Numbers of bats light-tagged, 1 9 7 9, with moon and sky conditions. GROUND OBSERVATIONS HELICOPTER OBSERVATIONS Date N Moon Sky Date N Moon Sky Myotis grisescens :-1atern ity: Heck 16 May 37 Dark Cl ear 16 Jul y 8 1 Dark Cloudy, fog Hol ton 28 May 25 Dark Clear 1 8 May 24 Dark Cloudy 3 July 70 ltt 2 C l ear 29 Jul y 79 J" Clear Bat #2 1 June 75 J" Clear Bachel or: Twenty-three Degree 5 Aug. 16 3/4 Clear Transient: O nyx 6 Aug. 35 Full Cl ear es 1 7 May 1 8 Dark Clear TOTAL R ELEASED 155 305 sodalis Bachelor: Great Scott 23 May 11 Dark C l ear 30 July 17 J" Cloudy *Also lighttagged on this date: 7 M. keeni i ; 4 M l u cifugus; Eptesicus fuscus; 1 Lasiuru s boreal i s 23

PAGE 28

Figure 14. Big brown bat from Great Scott Cave, Washington County. BAT 24 eight adults at a locality in Ontario ate 60 percent Trichoptera and 38 percent Diptera, whereas 12 adults in northern New York consumed 76 percent Diptera and 18 percent Trichoptera. Adults at four other localities in eastern Canada ate Trichoptera from 12 percent to 58 percent, Diptera from 39 percent to 78 percent, Belwood (1979) noted that nine adults and juveniles in Indiana ate ap proximately equal quantities of Diptera, Coleoptera and Lepidoptera. As sug gested by these authors and our data, M. lucifugus are opportunistic foragers, taking advantage of insect populations that vary both geographically and tem porally. Pipistrellus subf/avus One hundred forty-three pellets were obtained from 27 aduitPipistrellus subflavus (five samples from five nights in late June, July and early August). Trichoptera ranged from 33 percent to 88 percent, averaging 63 percent. Lepidoptera were less commonly eaten, ranging from 10 percent to 25 percent. Coleoptera were present in low numbers except for one small sample (4 August) in which they accounted for 55 percent Hymenoptera comprised 30 percent of the 21 June sample. In contrast, Whitaker (1972) in examining the diets of 23 pipistrelles from Indiana, found that they ate 30 percent Coleoptera, 29 percent Homoptera, 20 percent Diptera and 7 percent Lepidoptera. Like M. lucifugus, pipistrelles appear to be opportunistic. The contrast in diets between our data and Whitaker's suggests something more-that our bats are foraging near water. Like M. grisescens, they relied on aquatic insects. Although we lack light-tagging data for this species, we believe they forage high over streams (gray bats forage low), based on capture in the top of pulley-operated nets. Such vertical stratification would reduce competition between the two species, as in the case of M. sodalis and M. keenii, but clearly more extensive studies of the competitive relationships of these species pairs are needed. Figure 15. Movement of banded Myotis sodalis among five Meramec Area caves and to three distant hibernacula The numbers above or below each cave name are interpreted as in this example : 7 .6% of the 1,218 bats banded at Onyx Cave were recaptured at other caves, the rest were recaptured at Onyx. The figures in the circles are interpreted as in this example: 18.80/0 of the 333 bats banded at Bear Cave were recaptured at Copper Hollow Sinkhole; 5.2% of the 728 bats banded at Copper Hollow were recaptured at Bear Cave.

PAGE 29

KEY TO SYMBOLS Hibernating caves Summer recapture tes Figure 16. Dispersal of adult Myotis sodalis from hibernacula to summer and transient sites, and movement among four hibernacula. Eptesicus fuscus Two hundred thirteen pellets were examined from 25 adult and juvenile Eptesicus fuscus (seven samples from seven nights in June, July and August). We found big brown bats to be beetle strategists, as have other authors (see for example, Black, 1974; Whitaker, 1972; Belwood, 1979). Coleoptera averaged 76 percent, Homoptera 19 per cent and Lepidoptera only 4 percent. Thus the six common cave bats we studied in Missouri appear to partition the available prey populations, either by spatial partitioning or by concentrating on different insect taxa. The picture is far from clear, however, and the posi tion of non-cave species in these forag ing communities, especially the abundant Lasiurus borealis, is yet to be determined. MYOTIS SODALIS SUMMER HABITAT Between 10 June and 26 July, M. sodalis were netted at 10 localities in counties scattered over northern Missouri (Fig. 8). Although no colonies were found, the capture of lactating females and juveniles at these sites suggests that maternity colonies must have been nearby. Also, lactating females were captured flying into Pilot Knob Mine and Great Scott Cave in southern Missouri during June and early July. Humphrey et a!. (1977a) found that bats from a maternity colony in eastern In diana foraged mostly in riparian habitat as well as around solitary trees and forest edge on the flood plain near the stream (see Figs. 4 and 5 in Humphrey et a!., 1977a). As mentioned previously, riparian strips are the only kind of forest habitat available in most of northern Missouri, and thus our netting was car ried out in such habitat. The netting sites where we captured M. sodalis in northern Missouri had the following characteristics: Number of Sites 8 2 7 3 6 4 2 Description of Sites Thin riparian strips with croplands adjacent Forest patches adjacent Stream enclosed by over hanging vegetation Stream wide, open to the sky Water muddy, bottom muddy Water clear, bottom rocky Channelized, with regrowth of riparian trees From these data we can see that the typical stream offering habitat to M. sodalis in northern Missouri has only a 25 narrow strip of riparian vegetation and is surrounded by cultivated fields. The streams are typically narrow with overhanging vegetation (but note that these are the easiest situations in which to capture M. sodalis). Finally, the streams tend to be muddy, with muddy bottoms and steep muddy banks Two of the streams had been channelized, but riparian trees had regenerated. Trees with loose bark were abundant because of the recent demise of the American elm (Ulmus americana), victim of Dutch elm disease. We note that some streams where we failed to catch M. sodalis had similar characteristics. If female M. sodalis hibernating in Missouri dis persed uniformly throughout northern Missouri, southern Iowa and Illinois and lived in colonies averaging 100 bats each, there would be about 12 colonies per county. If this were anywhere close to the actual situation, it would be for tuitous if nets were set within a colony'S foraging area, which may occupy less than 1 km along a stream (Humphrey et a!., 1977a). In southern Missouri, mist netting activities failed to yield M. sodalis females or juveniles, but the capture of lactating females at the entrance of two hibernacula suggests that not all reproductively active females migrate northward. If so, these females are pro bably not foraging in riparian situations, but rather in forests, as with males (reported by LaVal et a!., 1977b; this paper) in southern Missouri. The capture of females in remnant forest patches away from streams in northern Missouri (Easterla and Watkins, 1969) and Iowa (J. Bowles, pers. comm.) sug gests that such habitat might have been more important to the bats before exten sive clearing of forests by early settlers. If true, it would appear that available summer habitat has been greatly re duced through most of the range of M. sodalis. Nevertheless, using the figures of one colony per km suitable riparian habitat and 12 colonies per county, the available habitat is certainly not being utilized fully. It is doubtful, then, that spatial limitations of available summer habitat are contributing significantly to the endangered status of M sodalis at this time. YEAR-ROUND ACTIVITY AT A MISSOURI CAVE Late summer and fall activity of bats at cave entrances has been documented adequately (see, for exam ple, Barbour and Davis, 1969; Hum phrey and Cope, 1976; Cope and Hum phrey, 1977; Fenton, 1969). LaVal et a!. (1977a) summarized swarming data on four species of Myotis in Missouri. Some workers have trapped or netted cave en-

PAGE 30

"" a> TABLE 14.--Behavioral activities of light-tagged Myotis grisescens and M. sodalis based on observations from ground. Table interpreted as in LaVal et al. (1977b). Myoti s gri sescens s soda 1 i s DARK MOON B RIGHT DARK MOON All Caves* Beck Nauss Holton Twenty-three Degree Onyx Great Scott N % N % N % N % N % N % N % Cross Country 43 31.2 4 17.4 6 30.0 27 100.0 6 37.5 21 60.0 11 100.0 Beeline Upstream 36 26. 1 4 17.4 3 15.0 1 2 9 1 9. 1 Beeline Downstream 76 55. 1 11 47.8 5 25.0 3 18.8 7 20.0 1 9. 1 Foraging Over Water 47 34. 1 6 26.0 5 14.3 Foraging In Riparian Trees 39 2 8 .3 6 26.0 2 10.0 6 37.5 1 2.9 Foraging In Hillside 3 18.8 7 20.0 4 36.4 And Ridgetop Trees Fora ging Above 5m 23 16. 7 4 17.4 Foraging Below 5m 14 10. 1 5 21. 7 Milling Around 32 23.2 9 39.1 3 15.0 13 52.0 6 37.5 7 20.0 3 27.3 Drinking 32 23.2 3 12.0 2 8.0 3 8.6 Returned To Cave 9 39. 1 2 10.0 6 24.0 3 18.8 13 37.1 1 9. 1 Hung Up In Tree 3 13.0 4 16.0 1 2 9 1 9. 1 Flew Towards River 8 34.8 *From LaVal et al. (1977b) based on 1976 data. Great Scott and Twenty-three Degree caves were included in that data, but not Beck, Holton and Mauss.

