Title: Population dynamics of the round-tailed muskrat (Neofiber alleni) in Florida sugarcane /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098269/00001
 Material Information
Title: Population dynamics of the round-tailed muskrat (Neofiber alleni) in Florida sugarcane /
Physical Description: ix, 204 leaves : ill. ; 28 cm.
Language: English
Creator: Lefebvre, Lynn Walsh, 1949-
Publication Date: 1982
Copyright Date: 1982
Subject: Neofiber alleni   ( lcsh )
Muskrat -- Florida   ( lcsh )
Sugarcane -- Diseases and pests -- Florida   ( lcsh )
Zoology thesis Ph. D
Dissertations, Academic -- Zoology -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (Ph. D.)--University of Florida, 1982.
Bibliography: Bibliography: leaves 128-131
Statement of Responsibility: by Lynn Walsh Lefebvre.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098269
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000352332
oclc - 09654247
notis - ABZ0298


This item has the following downloads:

populationdynami00lefe ( PDF )

Full Text








Dr. Nicholas R. Holler assisted in the planning of this study, in

obtaining field assistants, and in manuscript review. His advice and

insights have added immeasurably to the quality of this project. Paul

W. Lefebvre has been a devoted and expert assistant in all aspects of

the study. All of the photographs were taken by him. Dr. John L.

Seubert was instrumental in coordinating the conduct of this study as

both a U. S. Fish and Wildlife Service research project and a disserta-

tion subject. The assistance and encouragement of my advisor, Dr. John

H. Kaufmann, is deeply appreciated. My committee members, Dr. Katherine

C. Ewel and Dr. Carmine A. Lanciani,also provided needed advice and

moral support.

Mary C. Pittari, a 1979 graduate of the School of Forest Resources

and Conservation, University of Florida, assisted me very caoably in

most of the field work from June 1979 through January 1981. She was

alternately supported by the U. S. Fish and wildlifee Service and the

Florida Sugar Cane League. Two Cooperative Education students, Farris

M. Maddox and Thomas E. Sheppard, also provided field assistance.

Howard I. Kochman, U. S. Fish and Wildlife Service, performed

correlation and multiple step-wise regression analyses on snaptrapping

and burrow system data. His thoroughness and efficiency in completing

these analyses are greatly appreciated. Howard also advised me on the

correct taxonomic designation of snakes occurring in the study area.

N. Paige Groninger assisted in computerizing data. Drs. David L. Otis

and Richard M. Engeman, statisticians with the U. S. Fish and Wildlife

Service, performed nonparametric tests on Neofiber capture and produc-

tivity data.

Stanley L. Hooks, Director of Research for the Okeelanta Division

of Gulf and Western Food Products, Inc., and his assistant Jerry Powell

provided necessary field information and assisted in coordinating my

field activities with the harvesting and cultivating schedules. They

also provided many hours of enlightening discussion about the sugarcane

industry, as did Dr. Joseph R. Orsenigo, Director of Research for the

Florida Sugar Cane League. League General Manager F. Dalton Yancey and

Dr. Orsenigo coordinated the student grant provided by the Florida Sugar

Cane League to support Mary Pittari. Information and assistance pro-

vided by Carol M. Wade, Gloria G. Mullis, Christine M. Papaneri, and

Carlota M. Francis of the League are also greatly appreciated.

The assistance of the U. S. Fish and Wildlife Service's Denver

Wildlife Research Center library staff is gratefully acknowledged.

Mary F. Layman, formerly the Center's librarian, performed a very

helpful literature search, and Ruth M. Voorhees and Nancy K. Nibling

provided requested material. H. Randolph Perry, Patuxent Wildlife

Research Center, generously provided me with bibliographic material on


Jean B. Bourassa, U. S. Fish and Wildlife Service, provided most

of the radio transmitter collars. His advice and cooperativeness in

modifying the potting materials are much appreciated.

David G. Decker and Raymond E. Matteson, U. S. Fish and Wildlife

Service, provided technical assistance with field equipment, particu-

larly telemetry equipment.

Captain Donald K. Younker, Florida Department of Natural Resources,

made possible several visits to areas of Payne's Prairie inhabited by

Neofi ber.

Dr. Donald J. Forrester, Department of Preventive Medicine, College

of Veterinary Medicine, University of Florida, provided histopathology

analysis on Neofiber lung tissue, and Dr. Harvey L. Cromroy, Department

of Entomology and Nematology, University of Florida, identified mites

collected from Nleofiber.

Dr. Richard A. Kiltie and David E. Steffen provided helpful

comments on the manuscript. Discussions with Dr. Archie F. Carr, Jr.,

aided in interpreting field observations.

Last, but far from least, Luana A. Whitehead provided much adminis-

trative coordination and typed the several drafts and final copy.

Regina A. Hillman assisted in typing the first draft.

To all of these colleagues go my heartfelt thanks.



ACKNOWLEDGMENTS.............................................. iii

ABSTRACT..................................................... viii

INTRODUCTION.................... ............................ 1


Sugarcane Culture....................................... 6

Description of Study Fields............................. 8


Methods................................................ 17

Results................................................. 21


Methods............... ................................. 28

Results................................................. 36


Methods................................................ 57

Results................................................. 66

DISCUSSION.................................................. 109

CONCLUSIONS................... ............................. 122

MANAGEMENT IMPLICATIONS.................. .................... 125

LITERATURE CITED........................................... 128


PRODUCTS, INC ........................................ 132

BURROU SYSTEM LOCATIONS............................... .. 134

36 POSTHARVEST STUDY FIELDS.............................. 161

BURROW SYSTEM LOCATIONS ................................. 163

INDIVIDUALS............................................. 168



BIOGRAPHICAL SKETCH............................................... 204

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ROUND-TAILED MUSKRAT (Neofiber alleni)




Chairman: John H. Kaufmann
Major Department: Zoology

The round-tailed muskrat (Neofiber alleni) is one of several

species of rodents reported to damage sugarcane in south Florida.

Information on this microtine rodent's population dynamics was needed

to develop appropriate control measures and to determine if and when

they should be applied. Neofiber live in burrow systems in sugarcane

fields, and these systems can be readily located in recently harvested

fields. Snaptrapping, livetrapping, and radio telemetry were used to

study methods of population estimation, patterns of burrow system

occupancy, reproduction, survival, and interburrow system movements.

The sum of burrow system distances across rows, number of burrow systems

trapped, burrow system density (high or low), number of burrows plugged,

and number of burrows opened after 48 h were the best predictors of the

number of Neofiber captured per field, accounting for up to 94 percent

of the variation in observed captures. At a given time, average burrow

system occupancy was less than 2 individuals. Single adult males or


females, male/female pairs, a female with 1 or 2 young, or 1 or 2

subadults were the most common system occupants. Extended families

were encountered occasionally. Adult males frequently moved among

neighboring burrow systems, presumably looking for mates. Productivity

(number embryos + number juveniles per adult female) tended to be lower

or at least not greater in fields with low burrow system densities

(<10) than in high density fields (-15) burrow systems. The grower

whose fields were studied had recently initiated a 3-year rotation,

and Neofiber may be removed from fields before attaining their maximum

reproductive rate. Neofiber reproductive rate or survival of young, or

both, were low between May and November, the period of crop maturation.

The study area's Neofiber population appears to have declined in recent

years, possibly as a result of the shorter crop rotation, replanting

by blocks of fields, and the introduction of mechanical harvesting.

Control measures to reduce the population in this area are not warranted

at this time. Surveys for the presence of Neofiber burrow systems 1 to

2 months after fields are harvested are recommended. High local

populations could be effectively reduced by snaptrapping.


The round-tailed muskrat, Neofiber alleni, is a relatively large

microtine rodent, intermediate in size between the voles (Microtus spp.)

and the muskrat (Ondatra zibethicus). The range of the round-tailed

muskrat is almost entirely restricted to Florida; it also occurs in

the Okefenokee Swamp in Georgia.

Pleofiber is one of several species of rodents reported to damage

Florida sugarcane (Saccharum interspecific hybrids)(Samol 1972, Steffen

et al. 1981). Neofiber damages cane by gnawing on stalks and burrowing

under sugarcane rows, damaging and possibly feeding on cane roots

(Bourne in Porter 1953, Samol 1972). In 1974 the U. S. Fish and Wildlife

Service (USFWS) began a research program to define the problem of rodent

damage to Florida sugarcane. Initial studies included the collection of

population data on rodents and their damage to sugarcane in this region

(Walsh et al. 1976; Lefebvre, Ingram, and Yang 1978). The information

obtained is necessary in order to design and evaluate field tests of

control methods, which could include chemical, cultural, behavioral, or

mechanical techniques. However, these studies yielded little informa-

tion on Neofiber because it maintains a primarily subterranean existence

in sugarcane fields.

Elsewhere, the round-tailed muskrat typically is found in shallow

freshwater marshes. It builds shelters (houses) and feeding platforms

from marsh vegetation, and occasionally burrows into the mud during

periods of low water. In sugarcane, where the water level is maintained

at 0.60 to 0.75 m below the soil surface, the round-tailed muskrat is

apparently an obligate burrower and builds extensive tunnel systems.

Both in marshes and muck fields, IJeofiber appears to be colonial

(Fig. 1). Harper (1927) noted the clustering of houses within small

areas of Okefenokee Swamp prairies. Porter (1953) referred to the

interconnecting tunnels of a burrow system as a colony. Tilmant (1975)

designated a distinct cluster of houses, whether small (e.g., 5 houses)

or large (over 100 houses) as a colony in the Everglades National Park.

Birkenholz (1963) was more vague in his use of the term "colony". He

stated (p. 265) that even when all of Payne's Prairie (Alachua County,

Florida) was populated by Neofiber, ". the species existed as a

series of relatively isolated units or colonies." Steffen (1978)

referred to burrow systems in sugarcane fields as colonies under a

dictionary definition (Neilson 1959: 528) "A collection of organisms of

the same kind living in close association."

The round-tailed muskrat occurred in the Everglades Agricultural

Area south of Lake Okeechobee before the commercial sugar industry began

in 1929 (Steffen 1978). Howell (1920) identified the new subspecies

Neofiber alleni nigrescens from Ritta on the south shore of Lake

Okeechobee, and noted that its primarily fossorial behavior differed

from that of the other subspecies. The muskrat's presence in south

Florida sugarcane fields has been documented by several investigators

since 1937 (Steffen et al. 1981). Since 1974, Neofiber has been found

in abundance in several areas of the sugarcane growing region. In a

survey of recently harvested fields south of Lake Okeechobee in 1976,

Steffen et al. (1981) found that 40 percent of the fields in the

Western Division of the U. S. Sugar Corporation (13,072 ha) and 30

Fig. 1. feofiber houses on Payne's Prairie (above) and burrows within
a burrow system in a sugarcane field (below). The clustering
of houses and burrows within groups, and greater distances
between groups, suggest a colonial habit.


percent of the fields in the Mill Lot of the Okeelanta Division of Gulf

and Western Food Products, Inc. (8,094 ha) had Neofiber burrow systems.

Neofiber burrows have been evident for a number of years in fields

(approx. 26 ha) belonging to Lewis Friend Farms, near Canal Point on

the east side of Lake Okeechobee (J. R. Orsenigo 1979, pers. comm.).

Modesto Ulloa of Osceola Farms Co. showed USFWS personnel several

sugarcane fields with extensive rleofiber burrow systems, approximately

10 km southeast of Canal Point, in April 1981. He reported that 93 ha

of sugarcane, recently acquired by Osceola Farms, had a large number of

Neofiber burrows.

Monthly reproductive data were collected on round-tailed muskrats,

cotton rats (Sigmodon hispidus), and roof rats (Rattus rattus) trapped

in sugarcane fields over a two-year period (May 1974 through April 1976)

(Holler, Lefebvre, and Decker, unpubl. data). A total of 251 Neofiber

females was examined, and embryos of 50 pregnant females were counted.

Litter size averaged 2.1 embryos per pregnant female, and potential

annual productivity was estimated to be 3 young per adult female.

Neofiber potential productivity was lower than that of cotton and roof

rats (38 and 14 embryos/adult female-year, respectively). The average

litter size of captive Neofiber maintained at the Florida Field Station,

USFWS, was also 2.1 (183 young were produced in 86 litters). Birken-

holz (1963) found a mean of 2.3 embryos in 48 pregnant Neofiber females

from Payne's Prairie.

The USFWS's present efforts to develop a control strategy for

cotton rats and roof rats are directed toward obtaining efficacy data

on a toxic bait, to be applied aerially on fields. Because Neofiber

apparently spends little time above ground in sugarcane, an aerially


applied bait would probably be ineffective for control of this species.

Several aspects of Neofiber population dynamics needed investigation

to determine if and when control is needed, and what methods might be

most effective. The study described here was designed to determine 1)

the number, sex, and age of Neofiber occupying burrow systems in sugar-

cane, 2) how system numbers and size relate to Neofiber numbers within

fields, 3) if population size can be indexed by plugging burrows and

determining the number reopened per burrow system, 4) if there is a

relationship between population density and productivity, and 5) to

what extent there is movement by individuals among different burrow


Population data are needed not only to develop effective control

measures, but also to assess the status of the round-tailed muskrat as

an endemic Florida mammal. It is listed as a Species of Special

Concern by the Florida Committee on Rare and Endangered Plants and

Animals (Layne 1978). Neofiber has a low reproductive rate, which is

unusual for a microtine. Richmond and Stehn (1976) pointed out that

most of the microtine rodents can rapidly increase their numbers.

Moreover, the round-tailed muskrat's apparent dependence on a burrow-

ing existence in the sugarcane habitat may limit its capacity for

dispersal. The above factors should influence the decisions on what,

if any, control measures are needed.


South Florida Sugarcane Culture

Sugarcane is a grass, with a mature stalk culmm or stem) divided

into a number of segments (joints), each consisting of a node and an

internode. In very young cane, before the above-ground stem has formed,

the quantity of sucrose present is negligible (Barnes 1974). As the

stem elongates, limited amounts of sucrose are stored, but near matu-

rity, larger quantities are stored. At maturity, the stalk consists

of juice (essentially a sucrose solution containing simple sugars and

starches), small amountsof mineral salts, pith and rind fiber and an

outer covering of wax.

Most of the sugarcane in South Florida is grown on organic soils

embracing the eastern and southern shores of Lake Okeechoteein the

northern Everglades. The Florida sugar industry has experienced two

growth periods, from the mid-1920's until 1962, and from 1962 to date,

with the greatest expansion in the latter period. Approximately

145,000 ha have been harvested annually in recent years, making sugar-

cane second in .rea to citrus in Florida's agriculture.

A network of irrigation canals and field lateral ditches maintains

the water level in the canefields at about 0.6 to 0.75 m below the soil

surface, although fields may be flooded occasionally by heavy rains.

Fields are generally rectangles of 8, 16, or 32 ha and are laid out in

a regular pattern, with parallel rows of sugarcane 1.5 m (5 ft) apart,

oriented north-south or east-west. Some fields are divided longitudi-

nally into halves by shallow (approximately 1-2 m deep x 2 m wide)

ditches called middle ditches or field laterals.

Sugarcane planting usually commences in late August or September

and continues until December, after the fields have been thoroughly

prepared. Plant cane, as the first crop is called, grows for 12 to 18

months before harvest, while succeeding crops, called stubble or ratoon

crops, are generally harvested after a year of growth.

Florida sugarcane is harvested during the winter months (November-

March). When the cane is harvested, buds on the underground stalk (or

stool) give rise to the succeeding growth of cane known as a ratoon

(Barnes 1974). Thus a first ratoon crop has been cut once and will

provide the second crop, the second has been cut twice, etc. The

sugarcane grower whose fields were used in this study had recently

initiated a 3-year crop rotation, so that fields were replanted after

being allowed to ratoon twice (a plant crop plus two ratoon crops had

been harvested). Other growers may harvest as many as 8 or more crops

from some of their fields, although the industry average is probably 4

or 5. The number of crops to be taken after a single planting will

depend on production, and hence be influenced by frosts, freezes,

insect damage, rodent damage, weed competition, and other factors.

The day before a sugarcane field is scheduled to be cut it is

burned to reduce the amount of leaves (trash) taken to the mill, since

these interfere with the recovery of sucrose from the stalks. Leaf

removal is also necessary for hand cutting of cane.

Cutting of the cane is accomplished by hand with a cane knife or

by harvesting machines. These two methods result in different patterns

of trash distribution in the harvested fields (Fig. 2). Hand cutting

results in alternate windows of cut sugarcane stalks and trash (leaves,

cane tops). Each window covers two cane rows and the strip of ground

1.5 m wide between them. After the sugarcane stalks have been picked

up and taken to the mill, two rows and a strip of ground 3 to 4.5 m

wide are uncovered between trash windows. In machine harvesting,

trash consisting of partially shredded tops and leaves is scattered

more or less uniformly over the whole field. Thus the density of trash

over every other two rows in hand cut fields is at least twice as great

as trash density in machine cut fields. Steffen et al. (1981) found

that Neofiber burrow systems were more common in hand cut than machine

cut fields.

Description of Study Fields

All of the study fields were located in the Mill Lot of the

Okeelanta Sugar Division of Gulf and Western Food Products Co. (Figs.

3a-b). The Mill Lot, approximately 8 km south of South Bay, Palm

Beach County, Florida, is bordered on the east by U.S. Highway 27, on

the north by the Bolles Canal, and or the west by the Miami Canal.

A paved road and large canal divide the Mill Lot into north and

south halves. Smaller canals and unpaved roads run perpendicular to

the paved road, dividing the area into 14 blocks of 8 30-ha fields to

the north 9nd 15 blocks of 8 fields to the south of the road. Sugar-

cane rows in all fields run east-west. A buffer of at least one field

in 1979, and one half-field in 1980, separated study fields to the

north or south. Fields 22 BN and CN in 1979 (Fig. 3a) were considered

"halves" of the same field because they had no middle ditches.

Fig. 2. Machine cut (left) and hand cut (right) sugarcane fields
shortly after harvest.


.v CC

W .C-- ,l--
0 *,- U 0 ,---
- 0 0 ECJ
4-- CI-- Ln I- i

>4- O 4
- i- n 4- -Z U,

l O rLn CC

4J- 0 J '-
-- .= *-- O -
. .--
- -9- 4-
0 E J 0 -

L0 I0 4 -0

4-- .-
) *- --+- .-
>.E r o
-0 10 -. -- C>

= E .1 0

>i I-3 C- 0 0

i *- 4r L-

Se4-a 4- c

r- -.- 4--C
E 0 01j

O I4-)0 =
r 14-3 .9-- CO CJ

SWI OW 4-) 4-)U -
C !. w

4-J 1>C-
Sn ac o

0 '. W 4-' Cn E

.0 3 C i- T
-: E W *- C
4 3 *.- *- i- e
.- C U- 4-j V)
3 0 *e-- 00
C r- =I -

4-) QU Q 31 C -
0- .- C-- -0 31
j *L- :3 7O -D e0

-- *=- o-
.-- Cn U r- S-
*i- 1/ C -: 1 e 0
:E o **- U

rI) C !. Ul

*i- 3 10 C
1 0 l( r- D r-
.-- > U> V Q..
-o r-- E I

-le 4- *0 x L-





______________ _____-- _____________________-- U

__ ____ ^ ____ _ ___________ ______ --


4V) y

S 0) 0

0 0 0a
a -o

4- 0 4-
O *- O
-- -c I-

o 0 0

a U
a~g (a
Y4 4


>a 4-


O in

a) L.

0 0

.c- Q

* ,

c c

SN 8 1





0 -3
*I- r


lI 0

Sn u

C) a


> -)

0 1





2- "0



) --

A an




>-. 0 >,
i-) i-) *-

wU -T w
0 S.- -0
* ^1

S I /

Harvest season study fields were beginning their second or third

ratoon. In 1979, both halves of the same field were studied. In 1980,

halves of different fields were studied in order to increase the field

sample size. Most of the 1980 study fields had middle ditches, but in

those that did not, the number of rows was counted and half were

included in the study half-field. Fields were selected a minimum of

30 days and a maximum of 60 days after harvest. In 1979, all study

fields were hand cut while in 1980, 7 or 25 fields (28 percent) were

machine cut.

In both 1979 and 1980, third ratoon fields were harvested earlier

than second ratoon. In 1979, third ratoon fields were studied from 14

December to 16 February, second ratoon fields were studied from 15 March

to 22 April. In 1980, third ratoon fields were studied from 16

December to 22 January, second ratoon from 4 April to 15 May. Study

of third ratoon fields in 1980 was curtailed by heavy rain in late

January that flooded the fields.

Livetrapped study fields, which had been hand cut, were beginning

their second ratoon when selected in May, 1979 and 1980. Fields 14 CN

(north half) and 6 BN (north half) were harvested in November and

December 1979, respectively. Fields 8 GS (south half) and 24 CS (south

half) were harvested in November 1980 and February 1981, respectively.

Fields 14 CN and 6 BN both had middle ditches, while 8 GS and 24 CS did


Individual sugarcane fields and areas within fields are often

invaded by many other plant species. Probably the most important ones

to Neofiber are sedges (Cyperus spp.) Cvperus spp. are common

referred to as nutgrass. Sedges, commonly eaten by Neofiber,

occurred in most of the study fields and Neofiber burrow systems fre-

quently occurred in portions of fields covered with sedges. A few

fields were literally carpeted by sedges, such as 6 DN studied in 1979,

which was also riddled with Neofiber burrow systems. The association

between sedges and rIeofiber was not universal, but it has been observed

in other growers' fields.

Maidencane (Panicum hemitomon) was found on canal and middle ditch

banks, and fleofiber burrows were sometimes found in patches of this

grass. Maidencane was not observed in fields. Fall panicum (Panicum

dichotomiflorum) was abundant in one of the burrow systems in 1i C!.

Many other plants were common in the study area, both in fields and on

ditchbanks (Appendix I).

Several other small mammals occupy Neofiber burrow systems in

sugarcane fields: cotton rats (Sigmodon hispidus), roof rats (Rattus

rattus), and rice rats (Oryzomys palustris). Rabbits (Sylvilagus

palustris and S. floridanus) inhabited fields and ditchbanks, and

occasionally enlarged reofiber burrows for shelter. House mice (Mus

musculus) were occasionally seen and rarely trapped in the study fields,

and it is not known if these rodents ever inhabit Neofiber burrows.

M. musculus were trapped from their own distinctive burrows in sugar-

cane fields outside of the study area.

Bobcats (Lynx rufus) were probably the only significant mammalian

predator of Neofiber in the study area. Several were seen during the

day, their scats and footprints were commonly seen on field edges, and

scats were occasionally found in or near burrow systems. Feral house

cats (Felis domesticus) were rarely seen and their scats and footprints

were not observed.

Raccoons (Procyon lotor), otters (Lutra canadensis), opossums

(Didelphis virginiana), and armadillos (Dasypus novemcinctus) were

also known to occur in the study area.

Barn owls (Tyto alba) and northern harriers (marsh hawks, Circus

cyaneus) were probably the major avian predators of Neofiber in the

study area. Marsh hawks migrate through the sugar growing region

during the harvest season, when Neofiber burrow systems, exposed by

harvest, are most vulnerable to an avian predator. Burrowing owls

(Athene cunicularia) were observed nesting in old Neofiber burrows and

may prey upon young muskrats. Common grackles (Quiscalus quiscula) and

cattle egrets (Bubulcus ibis) may prey upon injured or disoriented

Neofiber after harvest. Numerous shorebirds and waterfowl were seen

in the study area, particularly during the harvest season, when fields

flooded with water discharged from the mill attracted migrating flocks.