PAGE 31

.*Roaring Spring I I I I I I I I I I I I I I I ___ .. ___ J Washington Co. \ r--------i I .----------------I Crawford Co. I ...... I I I \ \..., eTwenty-thr!e Degree __ 1-. __________ .., *Bat , I , , , KEY TO SYMBOLS Maternity caves Major transient and bachc.!or co .... es Minor 'ranslen'and bachelorcaves Stream netting sites I __ l _____________________ Dent Co. Figure 17. Location of summer caves used by Myotis grisesc ens (Merame c colony ) Also shown are sites where mist n ets were erected over streams in 1976. trances in the spring and early summer as well (see, for example, Humphrey and Cope, 1976; Mumford and Whitaker, 1975; Hall and Brenner, 1968). Mumford and Whitaker (1975) and Tuttle (1976b) netted and trapped on a few nights dur ing the winter, but few bats were cap tured. Great Scott Cave in Washington County, Missouri, is a major hiber naculum for M sodalis as well as a bachelor and transient roost for M. grisescens. Summer use of the cave by M. k eenii was discussed by Caire et al. (1979). Great Scott was trapped on 38 nights from 31 August 1975 through 9 October 1979. The earliest date netted was 1 March; the latest, 5 November. The 5 November data are not discussed here because of a disturbance that caused arousal of many hibernating bats inside the cave. However, small numbers of bats were still being trapped at other caves the same week of November; for example, 37 M. sodalis, 7 M grisescens 2 M lucifugus, 1 Pipistrellus and 16 Eptesicus were cap tured on 4 November at Onyx Cave. Trapping also took place at five other M. sodalis hibernacula (Bear, Copper Hollow, Onyx, Pilot Knob Mine and Ryden), but on fewer nights. Sporadic trapping at hibernacula not used by In diana bats resulted in few captures of that species, but other species were fre quently well-represented. Although no other cave was trapped as frequently as Great Scott, 23 other caves were trapped (rarely netted) on 149 nights, for an average of 6 .5 nights 27 per cave and ranging from one to 23. Results were tabulated in bats per hour since trapping times varied (Fig. 30). Activity tended to peak in late August and late September a fact born out by observations at other caves (LaVal et al., 1977a), but few other generalizations can be made since activi ty varied greatly from year to year. For example 19 June 1976 was a slow night but 20 June 1979 resulted in the greatest capture rate of any night trapped during the entire study. A species-by-species breakdown is perhaps more useful Myotis soda lis In three of four March-early April samples, female Indiana bats emerging from hibernation outnumbered males. Inexplicably, the reverse was true on 10 April 1979. At a larger hibernaculum, Pilot Knob Mine, females trapped on six April nights over a three-year period outnumbered males by ratios of 10:1 to 40:1; sample sizes ranged from 1,000 to 2,000. It appears that by the end of April most of the females have emerged, and males predominate in the samples Since many males emerged during March and April catch rate of males in the spring never seemed to equal that of females at emergence peak; for example, on 10 April 1976 Small numbers of males con tinued to be caught at the entrance dur ing May and June, some entering the cave, others day-roosting in the cave. However, in late June 1977 and 1979 and through July 1979 sizable numbers of males continued to use Great Scott Cave as compared to 1976 and 1978 Females began to reappear in small numbers in early August, peaking in late August. Numbers of males peaked in late August and again in late September-early October. Our highest capture rate at any cave was recorded at Pilot Knob Mine on 23 September 1978, where we caught bats at a rate ex ceeding 2,000 per hour Males were never outnumbered by females in fall samples and usually greatly exceeded them in numbers trapped, for reasons explained elsewhere in this report. Nevertheless, some females as well as males were still being captured through the first week of November. These observations are both supported and augmented by data presented earlier (LaVal et al., 1977a). Cope and Hum phrey (1977) also noted two peaks of fall activity, but these did not coincide precisely with ours. Likewise, their data for spring departure of females is similar to ours, but departure of our males peaked in late April rather than early May. Year to year variations in spring and fall capture rates no doubt reflect differences in timing of migratory movements due to a variety

PAGE 32

1 KEY TO SYMBOLS Hibernating caves Maramec area caves Summer caves Figure 18. Movement of Myotis grisescens between summer caves and hibernacula of factors, especially weather condi tions. However, high levels of activity in these seasons seem to be relatively predictable, as noted by Cope and Hum phrey (1977), in spite of fluctuations in catch rate. Our data also indicated a predictably low level of activity during most of May and early June. However, in 1979 capture rates during late June, July and August were much higher than in previous years. Weather was normal, and the hibernating population in Great Scott was somewhat lower than the previous year. At this time, we can offer no reasonable hypothesis to explain this type of apparently unusual concentra tion of male M sodalis. Myotis keenii As shown by Caire et al. (1979) and by our subsequent data, spring and fall movements peak in April and August respectively. We cannot be certain that these captures represent bats hiber nating in Great Scott and the other caves trapped because few M. keenii were found during hibernacula counts and none of these was banded. In any case, the number trapped during fall and spring was surpassed greatly by cap tures of males during June and early July (see Caire et aI., 1979, for discus sion). This pattern persisted for all four years trapped and must be important in some way to the survival of males of this species. Normally males exceeded females in all samples, except for 6 August 1977 at Pilot Knob Mine, when 84 females were captured as compared with only 63 males. Either females use these particular hibernacula less than males (preferring caves not used by large colonies of M. soda lis) or consis tent behavioral differences prevented us from trapping as high a percentage of the females present as the males. Myotis lucifugus Although this species was not especially common at any of our caves, enough were captured, especially at Pilot Knob, to demonstrate that activity was high only in August, September and April. Males predominated in the fall, females in the spring (when they often outnumbered males-as much as 10:1 at Pilot Knob Mine on 12 April 1978). In general, our observations agreed with those of Humphrey and Cope (1976), ex cept that they apparently were still catching sizable numbers during May when we caught very few. 28 NUMBERS OF BATS THAT MOVED 1 -5 ---6-15 ---16-35 ---36-75 Greal Scolf Twenty-three Degree OTHER SYMBOLS Movements to distant hib crnoculo 'Y Cove of bonding e Str com nctling sites TYPE O F CAV E US E Maternity Maior I ronsien t and bache lor Minor tronsient and bachelor Figure 19. Dispersal of adult female Myotis grisescens banded as adults during the maternity season at a maternity cave. The line widths and the symbols on this figure are used on all similar figures. Myotis grisescens As far as we can determine, gray bats hibernate only in small numbers in Great Scott and the other caves trapped. Gray bats captured at Great Scott in the summer are those associated with the resident bachelor colony. Those trapped in the spring and fall are transients working their way northward in the spring, southward in the fall. Since Great Scott is located away from the Meramec River, the main north-south route in this part of Missouri, transient activity is rather light and highly variable compared with some other caves in the area. However, a small number of males are consistently pre sent during the summer. These summer males are trap-wary so relatively few of those present (up to 5,000) are ever cap tured. Pilot Knob Mine, on the periphery of gray bat summer range in Missouri, is rarely visited by this species. The other hibernacula all serve as active transient caves, but except for Onyx, have little M. grisescens activity in the summer. Myotis leibii Only 21 of these bats were trapped at hibernacula. Sixteen (13 males; 3 females) of these diminutive bats were taken on 29 March, 7 April and 9 April

PAGE 33

TABLE 15. Behaviora1 a ctivities of light-tagged Myotis grisescens based on ob servations from aircraft. Distances are in kilometers. Percentag e s are as a percent of t otal number of observati ons. Bee1 ine F 1 i ght: O ver W a ter Over L and F o r aging: Bat #2 1 June N % 8 20.0 3 7.5 Mauss 3 J ul y 29 Ju1 v N % N % 9 34.6 Beck 16 Ju1 y N % 3 4 3.4 Ove r Wa t e r 27 67.5 5. 0 20 8 30. 8 8 27.6 In Riparian Trees In H illsi de A n d R idgetop Trees 20 1 3.B 7.4 60 7.7 T ype of F1 i ght Unde termined: Over R i ve r Over L and 11 37. 9 6 20 7 t1i 11 ing Aro u n d At Cave Gre a test Distance : Upstream Downstream N umber of Bat s Seen At Least Once 20.0 4 8 2 1 o 7.2 8 6 23 1 2 4.8* 12 1 9 2 7 *Distance fro m c a ve a l ong ro ute used by bat s partially ove r l and, parti ally over lake. at Pilot Knob Mine. On 6 August, two adult and two juvenile males were cap tured One male was trapped on 19 May at Great Scott. These data suggest that M. leibii leave hibernation early, in agreement with Barbour and Davis (1969), but reveal little else about the species' behavior. Pipistrellus subf/avus Pipistrelles were captured mainly during late April and May, and late July and August, suggesting that they are among the last bats to leave hibernation in the spring and the first to enter in the fall Observations in the caves confirmed this suspicion (also noted by other authors, including Barbour and Davis, 1969). Pipistrelles were rare in June early July samples as well as in March and October samples (supported also by the data of Mumford and Whitaker 1975). Moderate numbers were present in some September samples An excep tion to these generalizations was on 13 July 1978 at Round Spring Cave, 90 km south of the Meramec area Male pipistrelles were the most common species taken in the trap, accounting for 62 individuals Eptesicus fuscus Big brown bats were never trapped in large numbers at Great Scott or other caves. Ten to 20 individuals would have been considered a good catch Most bats were males captured in the spring and fall. Yet as indicated previously, E. fuscus is not uncommon in Missouri, oc curring in summer colonies exceeding 1,000 individuals. Generalizations The data from all 24 caves trapped tend to support the generalizations made regarding the six most commonly captured species. The phenomenon referred to as fall swarming was well documented in this study, and we sup port the postulations of Humphrey and Cope (1976) outlining the likely role of swarming. They (among others) defined swarming as a phenomenon in which large numbers of bats fly in and out of cave entrances from dusk to dawn, yet relatively few roost in the cave by day. However we note that M gris escens does roost in the cave by day and has not arrived there as a prelude to entering hibernation. Thus gray bat activity is not swarming activity, at least in the caves we trapped We also note that a number of authors have used swarming to refer to spring activity as well as fall activity All of our observations in Missouri lead 29 us to believe that bats caught during March April and in some species all or part of May, are bats emerging from hibernation and thus leaving the cave The phenomenon is thus qualitatively different from fall swarming and should be referred to by a different term We favor simply "spring emergence". SEASONAL CHANGES IN WEIGHT Seasonal weight changes are of in terest because they illustrate several im portant aspects of the life cycle of temperate zone insectivorous bats; in cluded are pre-hibernation fat deposi tion, weight loss during hibernation, weight gain after emergence from hiber nation, weight gain during pregnancy and lactation, and finally, sexual dimor phism in weight. The latter observation, which has revealed females as being both heavier and larger in most species, is probably related to the requirement that females carry heavy embryos and sometimes nursing young Or, as stated by Ralls (1976) in explaining her big mother hypothesis, "a bigger mother is often a better mother We attempted to weigh at least 30 empty-stomached adults of each sex

PAGE 34

G'.a l Scali Figure 2 0 Dispersal of female Myotis gri sescens banded as juveniles during the maternity season at a maternity cave. each time we trapped a cave. When variation was high, we weighed more if possible. Most of the weights were from bats trapped at M. sodalis hibernacula and M. grisescens summer caves. In ad dition, we weighed samples of hiber nating M. sodalis in early November 1975 and in early March 1976 ; we also included weights of some individuals mist-netted over streams. In the case of M. sodalis and M. grisescens, we used only large samples. For M. lucifugus and Pipistrellus we combined many small samples. We found little difference between weight samples collected on the same date in different years or at different caves. Therefore, we pooled all individual weight samples into bimonthly samples for purposes of graphic illustra tion. Samples of Eptesicus fuscus were too small to be useful for weight graphing Samples of M. keenii were large, but were dealt with adequately by Caire et al. (1979), who based their discussion on our 1975-77 data. The 1978-79 data would add little of significance to their conclusions. One-way analyses of variance were performed for bimonthly weight samples for each sex of the four species graphed. All were significant at the .001 level, with F values ranging from 15.59 to 175.73. Thus we are confident that the seasonal weight variations observed are real and not merely due to chance. T-tests were performed between sexes for each bimonthly sample, and will be discussed under the appropriate species Myotis soda/is Weights of 3,290 males and 2,180 females were used to construct the weight graphs of Fig. 31. No weights of non-pregnant females were available for late May and early June; the pregnant individuals weighed were all netted and, unfortunately, had full stomachs. Females outweighed males in all months except late June and early September (not statistically significant) The late August sample favoring females also was not significant All other samples were significantly different, with P values ranging from .01 < P < .02 to P < .001. Females achieved maximum weight in early October (8.90 g), males in late October (7.99 g). Minimum weights were recorded in late March, by which time females had lost an average of 2.47 g, males 2.22 g. 'Phese weight losses represent 27 8 percent of prehibernation weights for each sex. Hall (1962) recorded weight losses (early TABLE 16.--Bats mistnetted over stream s 1 976-7B. Cat c h rate i s in bats pe r net per hour ( a s us e d by LaVal an d LaVal. 1 979) Southern 1 Northern2 Northern2 Southern3 Missouri Mi sso u ri Mi ssouri ssouri 1 976 1 977 1 978 1 978 Seec i es N Rate N Rate N Rate N Rate Myotis soda1is 22 047 .069 .065 Myoti s 9ri sescens 201 .429 0 .037 38 .993 Myotis 1ucifugus 1 0 .021 .035 .01 8 0 Myoti s keeni i 17 .036 012 .009 0 Pipistrellus s u bf1avus 77 164 .069 10 .092 25 .653 Eptesi cus fuscus 1 5 .032 .104 8 .074 .052 Nyct iceius humera1is 53 113 17 196 .01 8 .052 Lasiurus borealis 267 .571 68 .783 69 .636 44 1. 1 50 L as iuru s cinereus 28 .060 1 0 .115 .01 8 0 noctivagans 2 .004 0 TOTAL 692 120 105 111 Number o f Nights Net t e d 23 1 3 Catch Rate. All Spec ies 1 .477 1. 38 1 0.968 2.90 Combined Catch Rate, 1. 152 Northern Mi ssou ri H' 1. 654 1.424 1.291 1. 402 1Captured i n 4 to 8 single. stream-level nets in three eastern Ozark counties. 2Captured in a single. 3 -tier. pulley-operated net approximatel y 7.5m high and 12m long, in various counties north of the H issouri R iv er 3Captured in a pulley -operated net as above in four wide l y -separated counties in the Ozarks. 30

PAGE 35

October to mid-April, in hibernation) of 24.2 percent for females, 22.9 percent for males. Both sexes gained weight slowly in the spring, more rapidly in the early summer, and then dropped slightly to a late summer plateau before begin ning a steady weight gain terminating in hibernation. Myotis grisescens Fig. 32 is based on weights of 2,542 adult males and 1,481 adult females. During May, June and July, the females shown on this figure were non reproductive and presumably yearlings (Tuttle, 1976b). Females were not weighed as pregnant until mid-May (606 weighed). Lactating females were en countered from early June to late July (386 weighed). In spite of the fact that yearling females were being compared with adult males from late May through late July, females were consistently heavier (P < .001) in all months except late May when males appeared to be slightly heavier (but not significantly so). Maximum weight of females came in late September (after which they migrated south from our study caves) at 12.61 g, whereas males weighed 1l.46 g in late October. Tuttle (1976b) reported that females at three summer caves in Tennessee weighed 1l.2 g, 1l.2 g and 12.5 g prior to migration, but did not give overwinter weight loss. In early April our females weighed 2.67 g less; males weighed in late March had lost 2.41 g. These weight losses amounted to 2l.2 percent for females, 2l.0 percent for males. Ideally, weight samples should have been taken at the hiber nacula since Tuttle (1976b) reported significant weight loss during migration. If our bats actually lost a significant amount of weight during migration (which is doubtful since the distance is relatively short), then the weight loss we reported would be somewhat high. We did weigh a sample of 29 hibernating bats of each sex at Marvel Cave on 1 April 1978 although we point out that few, if any, of the bats weighed elsewhere in this study hibernate at Marvel. Females weighed at Marvel were not significantly different from early April females weighed at summer caves; males weighed at Marvel, however, were significantly lighter (.001 < P < .01) than early April males at summer caves. Weights of both sexes were up in late April, down again in early May. Both fluctuated slightly-with a general upward trend-until late July and then took a sharp jump upward Both dropped during August. During September and October, prehibernation fat deposition was evident. Weights of pregnant females were notably greater Bal ;I 2 Great ... -------...,. Scoll Figure 21. Dispersal of male Myotis grisescens banded as juveniles during the maternity season at a maternity cave. than those of non-reproductive females, and lactating females were significantly heavier than non-lactating females only during June (Fig. 33). Myotis /ucifugus Weights were pooled for all 1,299 males weighed and all 559 females weighed (Fig. 34). Sample sizes for females were sometimes small, and weights were lacking for June and early July. Females outweighed males for all samples except early September, when only 14 were weighed; this result was not statistically significant (0.2 < P < 0.4). However, T-tests showed that females were significantly heavier (P< .001) only in April, early May and late September. It appears that the sexes differ little in weight in this species as contrasted with the other three for which we present weight data (as well as M. keenii, Caire et aI., 1979) Both sexes reached maximum weight in late October. Between then and March, females lost 2.16 g, males 2.25 g, or 23.6 percent for females and 25.5 percent for males. Fenton (1970) reported a 24.3 percent weight loss for hibernating females, 26.5 percent for males between October and April in Ontario. After an initial surge in weight in early April, female weights dropped in 31 May to a level which changed little until early September No doubt weights of pregnant and lactating individuals would be substantially higher if they were available. Fenton (1970) and Schowalter et al. (1979), for example, reported weights of pregnant in dividuals ranging over 10 g and as high as 14 g in late June. After an early April rise, male weights decreased drastically in late April then began a slow rise which peaked in late June. Male weights then dropped to the same level as female weights and remained there until September when both sexes began to deposit fat for hibernation. Pipistrellus subflavus All available weights were pooled for a total of 1,034 males, 287 females (Fig. 35) In addition, 22 pregnant females were weighed. In spite of small sample sizes, it was evident that females were consistently heavier than males. In fact, they were significantly heavier (P ranged from 02 < P < .05 to P < .001) for all samples tested. Peak weights were reached in late September No pip istrelles emerged in March, so the first spring weight samples are from early April. Females had lost 2.84 g (34.5 per cent), males 2.16 g (3l.0 percent). Male weights continued to decrease through late May, then fluctuated up and down, but gradually increased until late September when a sudden increase to maximum weight was recorded. Females showed a similar trend in the spring, with pregnant individuals only slightly heavier than non-pregnant bats in May. However, in early June pregnant females were 52 percent heavier than in late May; in late June they were 73 percent heavier than late May. Female weights declined during the re mainder of the summer before spurting upward in early September. Clearly recognizable juveniles were captured from late July through late September. During that period, female juveniles gained 2.43 g, males l.56 g (Fig. 36). During the same period, adult females gained only 2.08 g, males l. 78 g. In late July, young males weighed only 0.6 g less than adults; young females weighed l.17 g less. By late September, the juveniles of both sexes weighed 0.82 g less than adults. We doubt that juveniles equaled adults in weight by the time both entered hibernation in early October. Eptesicus fuscus Although we were unable to graph the seasonal weight changes of big brown bats, we secured enough pre hibernation weights (24 males; 16 females) and emergence weights (17 males; 9 females) to calculate weight loss during hibernation. Prehibernation