Great blue herons (Ardea herodias) and great egrets (Casmerodius albus)

were common in all seasons; both have been observed hunting Neofiber

on Payne's Prairie, Florida (Donald K. Younker, pers. comm.). Several

rodents, including thirteen-lined ground squirrels (Spermophilus

tridecemlineatus), Eastern chipmunks (Tamias striatus), and prairie

pocket gophers (Geomys bursarius), have been reported as prey items of

great blue herons (Peifer 1979).

By far the most commonly encountered snakes in canefields were

king snakes (Lampropeltis getulus). King snakes were several times

observed to enter or exit Neofiber burrows. Microtine rodents

(Microtus ochrogaster) were found to be the most important food item of

the prairie kingsnake (Lampropeltis calligaster), making up 48% by

volume of its diet (Fitch 1978). Rat snakes (Elaphe obsoleta) were

less frequently seen. Ribbon snakes (Thamnophis sauritus), water

snakes (Nerodia spp.)and cottonmouths (Aqkistrodon piscivorus) occurred

on ditch banks. Alligators (Alligator mississippiensis) and softshell

turtles (Trionyx ferox) occasionally were seen in canals. Alligators

prey upon Ondatra in Louisiana when muskrats are abundant (Valentine

et al., 1972). Alligator gar (Lepisosteus spatula) and largemouth bass

(Micropterus salmoides) also inhabited canals.

Fire ants (Solenopsis invicta and S. geminata) were extremely

common throughout the study area. Their nests frequently occurred in

Neofiber burrow systems, and care had to be taken not to set traps too

close to them.



Known-age Specimens

Twenty-five (13 females, 12 males) captive-born Neofiber were

weaned at 18 to 20 days and placed individually in 20 x 35 x 20 cm

stainless steel wire cages. They received Lab ChowP, Rabbit Chowv, and

water ad lib. They were anesthetized with Metofane (methoxyflurane)

and measured at one-week intervals until 50-60 days old, then at two-

week intervals, until they were sacrificed to provide known-age speci-

mens ranging in age from 30 to 220 days. Measurements included total

length, body length, tail, hind foot to the nearest mm, and weight to

the nearest g on an Ohaus triple-beam balance. Reproductive tracts

were removed and preserved in a 70% ethanol solution. Testis weights

were determined to the nearest mg, generally within a month of their

collection, and seminal vesicle development was noted. Measurements

were made on 10 additional known-age Neofiber that died while in cap-

tivity, reared under conditions similar to those described above but

some were raised with 1 or 2 cagemates.

Field Specimens

Three age/reproductive classes were determined on the basis of

known-age, laboratory-reared specimens and reproductive condition of

unknown-age, snaptrapped specimens. Juveniles were defined as 530

days old, adults as sexually mature, and subadults as animals older

than 30 days but not sexually mature.


Mean body measurements (total length, body length, and weight),

plus one standard deviation (SD), of known-age juveniles were used as

criteria to establish the upper boundary for the juvenile category for

snaptrapped specimens.

Preserved uteri from snaptrapped specimens were examined at 10-20X

for placental scars and to check counts of embryos made in the field.

Nonpregnant females were classified as parous if the uterus had at

least one scar (Fig. 4). Sexually mature snaptrapped females were

defined as those having embryos or placental scar(s).

Preserved testes were weighed to the nearest mg on a Mettler PIN

163 balance. The frequency distribution of field specimens' average

testis weights was used to determine the boundary between adult and

subadult testis weight.

Since it was noted that most males had either well-developed

seminal vesicles dilated with fluid, or small, empty seminal vesicles

(Fig. 5), seminal vesicle development was used to classify males with

average testis weights close to the adult/subadult boundary, adults

having dilated vesicles and subadults undilated or very slightly

dilated vesicles.

Mean body measurements (total length, body length, and weight)

minus or plus one standard deviation for adult male (-1 SD), adult

female (-1 SD), and juvenile (+1 SD) snaptrapped specimens were used to

determine upper and lower boundaries for the subadult category of live-

trapped specimens. If any 2 of the 3 measurements exceeded or fell

below a category boundary, the muskrat was classified accordingly. If

reproductive condition of livetrapped specimens was evident (obviously

pregnant or post-partum females, scrotal testes in males, or information

'II I I 7 1 1 1 1 7 I I I I I II I l II l I I II II II 11
raM ,I III IiI 'I" ''- 01 -I 1'-I '11= 1 10 1 a

( iicals 1 2 3
I I 1 1 2 1 . I .. I . I. I I .

Fiq. 4. Uteri of parous, nonpregnant (above) and nulliparous (below)
Neofi ber.

i i.

C-11 CIL..0

", .- , . . I .i I

... .i-- -- ... .... ..-... .*1: iIi 4LJ 4" I J l.JJ J I | '1 1 .1 I 1 ..11
- 1 4 5J I9 i8 .

.. .. .. .. .. .. .....- ... i, i ij.r I ... ... ..

I VASO F A -LL INC ra,o,AL..L Co.,.r. C,
.., 3 1

Fig. 5. Reproductive tracts of immature (above) and mature (below)
male Neofiber. Arrows indicate seminal vesicles.

from necropsies) it took precedence over body measurements in classify-

ing sexually mature muskrats; less than 5 percent of the field specimens

were classified on the basis of reproductive condition alone.


Because mean body length, weight, and total length did not differ

significantly for male and female known-age juveniles (P>0.10), the

combined-sex mean measurements plus 1 SD were used as the upper boundary

for the juvenile category (Table 1). If any 2 of these 3 measurements

from an unknown-age individual equalled or were lower than the values

given in Table 1 for known-age Neofiber, the individual was classed as

a juvenile.

Almost all snaptrapped males with an average testis weight below

200 mg showed very little development of the seminal vesicles. Also,

the relatively low incidence of specimens with average testis weights

between 100 and 200 mg (Fig. 6) made 200 mg a logical breaking point

that would result in minimal classification error. Thus males with an

average testis weight 200 mg were considered to be adults. Snap-

trapped females carrying embryos or with uterine scars indicating

previous pregnancy were considered to be adults. Mean weight, total

body length, and body length, minus 1 SD, of field-trapped sexually

mature Neofiber (Table 2) were used to determine the adult/subadult

boundary for both snaptrapped and livetrapped individuals. Sexually

mature males and mature, nonpregnant females had nearly the same body

measurements, but not weight (Table 2). Parous, nonpregnant females

weighed significantly less than adult males (P<0.01) and pregnantfemales



















o cy ai

r M in

W 0
) 0a
c, U.
ro 10

3 3
0 O

2. 'H



= U


1,0 i-* ^
(-f C- I

Mo n
ng ^

C\j <









f') L

WX '



a 60

i Adults

E 40 -

I -4 C -4 CIO j Cn C n C In

Average testis weight (ng)

Fig. 6. Frequency distribution of average testis weights of Neofiber
males snaptrapped from Florida sugarcane fields in 1979 and
1980. Males with an average testis weight 2 200 mg were
considered to be adults.












CM 0 0
L 0 C

C- It

,.O: r 0
X) rf-) Co



I 0


': IX


= IX



-n (n CM

30 0 '


1.0 mb CM
CM CM ro)
*^- OT en
f-* i i i-

cr> i i-i M O O

(P 0.01). Thus the mean measurements of nonpregnant females were used

to establish the adult category for females.

Measurements of captive males and females (Tables 3 and 4) indicate

that 2 of the 3 adult age class criteria were satisfied at 101 to 110

days by males and 81 to 90 days by females. This corresponds fairly

well with Birkenholz's (1963) determination of sexual maturity at 90 to

100 days. Females may actually mature more quickly than males. How-

ever, the number of captive animals measured was small, and mean male

measurements at 81 to 90 days were very close to satisfying adult

criteria (Table 3). Also, mean total length and body length measurements

of known-age females between 91 and 120 days old were just short of the

adult criteria (Table 4).

Juveniles in this study were defined as being 30 days old or less,

subadults as sexually immature and between 31 and 89 days old, and

adults as sexually mature and at least 90 days old.

Eight of 11 captive males, age 90 days to 1 year, had average

testis weights that exceeded 200 mg. Two males had average testis

weights over 190 mg, and also had dilated seminal vesicles. The elev-

enth male was 91 days old when sacrificed, and had an average testis

weight of only 37 mg. Two older known-age males (435 days and 4.4

years) had average testis weights considerably lower than 200 mg.

Several field specimens with large body measurements also had low

testis weights. Possibly older and/or isolated Neofiber undergo testic-

ular regression.

Hind foot length showed too much overlap between age classes to

be useful as an aging criterion (juveniles: 34-45 mm, subadults 39-48

mm; adults 40-50 mm).

C\J CM c\J CiJ 'I C\J CM C\J C\J %I

o c\ (n LO r I r-e L O

S m-i c j r- C) 0 0C

cJ '.- 0 *- :3 00 n r- - r-
q Lfn I r- rf- r- r- co co00
- -




u -o

,- (

01 -


4o o



i- -

E a
ra '






00 Ocl C14 00 %M .D krD .0 m' .0 1LO
(n (DC% CD (\j In In 'In Mi Mn fl
0) 0~i 0 0 C\J C\J (\J >("* (M >**)

M LO D U, L0 mU O ?- r Ln
C'iJ (' ,' -4

r- -. m C\j m
0o O 0 c\J m = .--
.,j r -j m ., .) m I ) M

r-.- C- c'.j C) 0 - Z3 In c3 I
S c'i M- M *Z- C\i C 0 M n M LO

0 : -, r- k-
0 CIo -4 C\j
N- NJ- c jj CMN

C(i Cl

cr Cj

0 0 0 0 r-

SLn o r- o -- -

M- r Ln o 00 cn C 0 C',J T
-4 -:I,

- -. c~'


c0l In O' ~ I'- 0

^ M r- :T* qr -:I, LO r_ ':T

C o e o LO <3 1 -
:j :Z :d :d c' :Z :Zr -:I-




U C,


a) )


0 C




2 0





0 r_


S Ix

-4 -4 -4

m-) 0o 0 m) 0

S C- o r cJ

LU. 0
Lot CM)

d- Lfl
C\ cli


o n
- -4

CM u*) Lo
00 CM CM

O 10 r-
e'i .-i C'.
-4 -4 1-4

0 o 00
o) C C)-

C'- ^ M C, M -f C, C,

m d -:I- m 1-4 -- m^ M

- 00 C0 Cco m C:: m O
-4 -4 CQ. e' ej i e'jQ '

0 0 0 00
IC C C0 D- d CI

n 0 a) -4 -4 -4
m Kd~l Lo 00 c"I

0 r- 1 1.0 k.-0 0o

CO C-- CP 0
us -^ i~~ r^ i~~ r~ c

no Lo

-4 n 74
,-4 -


o r - : a, n m
Cj 1-4 7-- I-4 -4 -4 -4 -4



Field Selection

Because burrow systems are much easier to find shortly after fields

have been harvested, an attempt was made to study burrow systems in as

many fields as possible between December and April in two years.

Second and third ratoon fields were surveyed using all-terrain

vehicles between 30 and 60 days after their harvest dates. Fields were

completely surveyed by starting 6 rows north or south of a corner,

travelling the length of the field, and then making successive passes

every 12 rows, so that a 6-row wide swath on each side of the vehicle

was searched on each pass. Active burrow systems were flagged and

counted. A burrow system was considered active if it contained at

least one burrow that showed signs of postharvest excavation by

Neofiber: usually a large mound of muck (Fig. 7) or at least some moist

muck pushed out of a burrow entrance. The minimum distance between the

nearest two burrows of systems classified as adjacent but separate was

15 m. Burrows with less than 15 m between them were included in the

same burrow system. This criterion was not completely arbitrary, as

it was based upon field observations which indicated that most distinct

groups of burrows were at least 15 m apart.