PAGE 36

Be t Figure 22. Movement of adult female Myotis grisescens banded at Saloon Cave (Meramec col ony), 1975-76 weights were 24.2 g for females, 22.5 g for males. Females lost 6.88 g (28.4 per cent) whereas males lost 7.6 g (33.8 per cent). The percentage for males would have been 31.8 percent if two suspiciously high weight values had been deleted. Myotis lelbii M. leibii bats were not encountered in sufficient numbers to provide infor mation on seasonal changes in weight, but weight data are presented here simply because so little data exists on the eastern population of this species. Thirteen males captured at Pilot Knob Mine on 29 March, 7 and 9 April had a mean weight of 4.15 g with a standard deviation of only 0.29. Three females from the same dates averaged 4.37 g with SD of 0 84. On 6 August, two males weighed 4.2 g and 4.0 g; two juvenile males weighed 3.8 g and 4.0 g. An adult male weighed 4.2 g on 19 May at Great Scott Cave. Summary and discussion The loss of weight during hiberna tion by the six Missouri cave bats studied is summarized in Table 17. One might have made three predictions about weight loss before viewing this table: (1) species of similar size would tend to have similar weight loss if strategies for hibernation were similar; (2) if species differed in size, the larger species would lose less weight percent agewise, assuming strategies were similar; (3) if strategies differed and sizes were similar, the species with the most efficient strategy would lose the least weight. Viewed in the perspective 32 of these predictions, we find some dif ferences easily explained and others not Myotis sodalis and M grisescens both hibernate in large clusters in cold sec tions of caves, a strategy we think should result in minimal weight loss (see Clawson et aI., 1980). The larger gray bats have lower weight loss (percentage wise) than the smaller Indiana bats, as predicted. Myotis lucifugus is in termediate in size between M. sodalis and M. gTisescens and hibernates singly or in small clusters in Missouri in the same temperature range as the former two species. Weight loss is intermediate. Myotis keenii hibernates in crevices, which are warmer than exposed cave ceilings (temperatures recorded near M. keenii hibernating in Great Scott Cave varied from 8C to 13C on 1 March 1979, at which time much colder temperatures existed elsewhere in the cave). As would thus be predicted, weight loss is significantly higher in this species. Pipistrelles hibernate singly in warm sections of caves. Again, weight loss is high, as would have been predicted. Eptesicus fuscus is a much heavier bat than the others, yet exhibits high weight loss. Big brown bats hiber nate in holes and crevices just inside cave entrances, where temperatures fluctuate widely. Although we have little additional data relating to the hiberna tion strategy of Eptesicus (see Barbour and Davis, 1969), it would appear to be a less efficient one than that of the other species studied. Eptwir.-"wo may compen sate for this apparent physiological shortcoming by emerging early in the spring to take advantage of large insects which have overwintered. On the other hand, our sample may not be represent ative of average weight loss in the species since many, if not most, winter at sites other than caves (Barbour and Davis, 1969). REPRODUCTION AND GROWTH Reproduction and growth of the six species of Missouri cave bats has been the object of numerous papers many of them summarized by Barbour and Davis (1969). More recently Humphrey et al. (1977a) have presented more detailed information for M. sodalis; Tuttle (1975, 1976a) and LaVal et al. (1977a) for M. gTisescens; and Humphrey and Cope (1976) and Schowalter et al. (1979) for M. lucifugus The only maternity sites studied in our project were M. grisescens materni ty caves. We obtained relatively little data from these caves because of the need to avoid possibly disastrous disrup tions to maternity efforts of this en-

PAGE 37

Figure 2 3 Movement of adult male Myotis grisescens banded at Saloon Cave (Meramec colony), 1975-76. Figure 2 4 Movement of adult male Myot is gri ses c ens banded a t Onyx Cave (Meramec colony), 1975-76 dangered species Thus, most of our data on reproduction and growth has resulted from our recording dates of capture and reproductive condition of adult females, and dates and weights of juveniles, whether trapped at caves or mist netted over streams. Myotis soda/is Obviously pregnant females were never encountered in this study, but un doubtedly we handled many such females in late April. Lactating in dividuals were captured as early as 11 June and as late as 6 July, but sample size was small. Post-lactating females were caught from 6 July through 6 August. The first volant young were captured on 12 July in the summer areas and 18 July at hibernacula, but few juveniles were captured until mid August. By early September it was no longer possible to separate juveniles from adults. Numbers of juveniles weighed were insufficient to provide useful information about growth Hum phrey et al. (1977a) reported lactating females from 25 June through 29 July 1974 and recruitment of juveniles from 20 July through 3 August They noted that reproductive events were about two weeks earlier in 1975 or about the same as we observed in Missouri. Observations of copulations were common in M. sodalis. The earliest observed was 16 September and the latest, 14 October. During late September and early October, dozens and sometimes hundreds of pairs were copulating noisily on ceilings just inside the entrance, especially at Great Scott Cave and Pilot Knob Mine. These copulating pairs seemed oblivious to our presence, and mating activity continued unabated while we were trapping at the entrances. Copulations also were ob served commonly in the bat trap bag MyoUs grisescens The reproductive chronology of gray bats during 1976 is outlined in Table 18. The data in this table from LaVal et al. (1977a) was discussed at length by those authors and was augmented by a table showing reproductive conditions of females in large samples trapped from 18 May through 17 August. Data from 1977-79 do not differ greatly Pregnant individuals were rec o gnized as early as 18 April 1979 and as late as 23 June 1978. Judging from the relatively large numbers of pregnant females observed in the third week of June 1978, it would appear that the period of parturition was extended in 1978 Even though a single newborn young was observed on 3 July 1976, hardly any pregnant females were captured after midJune of that year or in 1977 or 1979. Lactating fem a les were first caught during the first week of June of each year By 21 June 1978 some young were leaving their maternity cave in the evening exit flight. On 11 July 1978 juveniles were netted while foraging over a stream. Lactating females were captured as late as 7 August 1979. Based on aging of recaptured bats, we de c ided that we could not reliably separate juveniles from adults in Oc tober. In late September juvenile males weighed 1.43 g less than adults, and juvenile females weighed 2.17 g less 33 than adults Although we reported earlier (LaVal et aI., 1977a) that mean juvenile male weight equaled mean adult male weight on 14 October, we now doubt that this was true More likely, the mean weights were equal because of er rors in aging Juveniles in Tennessee had not equaled adult weights by the time they entered hibernation (Tuttle, 1976a) .. On 8 June, baby bats from newborn to several days old at Roaring Spring Cave weighed 5.17 g (Fig. 37). Three newborn bats weighed 2.8 g, 3 0 g and 3.2 g, similar to the 2.9 g mean reported by Tuttle (1975). Ten days later (18 June) the mean weight had risen only to 5 84 g as a result of continued births. However, on 30 June, with some young already volant, weight had increased sharply to 7.86 g At Bat Cave #2, a much smaller and more frequently disturbed colony, the baby bats only weighed 6 .98 g on 3 July of the same year A slower growth rate for this col ony would have been predicted from the findings of Tuttle (1975) All weights recorded after June (ex cepting 3 July, Bat Cave #2) were ob tained from cave entrance trapping. Weight increased through early July reflecting the fact that most of the young bats were on milk diets (although Saugey, 1978, noted that juveniles in Arkansas apparently were fed regur gitated food prior to weaning). After weaning weights dropped followed by fluctuating but mostly increasing weights until late September. Prior to 17 August weights of juvenile males were approximately the same as juvenile females (differences not statistically significant), so were pooled. From that date on, females were significantly heavier (P = .05 to P < .001). Tuttle (1976a) found that among different col onies the date on which statistically significant differences were first evident between the sexes varied from 26 July to 17 September. As noted by Tuttle (1975) use of col ony means, as opposed to banded cohort means, leads to an underestimation of average rate of individual growth, which he calculated at 57. 6 percent. This is due to continued parturition, the factor which resulted in little apparent weight gain between 8 and 18 June as noted above With few births after 18 June, weight gain data are more nearly representative of the actual growth rate The calculated growth rate for Roaring Spring Cave between 18 and 30 June was .17 g per day. With an ambient temperature of 18C on 18 June, growth rates at Roaring Spring were close to the value obtained by Tuttle (1975) at a colony of similar temperature, and very close to that predicted for a colony of

PAGE 38

TABLE 17.--Weight loss during hibernation of six species of Missouri bats. Prehibernation WEIGIiT LOSS Prehibernation We o j Actual Percent Actual Oail't Weight M M M F F F M sodalis 7.99 .014 2.22 27.8 27.8 2.47 .015 8.90 M'totis grisescens 11.46 .015 2.41 21.0 21.2 2.67 .016 12.61 1 uc Hugus 8 8 1 .015 2.25 25.5 23.6 2.16 .014 9.17 M'totis K eenii* 8.40 .019 3.00 35.7 31. 4 2.70 .017 8.60 PiQist rellu5 subflavus 6.96 .011 2.16 31.0 34.5 2.84 .014 8.22 EQtesicus fuscus 22.50 .041 7.60 33.8 28.4 6.88 .038 24.19 *Oata collected in this study, but previously published by Caire et al. (1979). TABLE 18.--Chronology of reproduction and growth of Myotis grisescens, 1976. Date 13 May June 1 8 June 30 June 1 July 3 July 10 July 14 July 16 July 5 August 14 October *See page 33 Event First palpable pregnancy observed First lactating female captured (first evidence of parturition) First faltering flights of juveniles observed F irst date on which majority of young capable of sustained flight First juvenile reached mean adult forearm length Last newborn young (last evidence of parturition) First juvenile flew from maternity site to another cave Mean juvenile forearm length equaled mean adult forearm length First juvenile netted while foraging Last lactating female caught (last evidence of nursing) Mean juvenile male weight equaled mean adult male weight (questionable because of possible errors in aging)* 34