Fields with fewer than 10 burrow systems per half (< 0.67 per ha)

were considered to have a low Neofiber density, and those with greater

than or equal to 15 per half-field (2 1.0 per ha) were considered to

Fig. 7. Mounds of freshly excavated muck at an entrance to a round-
tailed muskrat's burrow system in a recently harvested
sugarcane field.


have a high Neofiber density. These density criteria were based on the

mean density of burrow systems (0.69/ha) found by Steffen (1978) in his

1976 survey of the occurrence of Neofiber burrow systems in fields

belonging to the U. S. Sugar Corporation, located north and northwest

of the Okeelanta Mill Lot. Fields with densities greater than 9 but

less than 15 burrow systems per half-field were generally not used,

with one exception: 16 GN (south half), studied in 1979. This half-

field had 11 burrow systems, but the whole field averaged 15 systems,

and thus was considered a high density field.

Field selection was essentially dictated by the harvest schedule.

At any given time a limited number of fields wereavailable from which

to select study fields; thus,less than optimal choices occasionally

were made (e.g., 16 GNI). In no case was a field rejected that had a

higher burrow system density than those high density fields that were

selected. One high density field and one or more low density fields

were generally studied concurrently.

Burrow System Selection and Snaptrapping

All active burrow systems in low density fields were studied in

both 1979 and 1980. In 1979, burrow systems in high density fields

were selected for study by randomly selecting 1 or 2 groups of adjacent

burrow systems. Groups were determined from field survey maps, and

contained 3 to 12 burrow systems. Two days before trapping, burrows

in selected systems were plugged by kicking dirt into the entrances,

on a randomly selected half of both high and low density fields.

In 1980, half-fields rather than %whole fields were studied. Half-

fields were stratified into quarters, and 2 burrow systems per quarter

were randomly selected. If a quarter had only 2 systems, both were

studied. From 0 to 6 additional burrow systems (generally 1 per

quarter) were also studied because of their proximity to (less than 60

m from) randomly selected systems. Field 24 FS was the only exception:

all of this half-field's burrow systems were studied. Burrows in

selected systems in both high and low density fields (except 24 FS)

were plugged in the same manner as 1979 burrow systems.

Reopened or new burrows were trapped for 4 consecutive nights

using 9 x 18 cm Woodstream Corporation and McGill rat traps. Burrow

systems in which no burrows were reopened were not trapped. The number

of Neofiber captures per burrow system was recorded, and necropsies

were performed on all Neofiber specimens. The number of embryos per

pregnant female was counted, and reproductive tracts of both males and

females were preserved.

Burrow System Mapping

In 1979, distances between burrow systems within fields were deter-

mined by pacing. In 1980, burrow systems were mapped using a Rolatape(

660 measuring wheel, rolled by hand from system to system between two

rows, or attached to the rear luggage rack of an all-terrain vehicle.

Passes were made across high density fields, approximately 20 rows

apart, with stops at the northernmost burrow in each system. Rows

north or south of the row traveled to each burrow system were counted.

The distances between the farthest outlying burrows along cane rows

and across rows were determined for each system by pacing and counting

rows (Fig. 8). These x and y distances are indices to burrow system

size. The burrow systems in one 1980 low density half-field, 34 BN,



< x 0____

Cane row V0

Fig. 8. Schematic representation of Heofiber burrow system showing
distances x and y, which are considered to be indices to
burrow system size. Circles represent burrows.

were mapped but not trapped, and in one 1979 high density half-field,

28 CN, were trapped but not mapped.

Burrow System Distribution

Study half-fields were stratified (Fig. 9) and the number of

burrow systems within each stratum was counted. The stratification

was intended to show if there was a tendency for Neofiber burrow

systems to be distributed more frequently near the middle ditch

(Stratum II), in the center (Stratum IV),near the outer field edge

(Stratum III), or in the field ends bordered by roads or canals but

not close to either the middle ditch or a neighboring field (Stratum I).

If a stratum line fell on a burrow system, that system was not counted

unless 2/3 or more of the area it covered fell within one stratum.

Most of the burrow systems were counted.

Statistical Analysis

Differences in burrow system density by field stratum were tested

for by single classification analysis of variance (Sokal and Rohlf

1969); if a significant F-ratio was obtained, stratum means were tested

using Duncan's new multiple-range test (Steel and Torrie 1960).

The relationship among year, ratoon, and Neofiber density on total

captures per burrow system was analyzed by SASI/ PROC FUNCAT (functions

of categorical responses) which uses chi-square tests rather than

F-tests in a procedure analagous to ANOVA. Captures per burrow system

1/ Statistical Analysis System (Helwig and Council 1979).



-- -


0 S
: ,

1n 0.1





C) -1LA ,--

U- 0 0

'o -


.33 c)


were categorized as 0, 1, 2, 3, 4, or >5. Two models were fitted to

the data. The first model included the 3 main effects and all 4

possible interactions of the main effects; the second included the 3

main effects and only 1 significant interaction term.

An exploratory testing procedure was used to determine if differ-

ences in Neofiber productivity were related to year, ratoon, or density.

A Kruskal-Wallis one-way layout test (Sokal and Rohlf 1969) was per-

formed on each of 7 combinations of the data to test the main effects

and interactions. Although it is possible to encounter significant

test results by chance when multiple comparisons are made, the conser-

vativeness of the nonparametric procedure should alleviate this problem.

The effect of burrow plugging on Neofiber captures in 1979 was

analyzed by the SAS MRANK procedure, which is used to test nonpara-

metric hypotheses about the relationship between a set of independent

variables and a set of dependent variables. The effect of covariates,

such as burrow system size (x and y), could be eliminated. For each

hypothesis tested, MRANK prints a chi-square statistic and the proba-

bility of obtaining a larger value.

The 36 half-fields in which burrow systems were plugged before

trapping (12 in 1979 and 24 in 1980) were used to determine which

variables best predicted the number of Neofiber trapped in these fields.

Pearson correlation coefficients (Sokal and Rohlf 1969) were obtained

for 10 independent variables per field: number of burrow systems,

burrow systems trapped, burrows plugged (= total number burrows in

trapped systems), burrows opened, sum of the x-dimensions, sume of the

y-dimensions, sum of the trap effort (number of traps x number of nights

trapped), year, ratoon, and density (high or low) of burrow systems.

Linear regression was performed to determine the best 2-, 3-, 4-

10 variable models. One 2-variable and one 4 variable model were

selected for forward solution, stepwise regression. Predicted Neofiber

captures with prediction intervals for each field and distribution and

plots of residuals (deviations of observed from estimated values) were

also obtained. These analyses were all completed through the use of



Burrow System Distribution

High Density Fields

Burrow system density did not differ significantly among Strata II,

III, and IV, (P> 0.05). However, their density in Strata II and IV was

significantly greater (P <0.05) than in Stratum I (Table 5). Further-

more, no field had a very high density of burrow systems (_3 per ha)

in Stratum III, and a lower density in both II and IV. When Stratum

III had very high burrow system density, one or both of Strata II and

IV had very high densities (e.g., fields 6 DN, 34 CN-N, and 32 EN-N,

Appendix II).

Stratum I burrow system density appeared to have the least reliable

relationship with overall field density, which may be related to the

small size of this stratum. Eight of the 20 high density fields (40

percent) had no burrow systems in Stratum 1, and some of these 8 fields

had very high burrow system densities (>3 burrow systems oer ha' in

other strata (e.g., 26 DN-II, Appendix II).

Table 5. Mean number of Fleofiber burrow systems per hectare in 4 strata of
half-fields with high burrow system densities. Two of the 3 2nd
ratoon 1980 fields had no middle ditches, and thus no Stratum II.
These half-fields were not included in testing for differences
between half-field means.

Half-fields N Ia/ II III IV


3rd-b*1979 5 1.02 0.63 1.32 0.33 0.96 0.18 1.67 0.22

2nd.1979 6 1.25 0.67 3.38 0.81 2.38 0.78 2.73 0.57

3rd-1980 5 1.42 0.22 3.42 0.61 2.46 0.40 2.84 0.82

2nd-1980 3 1.40 0.84 1.13 0.54 1.60 0.40

All/ 17 1.16d 0.30 2.71ef/ 0.40 1.87df 0.34 2.43ef/'0.32


I = center of field ends
II = third of field near middle ditch
III = third of field near outer field edge
IV = field center


SMeans that do not share at least one letter in their superscript differed
significantly from each other (P<0.05; Duncan's new multiple-range test
[ Steel and Torrie 1960]).

Low Density Fields

There was no significant difference in burrow system densities

among strata in the 23 half-fields with middle ditches (P> 0.05)

(Table 6). Seven fields had no middle ditch. A low proportion (20

percent) of all 30 low density half-fields had burrow systems in

Stratum I, 57 percent had systems in Stratum IV, 73 percent in Stratum

III, and 70 percent (16 of 23) in Stratum II. In no field were there

burrow systems in all 4 strata, and in 17 fields (57 percent), burrow

systems were restricted to 1 or 2 strata.

Neofiber Captures per Burrow System

A total of 551 Neofiber was captured in 262 of 351 burrow systems

trapped. Thus there was an overall average of 2.1 Neofiber captured

per burrow system with at least 1 Neofiber capture, and an average of

1.6 leofiber captured per burrow system trapped. The range in captures

was 0 to 10 per burrow system. Undoubtedly Neofiber were present but

not trapped in some burrow systems. It is also possible that Neofiber

had abandoned some systems before the systems were trapped, or that

they were trapped in another burrow system (i.e., some Neofiber may

have visited more than one burrow system within the study period of a

given field).

In 10 systems that appeared to be active when plugged (4 in 1979,

6 in 1980), no burrows were reopened and no new burrows appeared in

48 h. Nine of these systems were relatively small (2 to 10 burrows,

: = 4.3). One system had 24 burrows. These systems were not trapped.

Burrow system density was the only single factor determined to be

correlated with the number of Neofiber captured per burrow system

Table 6. Mean number of Neofiber burrow systems per hectare in 4 strata of
half-fields with low densities of burrow systems.

Half-field N I- II III IV

3rdb-1979 6 0.05 0.05 0.34- 0.09 0.22 0.80 0.21 0.04

2nd-1979 6 0.20 0.20 0.42 0.05 0.50 0.21 0.05 0.05

3rd-1980 8 0.45 0.32 0.50 0.18 0.22 0.10 0.42 0.10

2nd-1980 10 0.24 0.16 0.22d/ 0.14 0.40 0.07 0.26 0.14

All 23 0.26 0.13 0.39 0.07 0.34 0.07 0.25 0.06

a/ Stratum I = center of field end


II = third of field near middle ditch
III = third of field near outer field edge
IV = field center

" ratoon

/ N = 4; 2 fields had no middle ditch.

/ N = 5; 5 fields had no middle ditch.

(P = 0.001),and the ratoon x: year interaction was also significant P =

0.012). In 1979, average captures per burrow system were lower in the

third than in the second ratoon, whereas in 1980, captures per system

were slightly greater in the third than in the second ratoon. Overall,

burrow systems in high density fields yielded an average of 1.75

Neofiber each, whereas systems from low density fields yielded an aver-

age of only 1.20 Neofiber each (Table 7).

The pattern of Neofiber captures per burrow system in low and high

density fields differed primarily in the number of systems with 4 or

more captures (Fig. 10). Low density fields had relatively few of these

(1 field had 1 burrow system from which 5 Neofiber were captured), and

relatively more systems from which only 1 Neofiber was captured.