PAGE 39

this size (3,600 young in 1976) Postflight weight gains of Meramec area juvenile bats were similar to those reported for cave 50 (Tuttle, 1976a). This would seem unexpected since cave 50 is much larger and warmer than Roaring Spring. The explanation is not that Roaring Spring is in some way a superior maternity cave. Rather, gray bats in Missouri simply weigh more than gray bats in Tennessee (M. D Tuttle, pers. comm.). We would anticipate that juvenile weights taken at a larger, warmer Missouri maternity cave would exceed those taken at Roaring Spring and exceed weights from any Tennessee cave. Copulations were rarely observed in M grisescens. We saw several pairs copulating, one as early as 20 September and one as late as 7 November. During censuses of hibernacula, no copulations were observed Myotis keenii Pregnant females were encountered only once on 11 May 1976. Lactating females were netted from 10 June to 11 August, suggesting that births are well spaced in this species. Kunz (1971) found lactating females only from 23 June to 28 July in Iowa. We netted our first juvenile in the summer range on 16 July, but juveniles had been trapped at hiber nacula as early as 10 July. Kunz (1971) noted that in Iowa no young bats were flying until 23 July. Reproductive events in this species may be earlier at more southerly latitudes although differences in weather conditions between the study periods could have accounted for the discrepancy observed. Although we z:emembered having seen M keenii copulating in the fall in the trap bag, our notes do not record such an observation. We did see a copulating pair at Great Scott Cave on 10 April 1976, however. Myotis /ucifugus We trapped pregnant females on 21 May 1978. On 10 June 1977 we had netted pregnant bats; lactating bats were taken on the same date Post lactating females were captured as early as 6 July and as late as 10 August. Volant young were trapped at caves as early as 23 July Copulations were noted only once, in the bat trap at a hiber naculum on 16 September 1976 See Humphrey and Cope (1976) for a review of reproduction in this species. Pipistrellus subflavus Pregnant females were trapped as early as 7 May and as late as 20 June. They were mist netted from 26 May through 29 June. Lactating bats were netted from 5 June through 24 July and post-lactating females from 16 July through 13 August. Juvenile bats were netted as early as 6 July, but the first ar rival of a juvenile at a hibernaculum was not until 5 August. Copulations were observed only once, on 5 October at a hibernaculum. Juveniles were captured in sufficient numbers to plot growth from late July through September (Fig. 36). During this period, females gained 2.43 g (a 49 percent increase), whereas males gained only 1.56 g (a 34 percent increase). The mean for females is perhaps unreliable due to small sample size. In late September, juvenile females were .82 g lighter than adult females; males were also .82 g lighter. These dif ferences are not statistically significant due to small sample sizes and high variance. Barbour and Davis (1969) sum marized what little is known about growth and reproduction in this species. During the course of this study we located two maternity groups in caves (Humphrey et aI., 1977b). Whitaker and Mumford (1972) as well as Cope and Humphrey (1972) observed colonies in Indiana barns. Eptesicus fuscus Pregnant big brown bats were cap tured only on 10 June 1977. Lactating females were netted on 10 June, 2 July and 5 July. Post-lactating females were caught on 12, 17 and 22 July Juveniles were netted on 12 and 17 July, but none was trapped at hibernacula until 1 August. No copulations were observed Two small groups of adults and juveniles were observed on cave ceilings. See Bar bour and Davis (1969) for details of reproduction in E. fuscus. Holton Figure 25. Move m en t of adult male My otis g risescens banded at B ec k Mauss and M o l es caves (Lake of th e Ozarks col o n y ) B ec k and Mauss are mate rnity caves. 35

PAGE 40

d \t Myotis grisescen5 banded at Boone, ;!d count' ",'on,) Ho,ton \" ",.""nit, ca'ie.

PAGE 41

Figure 27. Location of all currently active Myotis grisescens maternity caves (triangles). Stars represent hibernacula. 37

PAGE 42

Figure 2 8 Location of all currently active summer Myotis grisescens caves (triangles) except maternity. Stars represent hibernacula. 38

PAGE 43

KEY TO SYMBOL S Hibernating caves Summer recapture sites Figure 29. Dispersal of adult Myotis lucifugus from hibernacula to summer and transient sites The solid circle in Illinois is based on recapture of two bats banded at a maternity colony and recaptured at a hibernaculum. The other solid circles represent capture sites of single bats 39

PAGE 44

Figure 9 0 Ba t activit y at G reat Scott Cave, 1 975 79, as dete rm i n e d by shown In eac h p air o f ba r grap h s, m a l es are o n the l e f t fema l es o n th e entr a nce trapping. Only the five species listed on t h e figure l egen d are ri ght. Ca t c h r ates l ower t h a n five bats pe r h our a r e n ot sh own ACTIVITY AT GREAT SC.OTT CAVE 1975 AND 1977 500 450 P' pistnz. \Ius 5ubflavus ""tobs luc. \tu9uS 400 til Ml::otts 9r15Cl.5cR. 20 JUNE 7 AUG. 19DC.T. 1976 500 450 400 B 350 A. 1 S 300 p E 250 R H 2.00 0 U R. 150 100 50 0 I MAR. 10MI t 30A P R 5M(l.Y 2.2 2 8 {O JUNE JUNf. 7 AUG. 2.B AUG. 22 sm. 40

PAGE 45

1978 500 450 400 B A. 350 T S 300 p E 250 R \-l 2.00 0 LJ R 150 100 50 0 2.7 APR. 2l MIl,Y 22 JUNE 22 JULY 22. ALIG. 2.l 190C'\. 1979 500 450 B A 360 T 5300 p E.2.50 R \-\ 200 o u R 150 \00 50 41

PAGE 46

G R A H S WEIGHTS OF ADULT HYOTIS SODA LIS t, 8 Ii) <0 .,. .,. 7 (\I f'-. 0) /<> ... .,. (\I .,.JIj6 .,. is) is) f'-. <0 (\I (\I .,. (\I (\I 0) f'-. is) f'-. <0 0) 0) (\I <0 liS) Ii) (\I \ <0 f'-.I 'f' I I I March Apri l May June July August September October November Figur e 31. Weights of adult Myotis sodalis Males, with sample sizes, are represented by a solid line with pluses; females, with sample sizes, by a dashed line with diamonds No females weights are shown for late Mayor early June. WEIGHTS OF ADULT HYOTIS GRISESCENS 13 (r) is) , 12 Ii) (\I (\I :" G I \ t R I \ I .,. A II , (r) H \ S is) is)/ <0 '3.t (r) (r) ... f'-. / ...... IJ -/ (I,) t ) I (r) f'-. LJ) IB I ) is) 0) '
PAGE 47

UEISHTS Of REPROOUCTIVEL Y ACTIVE HYOTIS SRISESCENS 15r---------------------. 6 1:1 R A M S Ii II ' 9 f'-(f) / ..... G / /1 R / A 8 H / "
PAGE 48

G R A M S YEIGHTS OF ADULT PIPISTRELLUS SUBFLAVUS 9 8 7 6 5 (i). r:f VI t April May June July August September October FigU1'e 35. Weights of adult Pipistre llus subflavus. Males, with sample sizes, are represented by a solid line with pluses; non-reproductive females, with sample sizes, by a dashed line with diamonds; pregnant females, with sample sizes, by a dotted line with squares G R A S mrurs OF JUVOOLE PlPISTR8..lUS rulfUVUS JULY ID / / t' / (I) I (I) ID _I }.--(I) 6 / ID N 1 I I I I (I) AUGUST SEPTEMBER 44 Figur e 36. Weights of juvenile Pipistrellus su bfla vus late July through late September. Males, w i th samp l e sizes, are represented by a solid line with pluses; fema l es, with sample sizes, by a dashed line with diamonds

PAGE 49

WEIGHTS OF JUVENILE MYOTIS GRISESCENS 11.------------------------------------------------------, G R A M S ....... (f): .0 (g. 0). c1 ...... ...... N EJ. 11) .a 00 P (g .51) (g 00 0' ...... d 00 J> ,. 00 / / LI)/ (g I (f) ..... NI .EJ a (D (g 00 00 N Sept Figure 37 Weights of juvenile Myotis grisescens born in June, 1976. The combined weights of both sexes, with sample sizes, are repre sented by a dotted line with sq u ares; males, with sample sizes, are represented by a solid line with pluses; females, with sample sizes, are represented by a dashed line with diamonds. 45 Figure 38. Squared-ear Myoti s sodalis cap tured at Pilot Knob Mine, Iron County.