Population Sex and Age Structure

Slightly more females than males were trapped (232 vs 269). The

difference in captures of adult females and males was greater: 123 vs.

100, or a ratio of 1.23 adult females per male. Deviation from a 1:1

ratio was not significant (X2 = 2.37; P> 0.05).

Overall averages of 0.62 and 0.64 adult per burrow system trapped

were captured in low and high density fields, respectively. The aver-

age number of adult females and males captured per burrow system did not

appear to differ between low and high density fields: 0.32 (38/117) vs.

0.36 (85/234) adult female and 0.29 (34/117) vs. 0.28 (66/234) adult

male per burrow system from low and high density fields, respectively.

Thus, the difference in Neofiber captures per burrow system between

high and low density fields appears to be accounted for by the differ-

ence in the number of young subadultt + juvenile) Neofiber captured per

Table 7. Average number of Neofiber captured per burrow system trapped, by
ratoon x year x burrow system density combination.

High Burrow Low Burrow
Ratoon-Year System Density System Density

3rd.1979 1.32 60 0.91 22

2nd-1979 1.78 64 1.38 26

3rd-1980 2.08 66 1.41 32

2nd-1980 1.82 44 1.08 37

Overall 1.75 234 1.20 117

0 1 2

High Density

3 >4

Low Density

._ n

0 1 2 3

Number of Neofiber Captured

Fig. 10. Frequency distribution of burrow systems from which 0, 1, 2, 3,
or 4 Neofiber were captured in fields with high or low burrow
system densities.

S II L .

burrow system (Table 8). A total of 259 young (1.11 per burrow system)

was captured in high density fields, whereas only 69 (0.59 per burrow

system) were captured in low density fields.

In low density fields, the overall ratio of adult:subadult:juvenile

captures was 2.5:1.4:1.0, and the average number of juveniles captured

per adult female was 0.76. In high density fields, the age ratio was

1.3:1.2:1.0, and an average of 1.35 juveniles per adult female was



A total of 37 pregnant females (30 percent of all adult females)

was captured. Percent females pregnant was similar in the 4 high den-

sity ratoon x year combinations (28 to 36 percent), but ranged from 0

to 44 percent in low density ratoon x years (Table 9). Only 1 of 13

adult females (8 percent) trapped in 1979 low density fields was preg-

nant. Mean litter size was lower in 10 females from low density fields

(x = 1.50 embryos; SE = 0.167) than in 27 females from high density

fields (x = 1.80; SE = 0.103). However, this difference is not signi-

ficant (P> 0.05).

Mean number of embryos carried by pregnant females was 1.76 (SE =

0.09). Pregnant females usually carried 2 embryos, 1 in each horn of

the uterus (Fig. 11). Females carrying 2 embryos were significantly

larger in body length (x = 197 mm) than females carrying 1 embryo (x =

189 mm; P< 0.05).

Since the number of embryos per female was small (usually just 1

or 2), an index was used in testing for correlation of ratoon, density,

and year with female productivity:

S -40 9 Ci



F n
o 00

0 -_

> .

*- -Q

c II



O 0

Co) a
4 -
0 C-

a .-
a )
u a

*- --a




3- k.0












C\ r 7k Ln m

9- C U- r- C\ 0

Ln m e- n C '.0
-c'.J C~.j C'J

N CLO Njo' O Cin o
0 -i a -l r -C C\j C- -O CD

c". 0 ko '- t' '.0 rt .- -
CM Ut. c\j .0 rn 1.0 n l Ln




- C\i

m kD

Table 9. Percent adult female Neofiber pregnant and
by ratoon x density x year combination.

average litter size

Ratoon x Density x Year N Pregnant Litter Size
























123 30.1


L i Ll H I 7

Number and Position of Embrvos

Fig. 11. Frequency distribution of embryo positions in pregnant
Neofiber. L = left, R = right.

n + n.
Productivity Index (P.I.) = e + j

where n = number of embryos

n. = number of juveniles

N = number of adult females

in each field trapped (Table 10).

Density was highly significantly correlated with productivity

(P = 0.003). However, there were also significant interactions

between density and ratoon (P = 0.012), and density and year (P = 0.023).

Density appears to have been correlated primarily with productiv-

ity in third ratoon fields (Fig. 12a). This suggests that the low and

high density third ratoon fields differed in some aspects important to

Neofiber's production of offspring, whereas second ratoon low and high

density fields did not differ. Second ratoon low density fields may

simply be fields which Neofiber are just beginning to colonize. It is

also possible that second ratoon high density fields were not as suit-

able as third ratoon high density fields for production of young. The

density x year interaction resulted primarily from a difference in

average productivity index trend from 1979 to 1980 between third ratoon

high density fields (slightly upward) and the other ratoon x density

combinations (downward)(Fig. 12b). Overall, productivity in high den-

sity fields was the same in 1979 and 1980 fields, whereas productivity

in low density fields was lower in 1980 than in 1979 (Fig. 12c).

The Okeelanta Mill Lot Neofiber population clearly did not show a

higher rate of reproduction in low density fields in the 1979 and 1980

study periods. Neofiber may, however, be capable of reproductive

compensation. It is perhaps no coincidence that, among the 4 high

Table 10. Productivity indices for post-harvest
density x year. Productivity inde =
iNo. embryos + Nlo. juveniles captured
Number adult females captured





S 1.77

SE 1.13

































study fields by ratoon x














1. 1














1979 2nd ratoon

3rd ratoon
C,- -- ---

- o 0HD




Interaction between density and ratoon (a) and density and year
(b and c) effects on Neofiber average productivity indices.
Average Productivity Index = Number of embryos + Number of juveniles
Number adult females
captured per ratoon x year or density x year combination. LD = low
Neofiber density; HD = high Neofiber density.









-- -0 HD

1.5 -LD

1.0 ILD ---







Fig. 12.


density ratoon x years, the ratoon x year with the lowest average number

of juveniles captured per burrow system (second ratoon 1980) also had

the highest average number of embryos per burrow system (Table 11).

Capture Success in Plugged vs Unplugged Burrow Systems

Burrow systems yielded fewer Neofiber on the plugged halves of

1979 study fields than on the unplugged halves (1.24 vs 1.73 Neofiber

per burrow system, respectively). This difference was significant

when field to field variation was removed and when x and y (burrow

system length and width) were added as covariates (P<0.05). The x and

y values were not correlated with treatment (plugged or not plugged),

as would be expected because treatment was assigned randomly to halves.

Tests for correlations between Neofiber captures and various burrow

system and field parameters were thus conducted using only the plugged

burrow systems. Captures may have been lower from plugged burrow

systems because trap effort was lower or because plugging affected

Neofiber behavior, making individuals in plugged systems somewhat less

vulnerable to capture.

Relationship of Capture Success to

Burrow System and Field Statistics

The sum of burrow system y-dimetnsions (distance across rows) per

field (y-sum) was the single variable most highly correlated with

lNeofiber captures per field (Table 12), with an r2 value of 0.839.

The number of burrow systems trapped per field (trap-bs) was the next
most highly correlated (r = 0.803). Many of the other independent

variables were also correlated with rleofiber captures and with each

other. It is undesirable to use a large number of independent variables



- u*


o 3




0 L/


3 --

C .

0 LL.









c0 LD C~l

o o Q Q


m c1.1 C.

1-1 .- ,,n --I

x x x x

In t.0 co C0

01 01 C C
r- r- cO 00

CY C\j Cl C\j

Ln c'i

c C C c

oo ^- o




C 0
-0 S=



O- c
Q. O



l- 001 '


o c a

o f

4- .-
Q - i-

iu o

O -
C* '-
'- (U -

*- 30 a)

,- .-
a-- -
O t4=
01) C


O) C

U --

) -= -4
0 -



l- -o

o Q












v0 v0
C c
co r- 0) C.J
co 0 co
CO 0 e CO

i0 0 n

0 0 0 0

= *
cO Co r- c

0 0 0 0
6 6 C

c cc o-I cr ci cc
0C 0C 0 i 0n 0

C O O 6 6 6

0 0

%.D cc
o o

c s=

CO M-- cO

0 0 0 0

0 0') ID 0

00 0 0 0




3 c 3- 3
O i O- o- O

S- 3 -0 -3 5n
S a 0) -e I >
. Q. LO ') L -_ Nc\ -) 4 4-
a)*O a0) c E E o --
-O -O -o a 0 L
E '- E E CL. L 1 '- .o 4=
1 =- 0.. 30 I I '-- ) 0 0
S Z >- LJ >- Q

3- 0
CC0 0
0 --

c =
LO 00 0 00 00c 00
c00 c0 c0 c

0 0 0 0 0

o0 -

o oc
o 00

o o

0 0






C C a)
0 0 U

4-) x
a0 -

0) .- Q
l- 3s ;,

3 5. -
>-- >i .0

o o
L- L- -
= = L
J3 -O C

C^ c C
= = =a
If I) f
- cNJC'




4 -

a LL-
-0 C

.= r-

in regression analysis when they are intercorrelated and observations

are small (Kerlinger and Pedhazur 1973), as in this case. Moreover,

y-sum and trap-bs are relatively easy statistics to obtain, and there-

fore are the most practical and economical variables upon which to

base a prediction of Hleofiber numbers.

When all possible 2-variable models were tested, y-sum and density

yielded the highest r2 value (0.871), and the trap-bs and y-sum model
was a close second (r = 0.867). Trap-bs could be obtained with no

additional effort when y-sum is obtained, and for a larger sample of

fields would probably be a more reliable predictor because density

assumed only 2 values.

The regression equation obtained using y-sum and trap-bs as

predictors is y = 0.11 (y-sum) + 0.86 (trap-bs) 1.60, where 9 = the

estimated number of Neofiber captures. The y-intercept did not vary

significantly from zero (P = 0.16), therefore the equation was not

forced through the origin. The predicted values are reasonably close

to actual values (Appendix III), with only 3 fields having large

differences in predicted and actual captures: 6 HN, 8 ES, and 34 CN.

The actual number of Neofiber captures in these fields was just outside

of the prediction interval in each case. The distribution of the

residuals was close to normal, and the plot of residuals against

Neofiber captures was random in pattern, indicating a good fit of the

regression equation.

The best 3-variable model (number of burrows plugged, y-sum, and

density) yielded an r2 of 0.903, and the best 4-variable model (burrows

plugged, burrows opened, y-sum, and density) increased r2 to 0.937.

None of the 5 to 10-variable models yielded r2 values greater than 0.944.

Therefore, 4 variables are probably the most that should be used in a

regression equation to estimate Neofiber numbers, particularly in light

of the intercorrelations among predictors and relatively small sample

size. In some instances, e.g. for research purposes, the additional 6.5

percent of variation in leofiber captures explained by the 4-variable

regression model may be worth the added effort of counting and plugging

burrows and returning 48 h later to count reopened burrows. The

regression equation obtained by using burrows plugged, burrows opened,

y-sum and density is y = 0.10 (y-sum) + 10.36 (1 [low density field] or

2 [high density field]) + 0.21 (burrows opened) = 0.10 (burrows plugged)

- 8.53. As might be expected, the predicted Neofiber captures that

deviated the most from actual values in the 2-variable model were improved

in the 4-variable model (Appendix III). The Field 6 HN estimate improved

dramatically with essentially equal actual and predicted values. The

predicted value for Field 8 ES was only slightly improved, with an actual

value still just outside of the prediction interval. Residuals were

distributed normally and their plot-pattern was random.