PAGE 50

THE "SQUARED-EAR" ANOMALY To our knowledge, there are no reports in the literature of Myotis (or other bats) with abbreviated ears. In this study we encountered 15 bats with ears of half the normal length, sometimes with two square corners, sometimes gently curved as opposed to being pointed (Figs 38 and 39). Both ears were identical in every case, and there was no sign of scar tissue along the edge (several colleagues have suggested to us that the anomaly results from frostbite). We believe the squared-ear" anomaly is hereditary since it also occurs in tropical bats. In Costa Rica we captured one female M. albescens and one male M nigricans with squared-ears (both specimens are on deposit in the Museum of Natural History, University of Kan sas). In Missouri the anomaly was observed in the following male bats: nine M. lucifugus, three M sodalis, one M keenii and one M. grisescens. We cap tured only one female with this anomaly in Missouri, aM. grisescens. It would ap pear that "squared-ears" are sex-linked, based on our sample, and are much more common in M. lucifugus than in other species. The anomaly was not observed in bats of other genera in Missouri and we have not seen it in bats of other genera and families among many thousands handled in other areas of the U S., Mexico, Central America and Mrica (including thousands of Myotis tricolor) We cannot rule out the possibility of frostbite damage to ears, but we think it fails to account adequately for our observations. The survival value of squared-ears is certainly not evident. If it in any way interferes with the reception of ultrasound echos, then it probably would be subjected to negative selection, which would help account for its rarity (for example, it was encountered in only 0.015 percent of all M sodalis handled). STRATEGY FOR MANAGEMENT OF CAVE BATS IN MISSOURI Management of cave bats in Missouri must confront two interdepen dent issues : protection of habitat and education of the public; both must be undertaken if cave bats are to survive as a viable segment of our fauna. These issues are dealt with at some length in the Gray Bat Recovery Plan (in prepara tion), the Indiana Bat Recovery Plan (1976; to be revised shortly) and by Humphrey (1978) and Tuttle (1979a). Protection of habitat includes both winter habitat (caves) and summer habitat (caves for gray bats; foraging habitat for all species). Protection of bat caves As a result of the impetus provided by this study, Missouri has a substantial head start over most other states toward protecting cave habitat used by endangered bats. Table 19 shows levels of protection afforded to various caves in the state. A few caves not included on the list seem to be adequately protected by private owners, but we were refused permission to enter so could not assess their status. Many caves on the list are not protected adequately, but at some of these, efforts are being made to increase the level of protection. Still, it will take several years to bring all caves to the first and second protection level. Third and fourth level caves are in danger of losing their populations if additional steps to protect them are not taken soon. Of the 29 caves listed, 12 now have bat gates (Figs. 40 and 41), and two are fenced (Fig. 42). Nineteen are signposted, most of them with combina tion warning-interpretive signs supplied by the Missouri Department of Conser vation. For sign wording as well as details of gates and fences, see the Gray Bat Recovery Plan (which will be available soon from the U.S. Fish and Wildlife Service). Thirteen of the caves are active hibernacula for Indiana bats, affording some level of protection to almost all the M. sodalis hibernating in Missouri The four gray bat hibernacula listed give protection to virtually all M. grisescens in Missouri plus some from adjacent states. Nine active maternity caves are protected, leaving an equal number with no effective protection at this time. Fifteen active transient caves are protected, but most of these were protected for some other reason. Ten transient caves are unprotected, plus 11 more active caves of unknown status. Only two bachelor caves are protected, leaving at least eight without protection. Twelve protected caves are owned (plus Figure 89. Squared-ear Myotis nigricans captured at Monteverde, Costa Rica. 46

PAGE 51

.... ..., TABLE 19.--Missouri bat caves with some level of protection. M. sodalis M. grisescens Cave Counti H S Devi l's Icebox Boone Carrol Camden x* Mauss Camden Moles Camden River Camden x* Coa 1 ban k Ca rter Onyx Crawford x Saloon Crawford Bear Frankl in x Copper Hollow Franklin x Fisher Franklin Roaring Spring Franklin Beck Hickory Bl ackwe11 Hickory Pil ot Knob Mi ne Iron x x Coffin Laclede x Bat # 1 Mi 11 er Bat Ozark Brooks Pul as ki x Inca Pulaski x Ryden Pulaski x Bat Shannon x Chimney Shannon x Powder Mill Shannon x Round Spri ng Shannon Ma rve 1 Stone Tumbling Creek Taney Great Scott Washington x x Scotia Washington x @Frequent collapses at mine discourage *Abandoned by bats for this use. S = Summer B Bachelor Use H = Hibernaculum T = Transient Use M = Maternity Use H B T Gated Fenced Si gned x* x x* x x x x x x* x x x* x x x x x x x x x x x x x x x* x* x x x x x x x x x x x x x x x x x x x x x x x x x x x* x x x x x* x* x x x x x x x x x x x visitation, but threaten bats. Levels of Protection 1. Secure from disturbance 2. Minimal disturbance Protection Ownershil2 Level State 2 Private 3 Private 1 State 2 State 3 Federal 2 Federa 1 1 Federal 4 State 1 State 1 State 3 State (lease) 3 Federal 2 Federal 2 Private l@ State 1 State 1 Federa 1 2 Federal 2 State 3 State 2 Private 3 Federal 2 State 1 Federal 3 Private 2 Private 2 State 1 Private 2 3. 4. te di sturbance Serious disturbance

PAGE 52

one leased) by various state agencies, nine by federal agencies and seven are privately owned. The federal govern ment is currently negotiating to acquire one additional cave and to obtain a cooperative agreement to protect another. It is also negotiating for an easement on Bat Cave (Shannon Co.), which would allow it to construct a gate there. Both state and federal agencies plan to acquire and protect additional bat caves over the next few years. Although only fee acquisition can give an agency complete control over a cave, easements, leases and cooperative agreements may prove to be the answer when an owner does not wish to sell his cave, but will not or cannot protect it himself. The Missouri Department of Conservation has had good luck in buy ing caves it wanted to protect. Both easements and leases have been used successfully by various agencies, and two cooperative agreements are being negotiated. Many landowners who live in rural areas appreciate the need to pro tect bats, and a sizable number of these have agreed to protect their bat caves from disturbance. Often a resident land owner can provide more effective pro tection than any government agency. Note that most of the caves in Table 19 are used by other cave bats as well as the two endangered species and that management efforts intended to en hance cave habitat for gray and Indiana bats will also benefit the other species Protection of summer foraging habitat As pointed out by Humphrey (1978), Tuttle (1979a) and the recovery plans, protecting caves will be to no avail if summer foraging habitat is so degraded that it will not produce a food supply suf ficient to allow bats to increase or sus tain their populations. Although we know that pesticides are causing mor tality among gray bats in one part of the state (Clark et aI., 1978), we don't know how serious or how widespread this probl e m i s Habitat is also degraded by c hann e lization of streams, construction of r ese rvoirs, clearing for agriculture and urban developments, and pollution of air and water. All of these phenomena, symptoms of "progress," are occurring in Missouri as a result of population growth, development of technology, and demand for agricultural products. As pointed out by Humphrey (1978), it is unreasonable to expect the course of progress to be altered substantially in deference to endangered bats. Nevertheless, public agencies have become sensitive to the impact of their projects on endangered species in the last three years. Such powerful agencies as the Soil Conservation Service and Army Corps of Engineers recently have begun to investigate the impact of water control projects on endangered species of bats in Missouri and adjacent states where many of our bats spend part of the summer. In recent months these and other federal agencies as well as private corporations which must obtain federal permits for construction of utility rights of-way, plants, dams, etc., have ex pressed increased willingness to modify projects in order to reduce the impact on the foraging habitat of endangered bat species. We can expect to see a gradual improvement in pollution levels as more long-lived pesticides are banned and various federal and state anti-pollution laws are enforced. However, most of Missouri remains in private ownership, and until we see some fundamental changes in land use ethics coupled with a decrease in 'iemand for agricultural pro ducts, we can anticipate that individual landowners will exploit their land to the detriment of bat foraging habitat. There is no way government conservation agencies can acquire enough foraging habitat to guarantee the survival of bat populations. Instead, the byword must be vigilance against threats to major areas of foraging habitat, cooperation and encouragement of private and public landowners willing to modify pro jects to protect bat foraging habitat and continuing efforts to convince these peo ple that bats are worth protecting. Public education It is a well-known fact that bats have a bad public image in much of the world. Bats have been accused of many objec tionable acts, most of them based on oldwives tales and unfounded. More serious is the accusation that bats are major vectors of rabies. Although rabies has been found in most species of U.S. bats (Con stantine, 1979), it is uncommon and epidemics are rare. Real public health problems usually result from rabies in dogs, foxes and skunks rather than bats In Missouri since 1971, only 15.6 percent of 218 bats tested were rabid, and these 218 tested were bats suspected of being rabid. We have never seen a rabid bat. Only one case of a person being bitten by a rabid bat in Missouri came to our at tention during the last four years. The public, therefore, must first be con vinced bats are not bad, then convinced they are good. The facts are available to demonstrate the value of bats to the pUblic. For example, the 500,000 surviv ing gray bats in Missouri eat approx imately 400 tons of insects, many of them crop pests, in a single year. Gray bats account for only a small percentage of all the bats in Missouri. There are a number of ways such information can be disseminated to the pUblic. We have already mentioned signs which in Missouri carry interpretive messages explaining why the cave is protected. Naturalists who work for various public agencies in parks and Figure 40. Vertical bat gate installed at entrance of Great Scott Cave, Washington County, an Indiana bat hibe rnaculum. 48

PAGE 53

Figure 41. Cage-type gate protecting entrance to Bear Cave, Franklin County, an Indiana bat hibernaculum. Figure 42. Chain-link fence protecting Coffin Cave, Laclede County, a gray bat hibernaculum. Note that the top wires are not barbed to reduce the danger of bats becoming entangled. 49 recreation areas or who present pro grams in schools, other institutions and at organizational meetings can make a sizable contribution. These people, conservation and forestry agents, teachers and others with public contacts can make use of brochures Oike the "Fact Sheet on Bats" distributed by the Missouri Department of Conservation) and a slide show now being prepared by the Department. Naturalists employed by the Department have already begun an attempt to reverse the public image of bats. Articles in popular magazines can be helpful (for examples see LaVal and LaVal, 1978; Tuttle 1979b). Our studies have been the object of two TV news spots, which were good publicity for bats, but more comprehensive radio and TV coverage is needed. One of the most readily identifiable segments of the bat-conscious public, one that needs special emphasis in inter pretive efforts, is the ever-growing body of sport cavers or spelunkers. Many of these people belong to caving clubs or grottoes of the National Speleological Society (NSS). In Missouri, each grotto is represented in a statewide organiza tion, the Missouri Speleological Survey (MSS). Various publications are prepared by the NSS, MSS and the grot toes. All of these organizations are conservation-oriented, and leaders have expressed a desire to assist in our bat conservation efforts. These leaders and the membership they represent have a cumulative knowledge of caves that is vast, and their advice and cooperation should be sought repeatedly throughout our recovery efforts. Already, their help has been invaluable to us. Their members are probably best reached through their own publications, but our brochures and slide shows can be made available to them as well. The public, even if convinced of the basic benefits of having bats around, will lack the knowledge to attract bats in case there are few or no bats living nearby. Others would be unwilling to tolerate the mess sometimes created by bats roosting in their house or other buildings. We are experimenting with two possible solutions to these problems. One is the erection of bat roosting boxes. We have erected 53 of two different designs, but to date no bats have been attracted. Other biologists are also experimenting with bat boxes, and even tually this technique might prove workable. Meanwhile, in the summer of 1979 a group of private citizens designed and built a larger structure (Fig. 43) which they called a "bat refuge," at the Tea Lakes Wildlife Area near Rosebud, Missouri. The structure, which had been sprayed with a mixture of bat guano and

PAGE 54

Figur e 49. "Bat refuge" at Tea Lakes Wildlife Area near Rosebud, Gasconade County 50 water, promptly attracted 70 little brown bats (Myotis lucifugus) from a nearby picnic shelter roof The Missouri Department of Conservation is cur rently implementing plans to design, build and field test additional bat refuge structures, in order to be able to offer a viable alternative to extermination or exclusion of problem bats in buildings. In a study in Canada, Barclay et al. (1980) determined that the most effec tive means of controlling bats in buildings is to seal their access holes before bats return from hibernating caves in March and April. Attempts to exterminate bats by use of pesticides and other chemicals not only expose the occupants of the building to the toxic chemicals, but also have proven to be largely ineffective in discouraging bats, which tend to return after repeated ex termination efforts, most likely at tracted by the odor of urine and guano from the roost. Further, the use of pesticides may aggravate the situation health officials are seeking to eliminate because dying bats fall near treated buildings and thus are more likely to be encountered by humans. Also, sublethal doses of pesticides may activate latent viral infections, thereby increasing the incidence of rabies (Kunz et aI., 1977). UNPUBLISHED DATA In this study we amassed far more data than could be covered in this report. This data, which covers areas of study not included herein, is on deposit with the Missouri Department of Con servation It will be available to all biologists and wildlife managers. Some of the material was used in preparing the Gray Bat Recovery Plan.

PAGE 55

LITERATURE CITED Anthony, E. L. P., and T. H. Kunz 1977. Feeding strategies of the little brown bat, Myotis lucifugus, in southern New Hampshire. Ecology, 58:775-786 Barbour, R. W., and W. H. Davis. 1969. Bats of America. Univ. Press Kentucky, Lexington, 286 pp Barclay, R. M R., D. W. Thomas, and M. B. Fenton. 1980. Comparison of methods used for controlling bats in buildings J Wildl. Mgmt 44 : 502-506. Belwood, J J. 1979. Feeding ecology of an Indiana bat community with emphasis on the endangered Indiana bat, Myotis sodalis. Unpubl. M. S thesis, Univ. Florida, Gainesville, 103 pp Belwood, J J and M. B. Fenton. 1976. Variation in the diet of Myotis lucifugus (Chiroptera : Vespertilionidae) Canadian J. Zool., 54: 1674-1678 Black, H. L. 1974. A north temperate bat community. Structure and prey populations J Mamm 55:138-157 Black, H. L 1979 Precision in prey selection by the trident nosed bat (Cloeotis percivali) Mammalia, 43 : 53-57. Buchler, E. R. 1976. Prey selection by Myotis lucifugus (Chiroptera: Vespertilionidae). Amer. Nat., 110:619-629 Caire, W R. K. LaVal, M. L. LaVal, and R. L. Clawson 1979 Notes on the ecology of Myotis keenii (Chiroptera : Vespertilionidae) in eastern Missouri. Amer. Midland Nat., 102:404-407 Clark, D R., R K. LaVal, and D : M Swineford. 1978. Dieldrin-induced mortality in an endangered species, the gray bat (Myotis grisescens). Science, 199 : 1357-1359. Clawson, R. L., R. K. LaVal, M. L. LaVal, and W. Caire. 1980. Clustering behavior of hibernating Myotis sodalis in Missouri. J. Mamm., 61:245-253. Constantine, D. G. 1979. An up-dated list of rabies-infected bats in North America. J Wildl. Disease, 15:347-349 Cope, J. B. and S. R. Humphrey 1972. Reproduction of the bats Myotis keenii and Pipistrellus subflavus in Indiana. Bat Res. News, 13:9-10. Cope J. B., and S. R. Humphrey. 1977. Spring and autumn swarming behavior in the Indiana bat, Myotis s odalis J. Mamm., 58:93-95. Easterla, D. A., and L. C. Watkins. 1969. Pregnant Myotis sodalis in northwestern Missouri. J. Mamm., 50:372-373 Easterla, D. A., and L. C. Watkins. 1970 Breeding of Lasionycteris noctivagans and Nycticeius humeralis in southeastern Iowa Amer Midland Nat., 84: 254-255. Elder W H. and W J Gunier. 1978 Sex ratios and seasonal movements of gray bats (Myotis grisescens) in southwestern Missouri and adjacent states. Amer. Midland Nat., 99:463-472. Engel, J M., et al. 1976. Recovery plan for the Indiana bat. U.S Fish and Wildlife Service, Washington D C 34 pp. Fenton, M B. 1969. Summer activity of Myotis lucifugus (Chiroptera : Vespertilionidae) at hibernacula in Ontario and Quebec. Canadian J Zool., 47:597-602 Fenton, M. B. 1970. Population studies of Myotis lucifugus (Chiroptera : Vespertilionidae) in Ontario. Life Sci. Contrib Royal Ontario Mus., 77: 1-34. 51 Fenton, M. B., N. G. H. Boyle, T M. Harrison, and D J Oxley. 1977. Activity patterns, habitat use, and prey selection by some African insectivorous bats. Biotropica, 9:73-85 Gunier, W. J., and W H Elder. 1972. New records of Myotis leibii from Missouri Amer. Midland Nat. 89:489-490. Guthrie M J 1933. Notes on the seasonal movements and habits of some cave bats. J. Mamm 14:1-19 Hall, E. R., and K. R. Kelson. 1959. The mammals of North America The Ronald Press Co. New York, 1:1-546 + 79. Hall, J. S 1962. A life history and taxonomic study of the Indiana bat, Myotis sodalis. Reading Public Mus. and Art Gallery, Sci. Publ. 12: 1-68. Hall, J. S., and F. J Brenner. 1968. Summer netting of bats at a cave in Pennsylvania J Mamm., 49:779-781. Humphrey, S R 1975. Nursery roosts and cOp1munity diversity of Nearctic bats. J. Mamm., 56:321-346 Humphrey, S. R. 1978. Status, winter habitat, and management of the endangered Indiana bat, Myotis sodalis. Florida Sci 41:65-76 Humphrey, S. R., and J. B. Cope 1976 Population ecology of the little brown bat, Myotis lucifugus in Indiana and north-central Kentucky Spec. Publ., Amer. Soc. Mamm 4:1 81. Humphrey, S. R., and J B. Cope 1977. Survival rates of the endangered Indiana bat, Myotis sodalis J. Mamm., 58:32-36. Humphrey, S R., A. R Richter, and J. B. Cope. 1977a. Summer habitat and ecology of the endangered Indiana bat, Myotis sodalis. J. Mamm 58:334-346 Humphrey, S. R., R. K. LaVal, and R. L. Clawson. 1977b. Nursery populations of P ipistrellus subflavus (Chiroptera, Vespertilionidae) in Missouri. Trans. Illinois Acad Sci., 69:367 Keen, R., and H. B. Hitchcock. 1980 Survival and longevity of the little brown bat (Myotis lucifugus) in southeastern Ontario. J. Mamm., 61:1-7 Kunz, T. H. 1971. Reproduction of some vespertilionid bats in central Iowa. Amer. Midland Nat., 86:477-486. Kunz, T H 1974. Reproduction, growth and mortality of the vespertilionid bat, Eptesicus juscus, in Kansas. J Mamm. 55:1-13 Kunz, T. H., E. L. P Anthony, and W T Rumage III. 1977. Mortality of little brown bats following multiple pesticide applications. J. Wildl. Mgmt. 41: 476-483. LaVal, R. K. 1970. Banding returns and activity periods of some Costa Rican bats. Southwestern Nat., 15:1-10. LaVal R K., and H. S. Fitch. 1977. Structure, movements and reproduction in three Costa Rican bat communities. Occas. Papers Mus. Nat. Hist., Univ. Kansas 69: 1-28 LaVal, R K., R. L. Clawson W. Caire, L R. Wingate, and M. L LaVal. 1977a. An evaluation of the status of myotine bats in the proposed Meramec Park Lake and Union Lake project areas, Missouri. U S. Army Corps Engineers, St. Louis Dist 136 pp. LaVal, R. K., R. L. Clawson, M L LaVal, and W. Caire 1977b. Foraging behavior and nocturnal activity pat-

PAGE 56

terns of Missouri bats, with emphasis on the endan gered species Myotis grisescens and Myotis sodalis. J. Mamm., 58:592-599. LaVal R. K., and M. L. LaVal. 1978 Fact sheet on bats. Missouri Conservationist, 39:24-26. LaVal, R K., and M L LaVal. 1979. Notes on reproduction, behavior and abundance of the red bat, Lasiurus borealis. J. Mamm., 60:209-212. Mumford, R. E., and J. O. Whitaker 1975. Seasonal activity of bats at an Indiana cave. Proc. Indiana Acad. Sci., 84:500-507. Myers, R. F 1964 Ecology of three species of myotine bats in the Ozark Plateau. Unpubl. Ph.D. dissert., Univ. Missouri, Columbia, 210 pp. Paradiso, J. L., and A. M. Greenhall. 1967. Longevity records for American bats. Amer. Midland Nat., 78:251-252. Ralls, K. 1976 Mammals in which females are larger than males. Q. Rev. BioI., 51:245-276. Ross, A. 1967. Ecological aspects of the food habits of insectivorous bats. Proc. West. Found. Vert. Zool., 1:205-263. Saugey, E. A. 1978. Reproductive biology of the gray bat, Myotis grisescens, in north-central Arkansas. Unpubl. M.S. thesis, Arkansas State Univ., State College, 93 pp. Schowalter, D. B. and J. R. Gunson. 1979. Reproductive biology of the big brown bat (Eptesicus fuscus) in Alberta. Canadian Field Nat., 93:48-54. Schowalter, D. B., J. R. Gunson, and L D. Harder. 1979. Life history characteristics of little brown bats (Myotis lucifugus) in Alberta. Canadian Field Nat., 93:243-251. Schwartz, C. W., and E. R. Schwartz. 1959. The wild mammals of Missouri. Univ. Missouri Press, Columbia, 341 pp. Tuttle, M. D 1975. Population ecology of the gray bat (Myotis grisescens): Factors influencing early growth and development Occas. Papers Mus. Nat. Hist., Univ. Kansas, 36:1-24. Tuttle, M D. 1976a Population ecology of the gray bat (Myotis grisescens): Factors influencing growth and survival of newly volant young. Ecology, 57:587-595. Tuttle, M. D. 1976b Population ecology of the gray bat (Myotis grisescens): Philopatry, timing and patterns of movement, weight loss during migration, and seasonal adaptive strategies. Occas. Papers Mus Nat. Hist., Univ. Kansas, 54:1-38. Tuttle, M. D 1979a. Status, causes of decline, and manage ment of endangered gray bats. J. Wildl. Mgmt., 43:1-17. Tuttle, M. D. 1979b. Twilight for the gray bat. Nat. Parks Cons. Mag., 53:12-15. Tuttle, M. D., and D. E. Stevenson. 1977. An analysis of migration as a mortality factor in the gray bat based on public recoveries of banded bats. Amer. MidI. Nat., 97:235-240. van Zyll de Jong, C G. 1979. Distribution and systematic relationships of long-eared Myotis in western Canada. Canadian J. Zool., 57:987-994. Vineyard, J. D. 1979. Cave files report for January 1979. Missouri Speleo. Survey Liaison, 19:6. Whitaker, J. O. 1972. Food habits of bats from Indiana. Canadian J. Zool., 50:877-883. Whitaker, J. 0., and R. E Mumford. 1972. Notes on occur rence and reproduction of bats in Indiana. Proc. Indiana Acad. Sci., 81:376-383. ACKNOWLEDGMENTS A study of this scope would have been impossible without the assistance of many persons as well as a number of governmental agencies. First we would like to thank those who actually worked with us in the field for periods ranging from three to 18 months. Listed alphabetically, these include Virgil Brack, William Caire, Richard Clawson, Gene and Treva Gardner, Brian Miller, Richard Orr, Lloyd Wingate and Terry Zinno Other Department of Conserva tion employees who were especially helpful to us include Allen Brohn, Ken Sadler, John Wylie, James H. Wilson, Gordon Maupin, Joe Werner, Clifford Caldwell, Steve Sheriff and last, but definitely not least, Charles and Elizabeth Schwartz. Clawson was particularly helpful in permitting us to use trapping data from Onyx Cave, fall 1979, and in criticizing an early draft of this paper. Virgil Brack also was kind enough to review a draft of the paper. In addition, inumerable volunteers assisted us in the field at various stages of the study. Tom Baskett and Sandy Clark of the University of Missouri were indispensable while the study was associated with the University. Jerry Vineyard of the Department of Natural Resources served as our invaluable liaison with the Missouri Speleological Survey. Many cavers assisted us, but we single out Scott House, Jim Zollweg and Don Rimbach. John Brady of the Army Corps of Engineers, St. Louis District, was extremely helpful to us, as were Larry Visscher and Don Metz of the U.S. Fish and Wildlife Ser vice, Kansas City Area Office. We extend our grateful thanks to landowners statewide who allowed us to visit caves or net streams on their prop erty. We would like to single out Maude Scott of Sullivan; 52 the Mauss family of Climax Springs; the Mullen family of Steelville; the Wellman family of Lebanon; Leo Drey of St. Louis; Silver Dollar City; Alton Box Company and the Diabetes Association of Columbia. The following bat biologists were especially helpful to us: Merlin Tuttle discussed the biology of gray bats with us at great length, allowed us to use some of his unpublished data and read an early draft of this paper. Steve Humphrey talked with us many times about the biology of Indiana bats and joined us in the field to help organize the study in 1975 and to census hibernating bats in 1976. Richard Myers pro vided us with useful discussions of Missouri cave bats, allowed us to use some of his unpublished data and guided us into the chambers of Marvel Cave used by hibernating gray bats. John Bowles and his students helped us band thousands of M. sodalis and allowed us to use some of his unpublished data on M. sodalis in Iowa. Wilbur Gunier took us to Marvel Cave and shared some of his knowledge of gray bats in Missouri. Jack and Janet Twente worked in the laboratory with gray and Indiana bats we furnished to determine more about the phenomenon of hibernation, which is so important to the successful management of these endangered species. Don Clark carried out studies of pesticide levels in bats, guano and insects that we furnished to him. Tony Beason of KY3 News, Springfield, helped us bring our message of bat conservation to southwest Missouri viewers. Ken Carroll was our able and interested helicopter pilot in 1979. Tom Aley allowed us to use his Ozark Underground Laboratory for publicity purposes The St. Louis Chapter, Conservation Federation of Missouri,

PAGE 57

designed and built the "bat refuge" at Tea Lakes Wildlife Area and invited us to view the refuge and use the plans. The following government agencies assisted us in various ways. The Army Corps of Engineers (St. Louis District) financed the first 18., months of the project through the University of MissoUrl and took steps to protect several bat caves on their land. The Kansas City District cooperated to protect two bat caves threatened by Truman Reservoir. The U.S. Fish and Wil dlife Service (Denver Regional Office; Kansas City Area Office) provided constant assistance and paid two-thirds of the cost of the 1977-80 study through a cooperative agreement. The National Park Service, Ozark National Scenic Riverways, has taken steps to protect several bat caves on its land, has recently acquired Chimney Cave (a gray bat hibernaculum) and is negotiating an ease ment to protect Bat Cave (a hibernaculum for both en dangered species). The U.S. Forest Service, Mark Twain National Forest, has agreed to protect bat caves on its land and is sponsoring a survey of all caves it owns. The Missouri Speleological Survey, although not a state agency, has made available to us its cave catalog and its cave files, among other things. The Missouri Department of Natural 53 Resources has cooperated in protecting bat caves in state parks Finally, the Missouri Department of Conservation was a co-sponsor of the 1975-76 portion of our study and full sponsor of the 1977 -80 portion (financed in part under en dangered species project SE-l, "Improving the status of en dangered species in Missouri," through an agreement with the U.S Fish and Wildlife Service) Department employees who initiated the study, provided support at all levels while it was in progress and have been busily implementing its recommendations, deserve the highest possible commenda tion. To date, the Department has purchased or leased seven bat caves, gated five and fenced one. One important gray bat cave in Miller County was obtained specifically to protect gray bats even before the gray bat was listed as en dangered As a result of the Department's efforts, Missouri has forged far into the lead among all states in its efforts to protect endangered bats. This study was carried out under the authority of En dangered Species Permit No. PRT-8-31-C and a cooperative agreement between the U.S Fish and Wildlife Service and the State of Missouri.

PAGE 58

CZ'errestrial Vo.S THE MISSOURI DEPARTMENT OF CONSERVATION TERRESTRIAL SERIES The Terrestrial Series is published by the Missouri Department of Conservation to make available the results of original investigations of the terrestrial environment of Missouri. It was originally designated the Technical Bulletin Series. Numbers published to date include: 1. The Ecology and Management of the Wild Turkey in Missouri. Paul Dalke, A. Starker Leopold, and David L. Spencer. 1946. 86 pp. (Out of print.) 2. Some Aspects of Missouri Quail and Quail Hunting, 1938-1948. Rudolf Bennitt 1951. 51 pp. (Out of print.) 3. Canada Geese of the Eastern Prairie Population, With Special Reference to the Swan Lake Flock. Richard W. Vaught and Leo M. Kirsch. 1966. 91 pp. 4. Wildlife Foods and Understory Vegetation in Missouri's National Forests. Dean A. Murphy and Hewlette S. Crawford. 1970. 47 pp. 5. The Three-Toed Box Turtle in Central Missouri. Charles W. Schwartz and Elizabeth R. Schwartz. 1974. 28 pp. 6. Food and Nutrition of Cottontail Rabbits in Missouri. Leroy J. Korschgen. 1980. 16 pp. 7 Missouri Fur Harvests. Frank W Sampson. 1980. 8. Ecological Studies and Management of Missouri Bats, With Emphasis on Cave dwelling species. Richard K. LaVal and Margaret L. LaVal. 1980. 56 pp.