It may seem odd at first glance that the relationship between

Neofiber captures and burrows plugged was negative. However, this means

that for a given number of opened burrows, as the number of plugged

burrows increased reofiber captures declined. It does not seem unreason-

able that reofiber captures would decline as the proportion of burrows

opened to burrows plugged decreased. In many cases, reofiber may have

already abandoned large systems with many burrows by the time I trapped


The negative relationship between burrows plugged and Neofiber

captures probably resulted largely from the influence of 1979 third

ratoon burrow systems. The burrow systems of the other year x ratoon

combinations had mean numbers of burrows that ranged from 6.1 to 13.1,

whereas 1979 third ratoon systems in low and high density fields had

means of 22.0 and 19.0 burrows, respectively (Table 13). Yet the

latter burrow systems had relatively lower Neofiber captures (Table 13).

Burrow systems in 1979 third ratoon high density fields also had

a larger mean x-dimension (22.3 m) than systems in the other ratoon x

years (6.3 to 18.1 m)(Table 13). Extensive systems that undermined

long stretches of cane row, such as the one in the center of 16 GN-N

and in the west-central portion of 14 GN-S (Appendix I), were unique to

this ratoon x year. Large burrow systems in the other high density

ratoon x years, such as one near the middle ditch in 6 DN-S (Appendix

II), contained numerous groups of burrows less than 15 m apart.

The regression equations described above could be used to estimate

Neofiber density per field, provided that Neofiber captures from those

systems selectively plugged and trapped were representative of capture

success that might have been achieved had all burrow systems been

trapped. There is no reason to suspect that the sampled burrow

systems were not representative of all burrow systems. The grower or

researcher may want to use a correction factor to adjust for the effect

of plugging burrows, i.e., multiplying the field estimate by 1.4 (= 1. 24


< u



C n
C--" ,





U, c

*- I 0
0 ea

LA 0

2 --i

X- c

C 3

4- 0r

o E


** II

m/ /


r 00 MN

S C r-
In In r -

i In r- ON' In




a -

a )

r _1

*M 4--
E 0
= p





C '

Qn co C

M Cj M m

*- lq r~

00 CM 00

o I\ C-
'n a) co

0C I)
o in in
r- .3- co
CM. rn -

C\M In in
C0 co

D.0 In 'fl
In r- co

Mn r- co
0 co CMi

CM C k 0

oM cl CM

-^ o
C 0

a-' a7' 7

- ri CO in c
co - qr ao M co
4- C\M -:

CM co
:a ,

In e
CM -

CN -




-D c

In CM.
C- C-
-' 0


rn CM N -

0 0 0 0

co -

- In o0


.0 rn o

0 0


a-' ,_.

0 0

co ca
CaO aO

In '.

Tn o

U3 00

Id) U



Burrow System Selection and Mapping

After lay, the rapidly growing sugarcane hides the burrow systems,

and becomes increasingly difficult to penetrate. Thus, between May and

harvest in the late fall or winter, fewer fields could be studied.

Four second ratoon, high density half-fields (Appendix IV).

were selected for livetrapping, 2 in May 1979 (14 CN and 6 BN) and 2

in May 1980 (8 GS and 24 CS). Half-fields were stratified and burrow

systems were selected and mapped using the same methods described for

1980 postharvest study half-fields. Ten burrow systems in each of

Fields 14 CN, 6 BN, and 8 GS were selected and studied between May and

October in 1979 (14 CN and 6 BN) and 1980 (8 GS). These fields were

harvested in November. Eleven burrow systems in Field 24 CS were

studied between May 1980 and January 1981; this field was harvested in

February. Paths to these systems were maintained monthly from August

through harvest, when the maturing cane forms a tangle of stalks and

leaves. These paths (2 per study field) ran the length of each half-

field, and essentially divided it into thirds.

All burrows present in May and new burrows appearing each month

until harvest in the selected systems were staked, numbered, and mapped.

The area surrounding each system was searched to find any new burrows

within 15 m of already staked burrows. As the cane matured during the

fall, the risk of missing some new burrows in the dense vegetation was

greater. Already staked burrows were checked for fresh signs of

Neofiber excavation each month.

Burrow System Livetrapping

The selected burrow systems in each half-field were livetrapped

each month until harvest, using Japanese wire mesh live traps baited

with fresh apple chunks (Fig. 13). Traps were set at active-looking

old burrows (those found in previous monthss) and at all new burrows.

If no new burrows were found and no old ones looked active in a given

system, traps were placed at old burrows such that all parts of the

system were sampled. If no Neofiber weretrapped for 2 consecutive

months in a given system, trapping was discontinued in that system un-

less signs of activity reappeared.

Burrow systems in the two 1979 fields were split into 2 groups by

randomly selecting half of the systems on each path.(However, burrow

systems <30 m apart were always trapped on the same night.) Each

group in both fields was trapped on alternate nights for 3 consecutive

nights. All of the selected systems in the two 1980 fields were trapped

every other night for 6 consecutive nights. Thus there were 4 trap

nights per burrow system per month in 1979 and 3 nights in 1980.

Traps were set between 1600 h and dusk (later in summer than in

fall) and were checked between dawn and 1100 h (earlier in summer than

in fall).

All muskrats were anesthetized with Metofane'? on their first cap-

ture in order to measure them (total length, tail, hind foot). They

were weighed using a Chatillon 600 g capacity spring scale, and ear-

tagged with individually numbered Monel #1 fingerling tags in the right

Fig. 13. A Japanese live trap (Honolulu Sales, Ltd., Honolulu,


ear. All muskrats that were not radio-tagged were also toe-clipped for

individual identification. Only the tips (just below the nail) and not

entire digits were removed. Identification numberss, measurements,

weight, sex, capture location (burrow number), burrow system, and field

were recorded for each muskrat captured. Recaptured muskrats were not

anesthetized unless remeasured. Occasionally one wriggled out of my

grasp before both toe-clip and ear-tag numbers could be read, but on

only one occasion did a marked individual escape before either number

was determined. Neofiber were handled and released at the capture


Some of the study fields were live- or snaptrapped following har-

vest. Burrow systems in 14 CN were trapped 5 nights immediately after

harvest in November 1979 and 6 nights in December. This field was re-

examined for signs of Neofiber activity in April 1930, and several old

systems were trapped 7 nights in April. Field S GS was trapped 2

nights immediately after harvest in early November 1980, and 6 nights

later in November; it was flooded with water discharged from the mill

in early December. Field 6 BN was trapped 4 nights just before harvest

in November 1979, but was plowed under shortly after harvest. Field

24 CS was not trapped either immediately before or after harvest in

February 1981.


A total of 47 radio collars was placed on 59 Neofiber in the live-

trapped study fields and nearby fields (Fig. 14). Most of the radio

collars were put on adult muskrats; a few were put on subadults. Radio

collars (40) from the Denver Wildlife Research Center (DWRC)(Fig. 15)

Fig. 14. Fitting a radio collar to an anesthetized round-tailed


- t - " l

Oz 00

Fig. 15. Radio transmitter collars from the Denver Wildlife Research
Center (left) and AVMI Instrument Co. (right).


had modified whip antennas, made of 3 mm wide copper strap, that were

attached around muskrats' necks by pull-through beaded plastic straps.

Both straps were covered with heat-shrink tubing. Radio collars (7) from

AVMI Instrument Co. (AVM)(Fig. 15) were SM 1 rodent-type collars with con-

tinuous loop antennas made of plastic coated, stranded copper wire. The

antenna-collar is designed to fold back on itself to adjust to the

proper neck size, and is held in place by a piece of heat-shrink tubing.

All transmitters operated in the 164.425 to 164.700 MHz frequency

range, were powered by Hg-675 1.35 v batteries, and weighed 5 to 6 g.

The AVM transmitter package was protected by inner coatings of epoxy

and paraffin, and an outer layer of dental acrylic. The DWRC trans-

mitters were coated with various substances in an effort to improve

waterproofing. Twenty DWRC radios were used between August and December

1979, and 20 DWRC and 7 AVM radios were used between June 1980 and

January 1981.

Three-element hand-held Yagi antennas and 2 AVM LA-12 receivers

were used to locate radioed Neofiber approximately (e.g., to a burrow

system). Muskrats could be located to the nearest burrow by removing

the Yagi and placing the coaxial cable connector in burrow entrances,

gradually decreasing the range of signal reception by turning down the

receiver's gain control.

If an individual muskrat could not be located in the burrow system

in which it was previously radio-located, the 2 paths in each study

field were searched using 3-element Yagis. Attempts to find missing

radios were also made by driving around the field perimeter (except

along middle ditches), checking in all directions every 30 m. While

most of the AVM radios could probably be received in this manner (range

150 to 200 m in standing cane), many of the DWRC radios could probably

not be received (range 30-60 m in standing cane). The modified whip

antenna (a dipole antenna) has less gain than the continuous loop. This

may be true because the loop antenna can be tuned to maximize the extent

to which the rat's body acts as an antenna, or because the dipole is

inefficient when adapted to fit a rat's neck.

When the study fields had been harvested, 4-element Yagis were

mounted on either Honda ATC 110's or in a null-peak system on a truck,

and fields were searched by riding or driving transects, 30 m apart,

that ran the length of the fields. Stops were made at 30 m intervals

to check for all radios not accounted for. When 3 GS was flooded

following harvest in December 1930, we used a canoe and 4-element Yagis

to search for 3 radio-collared Neofiber that were known to be alive

before flooding, and to check for signs of Neofiber presence (houses

and feeding platforms).

Burrow Systems Trapped in Addition to Selected Systems

From June through harvest, 1979 and 1930, new burrow systems were

found in some of the livetrapped fields, either because of their prox-

imity to systems selected for study in May, or because a radio-tagged

Neofiber moved to a previously unmapped system. Burrows in new systems

were included in the monthly livetrapping. Burrow systems discovered

through radiotracking were trapped but burrows were not mapped. Some

burrow systems present in May were trapped in later months, as time

permitted, to increase the chance of recapturing a marked Neofiber that

might have moved to another system. All of the burrow systems trapped

in addition to the originally selected 41 burrow systems are referred

to as additional systems.


Female Neofiber that died in traps or while wearing radio collars

were examined internally to determine reproductive status. All muskrats

that died while wearing radio collars underwent necropsy to determine if

the collars were the probable cause of death, unless predation was clear-

ly the cause. Several muskrats not wearing collars, both marked and un-

marked, were found shortly after death above ground. These were also

examined internally to check for grossly visible abnormalities. Lung

tissue from one of these specimens was preserved and given to Dr. Donald

J. Forrester, Department of Preventive Medicine, University of Florida,

for pathology analysis. All Neofiber skulls found in trapped burrow

systems in 1980 study fields were collected.

Ectoparasite Collection

Ectoparasites were collected from 5 Neofiber and 5 Sigmodon cap-

tured in several burrow systems in 8 GS in October 1980. Rats were

anesthetized, laid on a 0.5 x 0.5 m plastic sheet, and dusted with 0.9

percent pyrethrins. Then debris from their fur was brushed onto the

sheet. Material collected on the sheet was transferred to a plastic

bag. Samples were examined within 24 h under a dissecting microscope

at 20X, and parasites were pooled by host species and preserved. Speci-

mens collected from Neofiber were submitted to Dr. Harvey Cromroy,

Department of Entomology and Nematology, and those from Sigmodon to

Thomas Rogers, School of Forest Resources and Conservation, University

of Florida, for identification.


rieofiber Captures and Burrowing Activity

The range in total number of individual Neofiber captured per

burrow system between lay and October was 2 to 13 in 1979 and 2 to 9 in

1980; the overall mean was 4.6 (Fig. 16). Mean Neofiber captures per

burrow system per month, including only those burrow systems with at

least one Neofiber capture per month, varied considerably in trend

between fields, but the means for the whole study periods were highly

consistent between fields: 1.72 to 1.80 (Table 14). The majority of

means did not differ from each other by more than 1 SE. Mean Neofiber

captures per burrow system within months over all fields showed a con-

sistent decline (although with overlapping confidence intervals) except

in September, when high field means for 8 GS and 24 CS boosted the over-

all mean. The number of burrow systems in which fleofiber were captured

declined between May and October, or fMay and January (Table 14).

The frequency distribution of rleofiber captures per burrow system-

month was similar in the 4 study fields (Fig. 17). A burrow system-

month is one burrow system trapped in one month. An average of 60

burrow system-months was trapped per field (range 57 to 64) between May

and October. Overall, burrow systems most frequently had 1 (35 percent

of a total of 239 burrow system-months), 0 (28 percent), or 2 (24 per-

cent) Neofiber captures per month. In only 30 (13 percent) of the

burrow system-months were 3 or more Neofiber captured. Total monthly

Neofiber captures per field varied in trend between fields (Fig. 18).

Fields 14 CNl and 8 GS showed a sharp decline in captures from May to

June, and a decline from September to October. Field 6 BN showed an

increase in captures from May to July, and a sharp decrease from July

4 r

14 CI

3 5 7
x = 4.8

S n1

1 3 5 7
x = 4.7

6 BN


9 11 13

8 GS


x = 4.3

24 CS

x =

Number of Individuals Captured

Number of individual Neofiber captured per burrow system in 4
Florida sugarcane fields. Ten burrow systems in Fields 14 CN and
10 in 6 BN were sampled from May through October 1979; 10 burrow
systems in Field 8 GS and 11 in 24 CS were sampled from May through
October 1980 and May 1980 through January 1981, respectively. Open
bars show total individuals captured between May and October;
hatched bars in 24 CS show total individuals captured between
November and January.

1 7

Fig. 16.



II 0




* A
- E

E 'I


o o



>) L


0.. L-/
m i

0 3

0 0


4- *u
Li L
>, 10

LO <
3 '"

0 o-

> C C6

0 0 0"

o Cr C

> u
0 *C) r
o *











.1 ^

> h



30 -
20 .


0 1 2 3 4 5







012 3 4 5 6

24 CS

0 1 2 3 4 56

0 1 2 34 5

Number of Neofiber captures

Fig. 17. Frequency distribution of Neofiber captures per burrow system as
percentages of total burrow system-months trapped. A burrow
system-month is one burrow system trapped in one month.

6 BN

14 C11

8 GS


*--- 14 CI
0----o 6 Eli


May Jun Jul Aug Sep Oct 1979

--- 3 GS
- --o 24 CS


/ oN0



i I I I I I I *
May Jun Jul Aug Sen Oct Nov Dec Jan
1930 1931

Fig. 13. Monthly Neofiber captures in 4 Florida sugarcane fields. Ten
burrow systems were livetrapped in Fields 14 CN, 6 BN, and
3 GS; 11 were trapped in 24 CS.



16 i





20 -1




1 T I _

' _

0 '/

to September. Field 24 CS captures fluctuated down and up with no

apparent overall trend.

Most of the burrow systems continued to show signs of Neofiber

activity through October 1979. Trapping was discontinued because of

inactivity in three systems for a total of 7 months: V in 14 CN (August-

October 1979), G in 8 GS (September-October 1980), and G in 24 CS

(September-October 1980)(Appendix IV a,c,d). Burrow systems varied

considerably in size and number of burrows, both when they were first

mapped in May, and when they were last examined before harvest. The

largest numbers of Neofiber were captured, not surprisingly, in the

largest burrow systems: 14 CN P (9 individuals), 6 BN Q (13), 8 GS A

(9), and 24 CS K (13)(Appendices IV and V). The relationship between

burrowing activity and Neofiber captures was not always as clearcut as

one might expect, however. In 6 BN (Fig. 19a), for example, Neofiber

captures and burrowing activity diverged sharply in August and September.

In the other 3 fields (Fig. 19 a and b), captures and activity were more

similar in trend, and all fields showed a decline in activity near the

end of the study periods. Overall, the correlation between number of

Neofiber captured per month and number of active burrows per month was

not significant (P>0.05).

The mean number of new burrows that appeared per burrow system per

month ranged from 1.7 to 13.5 (Fig. 20), with an overall mean of 6.8

from May to October. There was a decline from May to October or January

in new burrowing activity in all fields that was particularly marked in

the last month before harvest (October in 14 CN, 6 BN and 8 GS; January

in 24 CS). This was also the month in which the fewest Neofiber were

captured in all fields except 24 CS. The mean number of new burrows

14 Cri

/ \
/ \
/ ,,
\ / \
\ / N,
N. / N.
^^ ^. > .

Jun Jul Aug Sep

Oct 1979

0----0 Active burrows

___ [lumber individuals captures







120 -

S100 |


S 40

- 20

Oct 1979

Relationship between number of Neofiber captured and burrowing
activity per month in 2 Florida sugarcane fields studied in
1979. Number of active burrows includes burrows found in
previous months) that were still active and new burrows.

Fig. 19a.

24 CS





Jun Jul Aug Sep Oct

Nov Dec


\ 60


Jan '80-'81

Jan '80-'81

S--- Active Burrows

--- Neofiber Captured

8 GS

0 10




150 o
125 -


- 75


Oct '80

Fig. 19b. Relationship between number of Neofiber captured and burrowing
activity per month in 2 Florida sugarcane fields studied in 1930.
Number of active burrows includes burrows found in previous
months) that were still active and new burrows.



4 J


m 12




Fig. 21. Monthly means SE of new burrows per burrow system trapped
(9 to 11 per field) in 4 study fields. One field, 24 CS, was
studied in 3 additional months; the average numbers of new
burrows per system in this field is shown by shaded bars in

1 = 0

ov-ay an. un Jul Aug Sep Oct lov Dec Jan

Nlew burrows are those that have appeared since the previous mapping
period, except in m ay, which are burrows that were present when burrow
systems were initially mapped.

appearing monthly per system during the May through September periods

was 7.4; the mean for October was 3.7. Thus during the May through

September study periods, a monthly average of 4.1 burrows (7.4/1.8) was

produced per Neofiber captured, while in October, an average of 2.4

burrows (3.7/1.5) was produced per individual captured.

The mean number of burrows per system at the end of the study

periods were 56.3 (14 CIl), 34.6 (6 BN), 46.2 (8 GS), and 32.6 (24 CS).

Field 24 CS had the lowest mean despite having 3 extra months (November

through January) during which new burrows could have been added. A

large portion of this field was flooded in August and September, which

prevented burrowing in many systems during this period.

Population Sex and Age Structure

Because Fields 14 CN, 6 BN, and 8 GS were trapped immediately

before or after they were harvested in November, all Neofiber captures

from May through November, 1979 and 1980, were used to examine popula-

tion sex and age structure. When data from all fields were combined,

the ratio of adult and subadult males to females was nearly 1:1, but

varied considerably between fields (Table 15). The greatest divergences

from 1:1 occurred in 14 CN with 2.2 subadult males:1 subadult female

and 24 CS, with 2.2 adult females:1 adult male. The numbers involved,

however, are small, and the overall 1:1 ratio should probably be

considered to represent the true situation. The numbers of juveniles

captured per field were even smaller (Table 15). Total captures,

including those from additional burrow systems (p. 102) included 24

females, 18 males, and 1 of sex unknown.

Table 15. Total Neofiber captures, classified by sex and age, from 4
Florida sugarcane fields. Individuals trapped from May through
November 1979 in Fields 14 CN and 6 BN, and May through November
1980 in 8 GS and 24 CS. N = number of burrow systems trapped.

Field N Ad Ad Subado Subad Juvc Juv9 Total

14 CN 10 9 9 11 5 2 2 39,

6 BN 10 16 12 3 3 1 4 39

8 GS 10 14 12 4 7 3 2 42

24 CS 11 6 13 7 10 5 8 49

Total 41 45 46 25 25 11 16 169

- includes one juvenile of unknown sex

The numbers of young animals (subadults and juveniles) per adult

female also varied considerably between fields, and probably reflect

real field differences in reproductive output (embryos produced) or

success (young surviving long enough to be captured): 2.3 (14 CN), 2.3

(24 CS), 1.3 (3 GS), and 0.9 (6 BN).

The number of new adults appearing each month declined in all

fields between May and November; in 24 CS, 5 new adults appeared in

December (Fig. 21). The appearance of young Neofiber (subadults plus

juveniles) was more sporadic (Fig. 21). In 14 CN and 24 CS, a large

proportion of the young animals was captured in May and June. Another

pulse of young appeared in 24 CS in September and October, while only

a few more young appeared in 14 CN in August and September. Patterns

in 6 BN and 8 GS were less evident, perhaps because fewer young were

captured. Most of the young animals that appeared in these fields were

captured in August.

Adult associations

In most of the burrow systems trapped (36/41) at least one adult

male and female were captured over the 7-month study periods, and in

30 systems, adults of both sexes were captured within the same trapping

period. Two or more adult females were captured in 15 systems over the

study periods; 2 or more adult males were captured in 22 systems. How-

ever, within monthly trapping periods, only 5 burrow systems had simul-

taneous captures of 2 males, and 5 had simultaneous captures of 2

females. In only 2 cases did both same-sex adults continue to be

trapped in the same system; almost invariably one or both were not

recaptured or one was known to have moved to a different system. One

JI *9 r-n r= r-, =

May Jun

Jul Aug

May Jun

6 B11

F- I rL-za


+ Juv

Jul Aug Sep Oct flov '79

8 GS




9 *




Jun Jul Aug Sep Oct Nov Dec

Jan '81

Fig. 21. Number of new Neofiber captured and marked each month in 4 Florida
sugarcane fields.

14 Crl

Sep Oct

Nov '79

[lay Jun Jul Aug Sep Oct Nov '30

24 CS

exception was in 14 CN U, with adult males Nos. 14 and 19 (Appendix V).

This was a large burrow system (over 90 burrows), and males Nos. 14 and

19 were very likely to have been siblings, since they were similar-

sized subadults captured near each other with an adult female in May

1979. In burrow system P in the same field, adult female No. 48 was

captured over the same 4-month period in which subadult female No. 25

matured into an adult (Appendix V). Neither adult female was recaptured

in P following this period; female No. 48 was ultimately recaptured in

U in December 1979, following harvest. 14 CN P was also a large system,

and females Nos. 48 and 25 appear to have maintained separate ranges.

Adult female-juvenile associations

Juvenile captures from all burrow systems trapped were combined to

examine adult female-juvenile associations; primarily within-month

captures were considered, therefore identical periods of study for each

burrow system were unnecessary. Of 43 juveniles captured or located in

the 4 study fields, 34 (79 percent) were at or near (less than 5 m from)

an adult female's capture locationss. Seven juvenile females and 2

juvenile males were not captured near an adult female. Twenty-five

juveniles were captured at new burrows (ones appearing in the month of

capture or one month previously) and 13 were captured from older burrows,

despite the fact that more old burrows were available to them. Totals

of 541 new and 797 old burrows were available in the systems in which

juveniles were captured. More old burrows were available in the systems

in which juveniles were captured at new burrows (427 new vs 550 old),

and in the systems in which juveniles were captured at old burrows (183

vs 321 old). Thus, adult females appear to have had a preference for

rearing young in new burrows. Five juveniles were captured at burrows

mapped in May, thus the age of these burrows could not be determined.


Fields 14 CN and 24 CS had relatively high proportions of burrow

systems in which young Neofiber were captured (8/10 and 10/11, respec-

tively), and were known to survive to maturity (5/8 and 5/10). However,

only in 14 CN was a large proportion of young captured known to survive

to maturity (10/21)(Table 16). Offspring in 24 CS survived relatively

better than in 6 BN and 8 GS (6 of 24 captured by September survived to


Several young died in traps, and necropsy revealed that they had

no visible fat around internal organs, raising the question of whether

they might have been near starvation before entering the traps. Young

that were later recaptured as adults gained an average of 24 g (males)

and 21 g (females) per month during this transition, while young that

were recaptured but not as adults gained an average of 14 g (males) and

15 g (females) per month (Fig. 22). All but one of the young individ-

uals that were recaptured as adults (Fig. 22a) were subadults on

initial capture, whereas most of the young animals recaptured but not

as adults (Fig. 22b) were initially captured as juveniles. There was

considerable variation in weight changes of recaptured young, as indi-

cated by the large SE (Fig. 22). Three subadults and 1 juvenile lost

weight between captures, and were never recaptured as adults.

Clearly, some young viere retarded in their development. For

example, subadult male No. 39 (Appendix V) was captured in 6 BN R in

.lay, June, and July 1979, and weighed 135, 150, and 130 g, respectively.

Proportion of juvenile and
May and September 1979 and
hood in 4 Florida sugarcane

subadult Neofiber, captured between
1980, known to have survived to adult-

Field Subad Subad ? Juvo' Juv?

14 CN 7/11 3/5 0/2 0/2

6 BN 0/3 0/3 0/1 0/3

8 GS 0/4 1/7 0/2 0/2

24 CS 1/5 4/9 1/5 0/5

Total 8/23 8/24 1/10 0/12

0.35 0.33 0.10 0.00/

a One juvenile female captured in an additional burrow system was
captured as an adult.

Table 16.



30 c,- 9

30 24 (3.0)
20 --- -= ------ 21 -(6.4)
20 ,

1 10

Individual tleofiber
5 50
" 40



14 (7.1)


-20 Juvenile Subadult

Fig. 22. Weight gain or loss (g/mo) of individual juvenile and subadult
Neofiber a) recaptured as adults and b) recaptured but not as
adults in subsequent months. Mean weight gains are shown by
dashed lines, SE in parentheses. Neofiber were trapped in 4
Florida sugarcane fields between May and November, 1979 and 1980.

He also had a heavy mite (Acarina) infestation, particularly around his

eyes. Assuming that he was at least 30 days old at first capture (since

his body measurements met subadult criteria), he was at least 90 days

old when last captured in July, and could potentially have reached adult


Subadults were recaptured as adults after a mean of 1.9 months

(males) and 1.6 months (females), range 1-3 months. Juvenile male No.

116 (Appendix V: 24 CS C) was considered to have reached adulthood in

September 1980, when testicular development was visible externally,

and his weight reached 190 g, the maximum weight he attained. Assuming

he was approximately 30 days old when initially captured in rMay, he

matured in 150 days, which is longer than would be expected on the

basis of data from known-age, captive reofiber.

One other juvenile, female No. 176 (Appendix V) attained adult

size in just 2 months (approximate age 90 days). Because she was cap-

tured in an additional system, 8 GS N, she was not included in Table 15

and Fig. 22a. She gained 75 g per month for 2 months (male No. 116

gained an average of only 24 g per month) to reach an adult weight of

220 g.

Recaptured juveniles and subadults that were not recaptured as

adults were captured in an average of 2.3 months, range 1-4 months.

Eleven juveniles were recaptured, but not as adults (Fig. 22b). Nine

of these, captured in 2 consecutive months, had gained a mean of 17 g

(one lost weight), and still did not exceed 115 g at their second cap-

ture; 3 of the 9 were remeasured and their TBL's and BL's met subadult

criteria, even though their weights did not. Juvenile male No. 179

(Appendix V: 3 GS E), in contrast, gained 105 g between his first cap-

ture in August and last capture in October 1980.

A total of 3 Neofiber (Nos. 6, 10, 17 and 18 from 14 CN, lo. 36

from 6 BN, and Nos. 116, 128, and 130 from 24 CS; Appendix V) survived

the entire May to November study periods. One individual (male No. 29

from 14 Cl 0 and G) was known to have survived 9 mo (September 1979 to

April 1980).

Because the number of new adults that appeared after August

differed among fields (Fig. 21), recapture success of only those adults

initially captured from May through August and recaptured no later than

November was examined (Table 17). The overall mean number of trapping

periods (months) over which individual adult Neofiber were captured did

not differ significantly by sex (P>0.05). Field means (sexes combined)

ranged from 2.4 to 2.3 months, and were within 1 SE of each other.

However, a smaller proportion of monthly captures, particularly of

young Neofiber, were recaptures in Fields 6 BN and 8 GS than in Fields

14 CN and 24 CS (Fig. 23).


The fate of individual Neofiber wearing radio collars varied con-

siderably (Appendix VI). Some apparently tolerated the collars fairly

well (Fig. 24) and were radio-located over periods of several weeks or

even several months; others developed neck irritation within a week

after the collars were put on, and five deaths were attributed to wear-

ing of radio collars. In only one of these cases was my attachment of

the collar clearly at fault. Neck abrasions and lesions were also re-

ported on radio-collared nutria (Myocastor coypus) in Louisiana (Coreil

and Perry 1977). Adult female No. 300 was located at 14 CN K shortly

Table 17.

Mean number of trapping periods (months) over which adult male
and female Neofiber were livetrapped in 4 study fields. Means
include individuals initially trapped from May through August
1979 (Fields 14 CN and 6 BN), or May through August 1980 (Fields
3 GS and 24 CS).

Number of Months
Field Add Ad Overall

14 CN 6 2.8 0.95 3 2.9 0.92 14 2.3 0.64

6 BN 13 2.3 0.60 10 2.2 0.55 23 2.6 0.41

8 GS 11 2.2 0.40 12 2.7 0.50 31 2.4 0.32

24 CS 5 3.0 0.84 10 2.5 0.69 15 2.7 0.52

Overall 35 2.7 0.31 40 2.6 0.31 75 2.6 0.22







May Jun Jul Aug Sep Oct

Combined data, Fields 14 CN and 24 CS

Combined data, Fields 6 BN and 3 GS

-' ,
<\ c a

l i--i-Li

May Jun Jul

Aug Sep Oct

Monthly recapture success of adult and
juvenile) Neofiber in a study fields.
were trapped in 1979, Fields 24 CS and

young subadultt and
Fields 14 CN and 6 BN
3 GS were trapped in

- .L r I J / I L I* J I


1.00 -

0.30 -


0.40 -

0.20 .





0.40 -


Fiq. 23.



;JEi ;,



Fig. 24. Captive radio-collared muskrat digging a burrow.

after her death; her collar was too loose, and her foreleg was cut,

presumably by the beaded strap, because she had placed the leg between

the collar and her neck. She was emaciated and there were no other

signs of injury or disease to indicate that the collar was not the

primary factor causing death. This female was among the first muskrats

radio-collared, and care was taken not to make subsequent attachments

too loose. Neofiber males Nos. 405 and 493 died very shortly after

their collars were put on (3 or 4 days) and also showed no evidence of

predator or disease-related mortality. Male No. 89 was radio-collared

shortly after harvest in 14 CN, and was found to have a heavy mite

infestation and no body fat when he died two days later. Female No. 26

was located in several new burrows shortly before her death; she may

have been attempting to start a new burrow system. She also v'as

emaciated, and one of her forepaws was abraded, as if perhaps she had

tried to remove the collar.

It is possible that some of the radio-collared muskrats were

already stressed by such factors as lack of food, disease, or parasitism,

and that the collars, acting as additional stressors, rapidly acceler-

ated their deaths. Or perhaps there were considerable behavioral

differences between individuals, e.g., in burrowing activity, distances

moved, or interaction with conspecifics, These hypotheses would help

to explain why some individuals, such as female No. 164, were able to

wear radio collars for much longer periods with no apparent harm.

Three radio-collared Neofiber were found within hours after their

deaths, and they appeared to be the victims of predators. Three others

were killed by human activity (one mortality was probably caused by

harvest, one by postharvest discing, and one when a live-trap door

shut on the animal's neck). The remaining 9 muskrats that were known

to have died while wearing radios were too decomposed to permit deter-

mination of cause of death.

Radio collars probably affected the behavior of the muskrats wear-

ing them, but they enabled detection of movements that would otherwise

have been missed. Use of radio collars also revealed that some

Nleofiber were present but not trapped in studied burrow systems.

Most of the radio transmitter malfunction was caused by water

penetrating the protective outer coating. Several types of epoxy were

tried as insulation, and the most suitable one for the conditions

encountered in this study proved to be Epoxi-Patchn 309 (Hysol Division

of the Dexter Corp.). Many of the transmitters put on muskrats in

October 1980 in 8 GS and December 1980 in 24 CS had undergone a battery

change, and may have been prone to stop functioning earlier than antic-

ipated (e.g., the radios put on Nos. 21, 30, 46, 229, and 59: Appendix


Interburrow System Movements

Although the total numbers of individual adult males and females

captured were nearly equal (Table 15), and overall survival was similar

(Table 17), adult females tended to be recaptured in the same burrow

system over a longer period (x = 2.4 mo) than males (x = 1.9 mo).

However, this difference was not significant (P> 0.05). More male

than female movements to different burrow systems (31 vs 10) were

detected. Thus while both sexes were obviously capable of both long-

distance moves (Appendix IV) and long-term utilization of a single

system (Appendix V), in general males tended to do more of the former

and less of the latter than females. In 24 CS, however, 6 female and

7 male movements were detected. Two of the female movements (female

No. 206 from F to T and female No. 56 from K to W) would not have been

detected and a third (female No. 26 from D to S) might not have been

detected had these females not been wearing radio collars. Thus, given

such small numbers, it is impossible to determine if females in 24 CS

actually behaved differently than females in the other 3 fields, or if

relatively long moves are fairly typical of females, also, and by

chance were only detected in 24 CS.

Two of the 10 females known to have left a system were determined

to have returned to the same system; in both cases the movements

barely exceeded 15 m. One of the 10 was found dead in new burrows and

obviously could not have returned. An eleventh female (No. 142) was

found dead in 8 HS shortly after harvest of 8 GS (Appendix IVc). She

was not counted as one of the females that moved because I could not be

certain if she moved to 8 HS herself or was carried there by a scavenger

or predator. Eight of the 31 males that left a burrow system were

known to have returned to the same system; 6 of the 8 returned to

burrow systems in which an adult female was captured in the same trap-

ping period.

Five of the 10 females that moved went to previously undetected

burrow systems, and were not recaptured or relocated in their original

capture locations. Only 6 of 31 males moved to previously unmapped

burrow systems, and did not return to their original capture location

or another previously mapped burrow system. Five of the 6 were found

dead in new burrow system locations and could not have returned.

Fourteen adult females were radio-located in the same burrow

systems where they were originally captured in at least 2 consecutive

months (58 percent of all radio-collared adult females). Only 6 males

(20 percent of all radio-collared adult males) were located in the

same burrow systems in at least 2 consecutive months.

Thus male and female movements from a burrow system appear to

serve different functions. Males may be wandering among adjacent

burrow systems in search of receptive females, while female moves

appear more often to represent permanent dispersal to a new area.

Females Nos. 130, 26, 206, and 56 were very likely to have been the

founders of burrow systems 0, S, T, and W in 24 CS, respectively. A

particularly strong case can be made for female No. 130. As mentioned

earlier, maturing cane is a very dense growth, and it would be extreme-

ly difficult to find all of the burrows in a given area. However,

burrow system 0 appeared just east of D; a little further down the

same rows, and on the path we used every month to get to D (Appendix

IVd). Therefore, I am certain that 0 appeared between the June and

July 1980 study periods. There were just 6 burrows in 0 in July, and

only female No. 130 was captured here. Subadult male No. 169 (also

from D) appeared there in August, and a new adult male, No. 200, in


In two cases, a relatively short female movement may have been

associated with reproduction. Female No. 26 at burrow system S in

14 CN (Appendix V) moved several rows north of her previous range and

was recaptured at new burrows with a newborn Neofiber (presumably born

in the trap). Female No. 154 moved approximately 20 m from 8 GS K to

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs